|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001914467
007 cr mnu|||uuuuu
008 071024s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001984
Rivera, Luis A.
Destabilization and characterization of LiBH4/MgH2 complex hydride for hydrogen storage
h [electronic resource] /
by Luis A. Rivera.
[Tampa, Fla.] :
b University of South Florida,
ABSTRACT: The demands on Hydrogen fuel based technologies is ever increasing for substitution or replacing fossil fuel due to superior energy sustainability, national security and reduced greenhouse gas emissions. Currently, the polymer based proton exchange membrane fuel cell (PEMFC), is strongly considered for on-board hydrogen storage vehicles due to low temperature operation, efficiency and low environmental impact. However, the realization of PEMFC vehicles must overcome the portable hydrogen storage barrier. DOE and FreedomCAR technical hydrogen storage targets for the case of solid state hydrides are: (1) volumetric hydrogen density > 0.045 kgH2/L, (2) gravimetric hydrogen density > 6.0 wt%, (3) operating temperature < 150 degrees C, (4) lifetimes of 1000 cycles, and (5) a fast rate of H2 absorption and desorption. To meet these targets, we have focused on lithium borohydride systems; an alkali metal complex hydride with a high theoretical hydrogen capacity of 18 wt.%. It has been shown by Vajo et al. that adding MgH2, improves the cycling capacity of LiBH4. The pressure-composition-isotherms of the destabilized LiBH4 + MgH2 system show an extended plateau pressure around 4-5 bars at 350 degrees C with a good cyclic stability. The mentioned destabilizing mechanism was successfully utilized to synthesize the complex hydride mixture LiBH4 + 1/2MgH2 + Xmol% ZnCl2 catalyst (X=2, 4, 6, 8 and 10) by ball milling process. The added ZnCl2 exhibited some mild catalytic activity which resulted in a decomposition temperature reduction to 270 degrees C. X-ray powder diffraction profiles exhibit LiCl peaks whose intensity increases proportionately with increasing ZnCl2 indicating an interaction between catalyst and hydride system, possibly affecting the total weight percent of desorbed hydrogen. Thermal gravimetric analysis profiles for MgH2 + 5mol% nanoNi and LiBH4 + ZnCl2 + 3mol% nanoNi indicate that small concentrations of nano-nickel acts as an effective catalyst that reduces the mixture desorption temperature to around 225 degrees C and 88 degrees C, respectively. Future work will be focused on thermodynamic equilibrium studies (PCT) on the destabilized complex hydrides.
Thesis (M.S.)--University of South Florida, 2007.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 70 pages.
Advisor: John T. Wolan, Ph.D.
x Chemical Engineering
t USF Electronic Theses and Dissertations.
Destabilization and Ch aracterization of LiBH4/MgH2 Complex Hydride for Hydrogen Storage by Luis A. Rivera A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Chemical Engineering Department of Chemical Engineering College of Engineering University of South Florida Major Professor: John T. Wolan, Ph.D. Elias Stefanakos, Ph.D. Aydin K. Sunol, Ph.D. Sesha Srinivasan, Ph.D. Date of Approval: April 9, 2007 Keywords: Hydrogen Desorption, Lithium Borohydride, Magnesium Hydride, Dehydrogenation, Mechano-Chem ical Process, Dopants Copyright 2007, Luis A. Rivera
DEDICATION Dedicated to my family, my lovely wife Ileanexis, my daughter Adriana and my fellow engineers May God bless them.
ACKNOWLEDGEMENTS I would like to thank my advisors an d co-advisors Dr. J ohn Wolan, Dr. Elias Stefanakos, Dr. Aydin Sunol and Dr. Sesha Srinivasan for their support, guidance and advice provided over the course of my research experience here at USF. I would also like to thank Matt Smith, Rob Tufts, Jay Bieb er and Michael Jurcz yk for the analytical training and advice. I, gratefully acknowledge my friends Carla Webb, Cecil Coutinho, Jonathan Mbah, Eric Weaver, Diego Escobar and Aj ay Vidyasagar for their constant support throughout the learning process. I would also like to th ank Mr. Bernard Batson, Dr. Ashanti J. Pyrtle, Dr. Shekhar Bhansali, the National Science Foundations Bridge to Doctorate program and the US Department of Energys Hydrogen Fu el Initiative (HFI) program for the financial support. Last, but not least, I wish to acknowledge my family for their encouragements and moral support in every step of my academic career.
i TABLE OF CONTENTS LIST OF TABLES iii LIST OF FIGURES iv ABSTRACT vii CHAPTER 1 INTRODUCTION 1 1.1 Current Energy Situation 1 1.1.1 Oil Market 1 1.1.2 Global Warming and Greenhouse Effect 5 1.1.3 The Alternative 7 CHAPTER 2 INTRODUCTION TO COMPLEX METAL HYDRIDES 9 2.1 General Overview 9 2.2 Lithium Borohydride (LiBH4) 12 2.3 Magnesium Hydride (MgH2) 15 2.4 Zinc Borohydride (Zn(BH4)2) 16 CHAPTER 3 MATERIALS, EQUIPMEN TS AND APPROACH 17 3.1 Experimental Materials 17 3.2 General Approach 18 3.3 General Procedure 19 3.4 Equipments 22 3.4.1 Nitrogen Filled Glove Box 22 3.4.2 Ball Mill (BM) 25 3.4.3 Simultaneous DSC and TGA (SDT) 26 3.4.4 X-Ray Diffractrometer (XRD) 30 3.4.5 Pressure Composition Isotherms (PCI) Apparatus 32 3.4.6 Fourier Transform Infrared Spectrometer (FT-IR) 35 CHAPTER 4 RESULTS AND DISCUSSION 37 4.1 Undoped and Doped LiBH4 + ZnCl2 37 4.2 MgH2 + nano Ni 45 4.3 LiBH4 + MgH2 + Xmol% (ZnCl2 or TiCl3) 51
ii CHAPTER 5 CONCLUSION AND RECOMMENDATIONS 63 5.1 Conclusion and Recommendations 63 REFERENCES 65
iii LIST OF TABLES Table 1.1: World Oil Reserves by Country as of January 1, 2006 . 2 Table 2.1: Classifica tion of Metal Hydrides. 9 Table 2.2: DOE and FreedomCAR Hydr ogen Storage Team Technical Targets [23, 24]. 11 Table 2.3: Thermal Analysis of Lith ium Borohydride According to Fedneva et al. . 13 Table 2.4: Thermal Studies by Zuttel et al. on Lithium Borohydride [38, 39]. 13 Table 2.5: Thermal Studies by Orimo et al. on Lithium Borohydride . 13 Table 3.1: List of Materials. 17 Table 4.1: Thermogravimetric Analysis of Undoped and Doped 2LiBH4 + ZnCl2. 39 Table 4.2: Reduction Effect in the Theoretical Total H2 wt% by ZnCl2 Addition. 54
iv LIST OF FIGURES Figure 1.1: World Oil Prices, 1980-2030. 4 Figure 1.2: Price Comparison for a Ga llon of Regular Grade Gasoline in 2004 and 2005 . 4 Figure 1.3: Keeling Curve CO2 Measurement at Mauna Loa, Hawaii . 5 Figure 1.4: CO2 Concentration in Parts Per Mill ion Plotted Against Thousands of Years Before Present . 6 Figure 1.5: PEM Fuel Cell . 8 Figure 2.1: Hypothetical Hydrogen Economy Using NaBH4 as Hydrogen Storage . 10 Figure 2.2: Comparison of Volumetric and Gravimetri c Hydrogen Capacities for Some Metal Hydrides . 12 Figure 3.1: Procedure Followe d During the Investigation. 18 Figure 3.2: Ball Milling Bowl and the Specially Designed Lid with Schrader Valves. 20 Figure 3.3: XRD Sample Holder Covere d with Polyethylene Clear Plastic Wrap. 22 Figure 3.4: TA Instruments Diagram of Glove Box . 23 Figure 3.5: Picture of Glove Box. 24 Figure 3.6: Glove Box Sy stem Flow Diagram . 24 Figure 3.7: Picture of Ball Mill Equipment . 25 Figure 3.8: Cross-Sectional Diagram of the Planetary Ball Mill Movement . 26 Figure 3.9: Cross-Sectional Diagram of SDT Furnace . 27
v Figure 3.10: SDT Balance Housing . 27 Figure 3.11: Full Cross-Sectiona l Diagram of SDT-Q600 . 28 Figure 3.12: SDT-Q600 Photo. 29 Figure 3.13: SDT-Q600 Inside an Inert Atmosphere Glove Box. 29 Figure 3.14: Constructive (Left Picture) or Destructive (Right Picture) Interferences . 31 Figure 3.15: Schematic of an X-ray Diffractometer . 31 Figure 3.16: Xpert Diffractometer Picture. 33 Figure 3.17: PCT Diagram (Left) Associated with the Vant Hoff Plot (Right) . 34 Figure 3.18: PCTPro-2000 Hydrogen Sorption Apparatus Picture. 34 Figure 3.19: PCTPro-2000 Manifold Monitor Indicator. 35 Figure 3.20: Basic Diagram for an FT-IR Spectrometer . 36 Figure 4.1: SDT-Q600 TGA Profiles for Doped and Undoped LiBH4 + ZnCl2. 40 Figure 4.2: SDT-Q600 DSC Profiles for Doped and Undoped LiBH4 + ZnCl2. 41 Figure 4.3: XRD Profiles for Doped and Undoped 2LiBH4 + ZnCl2, Including a Hand Crushed Mixture. 42 Figure 4.4: XRD Profile for the Polyet hylene Clear Plastic Wrap (Thin Foil) Used to Protect the Samples. 43 Figure 4.5: XRD Profile of Pure ZnCl2. 44 Figure 4.6: SDT-Q600 TGA Profiles for Undoped MgH2 + BM9hrs and BM12hrs. 46 Figure 4.7: SDT-Q600 TGA Profiles for MgH2 + nanoNickel. 48 Figure 4.8: SDT-Q600 TGA Profiles for LiBH4/MgH2 + nanoNickel. 49 Figure 4.9: SDT-Q600 DS C Profiles for the Discussed Mixtures in Section 4.2. 50
vi Figure 4.10: SDT-Q600 TGA Profile s for LiBH4 + MgH2 + 2,4,6,8,10 mol% ZnCl2 Ball Milled for 2 Hours. 52 Figure 4.11: SDT-Q600 DSC Profiles for LiBH4 + MgH2 + 2,4,6,8,10 mol% ZnCl2 Ball Milled for 2 Hours. 53 Figure 4.12: Desorption Data Collected on a PCT for LiBH4 + MgH2 + 2mol% ZnCl2 Ball Milled 2 Hours Under H2 Ambient. Desorptions were Performed at Various Te mperatures: 1-3 cycles at 250oC; 4-6 Cycles at 300oC; 8-10 Cycles at 350oC. 55 Figure 4.13: PCT Desorption Plots at 250oC and 350oC for the Mixture LiBH4 + MgH2 + 2mol% ZnCl2 Ball Milled 2 Hours Under H2 Pressure. 57 Figure 4.14: XRD Profiles of LiBH4 + MgH2 + Xmol% ZnCl2 Ball Milled for 30 Minutes Under a H2 Gas Ambient. 58 Figure 4.15: XRD Profiles of LiBH4 + MgH2 + Xmol% TiCl3 Ball Milled for 30 Minutes Under a H2 Gas Ambient. 60 Figure 4.16: FT-IR Profile for LiBH4 + MgH2 + Xmol% ZnCl2 Ball Milled for 30 Minutes Under a H2 Gas Ambient. 61 Figure 4.17: FT-IR Profile for LiBH4 + MgH2 + Xmol% TiCl3 Ball Milled for 30 Minutes Under a H2 Gas Ambient. 62
vii DESTABILIZATION AND CHA RACTERIZATION OF LIBH4/MGH2 COMPLEX HYDRIDE FOR HYDROGEN STORAGE Luis A. Rivera ABSTRACT The demands on Hydrogen fuel based t echnologies is ever increasing for substitution or replacing fossil fuel due to supe rior energy sustainabil ity, national security and reduced greenhouse gas emissions. Curre ntly, the polymer based proton exchange membrane fuel cell (PEMFC), is strongly considered for on-board hydrogen storage vehicles due to low temperature operation, e fficiency and low environmental impact. However, the realization of PEMFC vehi cles must overcome the portable hydrogen storage barrier. DOE and FreedomCAR tec hnical hydrogen storage targets for the case of solid state hydrides are: (1) volumetric hydrogen density > 0.045 kgH2/L, (2) gravimetric hydrogen density > 6.0 wt%, (3) operating temperature < 150oC, (4) lifetimes of 1000 cycles, and (5) a fast rate of H2 absorption and desorption. To meet these targets, we have focu sed on lithium borohydride systems; an alkali metal complex hydride with a high theoretical hydrogen capacity of 18 wt.%. It has been shown by Vajo et al. that adding MgH2, improves the cycling capacity of LiBH4. The pressure-composition-isotherms of the destabilized LiBH4 + MgH2 system show an extended plateau pressure around 4-5 bars at 350C with a good cyclic stability. The mentioned destabilizing mechanism was successfu lly utilized to synthesize the complex
viii hydride mixture LiBH4 + MgH2 + Xmol% ZnCl2 catalyst (X=2, 4, 6, 8 and 10) by ball milling process. The added ZnCl2 exhibited some mild catalytic activity which resulted in a decomposition temperature reduction to 270C. X-ray powder diffraction profiles exhibit LiCl peaks whose in tensity increases proportionately with increasing ZnCl2 indicating an interaction between catalyst a nd hydride system, possibly affecting the total weight percent of desorbed hydrogen. Ther mal gravimetric analysis profiles for MgH2 + 5mol% nanoNi and LiBH4 + ZnCl2 + 3mol% nanoNi indicate that small concentrations of nano-nickel acts as an effectiv e catalyst that reduces the mixture desorption temperature to around 225oC and 88oC, respectively. Future work will be focused on thermodynamic equilibrium studies (PCT) on the destabilized complex hydrides.
1 CHAPTER 1 INTRODUCTION 1.1 Current Energy Situation Our current energy situation has been th e main motivator to pursue research in metal hydrides for on-board hydrogen storage. This situation has been created by several factors threatening to destabilize our economy, e nvironment and national energy security if not solved or at least tackled on time. A possible key factor is closely related to our oil dependency and its influence. However, there are multiple solutions that could reduce our oil dependency while at the same time re-stabilizing the economy, improving the environment and strengthening ou r national energy security. This section will briefly discuss how our oil dependency pl ays its role and possible alternatives. 1.1.1 Oil Market Note from the author: Most of the data presented in this section 1.1.1 Oil Market was taken from the Energy Information Admi nistration (EIA) which is the Official Energy Statistics from the United States government. For decades the world, mostly the United States, enjoyed the access to relatively inexpensive and abundant crude oil (fossil fuels) supply. In 2005, the United States energy consumption reached the 99.9 quadrillion Btu and a petroleum consumption of
2 21 million barrels per day. In addition, around 63% of the United States energy sources relied on fossil fuels to cover its demand in 2005 . According to EIA data , for the third quarter of 2006, the United States o il demand (around 20.80 millions barrels per day) has surpassed its oil supply (approx. 8.48 m illions barrels per da y) reaching deficits levels higher than 2 times its s upply. If the United States d ecides to use its own reserves, which are about 21.4 billions barr els (refer to Table 1.1) to supply its current demand, the reservoirs would be depleted in approximate ly 3 years, a possible reason to cover the deficit with foreign oil imports. Table 1.1: World Oil Reserves by Country as of January 1, 2006 . Country Oil Reserves (Billions Barrels) Saudi Arabia 264.3 Canada 178.8 Iran 132.5 Iraq 115.0 Kuwait 101.5 UAE 97.8 Venezuela 79.7 Russia 60.0 Libya 39.1 Nigeria 35.9 United States 21.4 China 18.3 Qatar 15.2 Mexico 12.9 Algeria 11.4 Brazil 11.2 Kazakhstan 9.0 Norway 7.7 Azerbaijan 7.0 India 5.8 Rest of World 68.1 World Total 1292.5
3 In a broader scope, a world supply of approximately 85.18 millions barrel per day barely suffices a total world consumption of approximately 84.22 m illions barrel per day . World reserves are estimated at 1292.5 b illions barrels (refer to Figure1.1) , if considering constant world consumption the reserves life-span would be around 42 years. Figure 1.1: World Oil Prices, 1980-2030. Comparison of IRAC (Imported Refiner Acquisition Cost) and Average Price of Impor ted Low-Sulfur, Light Crude Oil (ILSLCO) to U.S. Refiners . Estimated values for 2025, world oil c onsumption and total resources (the resources term counts: proved reserves, reserves growth and undiscovered reserves) were estimated at around 111 million barrels per day  and 2961.6 billion barrels , respectively. If by 2025, the estimate is re al with no other possible reserves and a constant consumption, the world oil resour ces would disappeared in approximately 73 years close to the year 2100.
4 Additionally, according to a report from the EIA , world oil prices have been dramatically increasing since 1995 and are proj ected to stabilize by 2015 (refer to Figure 1.1). The increase in crude oil prices has in creased U.S. retail gasoline prices which could reduce consumer purchase power and its accessibility (refer to Figure 1.2) . Figure 1.2: Price Comparison for a Gallon of Regular Grade Gasoline in 2004 and 2005 . Our dependency and the oil market have critically affected our national security and economy however, not limited to those men tioned. The constant use of fossil fuels has also affected our environment locally a nd globally. The combustion of fossil fuels produces carbon dioxide (CO2), a component directly linke d to the greenhouse gases and global warming. The next section will briefl y discuss the history and effect of carbon dioxide.
5 1.1.2 Global Warming and Greenhouse Effect Scientists tried to explain the relation ship between carbon dioxide effects of the global warming. The first steps to measure car bon dioxide levels in the atmosphere were still speculated until in 1960, when Keeling discovered that carbon di oxide levels in the atmosphere were increasing , (refer to Figure 1.3). Figure 1.3: Keeling Curve CO2 Measurement at Mauna Loa, Hawaii . However, the plot only indicates the increasing carbon diox ide levels in the atmosphere and not a direct relationship with temperature (greenhouse effect) and atmospheric carbon dioxide levels. Close to 1985, scientists studying Antarctic ice cores found a direct relationship of temperature with concentration of carbon dioxide levels for past ice age cycles (see Figure 1.4) .
6 Figure 1.4: CO2 Concentration in Parts Per Million Plotted Against Thousands of Years Before Present . This finding demonstrated that carbon di oxide takes an important role in our global climate. In addition, further investig ations have found that carbon dioxide levels are increasing . According to an EIA report, it has been projected that world carbon dioxide emissions increases from 21,223 m illion metric tons in 1990 to 43,676 million metric tons by 2030 . The continuous and increased use of carbon base energy sources, in these case fossil fuels, could fu rther catalyze the already affected global climate by the addition of more anthropogeni c carbon dioxide. Climate changes and limited oil availability threat to inflict worl dwide damage with irreversible consequences if solutions of a significant magnitude are not implemented. Several these solutions are non-carbon (fossil fuels) base power sources which are currently in use throughout the world, aiding the task reducing the use of fossil fuels however; these non-carbon base technologies also face limitations.
7 1.1.3 The Alternative Current technological advances are providi ng the world with the solution to our oil dependency. Technologies such as: nuclear energy, wind energy, hydropower and solar energy has been in use for several year s in many countries to supply the demand for energy, while reducing oil consumption and carbon dioxide generation. On the other hand, fully implementing these new technologies has proven to be a difficult task due to their intrinsic limitations , wh ich would not be discussed in this report. Recent global factors have motivated multiple countries, in cluding the United States, to consider PEM (Proton Exchange Membranes) fu el cells (see Figure 1.5) as future source of energy, in the what would be called the hydrogen economy due to their high efficiency, negligible air pollution and versatility [15, 16]. Despite the PEM positiv e qualities and the significant impact it will bri ng to the hydrogen economy, the hydrogen storage limitation must be solved in order to ach ieve the goal. In this report it is focused on the analysis and characterization of complex light we ight metal hydrides for on-board hydrogen storage. The main objectives of the research are based on the techni cal targets set by the US Department of Energy (DOE).
8 Figure 1.5: PEM Fuel Cell . The following chapter will provide a general overview of metal hydrides and the complex metal hydrides investigated for this report.
9 CHAPTER 2 INTRODUCTION TO COMPLEX METAL HYDRIDES 2.1 General Overview In a simple definition, metal hydrides are metallic elements (one or more metal elements) bonded with hydrogen. There are va rious types of metal hydrides which can be classified by the amount of different spec ies present in the molecule: binary, ternary and quaternary hydrides (re fer to Table 2.1). Table 2.1: Classificatio n of Metal Hydrides. Classification Example Binary  LiH, MgH2, etc. Ternary  LiBH4, Zn(BH4)2, etc. Quaternary LaMg2NiH7 , LiMg2RuH7 , etc. The general reaction for hydr ogen desorption (reaction 1) and absorption (reaction 2) for binary metal hydrides is as follows: ) ( 22g nH n M MH (1) n gMH H n M ) ( 22 (2)
10 Ternary and Quaternary metal hydrides fo llow complex reactions path which are specific for each compound. Due to different molecular combination as seen in Table 2.1, metal hydrides exhibit a wide arra y of characteristics and properties. Several of these materials are being i nvestigated as possible candidates for onboard hydrogen storage. An idea on how th e metal hydrides would be processed during the hydrogen economy is depicted in Fi gure 2.1 using sodium borohydride (NaBH4). Figure 2.1: Hypothetical Hydrogen Economy Using NaBH4 as Hydrogen Storage . There are key general requirements that the metal hydride candidate must meet before its selection : (1) Favorable thermodynamics. (2) Fast release and absorbing of hydrogen. (3) High gravimetric and volumetric hydrogen capacities. (4) Negligible change in hydrogen ab sorption/desorption storing capacity. The DOE (Department of Energy)  and FreedomCAR Hydrogen Storage Team  developed specific technical targets to consider a prospective compound as a hydrogen storage system. Partial data for the year 2015 is presented in Table 2.2.
11 Table 2.2: DOE and FreedomCAR Hydrogen St orage Team Technical Targets [23, 24]. Technical Targets Year 2015 Specific energy (MJ/kg) 10.8 Gravimetric capacity (kg H2/kg system) 0.09 Volumetric Capacity (kg H2/L system) 0.081 Energy density (MJ/liter) 9.72 System cost ($/kg system) 3 Operating temperature (oC) -20/50 Cycle life (cycles) 1500 Delivery pressure (bar) 2.5 Refueling rate (kg H2/min) 2.0 Several metal hydrides and compounds sati sfy some of the requirements on Table 2.2, but still lack in others. Investigations have been performed to store hydrogen in compressed tanks or as liquefied H2 , however application for on-board storage is limited due to low volumetric capacity and safe ty issues. Recent work has been done with methane reforming as a method to produce hydrogen. Unfortunate ly, this process is energy intensive . Metal hydrides compou nds located on the upper right hand side of the graph on Figure 2.2 are considered the best materials for hydrogen storage applications . Sodium aluminum hydride shows promise of filling the hydrogen storage technical needs however several investigations i ndicates a low hydrogen release at high temperatures and poor cyclic reversibil ity due to partial reactions [28 32]. For our research, doped and undoped mixtures of LiBH4 and MgH2 and additional ones (Zn(BH4)2 and dopants) were investig ated. The following sections will continue briefing about the background of the compounds and its different properties.
12 Figure 2.2: Comparison of Vo lumetric and Gravimetric H ydrogen Capacities for Some Metal Hydrides . 2.2 Lithium Borohydride (LiBH4) LiBH4 is one of the light weights and high hydr ogen storage capacity material. It has a theoretical gravimetric and volumet ric hydrogen storage capacities of 18.5 wt.% and 121 kgH2/m3, which surpasses the DOE and FreedomCAR targets (refer to Table 2.2). Initial investigations have demonstrated that the hyd rolysis reaction (3) [33-35] of lithium borohydride at room temperatures effectively releases hydrogen exothermically; however the process is irreversible with LiBO2 as byproduct. Hydrolysis reaction proceeds as follows: 2 2 2 44 2 H LiBO O H LiBH (3)
13 An alternate route that has been previously investigated is to thermally destabilize LiBH4. Lithium borohydride has a known melting point around 275oC  with the following thermal properties investigated by Fe dneva et al. , refer to Table 2.3: Table 2.3: Thermal Analysis of Lithium Borohydride Accord ing to Fedneva et al. . Temperature Range (oC) Description 108112 Endothermic peak. Structural transition. 268286 Fusion process with a slight weight loss. 380 Main weight loss due to H2 decomposition. 483492 Authors are not certain howe ver, it coincides with a weight loss. Recent thermal studies by Zuttel et al. [ 38, 39] and Orimo et al.  further expands the understanding regarding lithium borohydride behavior (see Table 2.4 and 2.5). Table 2.4: Thermal Studies by Zuttel et al. on Lithium Borohydride [38, 39]. Temperature Range (oC) Description 100 Structural transition with a slight weight loss. 270 Fusion phase. 320 First significant weight loss. 400500 Second significant weight loss. Table 2.5: Thermal Studies by Orim o et al. on Lithium Borohydride . Temperature Range (K) Description 380 (~107oC) Structural transition. 550 (~277oC) Melting phase. 600700 (~327427oC) Dehydriding reaction.
14 As an additional note, the structural transition of lithium borohydride has been found to be from orthorhombic  to hexagonal structure [42-44]. According to Stasinevich and Egorenko [ 45], the decomposition of alkali metal tetrahydroborides can proceed as reactions (4) or (5): 2 42 H B M MBH (4) 2 42 3 H B MH MBH (5) For lithium borohydride, the dehydriding reaction (6)  and rehydriding reaction (7)  can be generally described as: 2 42 3 H B LiH LiBH (6) 4 22 3 LiBH H B LiH (7) Theoretically the reaction (6) releases around 13.8wt% of hydrogen. According to Orimo et. al , the rehydrogenation of lithium borohydride was achieved however, the energy levels consumed during this pr ocess, 35Mpa of hydrogen and 873K, might not make it cost effective. Additional inves tigations performed with lithium borohydride, included the doping with SiO2 as a catalyst thus enab ling the decrease of LiBH4 dehydriding temperature to 300oC . A recent report indicates a dehydrogenationrehydrogenation cycle improvement and re ducing the reaction enthalpy of LiBH4 by the addition of MgH2 in a ratio of 2LiBH4 + MgH2 . The addition of MgH2 reversibly destabilizes LiBH4 which in consequence increases th e hydrogen equilibr ium pressure as reaction (8) :
15 2 2 2 42 2 1 2 1 H MgB LiH MgH LiBH (8) Part of the investigation was dedicate d to further understand the behavior of 2LiBH2 + MgH2. The report will discuss the signif icant results obtained while studying the system. The next section will br iefly introduce the compound used during the research MgH2. 2.3 Magnesium Hydride (MgH2) Magnesium hydride is a non-expensive binary hydride wi th a theoretical hydrogen capacity around 7.6wt% . It can effectivel y store hydrogen due to its thermodynamic stability however reaction ki netics are too slow and the decomposition temperature is high, approximately at 330oC . Different approaches to improve the reaction kinetics of MgH2 have been taken, including the mechanical milling. The mechanical milling of a hydrogen absorbing compound, in this case MgH2, under hydrogen pressure leads to hydrogen uptake, defects and changes in the surface [47-50] however the attention is more focused on the effective reduction of the desorption temperature by ball milling [51-55]. Experiments have taken place in order to catalytically improve MgH2 kinetics with Nb2O5 [56, 57]. More studies involve the addition of small amounts of Ti, V, Mn, Fe, Ni mechanically milled with MgH2 reporting good results [58-62]. A recent appr oach is the use of magnesium nickel alloys for hydrogen storage systems [63-65] the presence of nickel improves the hydriding and dehydridin g rates. The MgH2/nanoNickel system was investigated and the results presented in this repor t discussing the effects of nanoN ickel. In addition, mixture
16 of the system with dopants was also compared with Vajos et al.  in this document and will be discussed in Chapter 4. The next session will briefly introduce a studied compound Zn(BH4)2. 2.4 Zinc Borohydride (Zn(BH4)2) Zn(BH4)2 is a ternary complex metal borohydride with a decomposition temperature of around 85oC . Its theoretical hydroge n capacity is a bout 8.5wt% and it can be synthesized by metathesis reaction of NaBH4 and ZnCl2 in diethyl ether . A recent report from Eun Jeon et al.  i ndicates that zinc borohydride was successfully synthesized by ball milling zinc chloride and sodium borohydride without the use of a solvent, see reaction (9): NaCl BH Zn NaBH ZnCl 2 ) ( 22 4 4 2 (9) For this report, the po ssible formation of Zn(BH4)2 by ball milling LiBH4 and ZnCl2 would be explained, however additional investigations are needed to determine its presence.
17 CHAPTER 3 MATERIALS, EQUIPMEN TS AND APPROACH 3.1 Experimental Materials Table 3.1 provides a list of the materials used during th e investigation. These materials are used without further pu rification except drying inside the N2/vacuum glove box for O2 and moisture removal. Table 3.1: List of Materials. Name of material Purity Manufacturer ZnCl2 99.999% Sigma-Aldrich TiCl3 99.999% Sigma-Aldrich LiBH4 95% Alfa Aesar TiF3 99.999% Sigma-Aldrich nanoNi 99.99% QuantumSphere nanoNi 99.9% Sigma-Aldrich nanoZn 99+% Sigma-Aldrich MgH2 98% Alfa Aesar Polyethylene foil (thin foil) unknown Target Brands, Inc.
18 3.2 General Approach The general procedure followed for the inves tigation is represented in Figure 3.1. Figure 3.1: Procedure Followe d During the Investigation. The approach specified in Figure 3.1 would have a gene ral formula to distinguish if the mixture has not been altered with additional components. The formulas would be shown in the plot legend as follows: time BM C B A (10) time BM C B A (11) Determine mixture ratio of components Mix components inside N2 Glove Box Ball Milling Process N2 Glove Box Additional components? Yes No Characterization techniques
19 Formulas (10) and (11) indicates that components A, B and C were first mixed together and then ball milled (BM in the form ula is an abbreviation for ball milling) for a certain amount of time. However, if an a dditional component was added to the mixture and then ball milled again. The formulas would be presented in the plot legend as follows: time BM D time BM C B A ) ( (12) time BM D time BM C B A ) ( (13) Formulas (12) and (13) indicates that components A, B and C were first mixed together and then ball milled (BM in the form ula is an abbreviation for ball milling) for a certain amount of time, however an additi onal component D was added afterwards and then ball milled again. As an example, if the plot legend states: MgH2 BM9hrs + 10mol% nanoZn + BM2hrs. It indicates that MgH2 was previously ball milled for 9 hours then mixed with 10 mol% of nano-Zi nc and the new mixture ball milled for 2 hours. This method would enable the readers to understand the legend on each plot still, to avoid confusion, a full explanation as detailed before will be provided. 3.3 General Procedure All mixing, transfer and weighing was carri ed out in a nitrogen filled glove box to reduce the contact of oxygen and moisture with the samples. The mixtures were mechanically milled in a high energy Fritsch pulverisette planetary mono mill P6 using a stainless steel bowl (80 ml and 250ml) sealed with a specially de signed lid with two Schrader valves on opposite corners and a viton O-ring (refer to Figure 3.2).
20 Figure 3.2: Ball Milling Bowl and the Speci ally Designed Lid with Schrader Valves. A ball to powder weight ratio of 20:1 and a milling speed of 300 rpm were set to optimize the process over a varied range of milling times. Several milling durations were employed during the investigation ranging from 20 minutes to 2 hours. The ball milling time and procedure will be further discussed in details for each result in Chapter 4. Hydrogen flushing/purging was performed prior ball milling to every mixture in order to reduce the presence of oxygen and/or moisture inside the bowl duri ng the process. For ball milling times below 30 minutes the system was purged only once at the beginning, before starting the process however; for milling times higher than 30 minutes (i.e. 1 hour) the purging was performed every 30 minutes. After ball milling, the as-prepared mixture of complex hydrides were immediately tr ansferred to the glove box for further characterization analysis. Weight loss analyses were performed using a SDT-Q600 instrument from TA Instruments. For SDT-Q600 measurements the samples were loaded in alumina pans and set the general temperature ramp rate at around 5oC/min. All calibrations were performed
21 as per TA instruction manual. The Universal Analysis software V4.0C was employed to analyze the results obtained from both e quipments. An additional note, the to-bemeasure samples were handled inside the nitrogen filled glove box at all times except when purging with hydrogen. The measurement for isothermal volumet ric sorption was carried out by a HyEnergys PCT Sieverts type apparatus. Volume calibration was performed with and without the sample until a constant temperat ure with accuracy of 1 C was achieved. A Lab View software program was employed fo r data monitoring and recording. The measurement analyses were performed using Hy-Analysis macros in the Igor program. The powder X-ray diffraction (XRD) anal yses were carried out using a Philips Xpert diffractometer with CuK radiation of = 5.4060 As part of the Incident Beam Optics, a Fixed Divergence slit mo dule was used along with a 1o Fixed slit, a 10mm Beam mask and Soller slit of 0.04 rad. The accessories used for the Diffracted Beam Optics were: a Programmable Receiving Slit with fixed anti scatter slit, a 2o anti scatter slit, a monochromator and detector. Th e collected XRD patterns were analyzed employing the software PANalytical Xpert Hi ghscore software vers ion 1.0e for phase identification and crystalline size. A polyethyl ene clear plastic wrap (thin foil) was used to protect the samples from air and moistu re by wrapping the sample holder completely with the thin foil (see Figure 3.3). The thin foil shows diffraction peaks in the 2 range of 2128.
22 Figure 3.3: XRD Sample Holder Covered with Polyethylene Clear Plastic Wrap. The Perkin-Elmer one FT-IR spectrometer was used in order to study the B-H bonds or other possible chemical bonds in the 2LiBH4 / MgH2 mixtures, which would be discussed in Section 4.3. 3.4 Equipments 3.4.1 Nitrogen Filled Glove Box As a general definition, the glove box is a sealed container designed to maintain special atmospheric conditions while also a llowing the investigator to manipulate the objects or objects placed inside of it. Th e principal components of the glove box are (refer to Figure 3.4 and 3.5):
23 Figure 3.4: TA Instruments Diagram of Glove Box . (1) Gloves. (2) Vacuum pump. (3) Electronic sensors to monitor oxygen and moisture content. (4) Two antechambers or ports used to transf er materials in and out of the glove box; both ports can be opened from th e inside and outside the box. (5) The automatic gauge controls maintains the pressure inside the glove box, the system will increase or reduce pressure by injecting nitrogen or evacuating with vacuum pump, respectively according to the pr essure range set for operation. (6) The foot pedal, it allows the investigator to manually adjust the pressure inside the glove box. (7) The purification system removes and mainta ins low levels (less than 1ppm) of oxygen and moisture inside the gloves by continuously reprocessing (refer to Figure 3.6) the nitrogen through a molecular sieve of (Al2O3) alumina and copper catalyst.
24 Figure 3.5: Picture of Glove Box. Figure 3.6: Glove Box Sy stem Flow Diagram .
25 3.4.2 Ball Mill (BM) A high energy Fritsch pulverisette planetary mono mill P6 (see Figure 3.7) was used for mixing and synthesis of complex metal hydrides. Figure 3.7: Picture of Ball Mill Equipment . Other processes can be achieved using th e ball milling such as: particle size reduction and structural alterations of particles. The ball milling operation is as follows: the bowl is placed on the grinding platform (not e: the platform has a counter weight to minimize vibrations due to imbalances), secure d to the platform with the safety features and programmed to rotate either clockwise or counter clockwise. The grinding bowls rotational movement is opposite of the supporting disk movement (refer to Figure 3.8). The centrifugal force (up to 10 gs acceler ation) provides energy and motion to the grinding balls which are rolled halfway ar ound the bowl and then, due to centrifugal force, thrown across the bowl to impact the sample on the opposite side at high speed.
26 Figure 3.8: Cross-Sectiona l Diagram of the Planetary Ball Mill Movement . 3.4.3 Simultaneous DSC and TGA (SDT) The SDT measures heat flow (enthalpic ch anges) and weight changes related with transitions and reactions in materials. The equipment has the capability to assist in the differentiation of endothermic and exotherm ic processes with no weight loss (e.g., melting and crystallization) from those with weight changes (e.g., desorption). The operational temperature range for the mode l (SDT-Q600) ranges over the ambient to 1500oC. The furnace is a one piece alumin a sample tube surrounded by a platinum rhodium heater (refer to Fi gure 3.9) which can be set wi th heating rates of up to 100oC/min and 25oC/min for final temperatures of 1000oC and 1500oC, respectively. The balance beams are made of ceramic alumina w ith platinum liners platform at the furnace end. Platinum/Platinum-Rhodium thermocouples inside the ceramic beams from the platform to the meter mount provide the therma l measurements (refer to Figure 3.10). To prevent back-diffusion from the samples and contamination to the balance housing, the area is carefully purged with inert gas (in this case N2) (refer to Figure 3.11). The SDT-
27 Q600 was enclosed inside a glove box to redu ce samples exposure to air (refer to Figures 3.12 and 3.13). Figure 3.9: Cross-Sectional Diagram of SDT Furnace . Figure 3.10: SDT Balance Housing .
28 Figure 3.11: Full Cross-Secti onal Diagram of SDT-Q600 . The TGA portion of the instrument operate s as follows; the balance arms (ceramic beams) are maintained in a horizontal null pos ition while a position sensor sends an equal amount of light (supplied by a constant current infrared LED) to each of the photodiodes. If weight is lost or gained, the beam pos itions shifts causing an unequal amount of light to strike the photodiodes. The unequal amount of light is then transferred as a change in current which is proportional to the weight change. The DSC portion employs a single heat source (Platinum Rhodium heater) and two symmetrically located and identical sa mple platforms (ceramic beams) with thermocouples inside and along the beams lengt h. The platforms symmetry is necessary in order to uniformly apply heat. The anal ysis works converting the heat flow to the thermal equivalent of Ohms Law.
29 Figure 3.12: SDT-Q600 Photo. Figure 3.13: SDT-Q600 Inside an Inert Atmosphere Glove Box.
30 3.4.4 X-Ray Diffractometer (XRD) X-ray Powder Diffraction (XRD) is a technique which employs collimated monochromatic x-rays for char acterization and identification of crystalline structure. Additional uses include: qualitative and quantitative phase identification, identification of lattice parameters, thin film studies, etc. A Philips Xpert diffractometer with CuK radiation of = 5.4060 was employed for such purposes during the investigation. The cathode emits and accelerates th e electrons into the vacuum by high voltage while the anode collects them establishing a current flow The electrons impact the metal target (in this case copper) and produce X-rays, which are incident on the samp le. The diffraction takes place when X-rays are diffracted by th eir interaction with the atomic plane arrangement in the crystal. The basic principle of Bragg-Brentano geometry was used for the powder diffraction. According to Braggs Law, when X-rays are scattered from a crystal lattice constructively, not destructively (refer to Figure 3.14 and 3.15), peaks of scattered intensity are observed that satisfies the Bragg equation (13) given below: ) sin( 2 d n (13) Where is the angular position between the in cident and diffracted rays, d is the spacing between the planes of atom s lattice, n is an integer, is the wavelength of x-rays. A picture of the XRD equipment used as a ch aracterization tool is provided in Figure 3.16.
31 Figure 3.14: Constructive (Left Picture) or De structive (Right Pictur e) Interferences . Figure 3.15: Schematic of an X-ray Diffractometer .
32 Figure 3.16: Xpert Diffractometer Picture. 3.4.5 Pressure Composition Isotherms (PCI) Apparatus The PCI or PCT is an automated Sieverts type apparatus use to measure gas sorption properties of materials, in this case hydrogen sorption in metal hydrides. The model use during the investigation was the PCTPro-2000 by Hy-Energy LLC., USA. The model offers several features for measurements such as: PCI (Pressure Composition Isotherms), Gas Sorption Kinetic s, Heat of Formation, Cycl e-Life Kinetics, Cycle-Life PCT, Volume Calibration and Packing Density measurements . There are important preparation processes, where no data is coll ected, that must be performed prior the experiments in order to prevent l eaks, contamination, damage, etc.: (1) Purge Gas Lines when a new gas or bottle has been connected. (2) Purge System & Samples when a connection with a possible contaminant gas has been made. (3) Leak Check for every experiment.
33 As a general basic PCT operation, the reac tor with the sample is connected and a leak check is performed using Helium (99.9999% pure). Afterwards, the reactor volume is calibrated at constant temperature a nd to a set measured pressure. Hydrogen absorption is calculated by measuring the pres sure difference. Arrhenius plots can be obtain using the kinetic feature option by increa sing the temperature in intervals to obtain Log (sorption rate) vs. 1000/te mperature. In addition, va nt Hoff plot represents the thermodynamic equilibrium of the gas sorption process. The PCT feature is used to obtain the vant Hoff measurements. The enth alpy of formation, which is represented by the slope in the vant Hoff plot, can be calculated by the va nt Hoff equation: R S T R H Peq 1 ) ln( (14) The temperature is held constant while gas (pressure) is applied to the sample in aliquots, for each aliquot the system is allowe d to reach equilibrium. The process can be repeated until desired or until operational limits allows. If hydrogen is absorbed or desorbed by the sample, a plateau forms (s ee Figure 3.17) indicating hydrogen-sample interaction at that specific temperature and pressure. A picture of the model and monitor indicator can be observed in Figure 3.18 and 3.19.
34 Figure 3.17: PCT Diagram (Left) Associated with the Vant Hoff Plot (Right) . Figure 3.18: PCTPro-2000 Hydroge n Sorption Apparatus Picture.
35 Figure 3.19: PCTPro-2000 Manifold Monitor Indicator. 3.4.6 Fourier Transform Infra red Spectrometer (FT-IR) Fourier Transform Infrared (FT-IR) spectroscopy is an an alytical technique used to measure infrared intensity against wavelength of light. However, multiple regions are found in the infrared portion of the electromagnetic spectrum which is divided into three; the near(14000-4000 cm-1), mid(approx. 4000-400 cm-1) and farinfrared (approx. 400-10 cm-1) named for their relation to the visibl e spectrum. The Infrared spectroscopy is useful because chemical bonds interact wi th the matter by stretc hing, contracting and bending the chemical bonds at different energy levels. The far-infrared, has low energy and could be used for studies of rotational sp ectroscopy, the midinfrared for associated rotational-vibrational struct ure and nearinfrared for harmonic vibrations. In order to measure a sample, a beam of infrared light is passed through an interferometer creating constr uctive and destructive patterns of light beams which are then passed through the sample. By absorbi ng an amount of energy at each wavelength, a spectrum is created and recorded. From th is transmittance or absorbance spectrum may
36 be plotted showing at which wavelengths the sample absorbs the IR and as consequence, the chemicals bonds present. A Fourier transform spectrometer is a Mi chelson interferometer with a movable mirror. In its simplest form, a Fourier transform spectrometer might look like Figure 3.20. Figure 3.20: Basic Diagram fo r an FT-IR Spectrometer .
37 CHAPTER 4 RESULTS AND DISCUSSION 4.1 Undoped and Doped LiBH4 + ZnCl2 Figures 4.1 and 4.2 show the different thermal and weight loss profiles for undoped and doped (with MgH2, TiF3 and nanoNi) 2LiBH4 + ZnCl2, including an undoped 10 minutes hand mixed sample. The mixture of LiBH4 / ZnCl2 in a ratio of 2:1, was prepared by ball milling under hydrogen pr essure. As a clar ifying point to the reader, all LiBH4 / ZnCl2 mixtures, with the exception of the 10 minute hand mixed sample, were ball milled for 20 minutes. The profiles indicates the presence of a complex metal hydride, probably Zn(BH4)2 . Zinc borohydride Zn(BH4)2 has a theoretical hydrogen capacity of 8.5wt% and decomposition temperature around 1000C. It could be formed by ball milled mixture of 2LiBH4 + ZnCl2 however, further characterization is needed. Comparing both profiles in Figures 4. 1 and 4.2, the samples of 2LiBH4 + ZnCl2 doped with nanoNickel (QuantumSphere) s howed better kinetic s and thermodynamic effects at lower temperatures when comp ared to the undoped sample. The catalytic effectiveness of the mixture doped with 3mol% nanoNickel surpasses the other doped and undoped mixtures by reducing the desorption temperature from 1140C to 100oC. The measured weight loss for most mixtures, exce pt the hand mixed sample, is much higher
38 than the expected 8.5wt%, indicating a probable decompos ition of additional compounds besides of hydrogen. Gas analyses should be performed to determine gas composition. No significant effects were observed for TiF3 and MgH2 doped samples. A hand mixed sample was prepared usi ng a ceramic mortar and stirring for 10 minutes; the SDT-TGA plot s hows a weight loss of at least 9.0wt% at around 1250C. The small weight loss compared to other graphs might be due to an incomplete reaction of 2LiBH4 + ZnCl2. The SDT-DSC diagram indicates the presence of two endothermic peaks with onset at around 1120C and 1370C which could be associated with the remaining LiBH4 structural transition and decomposition of Zn(BH4)2, corroborating the incompleteness of the hand milled reaction. Table 4.1 represents the SDT-TGA analysis results for the undoped and doped 2LiBH4 + ZnCl2.
39 Table 4.1: Thermogravimetric An alysis of Undoped and Doped 2LiBH4 + ZnCl2. Sample Name On-set Temperature (C) Peak Temperature (C) Total weight loss (%) 2LiBH4 + ZnCl2 114.07 125.06 14.80 2LiBH4 + ZnCl2 + 1mol% TiF3 114.13 129.21 14.92 2LiBH4 + ZnCl2 + 2mol% TiF3 112.57 128.00 14.18 2LiBH4 + ZnCl2 + 2mol% MgH2 113.96 128.11 13.70 2LiBH4 + ZnCl2 Handmilled 138.69 150.50 9.435 2LiBH4 + ZnCl2 + 2mol% nanoNi 110.38 113.00 14.73 2LiBH4 + ZnCl2 + 1mol% nanoNi 114.38 116.90 13.54 2LiBH4 + ZnCl2 + 3mol% nanoNi 106.05 107.30 14.82 2LiBH4 + ZnCl2 + 4mol% nanoNi 106.65 109.31 12.67
40 Fi g ure 4.1: SDTQ 600 TGA Profiles for Do p ed and Undo p ed LiBH 4 + ZnCl 2
41 Fi g ure 4.2: SDTQ 600 DSC Profiles for Do p ed and Undo p ed LiBH 4 + ZnCl 2
42 2025303540455055606570 2 degreesIntensity (counts)X = 0; BM 20min X = 0; Handmilled 10min X = 2mol%MgH2; BM 20min X = 1mol%TiF3; BM 20min X = 2mol%TiF3; BM 20min X = 1mol%Ni; BM 20min X = 2mol%Ni; BM 20min* thin foil + LiBH4 MgH2 LiCl ZnCl2* + + + XRD of 2LiBH4 + ZnCl2 + Xmol%catalyst BM20min w H2 * *+ + Fi g ure 4.3: XRD Profiles for Do p ed and Undo p ed 2LiBH 4 + ZnCl 2 Includin g a Hand Crushed Mixture.
43 Figure 4.3 shows the XRD profiles for most SDT analyzed mixtures. The peaks locations were obtained and matched with: LiBH4 [38-40, 46], MgH2 [65, 78-81], LiCl [46, 82], thin foil (refer to Figure 4.4) and ZnCl2 (refer to Figure 4.5 and PANalytical Xpert Highscore software version 1.0e reference code 00-016-0850). 202530354045505560 2 degreesIntensity (counts ) (101) (103) (004) (112) (200) (105)(211) SS holder Thin Foil TiO2 wrapped in Thin FoilThin Foil Thin Foil Thin Foil Figure 4.4: XRD Profile for the Polyethylen e Clear Plastic Wrap (Thin Foil) Used to Protect the Samples. It can be corroborated the incomplete reaction for the 10 minute hand mixed sample with the peaks of the reactants present in the mixtur e. While comparing the hand mixed and ball milled samples analyses, two unknown peaks at around 200 and 20.50 indicate a possible relationship in the formation of Zn(BH4)2 in addition to the LiCl peaks observed indicating a LiBH4 reaction with ZnCl2, however further characterization is needed (see also section 4.3, 2LiBH4 + MgH2 + Xmol% ZnCl2 for comparison). The
44 presence of LiCl affects the total hydroge n capacity of the mixture by adding a dead weight which is measured by the SDT. No peaks for TiF3, MgH2 and Ni were observed possibly due to their low concentration. Figure 4.5: XRD Profile of Pure ZnCl2. XRD profile of ZnCl220 25 30 35 40 45 50 55 60 65 70 2?, degrees Intensity (counts) Thin Foil = TF TF TF (101) (102) (103) (112) (114) (105) (006) (202) (211) (106) (212) (213) (116) (107)
45 4.2 MgH2 + nanoNi As part of our investigation, several MgH2 + nanoX [X=Ni (QuantumSphere) or Zn] mixtures were prepared to study the e ffects of different nano-metal and dopants for the following mixtures: (1) 2LiBH4 + (MgH2BM9hrs + 10mol% nanoNi + BM2hr) + BM1hr. (2) 2LiBH4 + MgH2 + BM1hr. (3) MgH2 + 10mol% nanoNi + BM2hrs. (4) MgH2BM12hrs. (5) MgH2BM9hr + 10mol% nanoZn + BM2hr. (6) (2LiBH4 + MgH2BM12hr + BM1hr) + 10mol% nanoNi + BM1hr. (7) MgH2 BM9hrs. (8) MgH2BM12hr + 5mol% nanoNi + BM1hr. In Figure 4.6 and 4.9, portions of MgH2 were ball milled before mixing to lower the desorption temperature and study if fu rther improvements could be achieved by catalytic doping. Previous studies [51, 78, 83, 84] have shown that ball milling reduces MgH2 desorption temperature. The undoped MgH2 the sample ball milled for 12 hrs showed a slight reduction in the d ecomposition temperature at around 325oC when contrasted with the MgH2 sample ball milled 9 hrs, both results show lower desorption temperatures than the as-received MgH2 around 4280C .
46 Figure 4.6: SDT-Q600 TGA Profiles for Undoped MgH2 + BM9hrs and BM12hrs.
47 In Figure 4.7 and Figure 4.9, the profiles fo r mixtures (3), (5) and (8) indicate that samples doped with nanoNi showed a higher effectiveness in reducing the desorption temperature starting at around 225oC when compared with samples doped with nanoZn. The weight loss for doped and undoped MgH2 samples fluctuated at around 4wt% and 5wt% respectively. On the other hand, in Figure 4.8 and 4.9, the sample profiles for (1), (2) and (6) show desorption temperatures around 275oC, 350oC and 270oC, respectively. As observed, 2LiBH4 + MgH2 samples doped with nanoNickel exhibits lower dehydrogenation temperatures compared to th e undoped counterpart. In addition, total weight losses for the sample profiles (1), (2) and (6) were observed around 3.50wt%, 3.20wt% and 4.66wt%, respectively. The additi on of nanoNi influenced the desorption temperatures of MgH2 by reducing it. Further st udies which involve doping LiBH4 with nanoNickel will be carried out.
48 Figure 4.7: SDT-Q600 TGA Profiles for MgH2 + nanoNickel.
49 Figure 4.8: SDT-Q600 TGA Profiles for LiBH4/MgH2 + nanoNickel.
50 Figure 4.9: SDT-Q600 DSC Profiles for the Discussed Mixtures in Section 4.2.
51 4.3 LiBH4 + MgH2 + Xmol% (ZnCl2 or TiCl3) Figure 4.10 shows different SDT-TGA results for the pristine 95% LiBH4 and mixtures of LiBH4 + MgH2 + Xmol% ZnCl2 (X = 2,4,6,8,10) ball milled for 2 hours under H2 pressure. A slight weight loss of around 1.2% during the melting process of non-ball milled LiBH4 between the peaks 275-300oC was observed. A significant exothermic peak around 75oC with weight loss since the be ginning of the analysis  indicates a possible hydrolysis [33, 35] of LiBH4 due to moisture. A total weight loss around 13.6wt% was observed fo r commercial undoped LiBH4. In addition, a slight shift in the melting peak from approximately 283oC to 270oC was observed when contrasting DSC profiles (ref er to Figure 4.11) to the non-ball milled LiBH4. This effect could be more associated with the concentration of ZnCl2 than with the ball milling time and/or the addition of MgH2.
52 Figure 4.10: SDT-Q600 TGA Profiles for LiBH4 + MgH2 + 2,4,6,8,10 mol% ZnCl2 Ball Milled for 2 Hours.
53 Figure 4.11: SDT-Q600 DSC Profiles for LiBH4 + MgH2 + 2,4,6,8,10 mol% ZnCl2 Ball Milled for 2 Hours.
54 According to Figure 4.12, the desorption rate of LiBH4 + MgH2 + 2mol% ZnCl2 ball milled for 2 hours increases with incr easing temperature. Desorption rate at 350oC is around 5 times faster than the desorption rate at 300oC. The dramatic desorption difference when comparing the plots for 350oC and 300oC indicates a possible range to locate the significant desorption process, however these high temperatures are not within the DOE FreedomCAR technical targets. An additional observation, assuming the reaction products are: LiH + MgB2 + 2H2, the H2 total weight percent in the product mixture will decrease with increasing concentration of ZnCl2. Assuming the later product mixture, for the following ZnCl2 concentrations an approximate theoretical total H2 weight percentage would be obtained (refer to table 4.2). These weight percentages include the weight of ZnCl2 since it is not eliminated from the mixture. Additional analysis must be perfor med to the reaction products to determine gas composition. Table 4.2: Reduction Effect in the Theoretical Total H2 wt% by ZnCl2 Addition. Mol% ZnCl2 Total H2 wt% 2 10.34 4 9.359 6 8.549 8 7.867 10 7.286
55 Figure 4.12: Desorption Data Collected on a PCT for LiBH4 + MgH2 + 2mol% ZnCl2 Ball Milled 2 Hours Under H2 Ambient. Desorptions were Performed at Various Temperatures: 1-3 Cycles at 250oC; 4-6 Cycles at 300oC; 8-10 Cycles at 350oC.
56 The PCT diagram (Figure 4.13), shows tw o different desorption curves at 250oC and 350oC for LiBH4 + MgH2 + 2mol% ZnCl2 ball milled 2 hours under H2 pressure. There is no appearance of plateau pressure region for the 1st desorption cycle curve at 250oC which is an indication of no hydrogen being absorbed by the mixture at that temperature. As for the 11th desorption cycle curve at 350oC, a plateau pressure at around 4-5 bars was observed with a total desorp tion of 1.4 wt.%. It was proven the MgH2 effects in reducing the react ion enthalpy by 25 kJ/(mol H2) by destabilizing LiBH4 . The XRD profiles shown in Figure 4.14 correspond to the mixture of LiBH4 + MgH2 + Xmol% ZnCl2 ball milled under an ambient of H2 gas for 30 minutes. The presence of LiCl and Zn  after ball milling for 30 minutes indicates a reaction between ZnCl2 and LiBH4 taking place while the mixture is being pulverized. As stated before in section 4.1, the presence of Li Cl affects the overall hydrogen weight loss by adding its dead-weight to the mi xture, unless it is removed by purification. In addition, the peak corresponding to LiCl increases with increasing con centration of ZnCl2 while at the same time the relative intensit ies of peaks corresponding to MgH2 and LiBH4 decreases. No presence of MgB2 was found.
57 Figure 4.13: PCT Desorption Plots at 250oC and 350oC for the Mixture LiBH4 + MgH2 + 2mol% ZnCl2 Ball Milled 2 Hours Under H2 Pressure.
58 2025303540455055606570 2 degreesIntensit y ( counts ) X = 0 LiBH4+0.5MgH2+Xmol%ZnCl2 BM30min w H2+ +* thin foil + LiBH4 MgH2 LiCl Zn* + + + + X = 2 X = 4 X = 6 X = 8 X = 10 Figure 4.14: XRD Profiles of LiBH4 + MgH2 + Xmol% ZnCl2 Ball Milled for 30 Minutes Under a H2 Gas Ambient.
59 The XRD profiles shown on Figure 4.15 correspond to the mixture of LiBH4 + MgH2 + Xmol% TiCl3 ball milled under an ambient of H2 gas for 30 minutes. The presence of LiCl after ball milling for 30 minutes indicates a reaction between TiCl3 and LiBH4, possible reaction products could be Ti(BH4)3  and/or Ti(BH4)4 (see reference in ), however further analyses are needed Besides, the peak corresponding to LiCl increases with increasing concentration of TiCl3 while at the same time the peaks corresponding to MgH2 and LiBH4 decreases and no presence of MgB2 was found. The FT-IR profiles in Figures 4.16 and 4.17 indicate the B-H stretching for LiBH4 at around 2276 and 2213 cm-1 in addition the deformation bands for BH2 observed at 1118 and 1091 cm-1 [89, 90]. There were no other sign ificant peaks that could indicate the presence of additiona l mixing materials or form ation of new compounds.
60 XRD LiBH4+0.5MgH2+Xmol%TiCl3 BM30min w H2 gas2025303540455055606570 2 degreesIntensit y ( counts ) X=1 X=2 X=3 X=4+ + + + +* thin foil + LiBH4 MgH2 LiCl Figure 4.15: XRD Profiles of LiBH4 + MgH2 + Xmol% TiCl3 Ball Milled for 30 Minutes Under a H2 Gas Ambient.
61 450950145019502450295034503950 wavelength (cm-1)TransmittanceLiBH4MgH2LiBH4+0.5MgH22mol%ZnCl2FT-IR LiBH4 + 0.5MgH2 + Xmol%ZnCl24mol%ZnCl26mol%ZnCl28mol%ZnCl210mol%ZnCl B-H stretch B-H deform Figure 4.16: FT-IR Profile for LiBH4 + MgH2 + Xmol% ZnCl2 Ball Milled for 30 Minutes Under a H2 Gas Ambient.
62 FT-IR of LiBH4+0.5MgH2+Xmol%TiCl3 BM30min w H2 gas 450950145019502450295034503950 wavelenght (cm-1)TransmittanceX= X= X=3 X=4 B-H deforms B-H stretch Figure 4.17: FT-IR Profile for LiBH4 + MgH2 + Xmol% TiCl3 Ball Milled for 30 Minutes Under a H2 Gas Ambient.
63 CHAPTER 5 CONCLUSION AND RECOMMENDATIONS 5.1 Conclusion and Recommendations In summary, the undoped ball milled mixture of LiBH4 / ZnCl2 successfully reduced the desorption temperature close to 100oC. Doping with nano Nickel demonstrated to further decrease the dehydr ogenation temperature. In addition, it was found that LiBH4 + ZnCl2 + 3mol% nanoNi is the optimum mixture for maximum performance. The XRD profiles show LiCl peaks indicating a reaction between LiBH4 and ZnCl2. The weight losses were higher than 10wt%, if assumed that Zn(BH4)2 is formed during LiBH4 / ZnCl2 ball milling, then, the probabilities of additional compounds being released other than hydrogen ar e high. It is recommended, for this system, to investigate if ther e is the presence of Zn(BH4)2 and to determine gas composition during desorption. The ball milling process has shown to reduce the desorption temperature of MgH2. The addition of nanoNickel to MgH2 has further reduced the dehydrogenation temperature. If adding nanoNickel to the LiBH4 / MgH2 mixture, the system destabilizes which is indicated by a decrease in the decomposition temperature. However, it is recommended to perform gas composition and thermal equilibrium studies to understand the system.
64 The mixture of LiBH4 / MgH2 was successfully prepared by doping with different amounts of ZnCl2 and TiCl3 catalysts. DSC and TGA analysis shows a lower decomposition/melting temperature of LiBH4/MgH2 mixture in the presence of ZnCl2. The initial rate of hydrogen decomposition from LiBH4 + MgH2 + 2mol% ZnCl2 increases with an increase in temperature. Mo reover, the plateau pressure of 4-5 bars at 350oC indicates a lower energy level for dehydr ogenation-rehydrogenation cycling with a volumetric capacity of 3.0wt%, such capacity increases with cycling. The XRD profiles indicate a reaction occurr ing during the mechano-chemical mixing with ZnCl2 and TiCl3 showing the presence of LiCl as product. It is recommended to focus the investigat ion in identifying the possible materials formed besides of LiCl during ball m illing and gas composition during desorption process. Identifying the gas composition during desorption might provide useful information regarding the reac tion path and products. Furt her experimental analysis using pressure-composition-isotherms should al so be highly considered to identify reversibility.
65 REFERENCES 1. Energy Information Administration. September 2005. Table 1.3, Energy Consumption by Source, 1949-2005. 2005 Annual Energy Review. DOE/EIA0384(2005). Washington, DC. ( www.eia.doe.gov/emeu/aer ). 2. http://www.eia.doe.gov/emeu/ipsr/t21.xls Found on: February 3, 2007. 3. Worldwide Look at Reserves and Producti on, Oil & Gas Journal, Vol. 103, No. 47 (December 19, 2005), pp. 24-25. 4. Energy Information Administration, In ternational Energy Outlook 2006, DOE/EIA0484(2006) (Washington, DC, June 2006), p. 26. 5. Energy Information Administration, In ternational Energy Outlook 2006, DOE/EIA0484(2006) (Washington, DC, June 2006), p. 29. 6. Energy Information Administration, In ternational Energy Outlook 2006, DOE/EIA0484(2006) (Washington, DC, June 2006), p. 26. 7. Energy Information Administration, A Primer on Gasoline Prices DOE/EIA-X040 May 2006. ( www.eia.doe.gov ). 8. Keeling, C. D. Tellus 1960, 12 200-203. 9. Wilson, Carroll L., and William H. Matth ews, Eds. Inadvertent Climate Modification. Report of Conference, St udy of Man's Impact on Climate (SMIC), Stockholm. Cambridge, MA: MIT Press. p. 234 (1971). 10. Lorius, C.; Jouzel, J.; Ritz, C.; Merlivat, L.; Barkov, N. I.; Korotkevich, Y. S.; Kotlyakov, V. M. Nature 1985, 316 591-596. 11. Barnola, J. M.; Raynaud, D.; Korotkevich, Y. S.; Lorius, C. Nature 1987, 329 408414. 12. Genthon, G.; Barnola, J. M.; Raynaud, D.; Lorius, C.; Jouzel, J.; Barkov, N. I.; Korotkevich, Y. S.; Kotlyakov, V. M. Nature 1987, 329 414-418. 13. Energy Information Administration, In ternational Energy Outlook 2006, DOE/EIA0484(2006) (Washington, DC, June 2006), p. 73.
66 14. Dresselhaus, M. S.; Thomas, I. L. Nature 2001, 414 332-337. 15. Crabtree, G. W.; Dresselhaus, M. S.; Buchanan, M. V. Physics Today 2004, 57 3944. 16. Schlapbach, L.; Zuttel, A. Nature 2001, 414 353-358. 17. http://www.eere.energy.gov/hydrogenandfue lcells/fuelcells/fc_types.html 18. Grochala, W.; Edwards, P. Chem. Rev. 2004, 104 1283-1315. 19. Renaudin, G.; Gune, L.; Yvon, K. J. Alloys Compd 2003, 350 145-150. 20. Huang, B.; Yvon, K.; Fischer, P. J. Alloys Compd 1994, 210 243-246. 21. US Department of Energy, Basic Research Needs for the Hydrogen Economy: Report of the Basic Energy Sciences Worksh op on Hydrogen Production, Storage and Use May 13-15, 2003. p. 35. 22. US Department of Energy, Basic Research Needs for the Hydrogen Economy: Report of the Basic Energy Sciences Worksh op on Hydrogen Production, Storage and Use May 13-15, 2003. p. 31. 23. http://www1.eere.energy.gov/ hydrogenandfuelcells/stora ge/pdfs/targets_onboard_hyd ro_storage.pdf 24. U.S. Department of Energy, Hydrogen Storage Think Tank Report ; March 14th, 2003; Washington, D.C. 25. Dillon, A. C.; Jones, K. M.; Bekkedahl, T. A.; Kiang, C. H.; Bethune, D. S.; Heben, M. J. Nature 1997, 386 377-379. 26. Wang, S.; Li, G. Q.; Millar, G. J. Energy & Fuels 1996, 10 896-904. 27. US Department of Energy, Basic Research Needs for the Hydrogen Economy: Report of the Basic Energy Sciences Worksh op on Hydrogen Production, Storage and Use May 13-15, 2003. p. 41. 28. Bogdanovic, B.; Schwickardi, M. J. Alloys Compd 1997, 253-254 1-9. 29. Zaluska, A.; Zaluski, L.; Strom-Olsen, J. O. J. Alloys Compd 2000, 298 125-134. 30. Fichtner, M.; Engel, J.; Fuhr, O.; Kircher, O.; Rubner, O. Materials Science & Engineering, B: Solid-State Ma terials for Advanced Technology 2004, B108 42-47.
67 31. Srinivasan, S. S.; Brinks, H. W.; Hauback, B. C.; Sun, D.; Jensen, C. M. J. Alloys Compd. 2004, 377 283-289. 32. Sun, D.; Srinivasan, S. S.; Chen, G.; Jensen, C. M. J. Alloys Compd 2004, 373 265269. 33. Schlesinger, H. I.; Brown, H. C.; Finholt, A. E.; Gilbreath, J. R.; Hoekstra, H. R.; Hyde, E. K. J. Am. Chem. Soc 1953, 75 215-219. 34. Kong, V. C. Y.; Foulkes, F. R.; Kirk, D. W.; Hinatsu, J. T. Int. J. Hydrogen Energy 1999, 24 665-675. 35. Kojima, Y.; Kawai, Y.; Kimbara, M.; Nakanishi, H.; Matsumoto, S. Int. J. Hydrogen Energy 2004, 29 1213-1217. 36. Schlesinger, H. I.; Brown, H. C. J. Am. Chem. Soc 1940, 62 3429-35. 37. Fedneva, E. M.; Alpatova, V. L.; Mikheeva, V. I. Russ. J. Inorg. Chem 1964, 9 826827. 38. Zuttel, A.; Wenger, P.; Rentsch, S.; S udan, P.; Mauron, Ph.; Emmenegger, Ch. J. Power Sources 2003, 118 1-7. 39. Zuttel, A.; Rentsch, S.; Fischer, P.; Wenger, P.; Sudan, P.; Mauron, Ph.; Emmenegger, Ch. J. Alloys Compd 2003, 356-357 515-520. 40. Orimo, S.; Nakamori, Y.; Kitahara, G.; Miwa, K.; Ohba, N.; Towata, S.; Zuttel, A. J. Alloys Compd 2005, 404-406 427-430. 41. Harris, P. M.; Meibohm, E. B. J. Am. Chem. Soc 1947, 69 1231. 42. Gomes, S.; Hagemann, H.; Yvon, K. J. Alloys Compd 2002, 346 206. 43. Souli, J.-Ph.; Renaudin, G.; Cern, R.; Yvon, K. J. Alloys Compd 2002, 346 200. 44. Hagemann, H.; Gomes, S.; Renaudin, G.; Yvon, K. J. Alloys Compd 2004, 363 129. 45. Stasinevich, D. S.; Egorenko, G. A. Russ. J. Inorg. Chem 1968, 13 341-343. 46. Vajo, J. J.; Skeith, S. L.; Mertens, F. J. Phys. Chem. B 2005, 109 3719-3722. 47. Chen, Y.; Williams, J. S. J. Alloys Compd 1995, 217 181. 48. Tessier, P.; Enoki, H.; Bououdina, M.; Akida, E. J. Alloys Compd 1998, 268 285.
68 49. Huot, J.; Akida, E.; Takada, T. J. Alloys Compd 1995, 231 815. 50. Zaluska, A.; Zaluski, L.; Strm-Olsen, J. O. J. Alloys Compd 1999, 288 217. 51. Gennari, F. C.; Castro, F. J.; Urretavizcaya, G. J. Alloys Compd 2001, 321 46-53. 52. Zaluski, L.; Zaluska, A.; Tessier, P.; Stroem-Olsen, J. O.; Schulz, R. J. Alloys Compd 1995, 227 53-57. 53. Liang, G.; Boily, S.; Huot, J.; Van Neste, A.; Schulz, R. J. Alloys Compd 1998, 268 302-307. 54. Liang, G.; Boily, S.; Huot, J.; Van Neste, A.; Schulz, R. J. Alloys Compd 1998, 267 302-306. 55. Gross, K. J.; Spatz, P.; Zuttel, A.; Schlapbach, L. J. Alloys Compd 1996, 240 206213. 56. Barkhordarian, G.; Klassen, T.; Bormann, R. Scripta Mater 2003, 49 213. 57. Barkhordarian, G.; Klassen, T.; Bormann, R. J. Alloys Compd 2004, 364 242. 58. Liang, G.; Huot, J.; Boily, S.; Van Neste, A.; Schulz, R. J. Alloys Compd 1999, 291 295-299. 59. Liang, G.; Huot, J.; Boily, S.; Van Neste, A.; Schulz, R. J. Alloys Compd 1999, 292 247-252. 60. Liang, G.; Huot, J.; Boily, S.; Schulz, R. J. Alloys Compd 2000, 305 239-245. 61. Dehouche, Z.; Djaozandry, R.; Huot, J.; Boily, S.; Goyette, J.; Bose, T. K.; Schulz, R. J. Alloys Compd 2000, 305 264-271. 62. Huot, J.; Pelletier, J. F.; Liang, G.; Sutton, M.; Schulz, R. J. Alloys Compd 2002, 330-332 727-731. 63. Reilly, J. J.; Sandrock, G. D. Scientific American 1980, 242 98. 64. Douglass, D. L. Metall. Trans. A 1975, 6A 2179. 65. Cooper, D.; Wu, C. Y.; Yasensky, D.; Butt, D.; Cai, M. KONA 2005, 23 139-151. 66. Marks, T. J.; Kolb, J. R. Chem. Rev 1977, 77 263. 67. Jeon, E.; Cho, Y.W. J. Alloys Compd 2006, 422 273-275.
69 68. http://www.gloveboxes.com/glove-box-systems.htm 69. Diagram found inside TA Instruments System One User Manual, p. 33. 70. http://www.fritsch.de/entry.php?pag e=messe&ulang=2&uid=45f75a6dc3f32cecdb50 d7845143c1f7e2833c99278e794f 71. http://www.fritsch.de/data/manuals/ba_062000_1336_e.pdf Operating Instructions Planetary Mono Mill pulverisette 6. 72. Diagrams found on the SDT online help so ftware which is included with the SDT Q600 model. 73. http://www.3me.tudelft.nl/liv e/binaries/5f016600-69e9-4978-bbd3173bc0136f91/img/TGA%202.jpg 74. http://en.wikipedia.org/ wiki/Image:Loi_de_bragg.png 75. http://pubs.usgs.gov/of/2001/ of01-041/htmldocs/xrpd.htm 76. PCTPro-2000 Installation and Oper ation Manual, Hy-Energy LLC. 77. http://ocw.mit.edu/NR/rdonlyres/Che mistry/5-32Intermediate-ChemicalExperimentationSpring2 003/CA6722BA-7333-4A99-B4AB1F7B880080C6/0/Appendix_1_Qual_Instrumentation_03.pdf 78. Shang, C. X.; Guo, Z. X. J. Power Sources 2004, 129 73-80. 79. Hanada, N.; Ichikawa, T.; Fujii, H. J. Phys. Chem. B 2005, 109 7188-7194. 80. Huang, Z. G.; Guo, Z. P.; Calka, A.; Wexler, D.; Lukey, C.; Liu, H. K. J. Alloys Compd 2006, 422 299-304. 81. Jensen, T. R.; Andreasen, A.; Vegge, T.; Andreasen, J. W.; Stahl, K.; Pedersen, A. S.; Nielsen, M. M.; Molenbroek, A. M.; Besenbacher, F. Int. J. Hydrogen Energy 2006, 31 2052-2062. 82. Obtained from the software PANalytical X pert Highscore softwa re version 1.0e. Reference code: 00-004-0664. Primary reference: Swanson, Tatge., Natl. Bur. Stand. (U.S.), Circ. 539 I, 62, (1953). 83. Hanada, N.; Ichikawa, T.; Orimo, S.; Fujii, H. J. Alloys Compd 2004, 366 269-273. 84. Barkhordarian, G.; Klassen, T.; Bormann, R. J. Phys. Chem. B 2006, 110 1102011024.
70 85. Czujko, T.; Varin, R. A.; Wronski Z.; Zaranski, Z.; Durejko, T. J. Alloys Compd 2007, 427 291-299. 86. Slattery, D. K.; Hampton, M. D. Proceedings of the 2002 US DOE Hydrogen Program Review NREL/CP-610-32405. 87. Obtained from the software PANalytical X pert Highscore softwa re version 1.0e. Reference code: 00-004-0831. Primary reference: Swanson, Tatge., Natl. Bur. Stand. (U.S.), Circ. 539 I, 16, (1953). 88. Hoekstra, H. R.; Katz, J. J. J. Am. Chem. So c. 1949, 71 2488-92. 89. Chater, P. A.; David, W. I. F.; Johnson, S. R.; Edwards, P. P.; Anderson, P. A. Chem. Commun 2006, 2439. 90. Johnson, S. R.; Anderson, P. A.; Edwards, P. P.; Gameson, I.; Prendergast, J. W.; AlMamouri, M.; Book, D.; Harris, I. R.; Speight, J. D.; Walton, A. Chem. Comm 2005, 2823-2825.