Catalytic generation and storage of hydrogen from hydrolysis of sodium-borohydride under pressure. Application in a Hydrogen/Oxygen fuel cell

CATALYTIC GENERATION AND STORAGE OF HYDROGEN FROM HYDROLYSIS OF SODIUM-BOROHYDRIDE UNDER PRESSURE APPLICATION IN A HYDROGEN/OXYGEN FUEL CELL by Maria Josefina Figueira Ferreira A dissertation submitted to the Department of Chemical Engineering and to the Department of Mechanical Engineering, Faculty of Engineering from University of Porto, Portugal, in conformity with the requirements for the degree of Master in Fundamentals and Applications of Fluid Mechanics This dissertation was supervised by Dr. Alexandra Maria Pinheiro da Silva Ferreira Rodrigues Pinto, CEFT, Department of Chemical Engineering, Faculty of Engineering of the University of Porto, Portugal November 2009 ii iii To my son, Luís To the memory of my dearly loved grandfather, José da Silva iv v PREFACE This dissertation on hydrogen generation from catalytic hydrolysis of sodium borohydride has grown out from the necessity of finishing my Master course in Fundamentals and Applications of Fluid Mechanics, started on the year 2003, at FEUP. After a gap of four years, in early September 2007, I had the privilege of speaking with Dr. Carlos T. Pinho, the Head of the Master course, who offered me the opportunity of working on EDEN’s Project (http://www.h2eden.com) – Task PR07.33.01 – Hydrogen generation via chemical hydrides, on a 8.5 months scholarship, funded by the Agência de Inovação S.A. of Portugal. This was my first experience on the interesting field of clean energies! Dr. Alexandra Pinto, the scientific leader of Task PR07.33.01, with whom I’ve been working since 15 of October 2007, gently accepted my invitation to be the supervisor of this dissertation and I am deeply thankful for that. Hydrogen generation and storage from chemical hydrides, specifically from sodium borohydride, to fuel hydrogen/oxygen fuel cells is a vast and rapidly evolving research field since the late 1990s. This dissertation does not provide an exhaustive review; my ambition was to write a text that describes one year and half of intensive experimental work and that shows my delight in understanding the catalysed hydrolysis reaction of sodium borohydride (NaBH 4 ), thus closing the old objective of finishing my Master course. The start of this investigation turned out to be a hard work, because my previous experience on the subject was scarce; but I have been fortunate in having a supervisor who gave me all the support and freedom as the research evolved. Beginning with the aim of “producing and simultaneously storing molecular hydrogen due to solubility’s effects in the remaining solution after NaBH 4 hydrolysis completion in a batch reactor” (see Pinto et al., Int. J. Hydrogen Energy, 31 (2006) 1341-47), the work progresses following three main lines of kinetic experiments: alkali, alkali free and less Polar Organic Polymeric Solutions (lPOPS) hydrolysis of sodium borohydride under pressure. Therefore, three main individual chapters where written, in which the experimental results are shown and vi compared with published work. In this dissertation, most chapters start with a specific related literature review, and the need for a separate chapter of literature background in the mainly body text was not felt. In this sense, the information in each individual chapter can be understood without the need to read the precedents ones. It is my hope that the readers understand the lines beneath this dissertation and recognize the importance, which is particularly high in portable applications, of producing hydrogen via sodium borohydride hydrolysis, with the goal of feeding a polymer electrolyte membrane (PEM) fuel cell on demand. At last, a final word for Hydrogen: being the simplest and the most abundant atom both on the Universe and in Earth and having the highest energy to weight ratio (12.24×10 4 kJ/kg) of all the energy sources known today, are not prime reasons to change energetic paradigm? M. Josefina F. Ferreira vii ABSTRACT Ferreira, M.J.F., University of Porto, Faculty of Engineering, November 2009. Catalytic generation and storage of hydrogen from hydrolysis of sodium borohydride solutions under pressure – application in a hydrogen/oxygen fuel cell. Supervisor: Dr. Alexandra M.F.R. Pinto. The present dissertation aimed at studying the catalytic generation and storage of hydrogen from hydrolysis of sodium borohydride solutions under pressure. The catalytic hydrolysis of sodium borohydride (NaBH 4 ) was studied under pressure, up to 2.6 MPa, using a reused ruthenium nickel based powered catalyst, in three batch reactors with internal volumes of 0.645, 0.369 and 0.228 l, made of stainless-steel and positioned vertically. Three different types of sodium borohydride hydrolysis were considered, two of them in the presence of an inhibitor (sodium hydroxide), namely, alkali hydrolysis and also less Polar Organic Polymeric Solutions (lPOPS), and the other in absence of sodium hydroxide, designated by alkali free hydrolysis. The effects of temperature, sodium borohydride concentration, sodium hydroxide concentration, catalyst concentration and system pressure, on the hydrogen gas generation rate, were investigated. Particular importance was given to the effects of reactor bottom geometry on hydrogen generation yields, rates and induction times. Two different geometries were adopted for this purpose - a flat and a conical bottom shapes. The reactor with the conical shape significantly increases the reaction rates and decreases the lag time. Successive loadings of reactant solution with and without magnetic stirring were also presented to evaluate the capability of generating H 2 without recharging the reactor with fresh catalyst. A detailed characterization of ruthenium nickel based catalyst after 200 reutilizations is presented in terms of textural properties and of morphology. The results reveal some deterioration of the catalyst which leads to slower rates and lower hydrogen yields, but with similar lag time values. Analysis of reaction by-product by X-Ray Diffractometry revealed the presence of sodium metaborate in alkali hydrolysis and anhydrous borates in lPOPS hydrolysis. The formation of anhydrous borates is particularly important because it increases the gravimetric hydrogen storage density. During the entire experimental studies, the hydrogen generated by the catalytic hydrolysis of sodium borohydride was used to fuel a PEM fuel cell available at the CEFT research Laboratory to evaluate the capabilities of an integrated solution/application. viii ix RESUMO Ferreira, M.J.F., Universidade do Porto, Faculdade de Engenharia, Novembro 2009. Produção e contentorização de hidrogénio a partir da reacção de hidrólise catalizada de borohidreto de sódio sobre pressão – aplicação numa célula de combustível a hidrogénio/.oxigénio Supervisora: Professora Doutora Alexandra M.F.R. Pinto. O objectivo principal desta dissertação foi o estudo detalhado da produção e contentorização de hidrogénio a partir da reacção de hidrólise catalizada de borohidreto de sódio sobre pressão A hidrólise catalisada de borohidreto de sódio foi estudada sobre pressão, até 2.6 MPa, recorrendo a um catalisador em pó à base de níquel, impregnado com ruténio, não suportado, em três reactores de aço inox, com funcionamento por partidas e posicionamento vertical, com volumes internos 0.645, 0.369 and 0.228 l, respectivamente. Três tipos de hidrólise foram considerados – duas delas ocorrem na presença de um inibidor alcalino (hidróxido de sódio) e são designadas como hidrólise alcalina e como hidrólise de solução orgânica menos polar (lPOPS); a terceira ocorre sem a presença deste inibidor e é designada por hidrólise sem hidróxido. Os efeitos da temperatura, concentração do borohidreto de sódio, concentração de hidróxido de sódio, concentração de catalizador e pressão do sistema, na taxa de geração de hidrogénio molecular foram investigados. Foi dada particular atenção aos efeitos da geometria do fundo do reactor no rendimento, taxa e tempo de indução reacional. Duas geometrias diferentes para o fundo do reactor foram selecionadas com este propósitio: uma plana e a outra cónica. Alimentações sucessivas de solução reagente com e sem agitação magnética foram usadas para avaliar a actividade do catalizador. O catalizador de ruténio à base de níquel após 200 re-utilizações foi caracterizado em termos de textura e morfologia. Os resultados revelaram alguma deterioração do catalizador, traduzindo-se em taxas de produção de hidrogénio e rendimentos de reacção mais baixos, mantendo-se no entanto os tempos de indução. A análise do produto da reação por difracção de raios X revelou a presença de metaborato de sódio na hidrólise alcalina e de boratos de sódio anidridos na hidrólise de solução orgânica menos polar (lPOPS). O hidrogénio gerado por hidrólise catalítica de borohidreto de sódio foi usado para alimentar uma célula de combustível de membrana de permuta cateónica, disponível no laboratório do CEFT, para analisar a possibilidade de soluções/aplicações integradas. xvi xvii LIST OF FIGURES CHAPTER 1 Figure 1.1. Volumetric and gravimetric capacities for several possible H 2 storage technologies. [1.3] / 3 Figure 1.2. Status of hydrogen storage technologies. Capacities for both materials-only basis and total system are shown [1.9]. / 5 Figure 1.3. The first fuel cell, by William Grove (1839). / 7 Figure 1.4. (A) Schematic of a typical H 2 /O 2 PEMFC [1.13]. (B) Nafion® Polymer Electrolyte Membrane Structure and Characteristics. / 9 Figure 1.5. Front view and top view of the “MicroBoro Bus” (A – connection to the H 2 generator and storage tank; Bgas-washing bottle; CPEMFC single cell). / 10 CHAPTER 2 Figure 2.1. Photograph of the nickel based bimetallic catalyst in the form of a finely divided black powder. / 16 Figure 2.2. Photo Scanning electron microscope view of the synthesized catalyst powder (a) and associated elemental analysis EDAX (b). / 17 Figure 2.3. Nitrogen adsorption/desorption isotherms. / 18 Figure 2.4. Photograph of the main batch reactor with the two separate filling pieces used to change the internal free volume for NaBH 4 hydrolysis reactions. / 19 Figure 2.5. Schematic view of the reactors inside: reactor LG – flat bottom, and reactors MR and SR – conical bottom, with mention of the internal available free volume. / 20 Figure 2.6. Global picture of the experimental setup used for the kinetics studies of hydrogen production via catalytic hydrolysis of sodium borohydride, under pressure. (1) batch reactor; (2a-2b) data acquisition system; (3) “MicroBoro bus” with a PEM fuel cell; (4) thermostatic bath and (5) refrigeration unit. / 21 Figure 2.7. Survey picture of the: (a) main reaction vessel / LR; (b) accessories fixed on the outside lid of the reactor: 1reactor inlet bore valve, 2syringe, 3pressure probe, 4Bourdon manometer and 5reactor exhaustion needle valve; (c) ascendant view of the inside lid of the reactor: 1needle of the syringe, 2- “T.Bottom” k thermocouple and 3- “T.Top” k thermocouple; (d) top view of the inside lid of the reactor showing the left ‘holes’ after connecting the some accessories; (e) syringe with needle (150 mm length). / 22 CHAPTER 3 Figure 3.1. Hydrogen generation plot with a NaBH 4 concentration of 10%, an inhibitor concentration of 3%, at different temperatures. The reactions, for 20 cm 3 of reactant solution, were performed in the MR batch reactor (369 cm 3 ) with a proportion of Ru-Ni based catalyst/NaBH 4 : 0.4 g/g. / 35 Figure 3.2. Hydrogen generation plot with an inhibitor concentration of 3%, at 27 ºC, and with different NaBH 4 concentrations. The reactions where performed in LR batch reactor (646 cm 3 ) with a proportion of Ru-Ni based catalyst/NaBH 4 : 0.4 g/g. / 36 xviii Figure 3.3. Hydrogen generation with a NaBH 4 concentration of 10%, at 25 ºC, with different NaOH concentrations. The reactions, for 10 cm 3 of reactant solution, were performed in LR batch reactor (646 cm 3 ) with a proportion of Ru-Ni based catalyst/NaBH 4 : 0.4 g/g. / 37 Figure 3.4. Hydrogen generation with a NaBH 4 concentration of 10 wt%, an inhibitor concentration of 7wt%, at 45 ºC, for the three design reactors: LR (646 cm 3 ), MR (369 cm 3 ) and SR (229 cm 3 ). The reactions, for 20 cm 3 of reactant solution, were performed with a proportion of Ru-Ni based catalyst/NaBH 4 : 0.4 g/g. / 38 Figure 3.5. Hydrogen generation plots with a NaBH 4 concentration of 10%, an inhibitor concentration of 7%, at ≈ 26 ºC, for two batch reactors with different bottom shape. The reactions, for 10 cm 3 of reactant solution, were performed with a proportion of Ru-Ni based catalyst/NaBH 4 : 0.4 g/g. / 38 Figure 3.6. Five successive loadings of the reactant solution: 10 wt.% NaBH 4 , 7 wt.% NaOH, 83 wt.% H 2 O, with Ni-based/NaBH 4 : 0.4g/g, performed in the batch reactor LR (646 cm 3 ), showing the temperature profile inside the reactor at two specific points (bottom and top of reactor). / 39 Figure 3.7. Hydrogen generation in batch reactor LR (646 cm 3 ) of five successive loadings of the reactant solution: 10 wt.% NaBH 4 , 7 wt.% NaOH, 83 wt.% H 2 O, with Ru-Ni based catalyst/NaBH 4 : 0.4g/g. / 40 Figure 3.8. Hydrogen generation in batch reactor MR (369 cm 3 ) of seven successive loadings of the reactant solution: 10 wt.% NaBH 4 , 7 wt.% NaOH, 83 wt.% H 2 O, with Ru-Ni based catalyst/NaBH 4 : 0.4g/g. / 41 Figure 3.9. Two images of reaction “gaseous” by-product, after seven successive loadings of reactant solution (10 wt.% NaBH 4 , 7 wt.% NaOH, 83 wt.% H 2 O). The H 2 release shows the gas capability to agitate the nickel-based powdered catalyst presented at the reactor liquid phase. / 41 Figure 3.10. Hydrogen generation in batch reactor LR (646 cm 3 ) of eight successive loadings of the reactant solution: 10 wt.% NaBH 4 , 7 wt.% NaOH, 83 wt.% H 2 O, with Ru-Ni based catalyst/NaBH 4 : 0.4g/g (the 8 th loading was performed with magnetic stirring). / 42 Figure 3.11. Crystals pictures of the by-product classic hydrolysis of sodium borohydride, obtained by slow evaporation of a water solution at room temperature (≈ 25 ºC). / 43 Figure 3.12. Views of crystal structure of sodium metaborate dehydrate showing stacking layers formed by BO 3 triangles and by Na and water molecules: (a) view along the a-axis and (b) view along the caxis. Na in violet, O in red, B in pink and H in white. / 44 CHAPTER 4 Figure 4.1. Pictures showing the preparation of the reactant blend (catalyst plus NaBH 4 , both solids) to use in alkali free hydrolysis experiments. From left to right: Ru-Ni based catalyst, anhydrous sodium borohydride and the bottom of MR reactor with the mixture (stored in bottom conical shape). / 52 Figure 4.2. Hydrogen generation in the flat bottom batch reactor LR (646 cm 3 ) for experiments with Ru – Ni based catalyst/NaBH 4 : 0.4 g/g and H 2 O/NaBH 4 : 2-8 mol/mol. / 53 Figure 4.3. Hydrogen generation in the batch reactor LR (646 cm 3 ) with Ni-based/NaBH 4 : 0.4 g/g (27 and 28 times reused) and H 2 O/NaBH 4 : 2 mol/mol, after two successive loadings of pure water (of the same molar quantity). / 55 Figure 4.4. Hydrogen yield as a function of added water, H 2 O/NaBH 4 (mol/mol), in batch reactor LG (646 cm 3 ). / 55 Figure 4.5. Influence of the amount of catalyst/NaBH 4 (g/g) on hydrogen generation. / 56 Figure 4.6. Hydrogen generation in the studied three batch reactors - LR (646 cm 3 ), MR (369 cm 3 ) and SR (229 cm 3 ), with Ni-Ru based catalyst/NaBH 4 : 0.4 g/g (111, 150 and 151 times reused, respectively for LR, MR and SR) and H 2 O/NaBH 4 : 4 mol/mol. / 57 Figure 4.7. Influence of pressure on H 2 yield (Experiments done in three batch reactors at the temperature range of 289-295 K, with catalyst reused ≈150 times). / 57 xix Figure 4.8. Influence of reactor bottom shape on H 2 yield, rate and lag time (experiments performed in batch reactors LR (646 cm 3 ) and MR (369 cm 3 ), with flat and conical bottom shapes, respectively; at temperature range of 289-295 K, with catalyst reused ≈150 times). / 58 Figure 4.9. Photographs of MR reactor (with conical bottom shape) after an alkali free hydrolysis of NaBH 4 experiment. In the right, a view of the solid mixture of the by-product plus the catalyst. / 59 Figure 4.10. Temperature of the hydrolyzed NaBH 4 as a function of time by different pressures. / 60 Figure 4.11. Influence of temperature on hydrogen generation rate and yield. / 61 Figure 4.12. View along the c-axis of the crystal structure of sodium metaborate hydrate: (a) 4H 2 O, in MR at 0.69 MPa and (b) 2H 2 O, in SR at 1.26 MPa. The element color code, from black to light grey, is B, O, Na and H. / 64 CHAPTER 5 Figure 5.1 Hydrogen generation by a single loading of 20 cm 3 of reactant solution, at 45 ºC, with Ni-Ru based catalyst/NaBH 4 : 0.4 g/g: (a) MR batch reactor [0.369 L] and (b) SR batch reactor [0.229 L]. / 70 Figure 5.2. View along the a-axis of the crystal structure of sodium borate anhydrous, showing the BO 4 tetrahedral arrangement and the Na bound network. Na in violet, O in red and B in pink. / 73 Figure 5.3. View along the a-axis of the crystal structure of sodium borate anhydrous, showing the BO 3 triangular and the BO 4 tetrahedral arrangements sharing oxygen atoms and the Na bound network. Na in violet, O in red and B in pink. / 73 Figure 5.4. Gravimetric H 2 storage density (reactants basis) for the three systems studied in this chapter. / 75 CHAPTER 6 Figure 6.1. Nitrogen isotherms of powder Ru-Ni based catalyst 200 times reused. / 80 Figure 6.2. Pore size distributions of powder Ru-Ni based catalyst 200 times reused. / 80 Figure 6.3. SEM micrographs of powder Ru-Ni based catalyst, 200 times reused. Magnifications between 200 and 100 000X, increasing from (a) to (e). / 82 Figure 6.4. SEM image coupled with EDS spectrum of Ru-Ni based catalyst 200 times reused and a magnification of 200X. / 83 Figure 6.5. SEM image coupled with EDS spectrum of Ru-Ni based catalyst 200 times reused and a magnification of 1500X. / 84 Figure 6.6. SEM image coupled with EDS spectrum, at three different selected areas, of Ru-Ni based catalyst 200 times reused and a magnification of 10 000X. / 85 Figure 6.7. Photograph of Ru-Ni based catalyst 200 times reused pellet, with 10 mm in diameter, for use in XPS analysis. / 86 Figure 6.8. A survey XPS spectrum of Ru-Ni based catalyst 200 times reused. / 86 Figure 6.9. XPS spectra for Ru-Ni based powdered catalyst 200 times reused. / 87 Figure 6.10. Hydrogen generation plots with a NaBH 4 concentration of 10%, an inhibitor concentration of 7%, for two different stages of catalyst reutilization (28 times reused and 200 times reused). The reactions, for 10 cm 3 of reactant solution, were performed in LR batch reactor (646 cm 3 ) with a proportion of Ru-Ni based catalyst/NaBH 4 : 0.4 g/g, at room temperature of ≈ 25 ºC. / 89 xx xxi LIST OF TABLES CHAPTER 1 Table 1.1 – Combustion enthalpies for some common fuels [1.2]. / 2 Table 1.2 – FreedomCar requirements stated by the US DOE [1.4]. / 6 CHAPTER 2 Table 2.1 – Textural properties of catalyst powder used in this work before use. / 18 CHAPTER 3 Table 3.1 – Excess hydration factor, x, of aqueous solutions of sodium borohydride. / 28 Table 3.2 – Comparison of gravimetric efficiencies of various reactant preparations. / 29 CHAPTER 4 Table 4.1 - The FreedomCAR targets publish by US Department Of Energy (DOE). / 62 Table 4.2 - Influence of Ni-Ru based catalyst/NaBH 4 on gravimetric hydrogen density and hydrogen yield for H 2 O/NaBH 4 : 4 mol/mol. Prediction of the volumetric hydrogen density*. / 63 CHAPTER 5 Table 5.1 – Hydrogen yield for the studied reactant solutions in the two batch reactors and prevision of respective H 2 solubility effects, at 45 ºC. / 71 Table 5.2 – Values of conductivity and pH of the studied reactant solutions, before and after reaction completion. / 72 CHAPTER 6 Table 6.1 - Textural properties of powder Ni-Ru based catalyst 200 times reused sample. / 80 Table 6.2 - Specific conditions for XPS analysis of Ni-Ru based catalyst 200 times reused. / 86 Table 6.3 - Results of XPS analysis of Ni-Ru based catalyst 200 times reused. / 88 Table 6.4 - Results for yield, lag time and dtdP/ slope in hydrogen generation of 10 mL of 10wt% NaBH 4 and 7wt% NaOH solution, at two different stages of Ni-Ru based catalyst reutilization (for 28 and 200 times reused), in batch reactor LR (646 cm 3 ). / 89 xxii xxiii NOMENCLATURE yield, yield = n(H 2 ) exp / n(H 2 ) theoretical (n= number of moles) x , excess hydration factor V µ , micropore volume by t-method; V p , total pore volume, for P/P 0 =0.986. V , molar volume of the liquid t 1/2 , the half-life time, min T , absolute temperature, K S ext , (mesopore + macropore) surface areas by t-method; S BET ,BET surface area; r , the pore radius 0 /PP , relative pressure Greek letters φ , contact angle γ , its surface tension xxiv 1 CHAPTER 1 Introduction In this introductory chapter the topic and main objectives of this dissertation, entitled “Catalytic generation and storage of hydrogen from hydrolysis of sodium borohydride solutions under pressure – application in a hydrogen/oxygen fuel cell” are presented. A general introduction to hydrogen gas as an energy carrier, to sodium borohydride as hydrogen carrier and to polymer electrolyte membrane fuel cells is carried out. 1.1 GENERAL INTRODUCTION This dissertation investigates a potential improvement to hydrogen (H 2 ) generation and simultaneous storage, by catalytic hydrolysis of sodium borohydride (NaBH 4 ) under pressure, for supply fuel cells of the type PEM – Polymer Electrolyte Membrane. Due to the accelerate decrease of fossil fuels resources and the continuous growing of energy demand, the development of alternative and clean renewable energy carriers is of the utmost importance. In fact, the world may be even closer to running out of oil than usually admitted [1.1]. The transportation sector and other portable applications are areas where an alternative energy source may have a particularly great impact. Thus, employment of a concentrated energy (electricity) carrier is of vital interest and priority. In this context, hydrogen (H 2 ) is presented as an environmental friendly energy vector, since it may serve as an intermediate through which a primary energy source (for example, solar) can be effectively transmitted and consumed. 1.1.1 Hydrogen as energy carrier The hydrogen is the simplest and the most abundant atom both on the Universe and in Earth. The supreme important characteristic of hydrogen gas, the lightest and smallest molecule on our planet, is the fact that it has the highest energy to weight ratio, or CHAPTER 1 Introduction 8 Francis Bacon designed the first alkaline fuel cell. In the 1960s, NASA showed some of the potential applications for fuel cells by using them to provide power on space missions. Currently, five types of fuel cells are receiving attention for energy production in stationary or transportation applications. These are Alkaline Fuel Cells (AFC), Molten Carbonate Fuel Cells (MCFC), Phosphoric Acid Fuel Cells (PAFC), Solid Oxide Fuel Cells (SOFC), and Polymer Electrolyte Membrane Fuel Cells (PEMFC). All of these fuel cell types have been demonstrated as stationary power plants with the advantage of providing high quality on-site electrical power. The last decade has focused primarily on using PEMFC’s for transportation purposes. Fuel cells also have the benefit of producing constant amounts of electrical energy. Hydrogen is the ideal fuel for fuel cells. In all types of fuel cells mention in the above paragraph, hydrogen exhibits fast reaction kinetics and the only by product of the reaction when oxygen is used as the oxidant is water. 1.2.1 Polymer Electrolyte Membrane Fuel Cell (PEMFC) Polymer Electrolyte Membrane Fuel Cells (PEMFCs) have been the focus of a great deal of research in the last few years due to the increasing demand for clean power that can be supplied from a lightweight system. PEMFCs fall into this category due to their membrane electrode assemblies (MEA). The MEA is sandwich composed of three different layers. The first layer is the catalyst material, which consists of platinum particles (Pt) loaded into a carbon cloth support. The next layer is the PEM. Although there are a few different types of PEM materials, the industry standard is a copolymer of tetrafluoroethylene and perfluoro[2-(fluorosulfonylethoxy)propylviny] ether. This material has a trade name, Nafion®, and is made by DuPont [1.12]. The third layer is another Pt loaded catalyst layer usually of the same composition as the first catalyst layer, depending on the fuel types. The typical MEA is roughly 0.02’’ thick, and has negligible weight. Since the cell operates at temperatures below 100 ºC, the shell of the fuel cell can be made out of lightweight plastics. These characteristics make the PEMFC ideal for transportation power applications. Figure 1.4 shows a typical PEMFC [1.13] and the chemical structure for Nafion® Polymer Electrolyte Membrane Structure from DuPont. The circled portion within Figure 1.2 shows the MEA. The Polymer Electrolyte Membrane (PEM) is responsible for transporting the hydrogen ion produced through catalysis at the anode to the cathode, where it recombines with CHAPTER 1 Introduction 9 oxygen ions and electrons to form water. The transport of the hydrogen ion, a proton, is made possible by the structure of Nafion®. (A) (B) Figure 1.4. (A) Schematic of a typical H 2 /O 2 PEMFC [1.13]. (B) Nafion® Polymer Electrolyte Membrane Structure and Characteristics. The blue section is a Teflon-like, fluorocarbon backbone that has hundreds of repeating CF 2 -CF 2 -CF 2 - units. The brown section (vertical section of the molecular chain) is a side chain that connects the backbone to the sulfonic acid ions. The SO 3anions are permanently attached, but the H + ions become mobile upon hydration of the membrane. Thus, for a fuel cell to be functional, the fuel must be hydrated before entering the cell. For hydrogen, this is done by running the gas through a washing bottle. In the case of methanol or ethanol, the alcohol is in a water solution. The ion exchange capacity of Nafion® is 0.91meq/g (equivalent hydrogen mass per mass of dried Nafion®) [1.12]. In brief, the electrochemical reactions within an electrochemical cell are segregated to two half cells, where an electrochemical reaction occurs at each half cell. Therefore, the half cell reactions of a hydrogen/oxygen fuel cell, in an acidic electrolyte, are Anode: H 2 → 2H + + 2e - (1.3) CHAPTER 1 Introduction 10 Cathode: 2 1O 2 + 2H + + 2e - → H 2 O - (1.4) Overall: H 2 + 2 1O 2 → H 2 O - (1.5) The future of hydrogen as an energy carrier will be determined by the oil market and the price that society is willing to pay for an energy supply that is poor in carbon dioxide. The costs of hydrogen production are, for the time being, prohibitive of any introduction into the energy market. As soon as the world energy price permits, the hydrogen/electricity combination will be open to an interesting future. 1.2.2 The “MicroBoro Bus” – a didactic demonstration prototype The hydrogen generated in the experiments performed during this dissertation, was supplied to a PEMFC single cell that was housed in a bus like shaped mobile platform, used for didactic proposes – the “MicroBoro Bus” [1.14]. Figure 1.5 shows two pictures, a side view and a top view, of the Microboro Bus. The single cell was fed with hydrogen generated by the catalytic hydrolysis of NaBH 4 experiments. The developed low power PEM fuel cell uses an air breathing cathode and humidified hydrogen produced by the borohydride reactor. Natural convection to air feed the cathode is used when the platform is located on a stand and the energy conversion required is just the one needed to make the wheels of the platform turn. Forced convection, by means of a fan, supplies air to the cathode when the platform as a whole is programmed to move. The necessary energy to drive the fan is provided by the fuel cell [1.14]. Figure 1.5. Front view and top view of the “MicroBoro Bus” (A – connection to the H 2 generator and storage tank; Bgas-washing bottle; CPEMFC single cell). C B A CHAPTER 1 Introduction 11 1.3 OBJECTIVES AND ORGANIZATION OF THE DISSERTATION It is the objective of the present research to better understand the kinetic mechanism of catalysed hydrolysis of sodium borohydride for producing molecular hydrogen at elevated yields and rates, with high gravimetric and volumetric hydrogen densities. Hence, three main different series of kinetic experiments of hydrolysis of sodium borohydride under pressure, in the presence of reused nickel-ruthenium based catalyst, were carried out to achieve this goal. The present dissertation evaluates the following objectives: 1) Perform catalysed hydrolysis reactions of sodium borohydride in three different ways: - by joining an alkali to sodium borohydride solutions – alkali hydrolysis; - without joining an inhibitor – alkali free hydrolysis; - by joining an organic polymer or surfactant – lPOPS hydrolysis, to understand deeply the kinetic and chemistry of the hydrolysis and to investigate the influence of pressure, temperature and reactant concentrations on the reactions rates and yields; 2) Verify the influence of reactor bottom shape on hydrogen generation yields and rates; 3) Produce hydrogen by the concept of successive loadings of reactant solution; 4) Analyze the influence of magnetic stirring in refuelling hydrolysis; 5) Study the catalyst reutilization/durability and performance/capacity, after 200 times of being used; 6) Characterise the catalyst, in terms textural properties, morphology and XPS analysis, after a long period of reutilization. The organization of the dissertation involves eight chapters concerned with: Chapter 1, Introduction Introduces the research topic and present the main objectives of this dissertation, entitled “Catalytic generation and storage of hydrogen from hydrolysis of sodium borohydride solutions under pressure – application in a hydrogen/oxygen fuel cell”. Short reviews of hydrogen as a energy carrier, sodium borohydride as hydrogen carrier and PEM fuel cells are presented. Chapter 2, Experimental techniques and apparatus . In this chapter the materials used in all the experimental studies are described, including the nickel based bimetallic catalyst. For the latter, a textural and morphological characterization is provided. The experimental rig, with special emphasis given to the reactor design, and also a brief mention to reaction products analysis are offered. CHAPTER 1 Introduction 12 Chapter 3, Alkali hydrolysis/Results for sodium borohydride kinetic experiments . This chapter presents the results of alkali hydrolysis of sodium borohydride in the presence of a reused nickel based bimetallic catalyst. The effects of temperature, NaBH 4 concentration, NaOH concentration, and system pressure and reactor bottom shape on the hydrogen generation rate are investigated. Particular importance is given to the effect of successive loadings of reactant on catalyst activity, with and with out magnetic stirring. Selected XRD analysis for the reaction by-product is also presented. Chapter 4, Alkali free hydrolysis/Results for sodium borohydride kinetic experiments . This chapter shows the results obtained on alkali free hydrolysis of sodium borohydride in the presence of a reused nickel based bimetallic catalyst. A discussion on the stoichiometry of hydrolysis, in terms of the number of added water molecules to solid NaBH 4 , is carried out, based on the results of hydrogen generation rate and yields. Emphasis is given to reactor bottom design effects on the hydrogen production. The reaction by-products are analysed by XRD. Chapter 5, less Polar Organic Polymeric Solutions for sodium borohydride kinetic experiments . This chapter illustrates the results obtained on lPOPS hydrolysis of sodium borohydride plus small additions of an organic polymer and surfactant, in the presence of a reused nickel based bimetallic catalyst. With the aim of improve the solubility of H 2 in the remaining solution, a discussion of the results of hydrogen generation rate and yields is carried out. The reaction by-products are analysed by XRD. Chapter 6, Nickel based bimetallic catalyst characterization / results after 200 reutilizations . This chapter focuses on the characterization of the nickel based bimetallic catalyst after 200 reutilizations. Textural properties based on nitrogen adsorption isotherms; surface morphology by scanning electron microscopy (SEM) coupled with EDS spectroscopy and X-ray photoelectron spectroscopy (XPS) analysis, were used to characterized the powder catalyst. Chapter 7, Conclusion . This chapter is devoted to conclusions and suggestions for future work. Conclusions drawn from the results indicate that the reused ruthenium nickel based catalyst is capable of catalyzing the sodium borohydride solutions at sufficient rates, in the three schemes presented in this dissertation for comparison – alkali hydrolysis, alkali free hydrolysis and lPOPS hydrolysis of sodium borohydride for hydrogen generation under pressure. CHAPTER 1 Introduction 13 1.4 CONCLUSIONS High performance fuel cells rely on a source of hydrogen for operation. Distribution of hydrogen either in the gaseous or liquid state is energy intensive and requires extensive infrastructure development to become a safe and reliable source of energy. In this introductory chapter a short literature survey identified hydrogen gas as energy carrier and chemical hydrides as high density hydrogen carriers. Special emphasis was given to sodium borohydride which is stable in basic solution and can easily be manipulated. The inherent capacity for on-site hydrogen generation further enhances the projection for NaBH 4 as hydrogen vector for supplying hydrogen/oxygen fuel cells. Hence, a briefly outline of PEM fuel cells was also exposed. Finally, the feasibility of borohydride source hydrogen in an operational PEM fuel cell has been established. In truth, the hydrogen generated in the experiments performed during this dissertation, was supplied to a PEM single fuel cell that was housed in a bus like shaped mobile platform, used for didactic proposes – the “MicroBoro Bus” [1.14] 1.5 REFERENCES [1.1] ‘Crucial data distorted as global oil runs dry’, The Guardian weekly, 13-19 Nov. 2009. [1.2] ‘Catalytic generation of hydrogen from the hydrolysis of sodium borohydride. Application in a hydrogen/oxygen fuel cell. C.M. Kaufman, PhD Thesis, 1981. [1.3] http://www1.eere.energy.gov/hydrogenandfuelcells/storage/tech_status.html. [1.4] U.S. Department of Energy Hydrogen Program. Available from: http://www.hydrogen.energy.gov/ . [1.5] H.I. Schlesinger, H.C. Brown, A.E. Finholt, J.R. Gilbreath, H.E. Hoeskstra, K. Hyde, Sodium borohydride, its hydrolysis and its use as a reducing agent and in the generation of hydrogen, J. Am. Chem. Soc. 75 (1953) 215-219. [1.6] E. Marrero-Alfonso, J.R. Gray, T.A. Davis, M.A. Matthews, Hydrolysis of sodium borohydride with steam, Int. J. Hydrogen Energy 32 (2007) 4717-4722; Minimizing water utilization in hydrolysis of sodium borohydride: the role of sodium metaborate hydrates, Int. J. Hydrogen Energy 32 (2007) 4723-4730. [1.7] D.A. Lyttle, E.H. Jensen, W.A. Struck, A simple volumetric assay for sodium borohydride, Analytical Chem. 24 (1952) 1843-1844. [1.8] V.C.Y. Kong, F.R. Foulkes, D.W. Kirk, J.T. Hinatsu, Development of hydrogen storage for fuel cell generators. I – Hydrogen generation using hydrolysis hydrides, Int. J. Hydrogen Energy 124 (1999) 665-675. [1.9] A. Züttel, P. Wenger, S. Rentsch, P. udan, Ph. Mauron, Ch. Emmenegger, LiBH4 a new hydrogen storage material. J. Power Sources 118 (2003) 1-7. CHAPTER 1 Introduction 14 [ 1.10 ] U.B. Demirci, O. Akdim, P. Miele, Ten-year efforts and a no-go recommendation for sodium borohydride for on-board automotive hydrogen storage, Int. J. Hydrogen Energy 34 (2009) 2638-2645. [ 1.11 ] W.R. Grove, Phi. Mag, 14 (1839) 127. [ 1.12 ] DuPont, Nafion® perfluorinated polymer products, Product Information Guide, 2000. [ 1.13 ] European Fuel Cell GMBH, 2005. [1.14] Rangel C.M., Silva R.A., Pinto A.M.F.R. Fuel cells and on-demand hydrogen production: didactic demonstration prototype. Proceedings of the International Conference in Power Engineering and Electric Drives”, Eds. L.S. Martins and P. Santos, Setúbal, Portugal, September 12-14, 2007, paper 237. 15 CHAPTER 2 Experimental techniques and apparatus In this chapter the materials used in all the experimental studies are described, including the nickel based bimetallic catalyst, the experimental setup with special emphasis to the reactor design, and also a brief mention of reaction products analysis. 2.1 INTRODUCTION A detailed description of the materials used in the kinetic studies of the catalytic hydrolysis of sodium borohydride is presented, both for the reactant solutions and for the nickel-based bimetallic catalyst. The methods of preparation of the reactant solutions and/or pretreatment of the catalyst, the experimental rig design and detailed experimental plan for three different kinds of kinetics studies: alkali hydrolysis, alkali free hydrolysis and less Polar Organic Polymeric Solutions (lPOPS) are presented. The techniques used for reaction products analysis: pure molecular hydrogen and sodium borates by-product are also mentioned. 2.2 MATERIALS 2.2.1 Sodium borohydride Anhydrous sodium borohydride powder (96% purity) was provided by MERCK (No.: 1.06371.0100) and stored in a dessicator until use. In the experiments with the fuel (NaBH 4 ) in the liquid form, used in alkali hydrolysis and in lPOPS hydrolysis, the reactant solutions were stabilized by the addition of small quantities of sodium hydroxide pellets as a hydrolysis inhibitor. The sodium hydroxide CHAPTER 2 Experimental techniques and apparatus 16 (98% purity) was supplied by EKA (No.: 1310.73.2). Deionised water was used to prepare all the aqueous reactant solutions and, in all of them, the initial pH was approximately 14. In the experiments with alkali free hydrolysis of NaBH 4 for hydrogen generation, the anhydrous sodium borohydride powder was used in the solid state. 2.2.2 Nickel based bimetallic catalyst The catalyst used in all the experiments reported in this dissertation is a nickel based bimetallic catalyst, synthesized at LNEG – Laboratório Nacional de Energia e Geologia, Fuel Cells and Hydrogen Unit, Lisbon – Portugal. Catalyst preparation The catalyst, in the form of a finely divided black powder (see Figure 2.1), unsupported, was prepared from a mixture of precursors in deionised water - impregnating small quantities of ruthenium in the nickel salts (Riedel-de Haën) - by chemical reaction with 10 wt% stabilised borohydride solution (Rohm and Haas), as the reducing environment. When the reduction was complete the catalyst was appropriately decanted, washed, filtered, dried and heat-treated at 110ºC. The catalyst was kept in a dessicator until use. Figure 2.1. Photograph of the nickel based bimetallic catalyst in the form of a finely divided black powder. Catalyst samples were analyzed for textural properties. The determination of surface areas (S BET ), by N 2 physisorption was undertaken at 77K, using a Quantachrome Instruments Nova 4200e apparatus. Prior to the analysis, the samples (0.127 g) were degassed at 160 ºC for 3 h. Pore size distributions were obtained from the desorption branch of the isotherms using the Barrett, Joyner and Halenda (BJH) method. The micropore volumes and mesopore surface areas were determined by the t-method. CHAPTER 2 Experimental techniques and apparatus 17 Morphology and elemental composition analysis of the catalyst were done on a Scanning Electron Microscopy (SEM) coupled with EDS unit FEI Quanta 400 FEG ESEM/EDAX Genesis X4M operating at 15 kV in low vacuum mode (LVSEM for uncoated non conductive sample). The powder sample was prepared by simple dispersion over a double side adhesive carbon tape. X-Ray Photoelectron spectroscopy (XPS) of the catalyst before use, after the powder sample was made into a pellet of 10 mm diameter, was carried out on a VG Scientific ESCALAB 200 A spectrometer using Mg Kα (1,253.6 eV) as a radiation source. The photoelectrons were analysed at a takeoff angle of 0º. Catalyst characterization before used The catalyst used in this work is a powder containing Ni and Ru species with nanometric particle size, as evident in Figure 2.2a). The amount of ruthenium is small and was not detectable by EDAX, see Figure 2.2b). However, by XPS analysis less than 1 At% Ru was indicated. (a) Counts (a.u) (b) (a) Energy (keV) Figure 2.2. Photo Scanning electron microscope view of the synthesized catalyst powder (a) and associated elemental analysis EDAX (b). The nitrogen (N 2 ) adsorption and desorption data as a function of pressure (at liquid nitrogen temperature) are displayed in Figure 2.3. The results shows that de N 2 isotherm CHAPTER 2 Experimental techniques and apparatus 24 the bottom of the reactor. After sealing perfectly the reaction vessel, a stoichiometric amount of pure liquid water is rapidly injected into the reactor by means of a syringe. The main inspiration behind developing this experimental item was to produce pure hydrogen gas with very high gravimetric and volumetric densities. In a few words, the alkali free hydrolysis experimental work aims at studying the influence of: i) H 2 O/NaBH 4 (mol/mol) ratios between 2 and 8 mol; ii) reactor pressure; iii) reactor bottom shape; and iv) catalyst/NaBH 4 (g/g) ratios of 0.2 and 0.4 g, on hydrogen generation rates and yields. These experimental tests were performed without magnetic stirring and without temperature control. 2.4.3 less Polar Organic Polymeric Solutions (lPOPS) hydrolysis less Polar Organic Polymeric Solutions ( lPOPS ) hydrolysis of sodium borohydride for hydrogen generation under pressure embraces a series of experimental work done in the two batch reactors with conical bottom shape (MR and SM), in the presence of powder nickel-based bimetallic catalyst, reused about 160 times. The reactant solutions were produced by small additions of an organic polymer or surfactant to the stabilized aqueous solution of NaBH 4 . Also in this case, a proper quantity of powder reused Ni-Ru based catalyst, in the proportion of catalyst/NaBH 4 : 0.4 g/g, measured in an analytical balance, was stored in the bottom of the reactor. After sealing perfectly the reaction vessel, 20 mL of the reactant solution was rapidly injected into the reactor by means of a syringe. After reaction completion, magnetic stirring inside de reactor was allowed for 30 minutes to promote complete saturation of H 2 with the remaining by-product solution. All the experimental tests were performed with temperature control at 45 ºC. The motivation behind developing the lPOPS was to increase the non-polar hydrogenelectrolyte interactions and hence improve the affinity for H 2 storage in the liquid phase by solubility’s effects. Therefore, the lPOPS hydrolysis experiments aims at studying the influence of: i) the addition of 0.25 wt% CMC (Carboxil-Methyl-Cellulose) to the stabilized solution of 10 wt% NaBH 4 and 7 wt% NaOH; CHAPTER 2 Experimental techniques and apparatus 25 ii) the addition of 0.25 wt% SDS (Sodium-Dodecyl-Sulphate) to the stabilized solution of 10 wt% NaBH 4 and 7 wt% NaOH; and iii) the reactor pressure, on hydrogen generation rates and yields, conducive to H 2 solubility effects. 2.5 REACTION PRODUCTS ANALYSYS 2.5.1 Molecular hydrogen The main product of the hydrolysis reactions mentioned above is hydrogen gas. A simple verification that the gas feed to the PEMFC, when the reactor exhaustion needle valve is opened, is molecular hydrogen, we forced the gas to go all the way through a gas-washing bottle, filled with ultra pure water (conductivity of 1.48 µ S/cm at 25 ºC). After a period of noteworthy H 2 bubbling, the values of pH and conductivity of the remaining water inside the gas-washing bottle were checked. In general, no major variance in the values of pH and conductivity were found (typical final marks of 7.5 and 10 µ S/cm were found, respectively, at 25 ºC). Although the proper analytical method to verify the composition of the exhausted gas from the reaction vessel is the gas chromatography, it is reasonable to assume that the great majority of the gas captured during the experimental work, which is used to feed a PEM fuel cell hosted in a “Microboro bus” didactic prototype, is essentially hydrogen. 2.5.2 Sodium borates by-products The different course of hydrolysis reactions studied in this dissertation also produced certain by-products, namely sodium borates (or oxygen-containing compounds of boron). The by-product of the hydrogen generation reactions was dried by slow evaporation of a water solution at the environmental conditions vivid in the laboratory. Suitable crystals were then obtained and subsequently analyzed using one of the most common analytical techniques to identify solid samples: the X-Ray Diffractometry (XRD). In fact, the X-ray diffraction is primarily of value for the study of crystalline material. X rays are reflected off the surfaces of crystals, and by studying the patterns of reflection as the crystalline material is rotated in the path of the X rays, much information about the structure of the material can be obtained. CHAPTER 2 Experimental techniques and apparatus 26 The diffraction data were colleted at 293 K with a Gemini PX Ultra equipped with MoK α radiation and with CuK α radiation (facility of IBMC - Instituto de Biologia Molecular e Celular, University of Porto, Portugal). 2.6 CONCLUSIONS In this chapter the materials used in all the experimental work were described, including the nickel based bimetallic catalyst. The experimental setup gives special emphasis to the reactor design, made of stainless steel AISI 316. To study the catalytic hydrolysis of sodium borohydride under pressure, the O-ring used to seal the vessel was installed axially instead of a radially. Moreover, to follow the reaction temperature at two different cottas inside the vessel, two k thermocouples were positioned at distinct points: one near the bottom and the other very close to the top. To study the influence of reactor bottom shape on the hydrogen generation rates - this constitutes one of the originalities of the present dissertation, two different bottom geometries were designed – one flat and the other conical. The latter has the purpose of enabling non-dispersible effects during the contact of the catalyst with the reactant solution. The hydrolysis experiments were performed in the presence and in the absence of an alkali, namely, alkali hydrolysis and alkali free hydrolysis of sodium borohydride, respectively. Special attention is given to the refuelling experiments in order to validate the catalyst activity. Also, a small addition of an organic polymer or surfactant was added to the stabilized reactant solution with the objective of studying the H 2 solubility effects in the remaining solution inside the reactor after reaction completion. Finally, suitable crystal of the by-products of the hydrolysis of sodium borohydride, made by slow evaporation of a water content, where analyzed by XRD. 2.7 REFERENCES [2.1] C. San Marchi, B.P. Somerday, S.L. Robinson, Permeability, solubility and diffusivity of hydrogen isotopes in stainless steels at high gas pressures, Int. J. Hydrogen Energy 32 (2007) 100-116. [2.2] A.M.F.R. Pinto, D.S. Falcão, R.A. Silva, C.M. Rangel, Hydrogen generation and storage from hydrolysis of sodium borohydride in batch reactors, Int. J. Hydrogen Energy 31 (2006) 1341-1347. 27 CHAPTER 3 Alkali hydrolysis Results for sodium borohydride kinetic experiments This chapter presents the results of alkali hydrolysis of sodium borohydride in the presence of a reused nickel based bimetallic catalyst. The effects of temperature, NaBH 4 concentration, NaOH concentration, system pressure and reactor bottom shape on the hydrogen generation rate are investigated. Particular importance is given to the effect of successive loadings of reactant on the catalyst activity. Selected XRD analysis for the reaction by-product is also presented. 3.1 INTRODUCTION Sodium borohydride has been extensively studied as a hydrogen storage medium because it is readily available, relatively inexpensive, and more stable than other chemical hydrides (for instance, lithium or aluminium borohydrides) [3.1-3.7]. Almost all prior work uses an excess of liquid water, with a catalyst or promoter to achieve high yields and rates of hydrogen. NaBH 4 reacts with water to generate molecular hydrogen according to the hydrolysis reaction shown in Eq.(3.1): NaBH 4 + (2+x) H 2 O → NaBO 2 .xH 2 O + 4 H 2 + heat. (3.1) Ideal hydrolysis is attained for 0 = x [3.1], but in practice excess of water is required to pre-dissolve the hydride for storage or to keep the by-products in solution (accounting for the fact that the solid by-product, hydrated metaborate (NaBO 2 .xH 2 O), can exist with various degrees of hydration [3.8-3.10]). CHAPTER 3 Alkali hydrolysis 28 Table 3.1 show values of x (excess hydration factor) corresponding to aqueous solutions of sodium borohydride, ranging from 5 wt% to 35 wt% (approximately the solubility limit of NaBH 4 ), found in most hydrolysis schemes published in literature. Table 3.1 – Excess hydration factor, x, of aqueous solutions of sodium borohydride. NaBH 4 weight, % 5 7.5 10 20 30 35 x 37.9 23.9 16.9 6.4 2.9 1.9 From the values in Table 3.1 it is clear that solubility considerations alone cause chemical hydride systems to lose significant storage efficiency relative to the hydrogen content on a materials only basis. Hydrolysis is typically conducted in the aqueous phase at lower temperatures, and large quantities of excess water are required because of the low solubility of both NaBH 4 and the borate by-products in water [3.11]. In pure liquid water the reaction generates basic species that are thermodynamically and kinetically metastable. Metastability causes the reaction to cease at low yields of hydrogen even though the reaction is strongly favoured thermodynamically [3.12-3.14]. The metastable species require addition of mineral acid or use of a heterogeneous catalyst to drive the reaction to completion. Furthermore, the solid products of the reaction are an unknown mixture of hydrated sodium metaborates. Hydrated products capture water that is not reduced to hydrogen, further decreasing the efficiency of the reaction. Most systems that have been engineered to date utilize a route wherein both the hydride reactant as well as the product is dissolved in excess water. Because the solubility of the chemical hydride (55g NaBH 4 /100g H 2 O at 25 ºC) [3.15] is relatively low, and the solubility of the oxides is even lower (28g NaBO 2 /100g H 2 O at 25 ºC) [3.7], such systems inherently have a greatly reduced gravimetric efficiency (hydrogen storage capacity on a materials-only basis) compared to the ideal. Table 3.2, shows the effective value of x for different amounts of water, and the corresponding hydrogen storage capacity on a materials-only basis. Ideal hydrolysis requires no excess water. Complete dissolution of NaBH 4 at 25 ºC requires extra water corresponding to x=1.81. If sufficient water is added to keep the NaBO 2 in solution, the effective value of x is 11.04. CHAPTER 3 Alkali hydrolysis 29 Table 3.2 – Comparison of gravimetric efficiencies of various reactant preparations. Molar ratio, NaBH 4 /H 2 O x H 2 storage capacity, wt% Ideal hydrolysis 1 / 2 0 10.92 Saturated NaBH 4 solution, 25 ºC 1 / 3.81 1.81 7.57 NaBO 2 solubility limit, 25 ºC 1 / 13.04 11.04 2.96 8.5 wt% NaBH 4 plus 5 wt% NaOH 1 / 21.37 19.37 1.81 Furthermore, the addition of NaOH or other base to stabilize the solutions also affects the hydrogen storage density. Shang et al. [3.11] found that the optimum concentration, where the compounds are stable and dissolve in solution, is 8.5wt% NaBH 4 /5wt% NaOH resulting in a theoretical gravimetric storage efficiency of only 1.81wt% H 2 (x=19.37). Clearly, addition of water to achieve solubility of the reactants and/or products drives the storage efficiency down rapidly. Almost all research to date is based on conducting the reaction in an aqueous mixture, with large amounts of excess water. Despite favourable thermodynamics, as said before, aqueous phase reactions with pure water give conversions of less than 5%. It has been reported that the decrease in the initial rate of hydrogen and reaction yield is due to an increase in the pH of the solution, caused by the formation of the strongly basic metaborate ion [3.1]. This stabilization under high pH conditions has also been exploited to prevent premature reaction. Fundamental investigations on the hydrolysis of chemical hydrides, particularly lithium and sodium borohydrides, were performed by Schlesinger et al. [3.1] in the early 1950s. They observed that the rate of the aqueous hydrolysis reaction decreased with the formation of the basic sodium metaborate (BO 2- ) product, and concluded that the reaction rate is dependent on both temperature and pH. Kreevoy et al. [3.16] later described this dependence by the empirical rate law presented by Equation (3.2), )92.1034.0()log( 2/1 −−= TpHt , (3.2) where t 1/2 is the half-life of self-hydrolysis of NaBH 4 in minutes at a particular temperature T in K and pH is the solution pH value (in the absence of catalyst). That is to say, alkaline NaBH 4 solution at pH 14 can be kept for 430 days from self-decaying at room temperature. The reaction yield also depends heavily on the pH and temperature. Kojima et al. [3.7] reported that at room temperature (20-23 ºC) the overall reaction conversion stabilizes at CHAPTER 3 Alkali hydrolysis 30 7% as the basic borate by-product is formed. Moon et al. [3.17] investigated the pH and temperature effects on the rate of decomposition of more concentrated NaBH 4 solutions. Therefore, previous studies have shown that the uncatalyzed aqueous hydrolysis reaction of NaBH 4 proceeds at a very slow rate. In more concentrated solutions, the rate was diminished probably due to solubility and mobility inhibition by the NaBO 2 product [3.17]. In all solutions with inhibitor, the kinetics of H 2 generation was initially linear, but non-zero-order behaviour became evident after several hours. It was found that the addition of alkali stabilizers such as NaBO 2 , KOH and NaOH depressed the hydrogen evolution, a property utilized in many practical chemical hydride systems to promote stable long term storage. Indeed, to create a basic environment in which NaBH 4 hydrolysis is stifled, simple hydroxides such as NaOH and KOH should be added. However, the resulting alkaline solution continues to undergo slow reaction to release H 2 . Several studies have looked at the decomposition rate of these solutions at various NaOH and NaBH 4 concentrations and temperatures to gauge their shelf life. It is apparent that the rate of hydrolysis for dilute solutions is much different than that of concentrated solutions and that temperature has a dramatic effect on the stability of the solutions. The concentration of NaBH 4 in solution must be maximized to improve energy density but low enough to keep the NaBO 2 by-product in solution [3.11]. Minkina et al. [3.18] investigated the effects of temperature and NaOH concentration on the degradation of concentrated NaBH 4 solutions. They suggest that above 30 ºC, at least 5wt% NaOH must be added to slow the hydrolysis, which is consistent with Shang work who suggested working on systems of 8.5 wt% NaBH 4 and 5wt% NaOH solution at room temperature [3.11]. In order to release the hydrogen from these basic stabilized solutions, metal or acid catalysts are required [3.1]. Various studies have been performed on the catalyzed hydrolysis reaction of sodium borohydride [3.5-3.7,3.19]. It is documented in the literature [3.12, 3.14] that sodium borohydride undergoes hydrolysis in aqueous solution to give boric acid in more acidic solutions (pH<9) and borate in more basic solutions (pH>9), as shown in the following equations: BH 4- + H + + 3H 2 O → B(OH) 3- + 4H 2 , pH<9 (3.3) BH 4- + 4H 2 O → B(OH) 4- + 4H 2 , pH>9 (3.4) CHAPTER 3 Alkali hydrolysis 31 If sodium metaborate, a water soluble and environmentally benign solid, is assumed to be the sole product, the reaction can be written simply as in Equation (3.1). In fact, catalysts play such a vital role in hydrogen evolution from hydrolysis of NaBH 4 , that a huge number of substances have been attempted to be efficacious in increasing the rate of hydrogen evolution but never to yield in lowering the controllability of hydrogen generation. Several metal catalysts are effective reaction catalysts to enhance hydrolysis of the alkaline sodium borohydride solution. Among them, the catalysts with precious metals most used are: rhodium, platinum, ruthenium [3.5, 3.6, 3.20], platinum supported on LiCoO 2 , CoO and TiO 2 [3.7, 3.21, 3.22] or carbon [3.23], ruthenium supported on IRA400 anion resin [3.24] or alumina pellets [3.25], ruthenium nanoclusters [3.26], and Pt/Pd on carbon nanotubes (CNT) paper [3.27]. The most used high-performance catalysts containing non-noble metals are: nickel [3.1, 3.28], nickel and/or cobalt borides [3.12, 3.29, 3.30-3.37], cobalt boride supported on nickel foam [3.38, 3.39] or carbon [3.40], cobalt boride amorphous alloy powder and Pd/C [3.42], cobalt on γ -alumina [3.43], hydrogenphosphate stabilized nickel(0) nanoclusters [3.44], Co-Mn-B supported on nickel foam [3.45], and Co-P catalyst [3.46,3.47]. A metal alloy catalyst containing a less precious metal: Ru 60 Co 20 Fe 20 supported on activated carbon fibber, has been reported by Park et al. [3.48], who found high hydrogen release in the reaction given by Eq.(3.1). In the present work particularly interested is focused in the work published on Ru catalysts, preferably unsupported, for results comparison. Elsewhere, a comparative study of the catalysts mentioned in the previous paragraphs with the Ni-Ru based powdered catalyst will be published (but that study is not included in this dissertation). After this short review, in the following sections, the results obtained with the ruthenium nickel based powdered catalyst, unsupported, in the study of the hydrogen release from alkali hydrolysis of sodium borohydride (NaBH 4 ) will be presented. Reports on hydrogen rates and yields are given where a proportion of 0.4 g/g in catalyst/chemical hydride was found suitable, after preliminary kinetic tests. The effects of NaBH 4 and an inhibitor (sodium hydroxide, NaOH) concentration, as well as the process temperature, on the rate of hydrogen production were also analysed. CHAPTER 3 Alkali hydrolysis 32 3.2 EXPERIMENTAL PROCEDURE The alkali catalytic hydrolysis of sodium borohydride for hydrogen generation under pressure comprises a series of experiments done in the three batch reactors LR, MR and SR, with internal free volumes of 646 cm 3 , 369 cm 3 and 229 cm 3 , respectively described in the previous chapter. The experimental conditions of the investigations done on the effects of reaction temperature, NaBH 4 concentration, NaOH concentration, pressure of system and reactor bottom shape on hydrogen generation rates and yields, are given below, separately. The experimental method used for the particular case study of successive loadings of reactant solutions, to verify the activity of the Ni-based bimetallic catalyst in active refuelling, is also specified. The hydrogen yield in reaction (3.1) was calculated by the following equation Hydrogen yield = n(H 2 ) exp / n(H 2 ) theoretical , (3.5) where n(H 2 ) exp is the number of moles (mol) of generated H 2 and n(H 2 ) theoretical is the theoretical amount of generated H 2 assuming 100% conversion of NaBH 4 by applying the ideal gas law to the final volume of gas inside the reactor. 3.2.1 Effect of reaction temperature on hydrogen generation rate To investigate the effect of reaction temperature on hydrogen generation rate of the catalytic hydrolysis of sodium borohydride, a series of kinetic experiments were carried out in a extensive range of temperatures of: 19.5, 23, 30, 45 and 55 ºC, to satisfy part of real environmental conditions throughout the year (very low temperatures were not considered in this dissertation). The hydrolysis reactions were carried out in the batch reactor MR (internal volume equal to 369 cm 3 ), using 20 cm 3 of 10 wt% NaBH 4 and 3 wt% NaOH solution ( 3 g/cm 045.1 = ρ ). After a proper quantity of reused Ni-Ru based powdered catalyst ( ≈ 0.85 g) was stored in the bottom of the reactor and the reaction vessel was perfectly sealed, the batch reactor RM was placed in a thermostatic bath to reach and maintain constant the system temperature at each of the values of the set given above. Then, 20 cm 3 of the reactant CHAPTER 3 Alkali hydrolysis 33 solution was rapidly injected and the inlet valve closed (see legend of Fig.2.5b), with the reactor submersed in the water bath. The generation of hydrogen was followed using the data acquisition system until a constant pressure inside the reactor is reached. No magnetic stirring was used. In the experiments reported in this section, the Ni-Ru based powdered catalyst was reused between 157 and 161 times. 3.2.2 Effect of NaBH 4 concentration on hydrogen generation rate To explore the effect of sodium borohydride concentration on hydrogen generation rate of the catalytic hydrolysis of sodium borohydride, a series of kinetic experiments was conceived at seven different weight amounts of NaBH 4 : 5, 10, 15, 20, 25, 30 and 40 wt%, for a constant concentration of an inhibitor (NaOH) of 3 wt%. All the hydrolysis reactions were executed in the batch reactor LR (internal volume of 646 cm 3 ), using 20 cm 3 of reactant solution, at constant temperature of 27 ºC. The experimental procedure on this research item is similar to that described on the above section. No magnetic stirring was used. In the experiments run in this part of the analysis, the Ni-Ru based powdered catalyst was reused between 35 and 41 times. 3.2.3 Effect of NaOH concentration on hydrogen generation rate To examine the effect of the inhibitor sodium hydroxide concentration on the hydrogen generation rate of the catalytic hydrolysis of sodium borohydride, a sequence of kinetic experiments were conceived at six different weight amounts of NaOH: 1, 3, 7, 10, 20 and 30 wt%, for a constant concentration of NaBH 4 of 10 wt%. All the hydrolysis reactions were conducted in the batch reactor LR (internal free volume of 646 cm 3 ), using 10 cm 3 of reactant solution, at constant temperature of 25 ºC. The experimental procedure of this research topic is analogous to that described on 3.2.1. No magnetic stirring was used. In the experiments reported in this part of the analysis, the Ni-Ru based powdered catalyst was reused between 57 and 62 times. 3.2.4 Effect of pressure on hydrogen generation rate To study the influence of pressure on the hydrogen generation rate of the catalytic hydrolysis of sodium borohydride, three individual experiments were carried out in each CHAPTER 3 Alkali hydrolysis 40 The fluctuations detected in the reactor temperature along reaction (see Fig. 3.6), also indicate at which rates the hydrogen is being generated, i.e., a smooth pattern of the internal reactor temperature profile, both at the bottom and at its top, indicates a slowly hydrolysis kinetics; but, an asymptotically and random patterns in reactor temperature profiles, put in evidence the dynamic behaviour of reactants and products inside the reactor due to increased internal pressure. The plot in Figure 3.7 corroborates this assumption, showing slight higher hydrogen rates for the last loadings, in which the variation of temperature were more pronounced. 10 wt% NaBH4 and 7 wt% NaOH | five successive loadings 0 0,5 1 1,5 2 2,5 3 3,5 4 4,5 5 5,5 0 100 200 300 400 500 600 time / sec pressure / bar 1st load ; H2 yield=100% 2nd load ; H2 yield=100% 3rd load ; H2 yield=100% 4th load ; H2 yield=100% 5th load ; H2 yield=100% Figure 3.7. Hydrogen generation in batch reactor LR (646 cm 3 ) of five successive loadings of 10 mL each of the reactant solution: 10 wt.% NaBH 4 , 7 wt.% NaOH, 83 wt.% H 2 O, with Ni-Ru based catalyst/NaBH 4 : 0.4g/g. Figure 3.8 show a typical course of hydrogen generation, presented also in terms of the operating pressure as a function of time, for seven successive loadings of fuel solution, in batch reactor MR (369 cm 3 ). These results show an improvement in the performance of the used nickel based catalyst, comparing with similar trend published by Pinto et al. [3.33], the latter without ruthenium. In fact, the nano powdered Ni-Ru based catalyst used in the experiments reported on this dissertation, gives higher H 2 yield in a relatively short reaction time until the ‘plateau’ is achieved. CHAPTER 3 Alkali hydrolysis 41 0 2 4 6 8 10 12 0 100 200 300 400 500 time / s Pressure / bar 1st load ; H2 yield=100% 2nd load ; H2 yield=100% 3rd load ; H2 yield=100% 4th load ; H2 yield=100% 5th load ; H2 yield=100% Figure 3.8. Hydrogen generation in batch reactor MR (369 cm 3 ) of seven successive loadings of the reactant solution: 10 wt.% NaBH 4 , 7 wt.% NaOH, 83 wt.% H 2 O, with Ni-Ru based catalyst/NaBH 4 : 0.4g/g. As mentioned by Pinto et al. [3.33], the increasing pressure of the gas phase inside the reactor, due to hydrogen generation, forces at the same time, the hydrogen dissolution. This is an interesting finding since molecular hydrogen was generated but also stored in the liquid phase. Figure 3.9 shows the gaseous reaction products upshot after a quick exit from the reactor, at the end of the reaction of the seven successive loading (the H 2 generated in the previous successive loadings, was released to the atmosphere). Figure 3.9. Two images of reaction “gaseous” by-product, after seven successive loadings of reactant solution (10 wt.% NaBH 4 , 7 wt.% NaOH, 83 wt.% H 2 O). The H 2 release shows the gas capability to agitate the nickel-based powdered catalyst presented at the reactor liquid phase. As we can see, the “sparkling” by-product quickly dissipates the H 2 bubbles along the glass goblet, showing the gas capability to agitate the nickel-based bimetallic powdered CHAPTER 3 Alkali hydrolysis 42 catalyst presented at the reactor liquid phase. This attests that a significant amount of the H 2 generated inside the reactor was trapped in the liquid reactor phase, keeping it dissolved. Just after a slight aperture of the reactor exhaustion needle valve, due to pressure differences, the dissolved gas (H 2 plus air) is released. 3.3.6.1 Effect of agitation on hydrogen reaction rate In addition, and with the purpose of verifying the influence of agitation on hydrogen reaction rate during successive loadings of fuel, a Teflon TM magnetic stirrer was placed inside the reactor prior to the first loading injection. The series of experiments performed on this section, the Ni-Ru based powdered catalyst was reused between 226 and 233 times. Vigorous magnetic stirring was allowed during the formation of hydrogen by injection of the 8 th loading of reactant in a series of experiments by successive refuelling, to promote complete mass transfer (to overcome diffusion limitation). Figure 3.10, shows an eligible increase in yield and rate of H 2 , lowered by a minor reaction induction time (by comparing the results with the previous 6 th an 7 th loadings). This leads us to infer that mass transfer limitations exist between the reacting solution and the reused (233 times!) catalyst. 0 1 2 3 4 5 0 1000 2000 3000 4000 Reaction time /s Pressure /bar 1st load ; yield=89% 2nd load ; yield=89% 3rd load ; yield=88% 4th load ; yield=88% 5th load ; yield=89% 6th load ; yield=88% 7th load ; yield=90% 8th load ; yield=94% (MAGNETIC STIRRING) theoretical yields Figure 3.10. Hydrogen generation in batch reactor LR (646 cm 3 ) of eight successive loadings of 10 mL of the reactant solution: 10 wt.% NaBH 4 , 7 wt.% NaOH, 83 wt.% H 2 O, with Ni-Ru based catalyst/NaBH 4 : 0.4g/g (the 8 th loading was performed with magnetic stirring), at 25 ºC. CHAPTER 3 Alkali hydrolysis 43 3.4 REACTION BY-PRODUCT CHARACTERIZATION The by-products of the hydrogen generation reactions of all types of experiments reported on this chapter were analyzed by X-Ray Diffractometry. Suitable crystals – see Figure 3.11, were obtained by slow evaporation of a water solution at uncontrolled environmental conditions. Accordingly to publish literature, boric oxides come in white rhombic crystals or colourless, semi-transparent vitreous granules or flates, or hygroscopic lumps or powder (the crystals, usually forms a glass, have specific gravity’s of 1.84-2.46 and melt at 450 ºC) [3.49]. Figure 3.11. Crystals pictures of the by-product classic hydrolysis of sodium borohydride, obtained by slow evaporation of a water solution at room temperature ( ≈ 25 ºC). 3.4.1 Materials and methods The obtained crystals were found to be orthorhombic, space group Pca2 1 , cell volume V=364.62(4) Å 3 , a = 10.7252(7) Å, b = 5.2525(3) Å, c = 6.4724(4) Å (uncertainties in parentheses). There are four molecules per unit cell, calculated density 2.237 g/cm 3 . Diffraction data were collected at 293 K with a Gemini PX Ultra equipped with MoK α radiation ( λ =0.71073 Å). A total of 562 independent reflections were measured, of which 501 were observed (I>2 σ (I)). The structure was solved by direct methods using SHELXS97 [3.50] with atomic positions and displacement parameters refined with SHELXL-97 [3.51]. The non-hydrogen atoms were refined anisotropically and the hydrogen atoms were refined freely with isotropic displacement parameters. The refinement converged to R (all data) = 3.93% and wR 2 (all data) = 9.92%. 3.4.2 Results and discussion The crystal structure of the reaction (3.1) by-product reveals that the boron atoms are in a triangular configuration with three oxygen atoms, with B-O bond lengths between 1.27 and CHAPTER 3 Alkali hydrolysis 44 1.29 Å. The BO 3 triangular arrangement was already found in the crystal structure of orthorhombic [3.52] and monoclinic [3.53] metaboric acid although in the title crystal structure the oxygen atoms are bound to just one B atom. The BO 3 triangles form layers 3.238 Å apart that sandwich a layer of sodium and water molecules. For each boron atom there are three bounded oxygen atoms and one water oxygen. Half of the sodium atoms have their valence strength distributed between six Na-O bonds and the other half between seven Na-O bonds (as already determined in other sodium metaborate structures [3.54]). The water molecules are involved in a net of interactions with the sodium atoms. Figure 3.12, below, shows two different views of the crystal structure. (a) (b) Figure 3.12. Views of crystal structure of sodium metaborate dehydrate showing stacking layers formed by BO 3 triangles and by Na and water molecules: (a) view along the a -axis and (b) view along the c -axis. Na in violet, O in red, B in pink and H in white. We may conclude that the reaction (3.1) by-product, in all the types of experiments performed by alkali hydrolysis of NaBH 4 under pressure, is a sodium metaborate dehydrated, NaBO 2 .2H 2 O, as expected. Therefore, equation (3.1) can be re-written as follows: NaBH 4 + 4 H 2 O → NaBO 2 . 2 H 2 O + 4 H 2 + heat . (3.6) 3.5 CONCLUSIONS In this chapter the results of an extensive experimental work done with the objective of studying various effects in hydrogen generation rate, yield and lag time, by catalytic alkali hydrolysis of sodium borohydride, are reported. Briefly, the main conclusions are: CHAPTER 3 Alkali hydrolysis 45 - increasing the temperature of the reaction medium, increases the rate of H 2 generation; - the reaction rate increases with increasing the hydride concentration up to 20 wt% of NaBH 4 . Above this value, a slight decrease on both H 2 and yield was found; - the reaction rate is greatly enhanced by the increase of the NaOH concentration, ensuring good efficiency of hydrogen generation; - an average yield of ≈ 90% was found in H 2 generation; this is substantially lower than the maximum predicted hydrogen pressure, assuming 100% conversion and applying the ideal gas law. Pinto et al. [3.33] suggests that this fact is related with solubility effects in the liquid phase that remains inside the reactor; - the reactor bottom shape affects the course of hydrogen generation. It was found that the conical bottom geometry leads to an increase on reaction rate and yield, and slightest in H 2 time lag; - the Ni-Ru based catalyst has a good performance in refueling experiments. In fact, it was found that the catalyst tends to perform better, after the first loading of reactant (the amount of catalyst remains the same from the first to the last loading); - a vigorous stirring of the solution inside the batch reactor, during the experiments with successive fuel loadings, reveals minor reaction induction time and a slight higher yield in hydrogen generation. This indicates possible mass transfer limitations during the course of reaction between the fuel and the active sites of the catalyst. The x-ray diffractions analysis revealed a sodium metaborate dehydrated, NaBO 2 .2H 2 O, as the by-product of all experiments performed in the several subsections of this chapter. 3.6 REFERENCES [3.1] H. I. Schlesinger, H. C. Brown, A. E. Finholt, J. R. Gilbreath, H. R. Hoekstra, E. K. Hyde, Sodium Borohydride, Its Hydrolysis and Its Use as a Reducing Agent and in the Generation of Hydrogen, J. Am. Chem. Soc. 75(1953) 215-219. [3.2] V. C. Y. Kong, F. R. Foulkes, D. W. Kirk, J. T. Hinatsu, Development of Hydrogen Storage for Fuel Cell Generators. I: Hydrogen Generation Using Hydrolysis of Hydrides, Int. J. Hydrogen Energy 24, 7 (1999) 665-675. [3.3] W. H. Stockmayer, R. R. Miller, R. J. Zeto, Kinetics of Borohydride Hydrolysis. J. Phys. Chem. 65 (1961) 1076-1077. [3.4] R. Aiello, J. H. Sharp, M. A. Matthews, Production of Hydrogen from Chemical Hydrides via Hydrolysis with Steam, Int. J. Hydrogen Energy 24, 12 (1999) 1123-1130. CHAPTER 3 Alkali hydrolysis 46 [3.5] S. C. Amendola, S. L Sharp-Goldman, M. S. Janjua, M. T. Kelly, P. J. Petillo, M. Binder, An Ultrasafe Hydrogen Generator: Aqueous, Alkaline Borohydride Solutions and Ru Catalyst, J. Power Sources 85, 2 (2000) 186-189. [3.6] Amendola S. C., S. L. Sharp-Goldman, M. S. Janjua, N. C. Spencer, M. T. Kelly, P. J. Petillo, M. Binder, A Safe, Portable, Hydrogen Gas Generator Using Aqueous Borohydride Solution and Ru Catalyst, Int. J. Hydrogen Energy 25, 10 (2000) 969-975. [3.7] Y. Kojima, K.-I. Suzuki, K. Fukumoto, M. Sasaki, T. Yamamoto, Y. Kawai, H. Hayashi, Hydrogen Generation Using Sodium Borohydride Solution and Metal Catalyst Coated on Metal Oxide, Int. J. Hydrogen Energy 27,10 (2002) 1029-1034. [3.8] W.C. Blasdale, C.M. Slansky, The solubility curves of boric acid and the borates of sodium, J. Am. Chem. Soc. 61 (1939) 917-920. [3.9] E. Marrero-Alfonso, J.R. Gray, T.A. Davis, M.A. Matthews, Hydrolysis of sodium borohydride with steam, Int. J. Hydrogen Energy 32 (2007) 4717-4722. [3.10] E. Marrero-Alfonso, J.R. Gray, T.A. Davis, M.A. Matthews, Minimizing water utilization in hydrolysis of sodium borohydride: the role of sodium metaborate hydrates, Int. J. Hydrogen Energy 32 (2007) 4723-4730. [3.11] Y. Shang, R. Chen, Hydrogen storage via the hydrolysis of NaBH 4 basic solution: optimization of NaBH 4 concentration, Energy & Fuels 20 (2006) 2142-2148. [3.12] C. M. Kaufman, B. Sen, Hydrogen generation by hydrolysis of sodium tetrahydroborate: effects of acids and transition metals and their salts, J. Chem. Soc. Dalton 2 (1985) 307-313. [3.13] M. Kreevoy, J. Hutchins, H 2 BH 3 as an intermediate in tetrahydridoborate hydrolysis, J. Am. Chem. Soc. 94, 18 (1972) 6371-6376. [3.14] F. T. Wang, W. L. Jolly, Kinetic study of the intermediates in the hydrolysis of the hydroborate ion, Inorg. Chem. 11, 8 (1972) 1933-1941. [3.15] B. D. James, M. G. H. Wallbridge, Metal tetrahydroborates, Prog. Inorg. Chem. 11 (1970) 199-231. [3.16] M. M. Kreevoy, R. W. Jacobson, The rate of decomposition of sodium borohydride in basic aqueous solutions, Ventron Alembic 15 (1979) 2-3. [3.17] . G. Y. Moon, S. S. Lee, K. Y. Lee, S. H. Kim, K. H. Song, Behavior of hydrogen evolution of aqueous sodium borohydride solutions, J. Ind. Eng. Chem. 14, 1 (2008) 94-99. [3.18] V. G. Minkina, S. I. Shabunya, V. I. Kalinin, V. V. Martynenko, Stability of aqueous-alkaline sodium borohydride formulations. Russ. J. Appl. Chem. 81, 3 (2008) 380-385. [3.19] Y. Kojima, Y. Kawai, M. Kimbara, H. Nakanishi, S. Matsumoto, Hydrogen Generation by Hydrolysis Reaction of Lithium Borohydride, Int. J. Hydrogen Energy 29, 12 (2004) 1213-1217. [3.20] H.C. Brown, C.A. Brown, New highly active metal catalysts for the hydrolysis of borohydride, J. Am. Chem. Soc. 84 (1962) 1493-1494. [3.21] Y. Kojima, K.-I. Suzuki, K. Fukumoto, Y. Kawai, M. Kimbara, H. Nakanishi, S. Matsumoto, Development of 10 kW-scale hydrogen generation using chemical hydride, J. Power Sources 125 (2004) 22-26. CHAPTER 3 Alkali hydrolysis 47 [3.22] Y. Kojima, Y. Kawai, H. Nakanishi, S. Matsumoto, Compressed hydrogen generation using chemical hydride, J. Power Sources 135 (2004) 36-41. [3.23] C. Wu, H. Zhang, B. Yi, Hydrogen generation from catalytic hydrolysis of sodium borohydride for proton exchange membrane fuel cells, Cat. Today 93-95 (2004) 477-483. [3.24] P. Krishnan, T.H. Yang, W.Y. Lee, C.S. Kim, PtRu-LiCoO 2 an efficient catalyst for hydrogen generation from sodium borohydride solutions, J. Power Sources 143 (2005) 1723. [3.25] J.S. Zhang, W.N. Delgass, T.S. Fisher, J.P. Gore, Kinetics of Ru-catalyzed sodium borohydride hydrolysis, J. Power Sources 164 (2007) 772-781. [3.26] S. Özkar, M. Zahmakiran, Hydrogen generation from hydrolysis of sodium borohydride usisng Ru(0) nanoclusters as catalyst, J. Alloys Comp 404-406 (2005) 728731. [3.27] R. Peña-Alonso, A. Sicurelli, E. Callone, G. Carturan, R. Raj, A picoscale catalyst for hydrogen generation from NaBH 4 for fuel cells, J. Power Sources 165 (2007) 315-323. [3.28] J.H. Kim, K.T. Kim, Y.M. Kang, H.S. Kim, M.S. Song, Y.J. Lee, P.S. Lee, J.Y. Lee, Study on degradation of filamentary Ni catalyst on hydrolysis of sodium boroydride, J. Alloys and Compounds 379 (2004) 222-227. [3.29] A. Levy, J.B. Brown, C.J. Lyons, Catalyzed hydrolysis of sodium borohydride, Ind. Engng Chem. 52 (1960) 211-214. [3.30] D. Hua, Y. Hanxi, A. Xinping, C. Chuansin, Hydrogen production from catalytic hydrolysis of sodium borohydride solution using nickel borid catalyst, Int. J. Hydrogen Energy 28 (2003) 1095-2000. [3.31] S.U. Jeong, R.K. Kim, E.A. Cho, H.J. Kim, S.W. Nam, I.H. Oh, S.A. Hong, S.H. Kim, A study on hydrogen generation from NaBH 4 solution using the high-performance Co-B catalyst, J. Power Sources 144 (2005) 129-134. [3.32] B.H. Liu, Z.P. Li, S. Suda, Nickel and cobalt-based catalyst for hydrogen generation by hydrolysis of borohydride, J. Alloys and Compounds 415 (2006) 288-293. [3.33] A.M.F.R. Pinto, D.S. Falcão, R.A. Silva, C.M. Rangel, Hydrogen generation and storage from hydrolysis of sodium borohydride in batch reactors, Int. J. Hydrogen Energy 31 (2006) 1341-1347. [3.34] C.M. Rangel, R.A. Silva, A.M.F.R. Pinto, Fuel cells and on-demand hydrogen production: didactic demonstration prototype, Proceedings of the International Conference in Power Engineering and Electric Drives, Eds. L.S. Martins and P. Santos, Setúbal, Portugal, September 12-14, 2007, paper 237. [3.35] J.C. Ingersoll, N. Mani, J.C. Thenmozhiyal, A. Muthaiah, Catalytic hydrolysis of sodium borohydride by a novel nickel-cobalt-boride catalyst, J. Power Sources 173 (2007) 450-457. [3.36] S.U. Jeong, E.A. Cho, S.W. Nam, I.H. Oh, U.H. Jung, S.H. Kim, Effect of preparation method on Co-B catalytic activity for hydrogen generation from alkali NaBH 4 solution, Int. J. Hydrogen Energy 32 (2007) 1749-1754. [3.37] J.C. Walter, A. Zurawski, D. Mongomery, M. Thornburg, S. Revankar, Sodium borohydride hydrolysis kinetics comparison for nickel, cobalt and ruthenium boride catalyst, J. Power Sources 179 (2008) 335-339. CHAPTER 3 Alkali hydrolysis 48 [3.38] S.J. Kim, J. Lee, K.Y. Kong, C.R. Jung, I.G. Min, S.Y. Lee, H.J. Kim, S.W. Nam, T.H. Lim, Hydrogen generation system using borohydride for operation of a 400 W-sacle polymer electrolyte fuel cell stack, J. Power Sources 170 (2007) 412-418. [3.39] J. Lee, K.Y. Kong, C.R. Jung, E. Cho, S.P. Yoon, J. Han, T.G. Lee, S.W. Nam, A structured Co-B catalyst for hydrogen extraction from NaBH 4 solution, Cat. Today 120 (2007) 305-310. [3.40] J. Zhao, H. Ma, J. Chen, Improved hydrogen generation from alkaline NaBH 4 solution using carbon-supported Co-B a catalyst, Int. J. Hydrogen Energy 32 (2007) 47114716. [3.42] C. Zanchetta, B. Patton, G. Guella, A. Miotello, An integrated apparatus for production and measurement of molecular hydrogen, Meas. Sci. Tech. 18 (2007) N21-N26. [3.43] W. Ye, H. Zhang, D. Xu, L. Ma, B. Yi, Hydrogen generation utilizing alkaline sodium borohydride solution and supported cobalt catalyst, J. Power Sources 164 (2007) 544-548. [3.44] Ö. Metin, S. Özkar, Hydrogen generation from the hydrolysis of sodium borohydride by using water dispersible, hydrogenphosphate-stabilized nickel(0) nanoclusters as catalyst, Int. J. Hydrogen Energy 32 (2007) 1707-1715. [3.45] M. Mitov, R. Rashkov, N. Atanassov, A. Zielonka, Effects of nickel foam dimensions on catalytic activity of supported Co-Mn-B nanocomposites for hydrogen generation from stabilized borohydride solutions, J. Mater. Sci. (42) 2007 3367-3372. [3.46] K.W. Cho, H.S. Kwon, Effects of electrodeposited Co and Co-P catalysts on the hydrogen generation properties from hydrolysis of alkaline sodium borohydride solution, Cat. Today 120 (2007) 298-304. [3.47] K. Eom, K.W. Cho, H.S. Kwon, Effects of electroless deposition conditions on microstructures of cobalt-phosphorous catalysts and their hydrogen generation properties in alkaline sdium borohydride solution, J. Power Sources 180 (2008) 484-490. [3.48] J.H. Park, P. Shakkthivel, H.J. Kim, M.K. Han, J.H. Jang, Y.R. Kim, H.S. Kim, Y.G. Shul, Investigation of metal alloy catalyst for hydrogen release from sodium borohydride for polymer electrolyte membrane fuel cell application, Int. J. Hydrogen Energy 33 (2008) 1845-1852. [3.49] F. A. Cotton, G. Wilkinson, in “Advanced Inorganic Chemistry”, John Wiley & Sons, 5 th Edition, Chapter Six, p.162. [3.50] G.M. Sheldrick, SHELXS-97, Program for the solution of crystal structures; University of Göttingen: Germany 1997. [3.51] G.M. Sheldrick, SHELXL-97, Program for the refinement of crystal structures; University of Göttingen: Germany 1997. [3.52] H. Tazaki, The structure of orthorhombic metaboric acid, HBO 2 ( α ), J. Sci. Hiroshima Univ. A10, 37 (1940) 55-61. [3.53] W.H. Zachariasen, The crystal structure of monoclinic metaboric acid, Acta Cryst. 16 (1963) 385-389. [3.54] M. Marezio, H.A. Plettinger, W.H. Zachariasen, The bond lengths in the sodium metaborate structure, Acta Cryst. 16 (1963) 594-595. 49 CHAPTER 4 Alkali free hydrolysis Results for sodium borohydride kinetic experiments This chapter shows the results obtained on alkali free hydrolysis of sodium borohydride in the presence of a reused nickel based bimetallic catalyst. A discussion on the stoichiometry of hydrolysis, in terms of the number of added water molecules to solid NaBH 4 , is carried out, based on the results of hydrogen generation rate and yields. Emphasis is given to reactor bottom design effects on the hydrogen production. The reaction by-products are analysed by XRD. 4.1 INTRODUCTION The main inspiration behind developing this experimental item was to be capable of producing pure molecular hydrogen with very high gravimetric and volumetric densities. The gravimetric and volumetric energy densities attainable from hydrolysis of chemical hydrides depend in large measure on the amount of water required for the process. It is well known that hydrolysis of sodium borohydride produces hydrated metaborate by-products in which the degree of hydration of the metaborate is closely joined to the amount of water present in the reaction [4.1]. In the previous chapter, attention was focused on studying the catalytic stabilized hydrolysis of sodium borohydride in the presence of water excess, namely alkali hydrolysis of sodium borohydride, for hydrogen generation (this reaction generates basic species that are thermodynamically and kinetically metastable, causing the reaction to cease at low yields of hydrogen even though the reaction is strongly favoured thermodynamically [4.2]). The results revealed that when working at high pressures, and in the presence of a Ni-Ru based catalyst, even with water in excess, it is possible to produce CHAPTER 4 Alkali free hydrolysis 56 results of the present work - obtained at moderate pressures, high values of H 2 yield are achieved ( ≥ 80%) if H 2 O/NaBH 4 ≥ 4 mol/mol. 4.3.2 Effect of catalyst/NaBH 4 (g/g) on hydrogen generation rate Figure 4.5 shows the influence of the amount of catalyst used in hydrogen rate and lag time of an alkali free hydrolysis of sodium borohydride experiment, performed in batch reactor LR (646 cm 3 ). 0 1 2 3 4 5 6 0 5000 10000 15000 20000 25000 30000 Time / sec Pressure H 2 / bar Ru-Ni based/NaBH4: 0.2 g/g | H2 yield=78% Ru-Ni based/NaBH4: 0.4 g/g | H2 yield=78% Theoretical Yield Figure 4.5. Influence of the amount of catalyst/NaBH 4 (g/g) on hydrogen generation. Actually the amount of catalyst is another variable that influences the rate and the lag time. As can be seen in the plot of Fig.4.5, increasing the amount of the catalyst (catalyst/NaBH 4 : 0.2 g/g → 0.4 g/g) increases the dtdP/ slope in the linear zone (2.5E-4 bar/s or 0.037 L(H 2 )min -1 gcat -1 → 5.5E-4 bar/s or 0.041 L(H 2 )min -1 gcat -1 ) and decreases the time to reach the ‘plateau’, but, as expected, the pressure of produced hydrogen was similar (≈ 0.36 MPa and H 2 yield of 78 %) and independent of the temperature. 4.3.3 Effect of pressure on hydrogen generation yield and rate Figure 4.6 shows the rate of hydrogen generation in terms of H 2 pressure as a function of time, for the three batch reactor LR, MR and SR, at the stoichiometric amount of 4 moles of distilled water, using a proportion Ni-Ru based/NaBH 4 : 0.4 g/g. The values of H 2 yields achieved are also reported. CHAPTER 4 Alkali free hydrolysis 57 The plot in Figure 4.7 shows the H 2 generation yield versus pressure, for the three studied stoichiometric amounts of water, using Ni-Ru based catalyst (Ni-Ru based/NaBH 4 : 0.4 g/g), and also the results obtained by Kojima et al. [4.5], for comparison. 0 2 4 6 8 10 12 14 0 6000 12000 18000 Time / sec Pressure H 2 / bar LR; H2 Yield: 78% MR; H2 Yield: 86% SR; H2 Yield: 98% Figure 4.6. Hydrogen generation in the studied three batch reactors - LR (646 cm 3 ), MR (369 cm 3 ) and SR (229 cm 3 ), with Ni-Ru based catalyst/NaBH 4 : 0.4 g/g (111, 150 and 151 times reused, respectively for LR, MR and SR) and H 2 O/NaBH 4 : 4 mol/mol. 0 20 40 60 80 100 0 1 10 100 Pressure / MPa Hydrogen yield / % This work Kojima et al. (2004) Ni-based/NaBH 4 : 0.4 g/g H 2 O/NaBH 4 : 4 mol/mol Pt-LiCoO 2 /NaBH 4 : 0.5 g/g H 2 O/NaBH 4 : 2 mol/mol Figure 4.7. Influence of pressure on H 2 yield (Experiments done in three batch reactors at the temperature range of 289-295 K, with catalyst reused ≈150 times). As can be seen from the plots of Fig.4.6-4.7, the effects of pressure on H 2 yields and rates are remarkable. In truth, we may conclude that increasing the pressure, significantly increases H 2 rates and H 2 yields. CHAPTER 4 Alkali free hydrolysis 58 For the operating conditions presented on this dissertation, and for the special case of H 2 O/NaBH 4 : 4 mol/mol, performed at the moderate pressures between 0.36-1.26 MPa, values for H 2 yields up to 78-98% were obtained (see Fig.4.7). Kojima et al. [4.5] reported values of 80-93% for H 2 yields under much higher pressure levels (0.7-25 MPa), with H 2 O/NaBH 4 : 2 mol/mol (see Fig.4.7). It is worth to notice that the results presented in Fig.4.6 were performed using Ni-Ru based powdered catalyst reused ≈ 150 times, in batch reactors with flat and conical bottoms. As can be easily checked, the catalyst shows high activity under moderate pressure resulting in a significant increase in the amount of hydrogen generated. Kojima et al . [4.5] conclude the same trend (at H 2 O/NaBH 4 : 2 mol/mol) with a noble catalyst PtLiCoO 2 /NaBH 4 : 0.5 g/g at higher H 2 pressure, up to 25 MPa. 4.3.4 Effect of reactor bottom shape on hydrogen generation Looking back to the results plotted in Fig.4.6, presented in the previous section, it seems very important to study the effect of reactor bottom shape on H 2 generation rate and yield. The plot in Figure 4.8 show the H 2 generation trend for H 2 O/NaBH 4 : 2 mol/mol and Ru-Ni based/NaBH 4 : 0.4 g/g., in batch reactors LR (646 cm 3 ) and MR (369 cm 3 ), with flat and conical bottom shapes, respectively. 0 2 4 6 8 0 2000 4000 6000 8000 10000 12000 Time / sec Pressure H 2 / bar LR - FLAT bottom; H2 Yield: 78% MR - CONICAL bottom; H2 Yield: 86% 0 2 4 6 8 0 20 40 60 80 Figure 4.8. Influence of reactor bottom shape on H 2 yield, rate and lag time (experiments performed in batch reactors LR (646 cm 3 ) and MR (369 cm 3 ), with flat and conical bottom shapes, respectively; at temperature range of 289-295 K, with catalyst reused ≈150 times). As can be observed from the plot of Fig.4.8, the influence of reactor bottom shape on H 2 generation is notorious: the conical geometry greatly enhances the rate and yield of H 2 CHAPTER 4 Alkali free hydrolysis 59 generation. For the batch reactor MR, with conical bottom, a value of dtdP / slope in the linear zone of 5.7E-1 bar/s (24.2 L(H 2 )min -1 gcat -1 ) was found. In fact, a difference of 10 3 separates this value from that obtained in the reactor with a flat bottom shape (5.8E-4 bar/s or 0.043 L(H 2 )min -1 gcat -1 ), for the same H 2 reaction. An enlarged plot on the left side of Fig.4.8 put in evidence the sharply increase of H 2 pressure by time (confirmed by the high value in dtdP / slope mentioned in the previous paragraph), after a period of ≈ 36 seconds in lag time (the lag time is defined as the length of time required to observe an increase in the H 2 pressure). In reality, the reactor bottom shape influences also the H 2 lag time! Observing Fig.4.8 we may conclude that the conical reactor bottom geometry significantly decreases the lag time. A plausible explanation is that the reactor conical bottom shape enhances the contact between the catalyst and NaBH 4 powders, and the injected distilled water. The photographs in Figure 4.9, below, outline this aspect: the MR reactor (left side) when opened after the experiment which results are plotted in Fig.4.8, shows absence of liquid and looks empty. The photograph on the right side of Fig.4.9 shows the reaction by-product plus catalyst, in a form of a swelled aggregated porous solid, shaped by MR reactor conical bottom, which adhered to the T.Bottom k thermocouple, the latter fixed on the top of the reactor lid. Figure 4.9. Photographs of MR reactor (with conical bottom shape) after an alkali free hydrolysis of NaBH 4 experiment. In the right, a view of the solid mixture of the by-product plus the catalyst. The sharp increase of H 2 pressure with time shown in Fig. 4.8, which occurred in the MR batch reactor with a conical bottom, is in some way a consequence of also a sharp increase in the value of temperature of the hydrolyzed NaBH 4 . Figure 4.10 illustrates the variation of reaction temperature occurred during the hydrogen generation for the two experiments plotted in Fig 4.8. CHAPTER 4 Alkali free hydrolysis 60 250 300 350 400 450 0 200 400 600 800 1000 Time / sec Temperature / K H2 presure in LR: 0.36 MPa H2 presure in MR: 0.69 MPa H 2 O/NaBH 4 : 4 mol/mol Ru-Ni based/NaBH 4 : 0.4 g/g Figure 4.10. Temperature of the hydrolyzed NaBH 4 as a function of time by different pressures. Indeed, in the MR batch reactor, the occurrence of a ‘thermal runaway’ which was responsible for the increase of about 150 ºC in the reaction temperature, happened in a short time interval, less than one minute, and consequently, accelerated harshly the hydrolysis reaction. The temperature in the batch reactor (with a conical bottom), at 0.69 MPa, shows the maximum value of 428 K after the induction period, and decreases quickly with time. It may be conclude that when the temperature of hydrolyzed hydride increases the induction period decreased. A similar trend was found by Kojima et al. [4.5]. 4.3.5 Effect of temperature on hydrogen generation rate The influence of temperature on the velocity of hydrogen generation is put in evidence in Figure 4.11. As can be seen in the plot of Fig.4.11, the reaction rate is very sensitive to temperature. As expected, the rate rises with the increase in temperature and the hydrogen pressure demonstrates a near linear variation with reaction time. The influence of temperature is clearly shown by the increasing slope values on the linear region of the plots, for increasing values of the reaction temperature (from 15 to 55 ºC, the rate increases from 5.4E-4 to 4.4E-1 bar/s, or 0.021 L(H 2 )min -1 gcat -1 to 14.4 L(H 2 )min -1 gcat -1 ). Therefore, it seems that the reaction rate remains practically constant when, during the hydrolysis process, the NaBH 4 concentration decreases due to hydride and water consumption, demonstrating a zero-order reaction. A more detailed study of the effects of reactants concentration, operating pressure and temperature is necessary; in order to develop a CHAPTER 4 Alkali free hydrolysis 61 kinetic model that accurately reflects the reaction rate as a function of temperature and pressure. 0 2 4 6 8 10 12 0 12500 25000 time / s Pressure / bar 15 ºC | yield=87% 25 ºC | yield=89% 35 ºC | yield=83% 45 ºC | yield=75% 55 ºC | yield=89% Figure 4.11. Influence of temperature on hydrogen generation rate and yield. 4.3.6 Gravimetric and volumetric densities It is here worth to remember that the main purpose of the alkali free hydrolysis of NaBH 4 experiments reported in this dissertation is to produce pure hydrogen gas with very high gravimetric and volumetric densities. And the ability to produce anhydrous sodium borate is the key to increase the overall storage density of systems based on sodium borohydride as the storage media [4.1, 4.5]. Kojima et al. [4.5] reported that the structure of the by-product produced from the reaction of NaBH 4 and water using Pt-LiCoO 2 , is NaBO 2 .2H 2 O at atmospheric pressure, and anhydrous NaBO 2 at high pressure. They also mentioned that the temperature of the hydrolyzed NaBH 4 is higher in the catalyzed reaction at 0.10 MPa than at 0.68 MPa. A possible explanation can be related with the energy required for the addition of two water molecules in the heat of formation of NaBO 2 .2H 2 O. In some way, similar behaviour was found in the present work (see Figs. 4.8 and 4.10, above). In fact, if we look through the plot of Fig.4.6, the H 2 yield increases from 86% to 98% in the MR (369 cm 3 ) and SR (229 cm 3 ) batch reactors, both with conical bottom shape. Why? An inspection of the temperature profile inside of SR reactor shows a lower reaction temperature; maybe because the H 2 is not consumed to hydrate sodium borate by-product. It is therefore CHAPTER 4 Alkali free hydrolysis 62 important to investigate the type of by-products formed; and that will be done in Section 4.4. The US Department Of Energy (DOE) has published FreedomCAR requirements for automotive hydrogen storage systems (probably the most stringent targets yet articulated for hydrogen storage systems). Two of the principal technical targets, based on system mass and volume, are the gravimetric hydrogen density and the volumetric hydrogen density. Table 4.1 shows these FreedomCAR two specifications [4.4]. Table 4.1 - The FreedomCAR targets publish by US Department Of Energy (DOE). Year 2007 2010 2015 Gravimetric H 2 density / wt.% 4.5 6.0 9.0 Volumetric H 2 density/ kgH 2 /m 3 36 45 81 (Note that these specifications are based on system mass and volume, i.e., not only the storage material itself but also the reactors, tanks, valves, and all auxiliary equipment, which usually are call hardware.) Table 4.2 below, presents the predicted values for gravimetric hydrogen density and volumetric hydrogen density, based only on the storage material itself, for the systems studied in this chapter: H 2 O/NaBH 4 : 4 mol/mol and Ru-Ni based catalyst/NaBH 4 : 0.2-0.4 g/g. Hence, the influence of Ni-Ru based catalyst/NaBH 4 on gravimetric hydrogen density and on hydrogen yield is also checked. The density of the material (H 2 O/NaBH 4 : 4 mol/mol; Ni-Ru based/NaBH 4 : 0.2-0.4 g/g) was calculated by adding the densities of NaBH 4 (1.07 g/cm 3 ), water (1.00 g/cm 3 ) and Ni-Ru based catalyst (3.17 g/cm 3 ). It was found that our system of compressed hydrogen generation at moderate pressure of up to 1.26 MPa is currently capable of meeting 6.3 wt% and 70 kg H 2 /m 3 , respectively, for gravimetric hydrogen density and volumetric hydrogen density, based only on the storage material itself (and not on the storage system as a hole or including hardware ). Hence, it is not correct to compare the values obtained with those in Table 4.1, the 2010 FreedomCAR targets published by the US DOE. Can we ask: Is NaBH 4 a suitable hydrogen storage material for portable and/or niche application? It is difficult, indeed, to answer this question. To the best of the authors’ knowledge there is no paper or document clearly specifying the targets for portable applications. CHAPTER 4 Alkali free hydrolysis 63 Table 4.2 - Influence of Ni-Ru based catalyst/NaBH 4 on gravimetric hydrogen density and hydrogen yield for H 2 O/NaBH 4 : 4 mol/mol. Prediction of the volumetric hydrogen density*. Ni-based/NaBH 4 / g/g Gravimetric H 2 density / wt.% H 2 yield / % Volumetric H 2 density* / kgH 2 /m 3 Pressure / MPa 0.2 5.3 78 57 0.36 0.4 5.0 78 56 0.36 0.4 5.3 86 59 0.69 0.4 6.3 98 70 1.26 * It was assumed that the average value of the density of our materials, by addition relationship of densities of NaBH 4 (1.07 g/cm3), H2O (1.00 g/cm3) and Ni-based catalyst (3.17 g/cm3) is: 1.07 g/cm3, for H 2 O/NaBH 4 : 4 mol/mol, Ni-based catalyst/NaBH 4 : 0.2 g/g; and 1.11 g/cm3, for H 2 O/NaBH 4 : 4 mol/mol, Ni-based catalyst/NaBH 4 : 0.4 g/g Even for the purpose of a didactic demonstration, the MicroBoro Bus plus the H 2 generation system, presented here, delivers pure H 2 in high rates, during a good fraction of time/autonomy. Hence, it seems that NaBH 4 has potential for portable applications, or at least niche applications. In a recent paper, the values 3-7 wt% for gravimetric hydrogen storage capacities for NaBH 4 systems are mentioned [4.7]. It is important referring that sustainable method to recycle the borate by-products back to NaBH 4 would be a very important obstacle to overcome for the adoption of NaBH 4 as a hydrogen carrier of choice for portable H 2 fuel cell applications. 4.4 REACTION BY-PRODUCT CHARACTERIZATION The by-products of the alkali free hydrolysis experiments for hydrogen generation, reported on this chapter, were analyzed by X-Ray Diffractometry. As already referred suitable crystals, from the reactions performed under and above 1 MPa pressure, were obtained by slow evaporation of a water solution at uncontrolled room temperatures. The crystal structures were alike with respect to water content. 4.4.1 Materials and methods For the crystals coming through MR reactor experiments (at 0.69 MPa), XRD analysis revealed that they were monoclinic, space group C 2/ c , cell volume V =1478.8(2) Å 3 . Unit cell parameters a = 11.8899(9) Å, b = 10.6394(8) Å, c = 12.1989(9) Å, β = 106.609(8)º. Diffraction data were collected at 293 K with a Gemini PX Ultra equipped with MoK α radiation ( λ =0.71073 Å). The structure was solved by direct methods using SHELXS-97 [4.8] with atomic positions and displacement parameters refined with SHELXL-97 [4.9]. The non-hydrogen atoms were refined anisotropically and the hydrogen atoms were CHAPTER 4 Alkali free hydrolysis 64 refined freely with isotropic displacement parameters. The refinement converged to R (all data) = 5.30% and wR 2 (all data) = 11.87%. 4.4.2 Results and discussion The crystal structure reveals the B atoms in triangular and tetragonal configuration with O atoms. The Na atoms are coordinated by six water molecules. The molecular formula is Na 2 B 2 O 5 .4H 2 0. Less two water molecules were similarly find in the crystal structure in the SR batch reactor experiments (at 1.26 MPa). Figure 4.12, below, shows the two crystal structures discovered for the sodium metaborate hydrates. It may be concluded that the use of Ru-Ni based catalyst, that shows high activity in alkali free hydrolysis under high pressure, revelled not only high yields of hydrogen generation, as reported in the above sections, but also, for pressure > 1 MPa, a change in the structure of the by-product from Na 2 B 2 O 5 .4H 2 0 to two less water content. This is an important finding due to gravimetric hydrogen density and borates recycle. (a) (b) Figure 4.12. View along the c-axis of the crystal structure of sodium metaborate hydrate: (a) 4H 2 O, in MR at 0.69 MPa and (b) 2H 2 O, in SR at 1.26 MPa. The element color code, from black to light grey, is B, O, Na and H. 4.5 CONCLUSIONS The fundamental understanding of the alkali free hydrolysis of sodium borohydride obtained in this chapter suggests that it is possible to develop a hydrogen reactor/delivery CHAPTER 4 Alkali free hydrolysis 65 system that produces pure hydrogen in ≈ 100% yield, requires a suitable catalyst, does not involve strongly caustic solutions, and uses a minimum/stoichiometric amount of water. Chemical hydrides have the potential to meet certain H 2 storage targets only if the water consumption is minimized. A key thermophysical requirement is to minimize the “excess hydration factor”, as mentioned in this chapter. An alkali free hydrolysis under pressure may help to achieve this. Experimentally, we have shown for a system with H 2 O/NaBH 4 : 4 mol/mol and Ru-Ni based/NaBH 4 : 0.4 g/g, performed in a batch reactor with conical bottom shape (229 cm 3 ), that the H 2 produced reaches almost 100% conversion of NaBH 4 , delivered in fast rate and without significant lag time, when the operating pressure is > 1 MPa. 4.6 REFERENCES [4.1] Marrero-Alfonso E, Gray JR, Davis TA, Matthews MA. Minimizing water utilization in hydrolysis of sodium borohydride: the role of sodium metaborate hydrates. Int. J. Hydrogen Energy 32 (2007) 4723-4730. [4.2] C.M. Kaufman, S. Buddhadev, Hydrogen generation by hydrolysis of sodium tetrahydroborate: effects of acids and transition metals and their salts, J. Chem. Soc. Dalton Trans. (1985) 307-313. [4.3] H.I. Schlesinger, H.C. Brown, A.E. Finholt, J.R. Gilbreath, H.E. Hoeskstra, K. Hyde, Sodium borohydride, its hydrolysis and its use as a reducing agent and in the generation of hydrogen, J. Am. Chem. Soc. 75 (1953) 215-219. [4.4] U.S. Department of Energy Hydrogen Program. Available from: http://www.hydrogen.energy.gov/. [4.5] Y. Kojima, Y. Kawai, H. Nakanishi, S. Matsumoto, Compressed hydrogen generation using chemical hydride, J. Power Sources 135 (2004) 36-41. [4.6] A.M.F.R. Pinto, D.S. Falcão, R.A. Silva, C.M. Rangel, Hydrogen generation and storage from hydrolysis of sodium borohydride in batch reactors, Int. J. Hydrogen Energy 31 (2006) 1341-1347. [ 4.7 ] U.B. Demirci, O. Akdim, P. Miele, Ten-year efforts and a no-go recommendation for sodium borohydride for on-board automotive hydrogen storage, Int. J. Hydrogen Energy 34 (2009) 2638-2645. [4.8] G.M. Sheldrick, SHELXS-97, Program for the solution of crystal structures; University of Göttingen: Germany 1997. [4.9] G.M. Sheldrick, SHELXL-97, Program for the refinement of crystal structures; University of Göttingen: Germany 1997. CHAPTER 5 less Polar Organic Polymeric Solutions (lPOPS) hydrolysis 72 Table 5.2 – Values of conductivity and pH of the studied reactant solutions, before and after reaction completion. 20 mL reactant solution Conductivity pH Conductivity pH 10 wt% NaBH4, 7 wt% NaOH (mS/cm) (-) (mS/cm) (-) Classic - no polymer addition 80,3 13,4 64,5 13,4 plus 0.25 wt% CMC 80,5 13,6 67,5 13,4 plus 0.25 wt% SDS 87,1 13,4 23,0 13,5 temperature before the reaction after the reaction 298 K As can be seen in the previous table, a decrease of approximately 15 mS/cm in the value of conductivity was registered for both the reactant solutions: [10 wt% NaBH 4 + 7wt% NaOH + 83wt% H 2 O] and [10wt% NaBH 4 + 7wt% NaOH + 0.25wt% CMC + 82.75wt% H 2 O], before and after the reaction completion. A decrease of nearly more than four times was observed for the reactant solution with 0.25wt% SDS (23 mS/cm). This result was in some way expected due to the capability of SDS to form micelles. We believe that this aggregate, usually with its hydrophilic head oriented to the solvent (water) direction and hydrophobic tail leaning to its centre, had the capability to sequester H 2 to the micelle centre. Definitely that doesn’t seem to occur: the value of 0% of H 2 due to solubility effects (see Table 5.1) ‘speaks for itself’. 5.4 REACTION BY-PRODUCT CHARACTERIZATION By-products suitable crystals of the hydrogen generation reactions of all types of experiments reported on this chapter were analyzed by X-Ray Diffractometry. Very interesting results were obtained for the chemical structures of sodium borohydride hydrolysis by-product crystals: i) For the classic reaction (5.1), the crystal structure of the by-product is a metaborate dehydrate, NaBO 2 .2H 2 O, schematically presented in Figure 3.12 (see Chapter 3). ii) For the reagents presented in reaction (5.1) plus addition of 0.25wt% CMC, the crystal structure of the by-product is a borate anhydrous, NaBO 4 , schematically presented in Figure 5.2. The crystal structure reveals that the boron atoms are in a tetrahedral configuration with four oxygen atoms, with B-O bond lengths within the interval 1.466 and 1.492 Å. The B-O interatomic distances are around the expected value for a tetrahedral configuration 1.475 Å, which is appreciable higher than observed for triangular CHAPTER 5 less Polar Organic Polymeric Solutions (lPOPS) hydrolysis 73 configurations (1.365 Å) [5.8]. The sodium atoms have their valence distributed between five Na-O bonds (bond lengths within 2.329 and 2400 Å). The mean length of the Na-O bonds is 2.512 Å when the Na coordination number is seven and it is 2.450 Å when the coordination number is six [5.9]. It is reasonable to assume that the bond length and the Na coordination number decrease as the bond strength increases. Figure 5.2. View along the a-axis of the crystal structure of sodium borate anhydrous, showing the BO 4 tetrahedral arrangement and the Na bound network. Na in violet, O in red and B in pink. iii) Finally, for the reagents presented in reaction (5.1) plus addition of 0.25wt% SDS, the crystal structure of the by-product reveals that the oxygen atoms adopt a tetrahedral conformation around half of the boron atoms and a triangular conformation around the other half. Tetrahedral B-O bond lengths range from 1.450 and 1.500 Å and triangular B-O bond lengths from 1.355 and 1.387 Å, and are in agreement with the interatomic distances observed in other structures [5.8]. The BO 3 triangles and the BO 4 tetrahedra share oxygen atoms so as to produce endless zigzag chains. The sodium atoms form bonds with six oxygen atoms. The Na-O bond lengths are within 2.386 and 2448 Å, which is slightly lower than it is expected for a Na coordination number of six [5.9]. Therefore, we speculate that a borate anhydrous, formed by NaBO 3 and NaBO 4 shared molecules, schematically presented in Figure 5.3. Figure 5.3. View along the a-axis of the crystal structure of sodium borate anhydrous, showing the BO 3 triangular and the BO 4 tetrahedral arrangements sharing oxygen atoms and the Na bound network. Na in violet, O in red and B in pink. CHAPTER 5 less Polar Organic Polymeric Solutions (lPOPS) hydrolysis 74 At this point, we highlight that adding just 0.25 wt% of CMC or 0.25wt% of SDS to the reactants described in equation 1, leads to the formation of an anhydrous crystalline borate. A pertinent question can now be asked: what happens to water molecules usually associate to the classic catalytic hydrolysis of NaBH 4 by-product? Unexpectedly, it seems that both CMC polymer and SDS surfactant behave like desiccators of the NaBH 4 byproduct hydrolysis! Particularly in the latter one, the absence for H 2 solubility effects observed in the remaining solution with 0.25 wt% SDS (see Table 5.1) can now be explained by the capability of SDS to form micelles, enclosing water instead of hydrogen gas, in a way similar to a clathrate. Based on the above described achievements, it is thought that the results presented are significant findings in terms of: - NaBH 4 by-product hydrolysis recyclability. In fact, the inexistence of H 2 O in the crystalline borates mention in (ii) and (iii), marks effectively a cost reduction in the process of recycling these by-products back to NaBH 4 , comparing the same stage of the process but with the traditional NaBO 2 .2H 2 O. Actually, finding sustainable methods to recycle hydrolysis NaBH 4 by-product back to NaBH 4 , is a very important obstacle to overcome for the adoption of NaBH 4 as an hydrogen carrier of choice. Every step is important and our finding is, perhaps, a significant contribution; - Gravimetric H 2 storage density. Beyond doubt, the ability to produce anhydrous sodium borate (or metaborate) may be the key to increase the overall storage density of systems based on sodium borohydride as the storage media [5.1,5.2,5.6]. The gravimetric hydrogen storage density for the experiments reported in this work can be estimated considering the number of water molecules (2+x) attached to the by-products found in the crystals structures mentioned above. According to our experimental conditions, we may consider the quantity of water in the ratios H 2 O/NaBH 4 : 4, 2 and 2 mol/mol, respectively, for the crystals structures establish in (i), (ii) and (iii), as the minimum required amount (of water) necessary for reaction (5.1) completion. In this way, the usable specific energy from hydrogen, for the three sets of experiments reported in this work, can be evaluated on a reactant-only basis. The plots in Figure 5.4 show the results for the gravimetric H 2 storage density, for the three reactant systems studied. It can be seen that the gravimetric H 2 density is around 5.7 wt% for the classic reactant system and equal to 8.4 wt% and 8.7 wt%, respectively, for reactant systems with 0.25 wt% CMC and 0.25 wt% SDS. We believe that it is possible to augment the solubility of hydrogen gas in the liquid phase by-product (that remains inside the reactor under pressure) by small additions of CHAPTER 5 less Polar Organic Polymeric Solutions (lPOPS) hydrolysis 75 chemical materials, and this may be the key to increase the overall specific H 2 energy of systems based on NaBH 4 as hydrogen storage media. As far as the authors are aware, this issue has not been reported in the history of NaBH 4 as energy/hydrogen carrier [5.10]. 0 1 2 3 4 5 6 7 8 9 reactant solution Gravimetric H2 density / % "Classic" - no polymer addition LPOPS - 0.25 wt% CMC Surfactant - 0.25 wt% SDS Figure 5.4. Gravimetric H 2 storage density (reactants basis) for the three systems studied in this chapter. The present results also indicate a remarkable improvement from a previous work by the authors [5.7], in which, using the conventional excess of water method, performed also at moderate pressures (up to 2.6 MPa), a gravimetric hydrogen density of 4.3 % was achieved. 5.5 CONCLUSIONS The capability of the liquid by-product (NaBO 2 .2H 2 O, metaborate dehydrate) to store hydrogen gas by its dissolution was tested in this dissertation section. With that goal, small amounts of an organic polymer, Carboxyl Methyl Cellulose (CMC), and a surfactant, Sodium Dodecyl Sulphate (SDS), both in the form of fine powder, were added to the reactant solution before its injection in the reactor. The results obtained show that the gas dissolution in the liquid phase and the gravimetric hydrogen density increase with high H 2 pressures, and are additionally enhanced by the polarity change (measured by conductivimetry) of the remaining solution inside the reactor. As a consequence, anhydrous borates were produced in the presence of these additives (CMC or SDS). We CHAPTER 5 less Polar Organic Polymeric Solutions (lPOPS) hydrolysis 76 believe that this finding may be the key to increase the overall storage density of systems based on NaBH 4 as hydrogen carrier. To conclude, the results presented in this chapter showed that gravimetric H 2 density of the hydrolysis of sodium tetrahydroborate can be augmented to reach ≈ 9 wt%, by adding small amounts of an organic polymer or surfactant to the stabilized reactant solution. However, the eventual success of this new approach will depend upon the development of a simple method of converting borates (BO 2 , BO 3 , BO 4 groups) into tetrahydroborate. 5.6 REFERENCES [5.1] L. Schlapbach, A. Züttel, Hydrogen-storage materials for mobile applications, Nature 414 (2001) 353-358. [5.2] U.B. Demirci, O. Akdim, P. Miele, Ten-year efforts and a no-go recommendation for sodium borohydride for on-board automotive hydrogen storage, Int. J. Hydrogen Energy 34 (2009) 2638-2645. [5.3] U.S. Department of Energy Hydrogen Program. Available from: http://www.hydrogen.energy.gov/. [5.4] S.C. Amendola, S.L. Sharp-Goldman, M.S. Janjua, N.C. Spencer, M.T. Kelly, P.J. Petillo, M. Binder, A safe, portable, hydrogen gas generator using aqueous borohydride solution and Ru catalyst, J. Power Sources 85 (2000) 186-189. [5.5] H.I. Schlesinger, H.C. Brown, A.E. Finholt, J.R. Gilbreath, H.E. Hoeskstra, K. Hyde, Sodium borohydride, its hydrolysis and its use as a reducing agent and in the generation of hydrogen, J. Am. Chem. Soc. 75 (1953) 215-219. [5.6] E. Marrero-Alfonso, J.R. Gray, T.A. Davis, M.A. Matthews, Minimizing water utilization in hydrolysis of sodium borohydride: the role of sodium metaborate hydrates, Int. J. Hydrogen Energy 32 (2007) 4723-4730. [5.7] A.M.F.R. Pinto, D.S. Falcão, R.A. Silva, C.M. Rangel, Hydrogen generation and storage from hydrolysis of sodium borohydride in batch reactors, Int. J. Hydrogen Energy 31 (2006) 1341-1347. [5.8] W.H. Zachariasen, The crystal structure of monoclinic metaboric acid, Acta Cryst. 16 (1963) 385-389. [5.9] M. Marezio, H.A. Plettinger, W.H. Zachariasen, The bond lengths in the sodium metaborate structure, Acta Cryst. 16 (1963) 594-595. [5.10] U.B. Demirci, P. Miele, Sodium tetrahydroborate as energy/hydrogen carrier, its history, C. R. Chimie 12 (2009) 943-945. 77 CHAPTER 6 Nickel based bimetallic catalyst characterization Results after 200 reutilizations This chapter focuses in the characterization of the nickel based bimetallic catalyst after 200 reutilizations. Textural properties based on nitrogen adsorption isotherms; surface morphology by scanning electron microscopy (SEM) coupled with EDS spectroscopy and X-ray photoelectron spectroscopy (XPS) analysis, were used to characterized the powder catalyst. 6.1 INTRODUCTION Most published research papers in the area of this dissertation deal with catalytic material used for hydrolyzing sodium borohydride. It is well known that catalysts suffer from deactivation [6.1]. Nonetheless, very few papers studied the catalyst durability; among these, a good example is the work reported by Kim et al. [6.2]. The deterioration of the catalyst activity is a very important issue, since the catalyst is the key material for varying the amount of generated molecular hydrogen from NaBH 4 hydrolysis reactions. The US DOE set technical targets for durability and operability of catalysts, but the reasoning behind the targets set was not explained in their review [6.3]. It is the aim of this work to analyze the state of ‘health’ of the powder nickel based bimetallic catalyst after it was used two hundred times. In fact, the catalyst was used in several different types of kinetic experiments, as described in the previous chapters of this dissertation, and always revealed good activity (by the low values in lag time and high dtdP/ slopes in the linear zone). The fact that the catalyst was used more than 200 times motivated us to carry out a detailed study of the catalyst activity. In truth, the catalyst showed good working efficiency both under higher concentrations of hydride and inhibitor in the alkali NaBH 4 hydrolysis CHAPTER 6 Nickel based bimetallic catalyst characterization 78 with excess of water (even in the experiments with refuelling fresh NaBH 4 solution), and with alkali free hydrolysis with 2 < H 2 O/NaBH 4 , mol/mol < 4. It was also verified that the influence of reactor design (bottom shape) affects the catalyst activity; in particular, the reaction vessel with conical bottom shape promoted an intensification of the catalyst activity, by reducing the lag time of the reaction, in comparison with the flat bottom shape. 6.2 CATALYST CHARACTERIZATION The Ni-Ru based powdered catalyst after 200 times of reuse was characterized by textural properties based on nitrogen adsorption isotherms; by surface morphology using scanning electron microscopy (SEM) coupled with EDS spectroscopy and by X-ray photoelectron spectroscopy (XPS). A brief description of these instrumental analysis techniques and the results obtained are given in the following sections. 6.2.1 Textural properties Our concern here is with the internal surface area of the porous Ni-Ru based solid catalyst, and with the nature of the pores giving rise to this area. In most porous solids the external surface is a negligible fraction of the total surface, but if relevant it may readily be estimated by measuring the permeability of gas through a bed of the solid, or by sedimentation. The parameters of interest are then the internal surface area and the pore volume, from which an average pore radius may be obtained, and the pore size distribution. Knowledge of the pore size distribution is obtainable by mercury porosimetry. In this technique, mercury is forced under pressure into the pores, and the greater the pressure, the smaller the pores to which the mercury obtains access. By sensitively measuring the change in volume of the solid plus the mercury with increasing pressure, a picture of the pore size distribution is obtained. Useful information is also gained by measuring the physical adsorption of gases on porous solids. It is known that the monolayer capacity of a non-porous solid, measured by chemisorption, or by physical adsorption well above the boiling point of the adsorbing gas (usually N 2 ), can be easily translated into a surface area. However with porous solids, and using temperatures close to the boiling point of the gas, so that multilayer adsorption occurs, several forms of isotherms besides the Langmuir type can be observed. The Type IV is of great interest. This isotherm usually shows a hysteresis loop, that is, the isotherm does not follow the same path in the desorption as it does in adsorption. The reason for this CHAPTER 6 Nickel based bimetallic catalyst characterization 79 is that evaporation of condensed gas in fine pores does not occur as easily as its condensation; this is because a molecule evaporating from a highly curved meniscus has a higher probability of recondensing than one evaporating from a plane surface. This effect was discussed many years ago by Lord Kelvin who devised the following equation to describe the effect: φ γ cos 2 )/ln( 0 rRT V PP −= , (6.1) where V is molar volume of the liquid, γ its surface tension, r the pore radius and φ the contact angle (usually taken to be zero); the other symbols have their usual significance. The relative pressure at which condensation will occur in a pore of a given size can thus be determined, and the isotherm then used to obtain a pore size distribution. Hysteresis loops vary greatly in shape, and these too have also been classified in literature [6.4]. The textural characterization of the Ru-Ni based powdered catalyst (after 200 reutilizations), including the determination of surface areas (S BET ), was based on the N 2 adsorption isotherms, using a Quantachrome Instruments Nova 4200e apparatus. Prior to the analysis, the sample (0.127 g) was degassed at 160 ºC for 3 h. Nitrogen was used as adsorbate at liquid nitrogen temperature (77 K). Pore size distribution was obtained from the desorption branch of the isotherms using the Barrett, Joyner and Halenda (BJH) method. The micropore volumes and mesopore surface areas were determined by the tmethod. The adsorption and desorption isotherms of the powder Ni-Ru based catalyst 200 times reused are displayed in Figure 6.1. The result shows that the N 2 isotherm shape are not of type IV, at least for P/P 0 <0.986. Hence the validity of BJH method is discussable for this powder material. The plot on Figure 6.2 illustrates the pore size distribution. As shown, the pore radius peak shifts to a smaller size region, in the range 17-23 Å. CHAPTER 6 Nickel based bimetallic catalyst characterization 80 0 20 40 60 80 100 120 0 0,2 0,4 0,6 0,8 1 Relative pressure / P/P 0 Volume adsorbed / cm 3 STP g -1 Figure 6.1. Nitrogen isotherms of powder Ni-Ru based catalyst 200 times reused. 0,0E+00 4,0E-03 8,0E-03 1,2E-02 1,6E-02 2,0E-02 0 50 100 150 200 250 pore radius / Å dV(r) / cm 3 g -1 Å -1 Figure 6.2. Pore size distributions of powder Ni-Ru based catalyst 200 times reused. Table 6.1 summarizes the textural properties of powder Ni-Ru based catalyst 200 times reused. Table 6.1 - Textural properties of powder Ni-Ru based catalyst 200 times reused sample. S BET / m 2 g -1 V µ / cm 3 g -1 S ext / m 2 g -1 V p / cm 3 g -1 58 0 58 0.15 S BET = BET surface area; V µ = micropore volume by t-method; S ext = (mesopore + macropore) surface areas by t-method; V p = total pore volume, for P/P 0 =0.986. CHAPTER 6 Nickel based bimetallic catalyst characterization 81 It should be noted that the rate of a catalytic reaction should be proportional to the surface area of the catalyst, provided that the transfer of reactants to or of products from the surface is not the slow step. It is therefore normally desirable to get as much surface area as possible into a given volume. Analysing the data plotted in Fig. 6.2, we may conclude that the pores through the particles catalyst sample, assuming microspheres of uniform size, lies between 4 and 20 nm in diameter. This means that the pore size in our catalyst, after 200 reutilizations, belongs to the group of mesopores. Would pores below 2 nm in width, i.e., micropores be expectable? If micropores existed, can they be surface closed by effects of particle agglomeration? The values of surface areas, by BET and t methods, reveal 58 m 2 /g; surprisingly, this is not a large surface area for a fine powder catalyst [6.4]. In fact V µ = 0 cm 3 /g by t-method. 6.2.2 Morphology Surface morphological observations were obtained at CEMUP (Centro de Materiais da Universidade do Porto). The Scanning Electron Microscopy (SEM imaging) coupled with EDS (analysis) spectroscopy was performed using a unit FEI Quanta 400 FEG ESEM / EDAX Genesis X4M operating at 15 kV in low vacuum mode (LVSEM for uncoated non conductive sample). The powder sample was prepared by simple dispersion over a double side adhesive carbon tape. Figure 6.3 shows SEM micrographs of the powder Ni-Ru based catalyst 200 times reused, at magnifications up to 100 000X. SEM images at higher magnification showed distortions due to slight magnetic properties of the catalyst sample. In the micrographs of Fig.6.3, from (a) to (b), the catalyst sample in the bulk form seems that can be broken-up into smaller particles until reaching a consistency similar to that of a fine dust. Image 6.3(b) show a Ni-Ru based particle that is about 100 µ m in vertical length and a background of smaller ones of about 1-25 µ m in diameter. Going from Fig.6.3 (b) to (e), the magnetic properties of Ni-Ru based appear to cause some agglomeration of the smaller particles. Therefore, closer examination of the particles surface in Fig.6.3 (c) shows a finer reticulated structure, interconnecting more or less spherical particles. CHAPTER 6 Nickel based bimetallic catalyst characterization 88 Table 6.3 - Results of XPS analysis of Ni-Ru based catalyst 200 times reused. Element Peak range Area Sens. Factor At / % C 1s synthesis 13285 1 13.57 O1s synthesis 187038 2.93 65.20 Ni 2p3 synthesis 283316 14.61 19.81 Ru 3d5 synthesis 10269.3 7.39 1.42 At = total surface area. The XPS analysis showed that nickel (Ni 2p3) is the main metal surface constituent of our bimetallic powdered catalyst 200 times reused, 13.57 % At. Ruthenium (Ru 3d5) contribution is about 1.42 % At. The XPS data confirmed the increased in Ru concentration (Ru3d5) from 0.74 to 1.42 At %. A slight increase in area (At %) is suggested which is thought to be due to segregation of ruthenium as a result of the extensive use in the catalyzed hydrolysis. 6.3 CATALYST ACTIVITY ANALYSIS AFTER 200 REUTILIZATIONS This section intends to characterize the catalyst after 200 reutilizations. The state of ‘health’ of our Ni-Ru based powdered catalyst (200 times reused) is checked by evaluating its activity, in terms of yield, lag time and dtdP/ slopes in the linear zone, at two different stages of reutilization. Two experiments, of the topic of alkali hydrolysis of NaBH 4 , were chosen to illustrate our analysis. Plots in Figure 6.10 shows the hydrogen generation of 10 mL of a reactant solution with concentration of 10 wt% NaBH 4 and 7 wt% NaOH by weight, performed in reactor LR (646 cm 3 ) with a proportion of Ni-Ru based catalyst/NaBH 4 : 0.4 g/g ( ≈ 0.43 g), when the catalyst was 28 times and 200 times reused. No stirring was performed in the solutions inside the reactor. Both reactions were performed at room temperature, without temperature control: the reaction with catalyst reused 28 times was carried out in the summer of 2008 (at 27 ºC) and the reaction with catalyst reused 200 times in the middle of the spring of 2009 (25 ºC). Table 6.4 shows the values of lag time and dP/dt obtained from the curves presented in Figure 6.10. The results in Table 6.4 (and in Figure 6.10) revealed that our nickel-based bimetallic catalyst reused 200 times is ‘healthy’, although some deterioration in terms of H 2 yield and rate is evident (note that the two reactions plotted in Figure 6.10 employ a catalyst separated by 172 uses). CHAPTER 6 Nickel based bimetallic catalyst characterization 89 Table 6.4 - Results for yield, lag time and dtdP/ slope in hydrogen generation of 10 mL of 10wt% NaBH 4 and 7wt% NaOH solution, at two different stages of Ni-Ru based catalyst reutilization (for 28 and 200 times reused), in batch reactor LR (646 cm 3 ). Ni-Ru based catalyst yield / % lag time / s dtdP/ slope* / bar/s 28 times reused 100 20 2.48E-02 200 times reused 91 21 7.59E-02 * calculated in the linear zone. 0 1 2 3 4 5 0 200 400 600 800 time / s Pressure / bar 28 times reused catalyst 200 times reused catalyst Theoretical yield Figure 6.10. Hydrogen generation plots with a NaBH 4 concentration of 10%, an inhibitor concentration of 7%, for two different stages of catalyst reutilization (28 times reused and 200 times reused). The reactions, for 10 cm 3 of reactant solution, were performed in LR batch reactor (646 cm 3 ) with a proportion of Ni-Ru based catalyst/NaBH 4 : 0.4 g/g, at room temperature of ≈ 25 ºC. In terms of yield, the results found in Table 6.4 shows deterioration by a factor of 1.1, XX 20028 1.1 ηη = , this means that the average loss in yield per each utilization of the catalyst is approximately 0.6 %. In terms of hydrogen generation rate, a larger factor was found in the universe of 172 reutilizations: 3.3, i.e., XX dtdPdtdP 20028 )/(3.3)/( = in the linear zone; hence a loss of 1.9% per reutilization can be estimated for the rate value. With respect to lag time, identical values were found in the two experiments, indicating that the catalyst can be reused several times without affecting lag time. 6.4 CONCLUSIONS The study of catalyst durability on catalyzed hydrolysis of sodium borohydride is essential from an application point of view. Few works on this topic are available in the literature. In the present chapter, an effort was made to characterize the powder nickel-based bimetallic CHAPTER 6 Nickel based bimetallic catalyst characterization 90 catalyst, used in two different schemes of NaBH 4 hydrolyzing (alkali hydrolysis and alkali free hydrolysis) and after 200 reutilizations. The textural characterization of the Ru-Ni based powdered catalyst (200 reused) showed an N 2 isotherm shape different from type IV, at least for P/P 0 <0.986, evaluated by BJH method. The pore size distribution, obtained from the desorption branch of the isotherm, gave pore radius peak shifts in the range 17-23 Å, showing an abundant presence of mesopores in the catalyst. This corroborate with the values found for surface areas by BET and t methods of 58 m 2 /g. The t-method also revealed the absence of micropores volume in the 200 times reused powder catalyst. Surface morphology analysis by using scanning electron microscopy (SEM) coupled with EDS spectroscopy revealed that nickel is the main constituent of the powder Ni-Ru based catalyst 200 times reused. The XPS study showed that nickel (Ni 2p3) is the main metal surface element (13.57 % At.) and ruthenium (Ru 3d5) contributed with 1.4 % At. The catalyst activity was checked by confronting the values of yield, lag time and dtdP/ slopes (in the linear zone, of a graphical representation of reactor inside pressure as function of time). In two similar experiments for hydrogen generation by alkali hydrolysis of NaBH 4 , but performed at two different stages of catalyst reutilization (which differ from each other in 172 times), the results revealed some deterioration of the catalyst activity (the dtdP/ slopes are lower, as expected, after such a long time utilization). Finally, we may conclude that the powder Ni-Ru based catalyst, after 200 applications for hydrolyzing NaBH 4 , looses approximately 0.6 % in yield and 1.9% in rate, per reutilization, but still preserves the lag time required to observe an increase in H 2 pressure. For this reason, since no great major deterioration was found in the catalyst activity after it was used 200 times - in truth, it still works at very reasonable H 2 generation rates - we can conclude that its durability is rather long. A possible explanation for this active catalyst, comparing with similar work done by Pinto el al. [6.5], is the presence of 1.4 wt% of ruthenium, a noble Group VIII metal. 6.5 REFERENCES [6.1] J.H. Wee, A comparison of sodium borohydride as a fuel for proton exchange membrane fuel cells and for direct borohydride fuel cells , J. Power Sources 155 (2006) 329-339. [6.2] J.H. Kim, K.T. Kim, Y.M. Kang, H.S. Kim, M.S. Song, J.Y. Lee, Study on degradation of filamentary Ni catalyst on hydrolysis of sodium borohydride , J. Alloys Comp. 379 (2004) 222-227. CHAPTER 6 Nickel based bimetallic catalyst characterization 91 [6.3] U.S. Department of Energy Hydrogen Program. Available from: http://www.hydrogen.energy.gov/. [6.4] G.C. Bond, “Heterogeneous catalysis: principles and applications”, Oxford Chemistry Series, 1974. [6.5] A.M.F.R. Pinto, D.S. Falcão, R.A. Silva, C.M. Rangel, Hydrogen generation and storage from hydrolysis of sodium borohydride in batch reactors, Int. J. Hydrogen Energy 31 (2006) 1341-1347. CHAPTER 6 Nickel based bimetallic catalyst characterization 92 93 CHAPTER 7 Conclusion This chapter is devoted to conclusions and suggestions for future work. Conclusions drawn from the results indicate that the reused ruthenium nickel based catalyst is capable of catalyzing the sodium borohydride solutions at a sufficient rates, in the two schemes presented in this dissertation for comparison – alkali hydrolysis and alkali free hydrolysis of sodium borohydride for hydrogen generation under pressure. 7.1 CONCLUSIONS This section provides a summary of the primary research results. The sequence of appearance is the one of the body of the manuscript. Experimental techniques and apparatus Description of the materials used in all the experimental work, including the nickel based bimetallic catalyst, is given. The plan for studied the catalytic hydrolysis of sodium borohydride under pressure, in three batch reactors, with different bottom geometries, is available. To study the influence of reactor bottom shape on the hydrogen generation rates - this constitutes one of the originalities of the present dissertation, two different bottom geometries were designed – one flat and the other conical. The latter has the purpose of enabling non-dispersible effects during the contact of the catalyst with the reactant solution. Alkali hydrolysis, for sodium borohydride kinetic experiments An extensive experimental work was carried out with the objective of studying various effects in hydrogen generation rate, yield and lag time, by catalytic alkali hydrolysis of sodium borohydride. It was concluded that: CHAPTER 7 Conclusion 94 - increasing the temperature of the reaction medium, increases the rate of H 2 generation; - the reaction rate increases with increasing the hydride concentration up to 20 wt% of NaBH 4 . Above this value, a slight decrease on both H 2 and yield was found; - the reaction rate is greatly enhanced by the increase of the NaOH concentration, ensuring good efficiency of hydrogen generation; - the reactor bottom shape affects the course of hydrogen generation. It was found that the conical bottom geometry leads to an increase on reaction rate, in the absence of H 2 lag time; - the Ni-Ru based catalyst has a good performance in refueling experiments. In fact, it was found that the catalyst tends to perform better, with a higher H 2 rate, after the first loading of reactant (the amount of catalyst remains the same from the first to the last loading); - a vigorous stirring of the solution inside the batch reactor, during the experiments with successive fuel loadings, reveals minor reaction induction time and a slight higher yield in hydrogen generation. This indicates possible mass transfer limitations during the course of reaction between the fuel and the active sites of the catalyst. The x-ray diffractions analyses reveal a sodium metaborate dehydrated, NaBO 2 .2H 2 O, as the main by-product. Alkali free hydrolysis, for sodium borohydride kinetic experiments Chemical hydrides have the potential to meet certain H 2 storage targets only if the water consumption is minimized. The main purpose of the alkali free hydrolysis of NaBH 4 experiments reported in this dissertation was to produce pure hydrogen gas with very high gravimetric and volumetric densities. The alkali free hydrolysis under pressure, when the operating pressure is > 1 MPa, for a system with H 2 O/NaBH 4 : 4 mol/mol and Ni-Ru based/NaBH 4 : 0.4 g/g, performed in a batch reactor with conical bottom shape (229 cm 3 ), produced H 2 almost at 100% conversion of NaBH 4 , and hydrogen was delivered at a fast rate without significant lag time. In terms of gravimetric hydrogen density a value of 6.3% was found (this value does not take into account the hardware of the system). lPOPS hydrolysis, for sodium borohydride kinetic experiments The capability of the liquid by-product (NaBO 2 .2H 2 O, metaborate dehydrate) to store hydrogen gas by its dissolution was tested in this dissertation section. With that goal, small amounts of an organic polymer, Carboxyl Methyl Cellulose (CMC), and a surfactant, CHAPTER 7 Conclusion 95 Sodium Dodecyl Sulphate (SDS), both in the form of fine powder, were added to the reactant solution before its injection in the reactor. The results obtained show that the gas dissolution in the liquid phase and the gravimetric hydrogen density increase with high H 2 pressures, and are additionally enhanced by the polarity change (measured by conductivimetry) of the remaining solution inside the reactor. As a consequence, anhydrous borates were produced in the presence of these additives (CMC or SDS). We believe that this finding may be the key to increase the overall storage density of systems based on NaBH 4 as hydrogen carrier. However, the eventual success of this new approach will depend upon the development of a simple method of converting borates (BO 2 , BO 3 , BO 4 groups) into tetrahydroborate. Nickel based bimetallic catalyst characterization after 200 times reuse Catalyst durability on catalyzed hydrolysis of sodium borohydride is essential from an application point of view. The textural characterization of the Ru-Ni based powdered catalyst (200 reused) showed an N 2 isotherm shape different from type IV, at least for P/P 0 <0.986, evaluated by BJH method. The pore size distribution, obtained from the desorption branch of the isotherm, gave pore radius peak shifts in the range 17-23 Å, showing an abundant presence of mesopores in the catalyst. This corroborates with the values found for surface areas by BET and t methods of 58 m 2 /g. The t-method also revealed the absence of micropores volume in the 200 times reused powder catalyst. Surface morphology analysis by using scanning electron microscopy (SEM) coupled with EDS spectroscopy revealed that nickel is the main constituent of the powder Ru-Ni based catalyst 200 times reused. The XPS study showed that nickel (Ni 2p3) is the main metal surface element (13.57 % At.) and ruthenium (Ru 3d5) contributed with 1.4 % At. The catalyst activity was checked by confronting the values of yield, lag time and dtdP/ slopes (in the linear zone, of a graphical representation of reactor inside pressure as function of time). In two similar experiments for hydrogen generation by alkali hydrolysis of NaBH 4 , but performed at two different stages of catalyst reutilization (which differ from each other in 172 times), the results revealed some deterioration of the catalyst activity (the dtdP/ slopes are lower, as expected, after such a long time utilization). Finally, we may conclude that the powder Ni-Ru based catalyst, after 200 applications for hydrolyzing NaBH 4 , looses approximately 0.6 % in yield and 1.9% in rate, per reutilization, but still preserves the lag time required to observe an increase in H 2 pressure. CHAPTER 7 Conclusion 96 For this reason, since no great major deterioration was found in the catalyst activity after it was used 200 times - in truth, it still works at very reasonable H 2 generation rates - we can conclude that its durability is rather long. A possible explanation for this activity of the catalyst is the presence of 1.4 wt% of ruthenium, a noble Group VIII metal. 7.2 SUGGESTIONS FOR FUTURE WORK The following is a list of the objectives that would, perhaps, be most significant as further development of the present research: I – Development of a reaction model that reflects the reaction rate as a function of concentration, pressure, and temperature, model which is currently not available in the literature. In fact, a model capable of predicting the hydrolysis rate for hydrolyzing systems under increasing pressure, would be worthwhile principally for systems similar to that developed in this dissertation; II - Rate data is needed for NaBH 4 concentrations greater than 30 wt%. Systems intended to exceed the DOE 2010/2015 objective will require information about how solutions behave at high NaBH 4 concentrations. Additional information is also needed at low temperatures, less than 20 ºC, to address start-up issues associated with hydrogen generators or fuel cells operating in cold environments; III – Data on solubility of H 2 in aqueous borate systems is needed for storage molecular hydrogen in liquid phase. Working at high H 2 pressures, where the hydrogen generated is allowed to build up within the closed system, facilitate dissolution of hydrogen by solubility effects. Hence, it is possible envisage a system that generates and simultaneously stores molecular hydrogen in the reminiscent liquid phase inside the reactor. Indeed information about the amount of hydrogen dissolved in the solution with several concentrations of hydride and of inhibitor, which can be estimated trough Henry’s Law for the range of pressure and temperatures of operation, is necessary to carry out. IV – Knowing materials (chemicals) which enhance the solubility of hydrogen gas in liquid phase, under moderate pressures, is of great value for hydrogen storage in liquid phase.