A fuel cell is an electrochemical power source with advantages of both the combustion engine and the battery. Like a combustion engine, a fuel cell will run as long as it is provided fuel; and like a battery, fuel cells convert chemical energy directly to electrical energy. As an electrochemical power source, fuel cells are not subjected to the Carnot limitations of combustion (heat) engines. Fuel cells bear similarity to batteries, with which they share the electrochemical nature of the power generation process and to the engines that, unlike batteries, will work continuously consuming a fuel of some sort. A fuel cell operates quietly and efficiently and, when hydrogen is used as a fuel, it generates only power and water. Thus, a fuel cell is a so called ‘zero-emission engine’. In the past, several fuel cell concepts have been tested in the laboratory but the systems that are being potentially considered for commercial developments are: (i) Alkaline Fuel Cells (AFCs), (ii) Phosphoric Acid Fuel Cells (PAFCs), (iii) Polymer Electrolyte Fuel Cells (PEFCs), (iv) Solid Polymer Electrolyte Direct Methanol Fuel Cells (SPE-DMFCs), (v) Molten Carbonate Fuel Cells (MCFCs) and (vi) Solid Oxide Fuel Cells (SOFCs). Among the aforesaid systems, PEFCs that employ hydrogen as fuel are considered attractive power systems for quick start-up and ambient temperature operations. Ironically, however, hydrogen as fuel is not available freely in the nature. Accordingly, it has to be generated from a readily available hydrogen carrying fuel such as natural gas, which needs to be reformed. But, such a process leads to generation of hydrogen contaminated with carbon monoxide, which even at minuscule level is detrimental to the fuel cell performance. Pure hydrogen can be generated through water electrolysis but hydrogen thus generated needs to be stored as compressed/liquefied gas, which is cost-intensive. Therefore, certain hydrogen carrying organic fuels such as methanol, ethanol, propanol, ethylene glycol and diethyl ether have been considered for fueling PEFCs directly. Among these, methanol with hydrogen content of about 12.8 wt.% (specific energy = 6.1kWh kg-1) is the most attractive organic liquid. PEFCs using methanol directly as fuel are referred to as SPE-DMFCs. But SPE-DMFCs suffer from methanol crossover across the polymer electrolyte membrane, which affects the cathode performance and hence the fuel cell during its operation. SPE-DMFCs also have inherent limitations of low open-circuit-potential and low electrochemical-activity. An obvious solution to the aforesaid problems is to explore other promising hydrogen carrying fuels such as sodium borohydride (specific energy = 12kWh kg-1), which has a capacity value of 5.67Ah g-1 and a hydrogen content of about 11wt.%. Such fuel cells are called direct borohydride fuel cells (DBFCs). This thesis is directed to studies on SPE-DMFCs and DBFCs

Synopsis of the thesis entitled “Titanium Nitride-Based Electrode Materials for Oxidation of Small Molecules: Applications in Electrochemical Energy Systems” submitted by Muhammed Musthafa O. T under the supervision of Prof. S. Sampath at the Department of Inorganic and Physical Chemistry of the Indian Institute of Science for the Ph.D degree in the faculty of science. Fuel cells have been the focus of interest for many decades because of the ever increasing demands in energy. Towards this direction, there have been considerable efforts to find efficient electrocatalysts to oxidize small organic molecules (SOMs) such as methanol, ethanol, glycerol, hydrazine and borohydride that are of potential interest in direct fuel cells. Most studies revolve around platinum which is the best electrocatalyst known for the oxidation of many SOMs. However, platinum is extremely susceptible to carbon monoxide (CO) poisoning which is an intermediate in the electrooxidation of aliphatic alcohols. The best known catalyst, platinum-ruthenium alloy (PtRu), suffers from leaching of Ru during cycling resulting in decrease in efficiency in addition to loss of precious metal. Another important aspect of fuel cell catalyst degradation is corrosion of widely-used carbon support, under fuel cell conditions. Corrosion of carbon support weakens the adherence of catalyst particles on the support and in turn results in loss of catalyst and also in its easy oxidation. Carbon corrosion is also reported to decrease the electronic continuity of the catalyst layer. Hence, replacement of carbon support with durable material is required. The present research explores the use of non-carbonaceous, transition metal nitride for anchoring catalytic particles. The favorable physicochemical properties of titanium nitride (TiN) such as extreme hardness, excellent corrosion resistance in aggressive electrolytes, resistance to nearly all chemicals, salt and humidity, very good support for the adherence of fuel cell catalysts and excellent electronic conductivity motivated us to use this material for anchoring fuel cell catalysts such as Pt, PtRu and Pd. In the present studies, TiN coated on stainless steel (SS 304) surface is used as an electrode material. Catalysts such as Pt, Pd and PtRu are anchored on to TiN and used for the oxidation of methanol and ethanol in acidic as well as in alkaline media. Use of bare TiN is explored for the oxidation of sodium borohydride. The efficiency of TiN supported catalysts are compared with carbon supported ones. Preliminary studies on the use of TiN supported catalysts in fuel cells have been conducted as well. Figure 1 shows the topographic atomic force microscopic (AFM) image in combination with scanning Kelvin probe (SKP) image of platinized TiN (Pt-TiN) surface. Since Pt particles are metallic, they are expected to show lower work function values than that of TiN domains which is indeed observed in figure 1B where the location of Pt particles is shown as dip in the work function. Very interestingly, the interface of Pt-TiN possesses very different work function values confirming the existence of metal-support interaction and this is expected to have positive implications in fuel cell catalysis. Figure 1. Contact mode AFM (A) and the corresponding scanning Kelvin probe image (B) of Pt-TiN surface. Figure 2. Cyclic voltammograms of Pt-TiN and Pt-C electrodes in 0.5 M H2SO4 containing 0.5 M methanol at a scan rate of 10 mV/s. Loading of the catalyst used is 1 mg of Pt/cm2. The performance of Pt-TiN and PtRu-TiN are compared with the corresponding carbon supported catalysts (Pt-C, PtRu-C) for the electrooxidation of methanol. Figure 2 shows the voltammograms obtained on Pt-TiN and Pt-C in presence of acidified methanol. TiN supported catalyst performs better than carbon supported catalyst in terms of high currents at low over voltages (based on I-t measurements), long term stability and high exchange current densities (based on Tafel studies). The electrochemical characteristics of methanol oxidation on Pt-TiN and Pt-C catalysts are given in table 1. The current densities observed on TiN supported catalyst are almost three times higher than that of carbon supported catalyst confirming the promoting effect of TiN support towards methanol oxidation reaction. The performance of Pt-TiN electrocatalyst under fuel cell conditions reveals peak power densities close to 396 mW/cm2 at a current density of 375 mA/cm2, at 90C. Table 1. Characteristics of methanol oxidation on TiN and carbon supported catalysts in acidic medium. Material Onset Ep (mV) Ip EAA Ip Ip/Ib E=Ep-Eb potential (mA/mg (cm2/mg)b (mA/cm2 (mV) of Pt)a of Pt)c (mV) Pt-TiN 170 720 56 78.4 0.714 1.24 82 Pt-C 250 700 18 68.6 0.262 0.98 106 a Mass activity; Ip is the forward peak current and Ib is the reverse peak current; Ep and Eb are forward and reverse peak potentials. b Electrochemically active area (EAA) c Current density normalized for EAA Figure 3. In-situ FTIR spectra on bare TiN surface as a function of applied DC bias vs.SCE. The spectra are shown in regions of 1000 to 2000 cm-1 (A) and 2500 to 4000 cm-1 (B). Electrolyte used is 0.5 M methanol in 0.5 M H2SO4. Reference spectrum is obtained at 0 V. In-situ FTIR spectroelectrochemical measurements have been carried out to understand the intermediates and products formed during methanol oxidation. TiN surface is highly reflective and is quite amenable for reflectance IR studies. Figure 3 shows the potential dependant spectral characteristics of TiN in methanolic sulphuric acid. The bands observed at 1600 and 3600 cm-1 correspond to –OH bending and stretching vibrations of adsorbed water molecules. Interestingly, bands corresponding to adsorbed water are observed even at remarkably low over potentials of around 0.1 V vs. SCE where CO poisoning of Pt can be very severe. This experiment confirms the ability of inexpensive TiN to function like expensive Ru in fuel cell catalysis. Similar studies have been carried out for ethanol electrooxidation on TiN supported catalysts such as Pd, Pt and PtRu in acidic as well as alkaline conditions. Adherence of fuel cell catalyst on to TiN and carbon support is followed by cycling the electrode potential continuously as shown in figure 4. The adherence of Pd on TiN surface is very good and the stability tests reveal that Pd adheres and remains on TiN for a long time as compared to carbon support. Figure 4. Cyclic voltammograms of Pd-C (A) and Pd-TiN (B) in 1 M KOH at 100 mV/s. Pd loading used is 83 µg/cm2. In the chapter on borohydride oxidation, bare TiN electrode is used for the electrochemical oxidation of sodium borohydride. In direct borohydride fuel cells (DBFC), H2 evolution that occurs at low over voltages decreases the apparent number of electrons transferred and consequently the fuel cell efficiency. TiN has been shown to be a relatively H2 evolution-free electrocatalyst for borohydride oxidation (figure 5A). As shown in figure 5A, no H2 oxidation is observed (below -0.5 V) on TiN surface with increase in concentration of borohydride. This point to the fact that direct oxidation of borohydride is very favourable on TiN electrode and is confirmed by fuel cell measurements as shown in figure 5B. Non-platinum DBFCs using TiN as the anode (borohydride oxidation) and prussian blue supported carbon (PB-C) as the cathode (oxygen or hydrogen peroxide) electrocatalysts (figure 5B) reveal peak power density of 107 mW/cm2 for a current density 130 mA/cm2, at 80C. Figure 5. Cyclic voltammograms of TiN in 1 M NaOH containing varying concentrations of borohydride at a scan rate of 20 mV/s (A). Polarization studies of DBFC with TiN anode catalyst and PB-C (prussian blue supported on carbon) cathode catalyst (B). Anolyte is 0.79 M borohydride in 5 M NaOH and catholyte is 2.2 M acidified H2O2. The second aspect of the thesis is related to the use of TiN to prepare visible light active, nitrogen doped TiO2 (N-TiO2). This is carried out by electrochemical anodization of TiN in 0.5 M HNO3 at 1.4 V. The X-ray photoelectron spectroscopy (XPS) suggests the formation of oxide phase on anodized TiN surface (figure 6A) and is confirmed by reflectance UV-Visible spectroscopy. The visible light activity is used for the sunlight induced reduction of graphene oxide to reduced graphene oxide. As shown in the Raman spectra (figure 6B), a negative shift of the D and G band positions by about 20 cm-1 and the intensity ratio reversal after reduction confirms the formation of reduced graphene oxide on N-TiO2. Figure 6. (A) Ti (2p) region of XPS of fresh TiN and anodized TiN. Anodization has been carried out at 1.4 V vs. SCE in 0.5 M HNO3. (B) Raman spectra of exfoliated graphene oxide on anodized TiN before and after sunlight induced reduction. In summary, TiN has been shown to be an active support material for fuel cell catalysts in the present studies. The appendix details the basic electrochemical studies on TiN using various redox couples, electroploymerization of aniline and the formation of nanostructures on TiN surface. (For figures pl refer the abstract pdf file)

Materials for studies related to nanoscience and nanotechnology have gained tremendous attention owing to their unique physical, chemical and electronic properties. Among various anisotropic nanostructures, one dimensional (1D) materials have received immense interest in numerous fields ranging from catalysis to electronics. Imparting multi-functionality to nanostructures is one of the major areas of research in materials science. In this direction, use of nanosized materials in energy systems such as fuel cells has been the subject of focus to achieve improved performance. Tuning the morphology of nanostructures, alloying of catalysts, dispersing catalytic particles onto various supports (carbon nanotubes, carbon nanofibers, graphene, etc.) are some of the ways to address issues related to electrochemical energy systems. It is worth mentioning that highly stable and corrosion resistant electrodes are mandatory as electrochemical cells operate under aggressive environments. Additionally, carbon, which is often used as a support for catalysts, is prone to corrosion and is subsequently implicated in reduced performance due to poor adherence of catalyst particles and loss in electrochemically active area. Hence, there is a quest for the development of stable and durable electrocatalysts / supports for various studies including fuel cells. The present thesis is structured in exploring the multi-functional aspects of titanium carbide (TiC), an early transition metal carbide. TiC, a fascinating material, possesses many favorable properties such as extreme hardness, high melting point, good thermal and electrical conductivity. Its metal-like conductivity and extreme corrosion resistance prompted us to use this material for various electrical and electrochemical studies. The current study explores the versatility of TiC in bulk as well as nanostructured forms, in electrical and electrochemical studies towards sensing, electrocatalytic reactions and active supports. 1D TiC nanowires (TiC-NW) are prepared by simple solvothermal method without use of any template and are characterized using various physico-chemical techniques. The TiC-NW comprise of 1D nanostructures with several µm length and 40 ± 15 nm diameter (figure 1). Electrical properties of individual TiC-NW are probed by fabricating devices using focused ion beam deposition (FIB) technique. The results depict the metallic nature of TiC-NW (figure 2). Figure 1. (a) SEM, (b) TEM and (c) HRTEM images of TiC-NW prepared by solvothermal method. Figure 2. (a) SEM image and (b) I-V characteristics of TiC-NW - based device as a function of temperature. The contact pads are made of Pt. Subsequently, oxidized TiC nanowires are prepared by thermal annealing of TiC-NW, leading to carbon - doped TiO2 nanowires (C-TiO2-NW) (figure 3). Photodetectors are fabricated with isolated C-TiO2-NW and the device is found to respond to visible light (figure 3) radiation with very good responsivity (20.5 A/W) and external quantum efficiency (2.7 X 104). The characteristics are quite comparable with several reported visible light photodetectors based on chalcogenide semiconductors. Figure 3. (a) HRTEM, (b) EDAX, (c) Scanning TEM-DF images of C-TiO2-NW along with (d) Ti (e) O and (f) C mapping. (g) Current – voltage curves of single C-TiO2-NW recorded in dark (black) and in presence of visible light radiation (red) of intensity 57.7 mW/cm2 at 25oC. Inset of (g) shows the SEM image of the device (top) and schematic illustration of fabricated photodetector (bottom). The next chapter deals with the electrochemical performance of TiC demonstrated for studies involving oxygen reduction and borohydride oxidation reactions. Electrochemical oxygen reduction reaction (ORR) reveal that TiC-NW possess high activity for ORR and involves four electron process while it is a two electron reduction for bulk TiC particles (figure 4). The data has been substantiated by density functional theory (DFT) calculations that reveal different modes of adsorption of oxygen on bulk and nanowire morphologies. Stable performance is observed for several hundreds of cycles that confirm the robustness of TiC. The study also demonstrates excellent selectivity of TiC for ORR in presence of methanol and thus cross-over issue can be effectively addressed in direct methanol fuel cells. In the chapter on borohydride oxidation, bare TiC electrode is explored as a catalyst for the oxidation of borohydride. One of the major issues in direct borohydride fuel cells (DBFC) is the hydrolysis of borohydride that happens on almost all electrode materials leading to low efficiency. The present study reveals that TiC is a very good catalyst for borohydride oxidation with little or no hydrolysis of borohydride [figure 5 (a)] under the experimental conditions studied. Further, shape dependant activity of TiC has been studied and fuel cell performance is followed [figure 5 (b)]. Polarization data suggests that the performance of TiC is quite stable under fuel cell experimental conditions. Figure 4. (a) Linear sweep voltammograms for ORR recorded using (i) bulk TiC particles and (ii) TiC-NW in O2-saturated 0.5 M KOH at 1000 rpm. Scan rate used is 0.005 Vs-1. (b) Variation of number of electrons with DC bias. Black dots correspond to TiC bulk particles while red ones represent nanowires. Figure 5. (a) Cyclic voltammograms of borohydride oxidation on TiC coated GC electrode in 1 M NaOH containing 0.1 M NaBH4. Scan rate used is 0.05 Vs-1. (b) Fuel cell polarization data at 70oC for DBFC assembled with (i) bulk TiC particles and (ii) TiC-NW as anode catalysts and 40 wt% Pt/C as cathode. Anolyte is 2.1 M NaBH4 in 2.5 M NaOH, and catholyte is 2.2 M H2O2 in 1.5 M H2SO4. Anode loading is 1.5 mg cm-2 and cathode loading is 2 mg cm-2. The corrosion resistance nature of TiC lends itself amenable to be used as an active support for catalytic particles (Pt and Pd) for small molecules oxidation reactions. In the present study, electro-oxidation of methanol, ethanol and formic acid have been studied. As shown in figure 6 (a), the performance of Pd loaded TiC (Pd-TiC) is found to be higher than that of Pd loaded carbon (Pd-C) suggesting the active role of TiC. The catalytic activities of TiC-based supports are further improved by tuning their morphologies. Figure 6 (c) reveals that the activities are higher in case of Pd-TiC-NW than that of Pd-TiC. Figure 6. (a) Cyclic voltammograms of Pd-TiC and Pd-C for ethanol oxidation, (b) T EM image of Pd-TiC-NW and (c) voltammograms of Pd-TiC-NW in N2-saturated 1 M ethanol in 1 M KOH medium, scan rate used is 0.05 Vs-1. The next aspect explored, is based on the preparation of C-TiO2 and its use as a substrate for surface enhanced Raman spectroscopy (SERS). Carbon doped titanium dioxide is prepared by thermal annealing of TiC. It is observed that the amount of dopant (carbon content) is dependent on the experimental conditions used. SERS studies using 4¬mercaptobenzoic acid (4-MBA) as the analyte, indicates that C-TiO2 [figure 7 (a)] enhances Raman signals based on chemical interactions between the analyte and the substrate. Raman signal intensities can be tuned with the amount of carbon content in C¬TiO2. Enhancement factors are calculated to be (7.7 ± 1.2) x 103 (for 4-MBA) and (1.7 ± 1.2) x 103 (for 4-nitrothiophenol). The SERS substrates are found to be surface renewable using visible light, a simple strategy to re-use the substrate [figure 7 (b)]. The regeneration of SERS substrates is based on self cleaning action of TiO2 that produces highly reactive oxygen containing radicals known to degrade the molecules adsorbed on TiO2. Thus, the versatility of TiC has been demonstrated with various studies. In addition to using TiC-based materials, nanoparticles of Rh, Ir and Rh-Ir alloy structures have also been used for borohydride oxidation reaction. This is explained in the last section. In Appendix-I, preliminary studies on the preparation of TiC-polyaniline (PANI) composites using liquid-liquid interfacial polymerization is explained. Raman spectroscopy results suggest that the presence of TiC-NW makes PANI to assume preferential orientation in the polaronic (conducting) form. Appendix-II discusses the role of TiC-NW as a fluorescence quencher for CdS semiconductor nanoparticles.