Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Investigating the microwave-hydrogen peroxide treatment process for potential commercialization MacSween, Jeffrey Vanek 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_february_macsween_jeffrey.pdf [ 3.25MB ]
Metadata
JSON: 24-1.0216471.json
JSON-LD: 24-1.0216471-ld.json
RDF/XML (Pretty): 24-1.0216471-rdf.xml
RDF/JSON: 24-1.0216471-rdf.json
Turtle: 24-1.0216471-turtle.txt
N-Triples: 24-1.0216471-rdf-ntriples.txt
Original Record: 24-1.0216471-source.json
Full Text
24-1.0216471-fulltext.txt
Citation
24-1.0216471.ris

Full Text

  Investigating the Microwave-Hydrogen Peroxide Treatment Process for Potential Commercialization by Jeffrey Vanek MacSween  BASc, Queen‟s University, 2013  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in The Faculty of Graduate and Postdoctoral Studies (Civil Engineering)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  November 2015 © Jeffrey Vanek MacSween, 2015   ii  Abstract Two pilot-scale, dielectric heaters, that utilize different frequencies of electromagnetic radiation (EMR), were modified to treat organic slurries with the aid of hydrogen peroxide (H2O2).  Organic slurries investigated include: waste activated sludge (WAS), liquid dairy manure (DM), and palm oil mill effluent (POME).  Treatment efficacy was evaluated in terms of changes in the substrate‟s digestibility, available nutrients, and physical properties.  The first heating system operated at a microwave frequency of 915MHz (MW-H2O2) and was modified to attain treatment temperatures above 100°C.  Results showed that such high temperatures permitted the effective use of larger oxidant dosages, which was found to be beneficial for treating DM and useful for treating WAS, in select circumstances.  However, the additional equipment costs and difficulties encountered in operating a pressurized high temperature dialectic heating system may prove inhibiting at larger scales.  The second heating system used radiofrequency EMR at 27MHz (RF-H2O2) and was modified from its original batch configuration for continuous operation.   The continuous RF-H2O2 process, the first of its kind, was demonstrated to be a viable treatment technology that achieved comparable results to the better studied MW-H2O2 system for DM and WAS substrates.  A controlled comparison of the two dielectric heaters showed that their treatment efficacy was similar, but suggested that heating at 27MHz offers advantages with respect to its operation and the degradation of organics.  The effect of higher heating rates and a 60°C treatment regime on WAS were also investigated, revealing that the treatment efficacy of the MW/RF-H2O2 process was suppressed in both cases, particularly in regards to the release of nutrients.  Using original experimental results and information from the literature, a full-scale, RF-H2O2 process is presented for treating thickened WAS from one million people.  The proposed system recommends three 900kW 27MHz dielectric heaters, a final treatment temperature of 95°C, a H2O2 dosage of 0.35%(v/v) per percent of dry solids, a heat exchanger to preheat the substrate, and an optional holding tank to enhance the release of orthophosphates.    iii  Preface  This thesis contains original, unpublished work undertaken by the author.  All experiments presented below were planned in collaboration with Dr. Asha Srinivasan and Dr. Liao Ping.      iv  Table of Contents Abstract ......................................................................................................................................................... ii Preface ......................................................................................................................................................... iii Table of Contents ......................................................................................................................................... iv List of Tables ............................................................................................................................................... xi List of Figures ............................................................................................................................................ xiv List of Abbreviations ................................................................................................................................ xvii List of Symbols .......................................................................................................................................... xix Acknowledgements ................................................................................................................................... xxii 1.0 Introduction ....................................................................................................................................... 1 1.1 The Waste Water Problem ............................................................................................................ 1 1.1.1 Waste Activated Sludge ........................................................................................................ 1 1.1.2 Dairy Manure ........................................................................................................................ 2 1.1.3 Palm Oil Mill Effluent .......................................................................................................... 2 1.3 Basics of the MW/RF-H2O2 Process ............................................................................................. 3 1.4 Scope of Research ......................................................................................................................... 4 1.4.1 Knowledge Gap..................................................................................................................... 4 1.4.2 Project Goals ......................................................................................................................... 4 2.0 Literature Review .............................................................................................................................. 6 2.1 Theories and Principles of Dielectric Heating .............................................................................. 6 2.1.1 Electromagnetic Radiation .................................................................................................... 6 2.1.2 Dielectric Materials ............................................................................................................... 6 2.1.3 Heating Mechanisms ............................................................................................................. 9 2.1.4 Dielectric Interactions with Biological Systems ................................................................. 12 2.1.5 Effect of Temperature on Dielectric Properties .................................................................. 16 2.1.6 Impedance Matching ........................................................................................................... 16 2.2 Chemistry of the MW/RF-H2O2 Process ..................................................................................... 19 v  2.2.1 Reactivity of Hydrogen Peroxide ........................................................................................ 19 2.2.2 Effect of Dielectric Heating on Reaction Rates .................................................................. 25 2.3 Review of Previous Microwave and MW-H2O2 Research .......................................................... 26 2.3.1 Treatment of Sludge from Municipal Waste Waters .......................................................... 26 2.3.2 Treatment of Animal Wastes .............................................................................................. 29 2.4 Design of Industrial Dielectric Heaters ....................................................................................... 30 2.4.1 Microwave Frequencies ...................................................................................................... 30 2.4.2 Radiofrequencies ................................................................................................................. 34 2.4.3 Comparing Microwave and Radiofrequency Dielectric Heaters ........................................ 37 3.0 Methods and Materials .................................................................................................................... 39 3.1 Substrates Investigated ................................................................................................................ 39 3.1.1 Waste Activated Sludge ...................................................................................................... 39 3.1.2 Dairy Manure ...................................................................................................................... 40 3.1.3 Palm Oil Mill Effluent ........................................................................................................ 40 3.2 Physical Analysis ........................................................................................................................ 40 3.2.1 Extraction of Soluble Fraction ............................................................................................ 40 3.2.2 Total Solids Fractions (TS/TVS) ........................................................................................ 40 3.2.3 Suspended Solids Fractions (TSS/VSS) ............................................................................. 41 3.2.4 Particle Size Distribution (PSD) ......................................................................................... 41 3.2.5 Capillary Suction Time (CST) ............................................................................................ 41 3.3  Chemical Analysis ..................................................................................................................... 41 3.3.1 Chemical Oxygen Demand (TCOD/SCOD) ....................................................................... 41 3.3.2 Total Phosphorus (TP/STP) ................................................................................................ 41 3.3.3 Total Kjeldahl Nitrogen (TKN/STKN) ............................................................................... 42 3.3.4 Ammonia (NH3) .................................................................................................................. 42 3.3.5 Orthophosphates (OP) ......................................................................................................... 42 3.3.6 Polyphosphates (PP) ........................................................................................................... 42 vi  3.3.7 Volatile Fatty Acids (VFA) ................................................................................................. 43 3.3.8 Metals Analysis ................................................................................................................... 43 4.0 The MW/RF-H2O2 Experimental Setup .......................................................................................... 44 4.1 2.45GHz Microwave Dielectric Heater....................................................................................... 44 4.1.1 Description .......................................................................................................................... 44 4.1.2 Standard Operating Procedure ............................................................................................ 44 4.2 915MHz Microwave Dielectric Heater ....................................................................................... 45 4.2.1 Original Design for Low Temperature Operation ............................................................... 45 4.2.2 Modifications for High Temperature Operation ................................................................. 47 4.2.3  Standard Operating Procedure ............................................................................................ 50 4.2.4 Characterizing the Modified 915MHz Microwave System ................................................ 51 4.2.5 Recommendations for Future Modifications and Upgrades ............................................... 52 4.3 The 27MHz Radiofrequency Dielectric Heater .............................................................................. 53 4.3.1 Original Design for Batch Operation .................................................................................. 53 4.3.2 Modifications for Continuous Operation ............................................................................ 55 4.3.3  Standard Operating Procedure ............................................................................................ 58 4.3.4 Characterizing the Continuous 27MHz Radiofrequency System ....................................... 59 4.3.5 Recommendations for Future Modifications and Upgrades ............................................... 60 5.0 Waste Activated Sludge Experiments ............................................................................................. 61 5.1 Effect of MW-H2O2 Process on Dewaterability .......................................................................... 61 5.1.1 Results ................................................................................................................................. 61 5.1.2 Discussion on Digestion Indicators ..................................................................................... 62 5.1.3 Discussion on Dewaterability ............................................................................................. 63 5.2 Experiments on the 915MHz Microwave Heater at High Temperatures .................................... 65 5.2.1 Digestion Indicators ............................................................................................................ 65 5.2.2 Nutrient Release .................................................................................................................. 67 5.2.3 Physical Properties .............................................................................................................. 69 vii  5.2.4 Conclusions Pertaining to Commercial Applications ......................................................... 71 5.3 Experiments on the 27MHz Radiofrequency Dielectric Heater.................................................. 71 5.3.1 Effect of Hydrogen Peroxide .............................................................................................. 71 5.3.2 Low Temperature Study ...................................................................................................... 81 5.3.3 Conclusions Pertaining to Commercial Applications ......................................................... 84 5.4 Comparison Study ....................................................................................................................... 85 5.4.1 Summary of Results ............................................................................................................ 85 5.4.2 Discussion on Digestion Indicators ..................................................................................... 86 5.4.3 Discussion on Nutrient Release .......................................................................................... 87 5.4.4 Physical Properties .............................................................................................................. 87 5.4.5 Theoretical Explanation of Observations ............................................................................ 88 5.4.6 Major Conclusions .............................................................................................................. 89 5.5 Effect of Heating Rate on Pilot-scale Systems ........................................................................... 90 5.5.1 Digestion Indicators ............................................................................................................ 90 5.5.2 Nutrient Release .................................................................................................................. 91 5.5.3 Physical Properties .............................................................................................................. 94 5.5.4 Conclusions Pertaining to Commercial Applications ......................................................... 95 5.6 Behaviour of Phosphorus in the MW/RF-H2O2 Process ............................................................. 96 5.6.1 Distribution of Phosphorus in Waste Activated Sludge ...................................................... 96 5.6.2 Behaviour of Orthophosphates ............................................................................................ 98 5.6.3 Release of Polyphosphates ................................................................................................ 103 5.6.4 Maximizing Phosphorus Recovery ................................................................................... 103 5.7 Energy Efficiency of Dielectrically Heating Waste Activated Sludge ..................................... 106 5.7.1 915MHz Microwave Heater .............................................................................................. 106 5.7.2 27MHz Radiofrequency Heater ........................................................................................ 109 5.7.3 Comparing Dielectric Heaters for the Treatment of Waste Activated Sludge .................. 111 6.0 Dairy Manure Experiments ........................................................................................................... 112 viii  6.1 Preparing Dairy Manure for Treatment..................................................................................... 112 6.2 27MHz Radiofrequency Heater ................................................................................................ 114 6.2.1 Digestion Indicators .......................................................................................................... 115 6.2.2 Nutrient Release ................................................................................................................ 116 6.2.3 Comparison to Previous Experiments ............................................................................... 117 6.2.4 Conclusions ....................................................................................................................... 117 6.3 915MHz Microwave Heater...................................................................................................... 118 6.3.1 Digestion Indicators .......................................................................................................... 118 6.3.2 Nutrient Release ................................................................................................................ 120 6.3.3 Conclusions ....................................................................................................................... 121 6.4 Energy Efficiency of Dielectrically Heating Dairy Manure ..................................................... 121 6.4.1 915MHz Microwave Heater .............................................................................................. 121 6.4.2 27MHz Radiofrequency Heater ........................................................................................ 123 6.4.3 Comparison of Dielectric Heaters for the Treatment of Dairy Manure ............................ 124 7.0 Palm Oil Mill Effluent Experiments ............................................................................................. 125 7.1 Digestion Indicators .................................................................................................................. 125 7.2 Nutrient Release ........................................................................................................................ 127 7.3 Recommendations for Treatment Strategies and Future Experiments ...................................... 130 8.0 Design of a Full-scale RF-H2O2 Treatment System ...................................................................... 133 8.1 Design Criteria .......................................................................................................................... 133 8.2 Process Overview ...................................................................................................................... 133 8.3 Discussion of Proposed Design................................................................................................. 134 8.3.1 Heat Exchangers ............................................................................................................... 135 8.3.2 Dielectric Heaters .............................................................................................................. 135 8.3.3 Operating Conditions and Expected Treatment Results ................................................... 136 8.3.4 Conditional Modifications ................................................................................................ 137 8.3.5 Anaerobic Digester ........................................................................................................... 137 ix  8.3.6 Process Control Considerations ........................................................................................ 138 8.3.7 Concluding Remarks ......................................................................................................... 139 9.0 Summary of Work ......................................................................................................................... 140 9.1 Chemistry of the MW/RF-H2O2 Process ................................................................................... 140 9.2 Experiments with Waste Activated Sludge ............................................................................... 140 9.2.1 Effect of the MW-H2O2 Process on Physical Properties and Solid‟s Reduction .............. 140 9.2.2 Effect of High Temperatures and Oxidant Dosages Using a 915MHz Microwave Heater……………………………………………………………………………………………….140 9.2.3 Effect of the H2O2 Dosage Using a 27MHz Radiofrequency Heater ................................ 140 9.2.4 Feasibility of a Low Temperature Treatment Using a 27MHz Radiofrequency Heater ... 141 9.2.5 Comparing the Treatment Efficacy of a 915 and 27MHz Dielectric Heater ..................... 141 9.2.6 Effect of Heating Rate for Full-scale Applications ........................................................... 141 9.2.7 Behaviour of Phosphorus in the MW/RF-H2O2 Treatment Process .................................. 141 9.2.8 Energy Efficiency of the MW/RF-H2O2 Treatment Process ............................................. 142 9.3 Experiments with Liquid Dairy Manure ................................................................................... 142 9.3.1 Effect of Dairy Manure Preparation .................................................................................. 142 9.3.2 Effect of Hydrogen Peroxide‟s Injection Temperature Using a 27MHz Radiofrequency Heater……………………………………………………………………………………………….142 9.3.3 Effect of High Temperatures and Hydrogen Peroxide Dosages Using a 915MHz Microwave Heater ............................................................................................................................. 142 9.3.4 The Energy Efficiency of Treating Liquid Dairy Manure ................................................ 142 9.4 Exploratory Experiments with Palm Oil Mill Effluent ............................................................. 143 9.5 Conclusions ............................................................................................................................... 143 9.6 Recommendations for Future Work .......................................................................................... 143 9.6.1 Specific Areas Requiring Additional Research ................................................................. 143 9.6.2 Selecting an Industrial Dielectric Heater .......................................................................... 143 9.6.3 Design of a Full-scale RF-H2O2 Treatment System for Thickened Waste Activated Sludge……………………………………………………………………………………………….144 x  Bibliography ............................................................................................................................................. 145 APPENDIX A – Remaining Data for the 2.45GHz Microwave Experiments on Waste Activated Sludge .................................................................................................................................................................. 159 APPENDIX B – Complete Data for the High Temperature 915MHz Microwave Experiments on Waste Activated Sludge. ...................................................................................................................................... 160 APPENDIX C – Complete Data for the 27MHz Radiofrequency Experiments Investigating Hydrogen Peroxide on Waste Activated Sludge ....................................................................................... 162 APPENDIX D – Complete Data for the Low Temperature 27MHz Radiofrequency Experiment on Waste Activated Sludge ....................................................................................................................................... 166 APPENDIX E – Complete Data for the Low Temperature 915MHz Microwave Experiment on Waste Activated Sludge ....................................................................................................................................... 167 APPENDIX F – Complete Data for the Fast Heating Rate 915MHz Microwave Experiment on Waste Activated Sludge ....................................................................................................................................... 168 APPENDIX G – Complete Data for the 27MHz Radiofrequency Experiments on Dairy Manure .......... 169 APPENDIX H – Complete Data for the 915MHz Microwave Experiments on Dairy Manure ............... 171 APPENDIX I – Complete Data for the 2.45GHz Microwave Experiments on Palm Oil Mill Effluent ... 173 APPENDIX J – Summary of Assumptions and Calculations Used to Design the Full-scale RF-H2O2 System ....................................................................................................................................................... 174     xi  List of Tables Table 1 – Dielectric properties of activated sludge samples at 27 and 915MHz ........................................ 15 Table 2 – Summary of reactions between unactivated H2O2 and common ................................................. 20 Table 3 – Compilation of kinetic data from aqueous H2O2 decomposition studies .................................... 22 Table 4 – Examples of beneficial results from MW pretreating sewage sludge prior to anaerobic digestion .................................................................................................................................................................... 27 Table 5 – Examples of beneficial results relating to the release and recovery of nutrients from sewage sludge .......................................................................................................................................................... 28 Table 6 – A comparison of common microwave applicator designs .......................................................... 33 Table 7 – Comparison of MW and RF components used for dielectric heating ......................................... 37 Table 8 – Detection limits for the ICP metal analysis ................................................................................ 43 Table 9 – Control valves used in the modified 915MHz MW system and their function .......................... 49 Table 10 – Height of tuning rods outside of the 915MHz MW‟s waveguide for WAS and DM substrates .................................................................................................................................................................... 52 Table 11 – The continuous RF heater‟s electronic settings for WAS and DM experiments ...................... 59 Table 12 – Experimental conditions and select results for experiments on WAS using the 2.45GHz batch MW ............................................................................................................................................................. 62 Table 13 – Experimental conditions for high temperature WAS runs using the 915MHz MW Heater ..... 65 Table 14 – Ratio of species required for struvite formation from WAS experiments using the 915MHz MW heater, where numbers in brackets reflect the experiment‟s respective raw sample .......................... 68 Table 15 – Select physical sludge properties from high temperature WAS experiments using the 915MHz MW heater, where numbers in brackets reflect the experiment‟s respective raw sample .......................... 69 Table 16 – Experimental conditions for WAS experiments investigating the effect of H2O2 using the 27MHz RF heater ........................................................................................................................................ 71 Table 17 – Struvite molar ratios from WAS experiments investigating the effect of H2O2 at 90°C using the 27MHz RF heater .................................................................................................................................. 78 Table 18 – Select physical properties from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater ........................................................................................................................................ 79 Table 19 – Experimental conditions for a low temperature WAS run using the 27MHz RF heater .......... 81 Table 20 – Comparison of raw samples from the low temperature and low H2O experiments using the 27MHz RF heater on WAS ......................................................................................................................... 82 Table 21 – Select digestion parameters for the low temperature WAS experiment using the 27MHz RF heater ........................................................................................................................................................... 82 Table 22 – Experimental conditions for the dielectric heater comparison study using WAS ..................... 85 xii  Table 23 – Statistical test results from the dielectric heater comparison experiment using WAS .............. 86 Table 24 – Experimental conditions for the heating rate comparison study using a 915MHz MW heater on WAS ............................................................................................................................................................ 90 Table 25 – COD results from the heating rate comparison study using a 915MHz MW heater on WAS .. 91 Table 26 – TS and TSS results from the heating rate comparison study using a 915MHz MW heater on WAS ............................................................................................................................................................ 95 Table 27 – Comparison of the changes in OP and PP between 60 and 75°C from WAS experiments presented in this work ............................................................................................................................... 100 Table 28 – Stability constants and thermodynamic data for select aqueous metal phosphate complexes 101 Table 29 – Recommended MW/RF-H2O2 treatment conditions for various P recovery strategies .......... 106 Table 30 – Representative efficiency values for the 915MHz MW heater treating WAS at high temperatures .............................................................................................................................................. 107 Table 31 – Raw samples from all DM experiments and their method of preparation .............................. 112 Table 32 – Select chemical and physical properties of all untreated DM samples ................................... 112 Table 33 – Experimental conditions for DM runs using the 27MHz RF heater ....................................... 115 Table 34 – Comparison of nutrient releases from DM experiments using the 27MHz RF heater ............ 116 Table 35 – Experimental conditions for DM runs using the 915MHz MW heater ................................... 118 Table 36 – Nutrient results from DM experiments using the 915MHz MW heater ................................. 120 Table 37 – Experimental conditions for POME runs using the 2.45GHz batch MW ............................... 125 Table 38 – Important struvite ratios from POME experiments using the batch 2.45GHz MW heater ..... 130 Table 39 – Design criteria for the proposed full-scale RF-H2O2 system treating TWAS ......................... 133 Table 40 – Summary of stream characteristics for the proposed full-scale RF-H2O2 system treating TWAS .................................................................................................................................................................. 134 Table 41 – Complete TS, TVS, TSS, VSS, and TCOD data from experiments on WAS using the 2.45GHz batch MW.................................................................................................................................................. 159 Table 42 – Complete data set from high temperature WAS experiments using the 915MHz MW heater160 Table 43 – Complete data set from a no H2O2 experiment on WAS using the 27MHz RF heater ........... 162 Table 44 – Complete data set from a high H2O2 dosage experiment on WAS using the 27MHz RF heater .................................................................................................................................................................. 163 Table 45 – Complete data set from a medium H2O2 dosage experiment on WAS using the 27MHz RF heater ......................................................................................................................................................... 164 Table 46 – Complete data set from a low H2O2 dosage experiment on WAS using the 27MHz RF heater .................................................................................................................................................................. 165 xiii  Table 47 – Complete data set from a low temperature experiment on WAS using the 27MHz RF heater .................................................................................................................................................................. 166 Table 48 – Complete data set from a 90°C WAS experiment using the 915MHz MW heater ................. 167 Table 49 – Complete data set from a fast heating rate WAS experiment using the 915MHz MW heater 168 Table 50 – Complete data set from a 25°C H2O2 addition experiment on DM using the 27MHz RF heater .................................................................................................................................................................. 169 Table 51 – Complete data set from a 60°C H2O2 addition experiment on DM using the 27MHz RF heater .................................................................................................................................................................. 170 Table 52 – Complete data set from a 110°C DM experiment using the 915MHz MW heater ................. 171 Table 53 – Complete data set from a 130°C DM experiment using the 915MHz MW heater ................. 172 Table 54 – Complete data set from POME experiments using the 2.45GHz batch MW heater ............... 173 Table 55 – Summary of assumptions and parameters used to design a full-scale RF-H2O2 system ......... 174   xiv  List of Figures Figure 1 – Graphical representation of the permittivity equation ................................................................. 7 Figure 2 – Dipolar rotation of a water molecule from an oscillating electric field ..................................... 10 Figure 3 – Chloride and sodium ions undergoing ionic conductance due to an oscillating electric field ... 11 Figure 4 – Interfacial polarization of a heterogeneous material: A) with no electric field, and B) with a static electric field ....................................................................................................................................... 12 Figure 5 – Idealized dielectric dispersion spectra of a biological system ©Markx and Davey, 1999 ........ 14 Figure 6 – A staggered through-field applicator for RF heating systems ................................................... 36 Figure 7 – Process flow diagram of the pilot-scale WWTP that provided the WAS samples .................... 39 Figure 8 – 2.45GHz MW oven: A) outside view, and B) view of vessels, where i) sample vessel, ii) sheath, iii) cap, iv) collar, v) spacers, and vi) vessel holder ....................................................................... 44 Figure 9 – Model of the pilot-scale 915MHz MW heater, where: A) generator, B) waveguide, C) applicator, D) silicone tubing, E) tuning screws, F) water cooling system, G) isolator, H) holding tank, I) primary pump, J) H2O2 reservoir, and K) H2O2 pump ................................................................................ 46 Figure 10 – Process flow diagram of the 915MHz MW modified for operation at temperatures above 100°C .......................................................................................................................................................... 47 Figure 11 – Temperature profiles of salt water heated with the modified 915MHz MW oven: A) 6L/min, and B) 8.4L/min .......................................................................................................................................... 51 Figure 12 – Model of the original batch mode 27MHz RF heater: A) generator, B) co-axial cable, C) outer shell and ground electrode, and D) active electrode ................................................................................... 54 Figure 13 – Contour map describing the change in heating rate relative to the centre of the batch RF heater‟s active electrode .............................................................................................................................. 55 Figure 14 – Upgraded electrode configuration for the continuously operated 27MHz RF heater .............. 56 Figure 15 – Process flow diagram of the continuous 27MHz RF heating system ...................................... 56 Figure 16 – PSD functions for WAS experiments using the 2.45GHz batch MW ..................................... 64 Figure 17 – COD results from high temperature WAS experiments using the 915MHz MW heater, where numbers in brackets reflect the experiment‟s respective raw sample ......................................................... 66 Figure 18 – STP speciation from high temperature WAS experiments using the 915MHz MW heater, where numbers in brackets reflect the experiment‟s respective raw sample .............................................. 67 Figure 19 – Solubilization of Ca, K, and Mg from high temperature WAS experiments using the 915MHz MW heater, where numbers in brackets reflect the experiment‟s respective raw sample .......................... 68 Figure 20 – Select PSD functions from high temperature WAS experiments using the 915MHz MW heater, where numbers in brackets reflect the experiment‟s respective raw sample ................................... 70 xv  Figure 21 – Percent of SCOD from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater ..................................................................................................................................................... 72 Figure 22 – Effect of H2O2 dosage on SCOD at 90°C from WAS experiments using the 27MHz RF heater .................................................................................................................................................................... 73 Figure 23 – VFA results from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater ........................................................................................................................................................... 74 Figure 24 – OP results from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater ........................................................................................................................................................... 75 Figure 25 – TP speciation at 90°C from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater ........................................................................................................................................ 75 Figure 26 – Differences in TP and STP measurements between the colourimetric and ICP method for WAS experiments investigating the effect of H2O2 using the 27MHz RF heater ....................................... 77 Figure 27 – Release of NH3 from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater ........................................................................................................................................................... 78 Figure 28 – Select PSD functions at 90°C from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater .................................................................................................................................. 80 Figure 29 – Contributions to SCOD gains from the low temperature RF run and an estimated run at 90°C for WAS ...................................................................................................................................................... 83 Figure 30 – OP release from the low temperature WAS experiment using the 27MHz RF heater ............ 84 Figure 31 – PSD functions at 90°C from the dielectric heater comparison experiment using WAS .......... 88 Figure 32 – VFA results from the heating rate comparison study using a 915MHz MW heater on WAS . 91 Figure 33 – OP results from the heating rate comparison study using a 915MHz MW heater on WAS .... 92 Figure 34 – STP speciation from the heating rate comparison study at 90°C using a 915MHz MW heater on WAS ....................................................................................................................................................... 93 Figure 35 – STKN and NH3 results from the heating rate comparison study using a 915MHz MW heater on WAS ....................................................................................................................................................... 93 Figure 36 – Classifications of PPs, where: A) linear (triphosphoric acid), B) cyclic (trimetaphosphate), and C) branched .......................................................................................................................................... 97 Figure 37 – Temperature profile of the 110°C MW-H2O2 run ................................................................. 106 Figure 38 – The RP throughout select 915MHz MW experiments using WAS, where: A) addition of H2O2, B) low temperature tune, C) post-H2O2 tune, and D) high temperature tune ................................. 108 Figure 39 – Representative temperature profiles from the 27MHz RF low H2O2 experiment using WAS .................................................................................................................................................................. 109 Figure 40 – Representative power data from the 27MHz RF low H2O2 experiment using WAS ............ 110 xvi  Figure 41 – The total concentrations of select nutrients in acidified DM before and after screening ...... 114 Figure 42 – SCOD from DM experiments using the 27MHz RF heater................................................... 115 Figure 43 – COD from DM experiments at final treatment temperatures, where the code in brackets reflect the experiment‟s respective raw sample ........................................................................................ 119 Figure 44 – Temperature profiles from DM experiments using the 915MHz MW heater ....................... 122 Figure 45 – RP fraction throughout DM experiments using the 915MHz MW heater ............................. 122 Figure 46 – Representative temperature profile from the DM experiment using the 27MHz RF heater with H2O2 added at 60°C................................................................................................................................... 123 Figure 47 – Representative power data from the DM experiment using the 27MHz RF heater with H2O2 added at 60°C ............................................................................................................................................ 124 Figure 48 – COD from POME experiments using the batch 2.45GHz MW heater .................................. 126 Figure 49 – VFA results from POME experiments using the batch 2.45GHz MW heater ....................... 127 Figure 50 – TP fractions from POME experiments using the batch 2.45GHz MW heater....................... 128 Figure 51 – Fraction of STKN from POME experiments using the batch 2.45GHz MW heater ............. 129 Figure 52 – Soluble Ca, K, and Mg fractions from POME experiments using the batch 2.45GHz MW heater ......................................................................................................................................................... 130 Figure 53 – Process flow diagram for the proposed full-scale RF-H2O2 system treating TWAS ............. 134 Figure 54 – Possible electrode dimensions that are capable of heating the required TWAS flow rate .... 177    xvii  List of Abbreviations ATP:  Adenosine triphosphate BDL:  Below detection limit CST:  Capillary suction time DM:  Dairy manure  EBPR:  Enhanced biological phosphorus removal EDTA:  Ethylenediaminetetraacetic acid EMR:   Electromagnetic radiation EPA:  United States Environmental Protection Agency EPS:    Extracellular polymeric substance FP:  Forward power H2O2:  Hydrogen peroxide MBR:  Membrane bioreactor  MW:    Microwave MW-H2O2: Microwave hydrogen peroxide treatment process *OH:  Hydroxyl radical OP:  Orthophosphates O3:  Ozone PAO:    Phosphorus accumulating organisms PCC:   Pearson product-moment correlation coefficient PHA:  Polyhydroxyalkanoates  POME:  Palm oil mill effluent  PP:    Polyphosphates PSD:    Particle size distribution PTFE:  Polytetrafluoroethylene RF:   Radiofrequency RF-H2O2:  Radiofrequency hydrogen peroxide treatment process RP:  Reverse power xviii  SCOD:   Soluble chemical oxygen demand SCR:    Semiconductor-controlled rectifier SRT:  Solids retention time STP:    Soluble total phosphorus  SWR:  Standing wave ratio TCOD:   Total chemical oxygen demand TKN:    Total Kjeldahl nitrogen  TP:    Total phosphorus  TS:    Total solids TSS:    Total suspended solids TVS:    Total volatile solids TWAS:  Thickened waste activated sludge VFA:    Volatile fatty acids VFD:  Variable frequency device  VSS:    Volatile suspended solids VSWR:  Voltage standing wave ratio WAS:   Waste activated sludge WASSTRIP: Waste activated sludge stripping to remove internal phosphorus WWTP: Waste water treatment plant   xix  List of Symbols A  Pre-exponential factor a  Width of a waveguide  Ac  Area C  Capacitance c  Speed of light Cp  Constant pressure heat capacity d  Hydrogen peroxide dosage δ  Dielectric loss angle dp  Penetration depth Ea  Activation energy E  Electric field E0  Initial electric field E0,max  Maximum allowable initial field ε  Permittivity, magnitude   ε0  Permittivity of free space εr  Relative permittivity, magnitude ε‟  Energy storage term, real component of dielectric constant,  ε”  Dielectric loss term, complex component of dielectric constant ε”d  Dielectric loss term due to bound polarization phenomena  ηf  Over capacity factor ηRF  Energy efficiency of a RF heater f  Frequency  G  Conductance of a transmission line  Γ  Wave reflection coefficient  I  Current Ip  Intensity of power density j  Imaginary number xx  K  Equilibrium constant kd  Decomposition rate constant for hydrogen peroxide µ   Magnetic permeability µ0  Magnetic permeability of free space Pd  Dissipated power Pr  Fraction of reflected power Prated  Nominal power rating of equipment Ps  Stored power PT  Total power Pt:  Transmitted power Q  Flow rate q  Heating rate R  Universal gas constant  Rr  Real resistance ρ  Density ζ  Conductivity T  Temperature T0  Initial temperature  Tf  Final temperature  η  Retention time  TSSr  Reduction in total suspended solids V  Voltage  v  Volume vp  Phase velocity X  Conversion of reactant Z  Impedance  z  Distance between electrodes Z0  Characteristic impedance xxi  ZL  Impedance of load ZS  Impedance of source  xxii  Acknowledgements  I take this space to sincerely thank the many people that have helped me throughout my work: Sam Bailey for teaching me how to use the various analysis tests and equipment; Dr. Asha Srinivasan and Dr. Liao Ping for their help with designing and running experiments, as well as offering their experience when analysing results; Paula Parkinson and Timothy Ma for their invaluable laboratory advice and analysing all of my samples; Cristina Oliveira and Sarah Ning for their patience and assistance with running experiments; Rony Das, Chaoyang Yue, and the rest of the pilot plant crew for allowing me to collect samples; Glenn Blaker and Terry Enegren for their guidance and expertise when modifying the dielectric heaters; Doug Hudniuk for his help in the workshop and building parts of my experimental apparatus; and to Dr. Sergey Lobanov for his help translating papers. I would like to express my deepest gratitude to my supervisor, Dr. Victor Lo, for his ceaseless support and guidance; as well as to Dr. Donald Mavinic for kindly lending his time to review this work. I would also like to thank the many professors and teachers who are responsible for everything that I have learned, and all of my family and friends for their support throughout this program.    1  1.0 Introduction 1.1 The Waste Water Problem The inspiration for this thesis and related research comes from the fact that despite decades of technological advances and focused research, the majority of waste water treatment process still result in residual by-products that require disposal.  This is particularly true of organic wastes that are treated with biological systems.  As efficient and capable microorganisms are, they have strict limits in their ability to breakdown and uptake pollutants, while demanding stringent living conditions.   In light of this, there are two new paradigms that are starting to shift the way we view water borne pollution.  The first is that waste is inherently a sign of inefficiency and any system creating waste should therefore be optimized to prevent its formation in the first place. In some cases however, notably sewage, waste is an unavoidable consequence of human activity.  For these instances, the second paradigm is perhaps more fruitful; turn the waste into a resource that can be used elsewhere.  The MW-H2O2 technology developed over the past 10 years in this lab and advanced in the following pages incorporates elements of both these treatment strategies.   Embracing the waste-to-resource philosophy insists on a new point of view when it comes to tackling society‟s waste challenges; one that is holistic and considers the waste, and the process producing it, in the larger social, economic, and environmental world that it is part of.  If we are to make resources from our leftovers, their ability to be useful will multiply as society‟s various waste streams are combined and coordinated to maximize their potential.  With this in mind, the MW-H2O2 will be considered in context of other complimentary processes that can enhance its potential.   For the work presented below, treatment using the MW-H2O2 process will focus on waste activated sludge and dairy manure. 1.1.1 Waste Activated Sludge There are 4 000 wastewater treatment plants (WWTP) in Canada (Canadian Council of Ministers of the Environment, 2012), producing 660 000 dry tonnes of residual sludge a year (Robins Environmental, 2013).  Of these WWTPs, it is estimated that 25% use an activated sludge process as part of their treatment train (Canadian Water and Wastewater Association, 2001).  An activated sludge process is characterized by an aerobic bioreactor containing suspended biomass, followed by a sediment tank or other liquid-solid separator, where a fraction of the solid portion is recycled back into the bioreactor (Tchobanoglous, Burton, and Stensel, 2003).  This solid portion is referred to as the waste activated sludge, or WAS, and requires further stabilization to produce biosolids that can be safely disposed of.  2  Converting WAS into biosolids is typically accomplished through anaerobic or aerobic digestion, which often accounts for over 50% of a WWTP‟s total cost (Appels et al., 2008).   The most common method for disposing biosolids in Canada, at 41%, is to apply them to agricultural lands as fertilizer (Canadian Water and Wastewater Association, 2001).  However, increasing regulations have limited such applications, as well as the use of other disposal options, including incineration and landfilling; see the report by the Canadian Council of Ministers of the Environment (2010) for a thorough overview of biosolids regulations in Canada.  The rising disposal standards are a consequence of concerns relating to pathogens, heavy metals, pharmaceutical products, nutrients, and the release of odours (Canadian Council of Ministers of the Environment, 2012). 1.1.2 Dairy Manure Manure produced by dairy cows can be broadly described as a complex mixture, containing in order of increasing amounts: carbohydrates (ex. cellulose), proteins, lipids, lignin, volatile fatty acids, and inorganic residuals (Sommer et al., 2013).  Many of these inorganic residuals are nutrients, such as P, K, N, Ca, K, and Mg, as well as variable amounts of silica from bedding sand.   There are 959 300 dairy producing cows in Canada, with farms having average heard sizes of 80 lactating cows (Statistics Canada, 2014).  Of the approximately 22MT of dairy manure (or DM) produced annually in Canada (Statistics Canada, 2008), ~65% of dairy farmers do not treat their stored manure at all, while the most popular treatment process is composting (Beaulieu, 2004). In many cases, too much manure is produced for a farm to utilize on its lands, or the manure is not managed properly.  This can lead to dire environmental impacts, including: the release of greenhouse gases, especially methane if kept under anaerobic conditions; nutrient runoff causing eutrophication in the surrounding ecosystem; contamination of water supplies from pathogens; the loss of limited nutrients, particularly phosphorus; and odorous gases from the release of ammonia or sulphur oxides (Brandjes et al., 1996; Kumar et al., 2013; Sommer et al., 2013). 1.1.3 Palm Oil Mill Effluent Palm oil is an increasingly popular food grade oil that is derived from the mesocarp of oil palm plants.  The top producers are Indonesia and Malaysia, who are responsible for 44 and 43% of the world‟s palm oil supplies, respectively (Gobi and Vadivelu, 2013).  To recover the oil from the plant, a complex extraction process is required that generates substantial amounts of waste water, called palm oil mill effluent (POME).  This waste water comes from three parts of the extraction process, including steam used to sterilize the fruit bunches, water from a hydroclone that separates crushed shells and kernels, and sludge from a crude oil separator; these streams contribute to POME in a ratio of approximately 9:15:1, 3  respectively (Wu et al., 2010).  The combined waste streams produce an acidic effluent that is discharged at 80 to 90°C. For every tonne of crude palm oil produced, it is estimated that 3.5 tonnes of POME are generated (Madaki and Seng, 2013).  This large waste inefficiency results in over 50 million cubic metres of POME released every year in Malaysia alone, a figure that will continue to rise with expected increases in palm oil production.  Conventional POME treatment consists of facultative ponds for approximately 85% of mills due to their low cost and maintenance needs; however, this is largely regarded as an ineffective treatment process (Madaki and Seng, 2013; Wu et al., 2009; Wu et al., 2010).  As a result, waterways near palm oil mills are often heavily polluted and gross amounts of greenhouse gases escape from the ponds‟ large surface areas.  POME can be characterized as a colloidal system of plant derived matter comprised of the following solid fractions (Wu et al., 2009): carbohydrates (~30%), particularly cellulose and lignin; a nitrogen free extract (~25%); ash (~15%); proteins (~13%); and lipids (~10%).  The lipids consist primarily of long chain fatty acids, while the nitrogen free extract is associated with a stable oil emulsion.  Significant quantities of nutrients are also present, mainly P, followed by K, Ca, and Mg in lesser amounts.  1.3 Basics of the MW/RF-H2O2 Process The microwave hydrogen peroxide (MW-H2O2) process, as its name suggests, is the combination of a microwave dielectric oven and hydrogen peroxide to treat organic slurries.  The microwave oven, operating between 300MHz and 300GHz, rapidly and efficiently heats the slurry, while the hydrogen peroxide acts as a powerful oxidant that degrades the organic solids.  When used together, they act synergistically and their effects are multiplied.  This thermo-chemical process is, therefore, not limited by the biochemistry of microorganisms.  In principle, any level of treatment can be achieved if the appropriate temperature-oxidant dosage is utilized.  The basic mechanisms of treatment are the breakdown of organic molecules into smaller constituents and their subsequent complete oxidization into carbon dioxide, referred to as mineralization.  During treatment, resources can also be recovered, such as bioavailable nutrients, biofuel, and various chemical feedstocks for use in other industrial processes.  The MW-H2O2 process has been well studied for many years, but a novel modification investigated in this thesis is the use of a radiofrequency dielectric heater (1-300MHz) to replace the common MW heater.  To differentiate this new technology from its parent, it is henceforth referred to as the RF-H2O2 process.   4  1.4 Scope of Research 1.4.1 Knowledge Gap The current state of published literature on the MW-H2O2 treatment process reveals a lack of research on pilot-scale systems, particularly at high temperatures.  Consequently, it has yet to be conclusively shown that the MW-H2O2 process will be feasible in a commercial setting, thereby preventing their implementation at such scales.  More fundamental concepts of the process‟s mechanisms, such as the reactivity of H2O2 and the dielectric properties of organic slurries, have been inadequately investigated to date.  Lastly, aside from a few singular studies, research using dielectric heaters at frequencies other than 2.45GHz has not been reported.  The unknowns identified above were set into three major goals, as detailed below. 1.4.2 Project Goals 1.4.2.1 Goal #1 Modify a 915MHz microwave (MW) pilot-scale dielectric heater to allow for experiments above 100°C.  This would validate the „proof-of-concept‟ for high temperature treatment at commercial scales by comparing results to previous bench scale experiments.  Other, more specific objectives with the modified 915MHz heater include: 1) Determine if dewaterability can be improved at higher temperatures and/or oxidant dosages 2) Identify any operational or efficiency challenges when treating above 100°C 3) Conclude if high temperature conditions should be used in full-scale applications 1.4.2.2 Goal #2 Modify a batch loaded 27MHz radiofrequency (RF) pilot-scale dielectric heater for continuous operation to simulate a full-scale system.  Similarly, this would validate the „proof-of-concept‟ for treating organic slurries at commercial scales using lower frequencies, such as radio waves.  Other, more specific objectives with the modified 27MHz heater include: 1) Determine the effect of H2O2 dosage on treatment effectiveness  2) Identify any operational or efficiency challenges when treating at lower frequencies 3) Resolve whether the temperature at which H2O2 is added affects the treatment of DM 5  1.4.2.3 Goal #3 Combine all relevant knowledge to design a full-scale MW/RF-H2O2 treatment process for WAS and make recommendations pertaining to its operation.  This will be aided by fulfilling the following objectives: 1) Compare the 27MHz and 915MHz heaters to determine if either offers an advantage in terms of treatment effectiveness or operation 2) Recommend an optimal treatment temperature range, such as: above 100°C, just below 100°C, or much below 100°C 3) Determine if faster heating rates, as would likely be used in full-scale systems, impact the treatment effectiveness of the MW/RF-H2O2 process 4) Maximize the release of orthophosphates to improve struvite P recovery   6  2.0 Literature Review 2.1 Theories and Principles of Dielectric Heating In a sentence, dielectric heating is the use of electromagnetic radiation (EMR) to heat dielectric materials.   2.1.1 Electromagnetic Radiation Electromagnetic radiation is attributed to the release of photons from accelerating charged particles that propagate through space with a set frequency, a characteristic that arise from their wave-particle duality.  This duality allows EMR to be described analogously to waves, such as those formed on bodies of water by wind.  Electromagnetic waves however, have two components, a magnetic and an electric element that propagate 90° (transversely) to each other.  These two components are inseparable and furthermore, travel in phase with one another, meaning that the peak of the electric field corresponds to the peak of the magnetic counterpart.  The characteristic frequency of each released photon has a corresponding quantum of energy.  Longer frequencies have shorter wavelengths and less energy per photon.  However, the total amount of energy transferred by an electromagnetic wave can be ascribed to its amplitude, equivalent to a larger ocean wave that carries more energy due to its increased momentum.  Two types of dielectric heaters are used in this work with each utilizing a different frequency of the electromagnetic spectrum.  The more familiar is the microwave frequency range, typically defined as covering 300MHz to 300GHz (Marra, 2012).  The second heater operates in the radiofrequency (RF) range, generally said to cover any frequency below 300MHz.  The magnetic and electric fields that make up EMR can interact with matter in a variety of ways.  If the matter in question is a dielectric material, then heating can occur as explained in the proceeding section. 2.1.2 Dielectric Materials Dielectric materials are any substances that are electrical insulators but capable of being polarized, such as water.  Materials can interact with EMR either through the magnetic or electric field by reflecting, transmitting, or absorbing the field‟s energy.  How a material behaves in the presence of these fields is characterized by its permittivity, ε, and permeability, µ, for the electrical and magnetic fields, respectively.  The equations presented below will cover the basic relationships between EMR and dielectric materials and can be found in most texts covering electromagnetic theory, such as Orfanidis (2014).  The ε and µ can be described as the resistance a material offers to each field and can be related to the speed, or phase velocity vp, at which EMR will travel through said material (Equation 1): 7      √        (1) The effect of EMR‟s magnetic component is negligible for non-magnetic materials and their permeability is nearly equal to µ0, the permeability of free space (Venkatesh and Raghavan, 2004).   Instead, it is the EMR‟s electric field that primarily interacts with the material and thus, its permittivity largely dictates how it will be heated.  The equation of this governing property is shown below in Equation 2:               (2) The permittivity, called the dielectric constant in older texts, is a complex property made up of a real (ε‟) and imaginary part (ε”), where j is √  .  The ε‟ is known as the energy storage term and represents how much potential energy is stored in the material as it becomes polarized.  This stored energy is released once the molecules relax in the absence of EMR.  More importantly for heating applications is ε”, often referred to as the dielectric loss term.  It indicates how much energy the material will absorb from EMR and be dissipated as heat, opposed to being reflected or transmitted.  Therefore, materials with large ε” values will be heated more readily in the presence of equivalent electric fields.  Both terms are also highly dependent on temperature and frequency.  The relationship shown in Equation 2 can be represented graphically, as displayed in Figure 1.  Figure 1 – Graphical representation of the permittivity equation  From the above figure, ℝ is the set of real numbers and ℂ is the set of complex numbers, making the permittivity the magnitude of the ε‟ and ε” vectors.  The ratio of the permittivity‟s vector components is often reported as the loss tangent, tan(δ), as shown in Equation 3 and highlighted in Figure 1.     ( )             (3) 8  The loss tangent can be thought of as a quality factor that describes the ratio of energy that is dispersed as heat versus being absorbed as potential energy.  The dielectric loss term can be further described as the sum of several heating mechanisms (Equation 4).                      (4) Where f is frequency,      represents the losses due to several bound polarization phenomena, and ζ is the unbound ionic conductance that also contributes to heating.  The mechanics of these heating mechanisms are explained in the proceeding section.  In practice, these separate heat loss mechanisms are often not considered separately and only ε” is measured.  As mentioned before, when EMR encounters a dielectric material, a fraction of its energy may reflect backwards (Pr), become stored within the material (Ps), or be dissipated as heat (Pd).  The portion of power reflected (Pr) is governed by the difference in impedances between the incoming EMR and the material; see Section 2.1.6 for a complete treaty.  The amount of stored power (Ps) can be quantified using Equation 5.      √         √ √     ( )    ( )(        √   √     ( )    ( ) )      (5) Where Ac is the cross-section of the materials; z is its thickness; and E02 is the initial electric field strength at the surface of the material.  The amount of power absorbed as heat (Pd) is of course dependent on the material‟s volume and dielectric loss term, as related in Equation 6:    | |               (6) In Equation 6, v is the volume of substrate and E is the root-mean square (i.e. average) of the electric field, assumed to be constant.  To summarize, as EMR propagates through the material, it continuously loses energy as it is dissipated as heat or becomes stored in the electrical field, until either all of its energy is absorbed (i.e attenuated) or some residual EMR is able to transmit completely through (Pt).  A useful parameter to quantify this propagation is the penetration depth, dP, shown in Equation 7.  It is defined as the distance through a medium at which only 1/e (~%37) of the transmitted EMR‟s energy remains, the rest being absorbed.          √     √√  (     )⁄        (7) Where c is the universal constant for the speed of light in a vacuum and ε0 is the permittivity of free space. Equation 7 shows that the dp generally increases at lower f, lower ε”, and higher ε‟.  Thus, if the 9  material being heated is thicker than the penetration depth, then the E from Equation 6 can no longer be assumed as constant (Hill and Jennings, 1993).  In such cases, the E field can be approximated to decay with distance, z, from the surface of the material as a function of the penetration depth; shown in Equation 8 by Hill and Pincombe (1992). | |          ⁄          (8) E0 is the initial electric field strength at the surface of the substrate.  This decay function assumes that the substrate‟s dielectric properties are constant with space and time.  It is important to recall that there is an absolute upper theoretical efficiency for the heating of dielectric materials, governed by the ratio of Ps to Pd as dictated by their dielectric properties.  Therefore, even if the EMR power is transferred perfectly to the material, a portion of it will be stored in the electric field and not contribute to heating.  Depending on this ratio, some materials are more appropriate for dielectric heating than others.   2.1.3 Heating Mechanisms 2.1.3.1 Dipolar Rotation Dipolar rotations, also referred to as γ-dispersions, occurs to polarizable molecules, such as water, and to a lesser extent, molecules that can have an induced dipole, thereby affecting all materials to some degree.  From Equation 4, dipolar rotations are the primary heating mechanism that contributes to the      term for homogeneous materials.  In the presence of an oscillating electric field, polarized molecules will orient themselves to allow their electronegative atoms, dense with electrons, to face the positive region of the electric field (Marra, 2012).  Because the electric field is oscillating at the EMR‟s characteristic frequency, the molecules are constantly re-orientating themselves to ensure that their negatively charged parts are always facing the ever moving electric field‟s positive region.  This molecular dance is illustrated in Figure 2 for a water molecule.  Such rotation of molecules results in intermolecular friction that heats the material.    10   Figure 2 – Dipolar rotation of a water molecule from an oscillating electric field Furthermore, as the frequency increases, the molecules will re-orientate faster to keep up with the rapidly changing electric field.  At a certain frequency however, the rotating molecules will no longer be able to match the electric field‟s oscillations, resulting in less friction.  The frequency at which this threshold occurs is called the Debye frequency, and represents the most efficient point for heating a material dielectrically via dipolar rotations.  By plotting      with frequency, the maximum peak would mark the Debye frequency.  Dipolar rotations are dominant in the microwave frequencies, particularly for water, which reaches its maximum at 19.5GHz (Venkatesh and Raghavan, 2004). 2.1.3.2 Ionic Conductance  Ionic conduction is similar to that of dipolar rotation, but occurs when cations and anions dissolved in a material attempt to move towards the negative and positive regions of an alternating electric field, respectively (Marra, 2012).  This phenomenon is represented by the     ⁄  term in Equation 4.  Due to the oscillating electric field, these regions are constantly changing, forcing the ions to switch their direction in an effort to follow them, as depicted in Figure 3. 11   Figure 3 – Chloride and sodium ions undergoing ionic conductance due to an oscillating electric field This reciprocating movement also creates internal friction as the layer of solvated water molecules surrounding the ions get dragged behind them, thereby raising the material‟s temperature.  The extent of heating will be greater for materials containing higher concentration of ions that are more mobile and more charged (Schiffmann, 2006).  Ion conductance is dominant in the radiofrequency ranges and typically increases linearly as the frequency decreases.  Above approximately 1GHz, the effects of ionic conductance are no longer observable and dipolar rotations take precedence (Tang et al., 2005). 2.1.3.3 Interfacial Polarization  Interfacial polarization, alternatively known as the Maxwell/Wagner/Sillars (MWS) polarization or β-dispersions, occurs in heterogeneous mixtures, where charges accumulate at the interphase of components with different dielectric properties (Kremer and Schonhals, 2003).  More specifically, changes in the material‟s conductivity will result in a build-up of charge along the boundary when an external electric field is applied, as illustrated in Figure 4.  12    Figure 4 – Interfacial polarization of a heterogeneous material: A) with no electric field, and B) with a static electric field In essence, these charge accumulations form micro capacitors that primarily affect the ε‟ term by increasing the amount of energy that can be stored in the material, although they also contribute to ε” via the      term in Equation 4.  The effects of ionic conductance typically overshadow this contribution, especially for highly conductive mixtures.  However, interfacial polarization is believed to play a significant role in biological systems, such as tissues and blood (Kuang and Nelson, 1998). Similar to ionic conductance, this process is seen at radiofrequencies in the general range of 10KHz to 10MHz.   2.1.4 Dielectric Interactions with Biological Systems Biological and living systems are difficult to model and describe dielectrically due to their highly heterogeneous nature and tendency to change with time.  Although dielectric data has been reported in the literature for various biological substrates, data is specifically limited for WAS and DM.  What is known about these substrates and how the previously described heating mechanisms interact with biological systems in general, will be reviewed in this section. 2.1.4.1 Cells and Tissues Interfacial polarization (Section 2.1.3.3) occurs in biological systems due to moving ions that become blocked by cell membranes, allowing a build-up of charge.  In contrast, ions that are in the bulk intercellular media, or able to pass through the cell‟s membranes are considered to be part of the ionic conductance phenomena (Section 2.1.3.2).  Cells themselves are electrically non-conducting due to their membranes, but their cytoplasm is known to be rich in ions and thus, highly conducting.  In fact, even the movement of ions within cells (before being blocked by membranes) would contribute to ionic conductance.  This makes it highly likely that the ionic conductance process is able to dissipate thermal energy inside the cell.  If this is the case, it would be the likely explanation for the superior cell lysing A) B)  E=0   E  13  results observed with MW system‟s compared to that of conventional heating methods (Bobowski et al., 2012; Hong et al., 2004).  The amount of free ions in the biological system is therefore an important factor, where a small change in their concentration can have a large impact on their dielectric loss factor (Kuang and Nelson, 1998).  Conversely, ions that are bound and not able to move would not significantly affect the sample‟s dielectric properties.  Because ionic conductance dominates in the RF range, it is in practice difficult to distinguish the relative contribution between this process and that of interfacial dispersion to ε”.   In the RF range, the cell density, microstructure, and amount of extracellular polymeric substance (EPS) would largely dictate which heating mechanism prevails and to what extent.  Several studies, summarized by Kuang and Nelson (1998), have looked at the impact of water content on the magnitude of interfacial dispersions.  For wheat and plant leaves, it was observed that lower water contents would reduce this mechanism. An explanation was proposed claiming that less water decreased the thickness of the electrolyte and thereby reduced the amount of charge that could accumulate.  In terms of WAS, this suggests that for very high solids contents, heating via interfacial polarization may be suppressed.  Other experiments conducted on blood, which contains fewer membranes, water content had a negligible impact.  For very high blood concentrations though, interfacial dispersions were increased due to the additional blocking of otherwise free ions.  This system would be somewhat analogous to DM, which can be described as a system of loosely bound particles due to the absence of a significant EPS fraction.  In this case, higher solids content may actually improve interfacial dispersions.   Lysing cells will also substantially reduce interfacial dispersions as the lack of intact membranes can no longer block the movement of ions.  In fact, it has been proposed that the resulting decrease in ε can be measured and used as an indicator for cell viability in WAS samples (Pajoum-Shariati et al., 2013).  It is reasonable to assume however, that the breakdown of cell structures must grant ions greater mobility, and in turn, increase the sample‟s ionic conductance, thereby partially offsetting the loss of interfacial dispersions.  The implications for the MW/RF-H2O2 process are that as the cells become lysed during a treatment run, the dielectric properties of WAS may become unfavourable for heating.  Additionally, it has been suggested that interfacial dispersions induced by an alternating electric field can further lyse cells by imposing an electric potential large enough to induce dielectric breakdown (Markx and Davey, 1999).  Consequently, a current would be allowed to flow freely through the cell in order to negate the voltage gradient, while damaging the cell‟s internal structures in the process.  In the microwave frequency range (i.e. generally above 1GHz), dipolar rotations govern the dielectric loss term.  Since water has a strong response to dipolar rotations and is the most common polar molecule in 14  biological systems, the sample‟s water content will have a large impact on its ε” term; higher water content results in a larger loss factor.  Since the water is selectively heated, the biological components must then be heated via conductance or convection, suggesting no advantage compared to conventional heating methods (Bobowski et al., 2012).  Whether the water is free to rotate or bound will also impact the dielectric properties.  It should be noted that cells by themselves do not inhibit the movement of water as they are semi-permeable to water molecules and allow their transport based on osmotic pressure differences (Kuang and Nelson, 1998).  Thus, it is the EPS that largely dictates the fraction of bound water.  Bound water‟s loss factor will peak at a lower frequency than that of free water (i.e. less than 19.5GHz), and will have a suppressed dielectric loss factor that reduces its ability to be dielectrically  heated (Barba and D‟Amore, 2002).  This distinction is recognized by referring to the dielectric properties of bound water as δ-dispersions.  Other studies have presented strong evidence that δ-dispersions are especially caused by protein bound water (Pethig, 1984).  Furthermore, proteins themselves are selectively heated due to their ability to be polarized, and any proteins on the surface of the cell membrane would also experience concentrated heating effects.  This may provide another explanation for the enhanced cell lysing results observed with MW heaters, as previously discussed.  To summarize the above dielectric heating mechanisms,  an idealized spectra reproduced with permission from Markx and Davey (1999) is provided in Figure 5, showing the ε” contributions from  interfacial polarization (β),bound water dipolar rotations (δ), and free water dipolar rotations (γ).  Figure 5 – Idealized dielectric dispersion spectra of a biological system ©Markx and Davey, 1999 2.1.4.2 WAS with High Solids Content One study by Bobowski et al. (2012) was found in the literature that reported the dielectric properties of thickened waste activated sludge (TWAS) and sludge cake, with TS values of 4.5 and 18%, respectively.  15  Examining the published ε‟ data, the TWAS was equivalent to that of pure water at microwave frequencies, including 915MHz, while the sludge cake with its higher solids content had a strong negative deviation.  At the lower radiofrequencies, both substrates had significant positive deviations relative to that of pure water due to the rise of interfacial polarizations.  Similarly, the sludge cake exhibited a larger change than the TWAS from its higher solids content. The          ⁄  terms were measured and reported as one value, and showed once again that the TWAS behaved nearly identically to that of pure water in the microwave range.  For the sludge cake sample, it yielded a negative deviation from water due to the greater proportion of bound water, creating δ-dispersions.  At 915MHz however, the overall ε” was actually increased as this frequency is at the lower end of the microwave range.  Importantly, this means that the higher the solids content of the substrate, the faster it will be heated dielectrically at 915MHz.  In the RF range, both substrates‟ total ε” functions linearly increased with decreasing frequency, matching well with a fitted     ⁄  model, suggesting that the ionic conductance was primarily responsible for the observed trend.  It should be noted that the effect of surface polarization may also be contributing to the      component, particularly for the sludge cake sample, which followed a weaker linear trend compared to the TWAS.  The substrate with higher solids content yielded the larger lost value, presumably due to greater ion concentrations, although enhanced surface polarizations cannot be discounted.  The approximate dielectric properties are summarized below in Table 1 for each EMR frequency investigated in this work (Bobowski et al., 2012). Table 1 – Dielectric properties of activated sludge samples at 27 and 915MHz Substrate Frequency (MHz) εr' (1) εr” (1) tan(δ) (1) dp (cm) TWAS 27 85 220 2.6 10 915 80 12 0.15 4 Sludge Cake 27 100 450 4.5 7 915 65 22 0.3 2 Note:  permittivity values are relative to the permittivity of free space, and therefore unit-less  It is worth reiterating that the loss term is higher at 27MHz, as clearly shown in Table 1, meaning that the substrates are substantially easier to dielectrically heat at lower frequencies.  Additionally, the loss ratio (tanδ) is larger at 27MHz, suggesting efficiency gains at this frequency as a higher fraction of the absorbed energy will be converted into heat, rather than stored in polarized molecules.  The last column in Table 1 reports the EMR‟s calculated penetration depth using Equation 7.  Although larger penetration depths are estimated at 27MHz, both frequencies‟ penetration depths are low due to the lossy nature of the substrates. 16  2.1.5 Effect of Temperature on Dielectric Properties To the best of the author‟s knowledge, no studies have yet been reported that investigate the effect of temperature on the dielectric properties of WAS or DM, an area that requires attention due to the well-known temperature dependency of ε” and the large temperature ranges involved in the MW/RF-H2O2 process.  However, the temperature dependency of water‟s dielectric properties is well understood and can be used as a guide, particularly in the microwave range where the dipolar rotation loss mechanism dominates.  Presumably, the temperature-behaviour of water is likely applicable to WAS and DM with low solids contents that yield a high fraction of unbound water.  From data presented by Barba and D‟Amore (2002), it can be seen how ε‟ is reduced at lower frequencies, including 27 and 915MHz, with increasing temperature.  Similarly, the ε” peak in the microwave frequency range shifts towards higher frequencies and decreases in magnitude as the temperature is raised.  At 915MHz, the result is a decreasing loss factor at higher temperatures.  Bound water (δ-dispersions) will likely cause an increase in a substrate‟s ε‟ but a decrease in its ε” values at 915MHz.  This is expected because as the substrate is heated, the bound water obtains a greater degree of mobility, allowing it to behave more like free water.  To summarize in the context of dielectric heating at 915MHz, the substrate will most likely be harder to heat at higher temperatures due to the behaviour of bound and unbound water.   In the RF range, governed by ionic conductance and interfacial dispersions, pure water can no longer be used as a guide as it is nearly lossless in this bandwidth.  Instead, well studied salt solutions (NaCl) can be used to make predictions for the temperature dependency of waste water‟s dielectric properties.  Examining data presented by Barba and D‟Amore (2002) reveals that the overall ε” increases with higher temperatures due to enhanced ionic conductance dispersions; this is in stark contrast to the behaviour of polarization mechanisms (εd”).  It can be hypothesized that as the substrate is heated, ions that were previously bound or restricted in their movements would become increasingly mobile and start to contribute to ionic conductance losses.  In summary, the ability to heat waste waters at 27MHz is likely to improve as the temperature rises. 2.1.6 Impedance Matching This section will present the basics of impedance theory in the context of dielectric heating applications and their design, summarizing from Orfanidis (2014). 2.1.6.1 Theory of Impedance In order for EMR energy to be transmitted into and absorbed by a dielectric material, the impedance of the incoming EMR must be equal to that of the load.  Impedance is a complex value that quantifies the 17  resistance offered to any property exhibiting wavelike behaviour.  In the context of dielectric heating, the impedance (Z) can be mathematically defined for an EMR wave propagating through a dielectric material (Equation 9) or for an alternating RF electrical current in a conductor (Equation 10), both having units of ohms.    √         (9)          (          )     (10) For Equation 10, V and I are the sinusoidal voltage and current, respectively; Rr is the real resistive part of a circuit; the imaginary term is referred to as the reactance and is made up of the inductance (L) and capacitance (C).  L quantifies the opposition to a changing current due to the conductor‟s magnetic properties, while C represents the amount of electrical charge that becomes stored.  In practice, EMR is transferred as waves through a waveguide in the MW range, and as an alternating current in a transmission line for the RF range, hence the two sets of equations.  The characteristic impedances (Z0) of both these components are provided in Equations 11 and 12, respectively.       √  (    √  )      (11) Equation 11 is for a waveguide operating with a TE10 mode, the most common configuration, where a is the width of the waveguide.    √                     (12) In Equation 12, G is the conductance of the transmission line‟s material. 2.1.6.2 Impedance and Energy Efficiency As mentioned above, if the load, transferring equipment, or generator, have different impedances, a percentage of the energy will be reflected back towards the source, producing a standing wave. This reflected power not only represents an energy loss, but can also damage equipment if not directed away from the generator.  The difference in impedances dictates the reflection coefficient, Γ, which is the ratio of the reflected wave‟s amplitude to that of the incident wave‟s (Equation 13).               (13) 18  ZL and ZS are the impedances of the load and source, respectively.  However, the impedances themselves are rarely directly measured.  Instead, it is more practical to measure the standing wave ratio (SWR), defined as the ratio of the resulting standing wave‟s maximum amplitude (antinode) to its minimum (node).  The SWR in turn can be related to the magnitude of Γ using Equation 14. | |                  (14) Equation 14 applies to both RF and MW systems, except for RF AC currents, where the voltage standing wave ratio (VSWR) is used in place of the SWR.  The VSWR is analogously defined as the ratio of the current‟s maximum AC voltage to its minimum.  Most importantly, the square of the SWR (or VSWR) can be used to determine the fraction of reflected power (Pr), and hence, the energy efficiency of the applicator (Equation 15).    | |  (          )             (15)  Again, Equation 15 applies equally to both RF electrical and MW systems, except that VSWR is used in the former.  2.1.6.3 Tuning The generator and waveguide (transmission line for RF systems) have static impedances, allowing these components to be designed such that their impedances are equivalent.  To match the impedance of the applicator and load to the rest of the system, also referred to as „tuning‟, a capacitance (C) is added such that the applicator‟s impedance is the complex conjugate of the generator‟s (Metaxas and Meredith, 1983), via Equations 9 to 12.  For most loads, the impedance is non-constant due to its changing dielectric properties; thus, a variable capacitance device must be used to maintain acceptable efficiencies.  In a waveguide, capacitance is often effected by inserting large metal screws (see Section 2.4.1.4); for RF heating systems, it is most common to use a parallel plate capacitor separated by a variable distance that is wired into the transmission line (see Section 2.4.2.4).  Industrial dielectric heaters may have their SWR, or VSWR, constantly monitored and the capacitance automatically adjusted accordingly to compensate for the changing impedances.  There are several other impedance matching networks or electronic configurations worthy of mentioning: quarter wavelength transformers, stub and line networks, and lumped L networks (Sorrentino and Bianchi, 2010).  However, they all work on the same principal, that is, by either changing the system‟s inductance or capacitance.  19  2.2 Chemistry of the MW/RF-H2O2 Process 2.2.1 Reactivity of Hydrogen Peroxide Hydrogen peroxide (H2O2) is a powerful reactant that is capable of attacking and degrading many organic substances.  Although it has a high oxidant potential of 1.76V (Jones, 1999), by itself, it actually behaves as a weak oxidant and in fact can act as a reductant depending on the co-reactants involved (Walter et al., 1955).  To achieve its full oxidation potential, the H2O2 needs to be activated by a catalysis in order to form radical chemical species, primarily the hydroxyl radical (*OH).  The *OH is an even greater more powerful oxidant, capable of rapidly and non-selectively degrading most organic molecules.  Thus, the organic reactions that H2O2 is capable of can be classified into two broad categories:  1) Direct attack by the H2O2 molecule or one of its deprotonated species (Section 2.2.1.1) 2) Activation of H2O2, typically by a catalyst, to produce *OHs (Section 2.2.1.1) The mechanisms for either of the above classifications typically require the H2O2 to transform into a more reactive form, such as OH+, HO2*, or HO2- (Walter et al., 1955).  This means that H2O2‟s reactivity is dependent on its ability to act as a weak acid (kd = 1.78x10-12), which dissociates according to Equation 16 (Zeronian and Inglesby, 1995). H2O2  H+ + HO2-                       (16) Higher temperatures will increase the dissociation of H2O2, although the pH of the solution will ostensibly have a larger effect on the extent of dissociation (Zeronian and Inglesby, 1995).  Under basic conditions, the HO2- species is favoured, whereas acidic conditions promote H2O2 and the OH+ species via Equations 17 and 18 (Mikutta et al., 2005).  H2O2 + HO-  H2O + HO2-      (17) H2O2 + H+  H2O + OH+     (18)  In many microorganisms, H2O2 is prevalent in minute dosages and plays a role in various biological processes.  Consequently, the enzyme catalase is also ubiquitous in nearly all plants and animals, which rapidly decomposes H2O2 (kd = 4x107M-1) directly into O2 and H2O (Mikutta et al., 2005).  This catalase, presumably present in WAS, has the potential to significantly reduce the efficiency of H2O2. 2.2.1.1 Direct Organic Reactions H2O2 can react directly with a variety of organic substances in the absence of any catalyst.  Among the first processes to take advantage of this property is the bleaching of textiles and cotton, whereby the 20  addition of H2O2 destroys conjugated chromophoric groups that cause discolouration.  Mechanistically, H2O2 can act as an electrophile due to its highly polarizable peroxide (O-O) bond, or behave as a nucleophile, particularly when it becomes deprotonated in high pH environments (Jones, 1999).  Unactivated reactions with common functional groups and their resulting products are summarized in Table 2. Table 2 – Summary of reactions between unactivated H2O2 and common  Substrate`s Functional Group Characteristics of Reaction Product Alkanesa Slow Oxidation  Alkenesb Nucleophilic attack by HO2- or attack after the formation of peroxy acids (see carboxylic acids) Epoxies  Carboxylic acidsb  - Peroxy Acids Aldehydesb Nucleophilic attack Hydroxyl alkyl peroxides and other acids Nitrilesb - Amides Disulphidesb - Thiols and Sulfenic acids Heterocyclicb Rapid Oxidation of double bonds Carbonylsa Nucleophilic attack Hydroperoxides a) From Jones (1999) b) From Walter et al. (1955)  Of importance for the MW/RF-H2O2 process is the degradation of proteins.  This occurs through a complex series of reactions that is dependent on the protein`s structure and side chain functional groups.  Typical end products include ammonia, as well as various ketones and aldehydes (Walter et al., 1955).  In the case of bleaching, the exact mechanism is still debated in the literature.  Strong evidence has been presented indicating that the HO2- species is responsible at basic pHs, while H2O2 itself reacts directly with chromophore groups under acidic conditions (Spiro and Griffith, 1997).  Cellulose can also be degraded by H2O2 via HO2- nucleophilic attack (Zeronian and Inglesby, 1995).  Generally, these direct reactions proceed more slowly than those resulting from activated H2O2, as discussed in the proceeding section.  2.2.1.2 Formation of Hydroxyl Radicals from Hydrogen Peroxide The decomposition of H2O2 to produce *OH intermediates is a well-accepted phenomena.  Shown in Equation 19, the decomposition mechanism requires the peroxide bond to be evenly broken apart to create two *OHs, with a rate constant of kd (Walter et al., 1955).   (M) + H2O2  *OH + *OH + (M)       (19) 21  M from Equation 19 is included for liquid reactions as a collision partner, typically assumed to be a water molecule.  The peroxide bond has a dissociation energy of 213kJ/mol, making it weaker than C-C, O-H, or C-H bonds (Jones, 1999).  However, a catalysis or external source of energy is required to overcome the peroxide bond‟s activation energy, for example: high temperatures, ultraviolet light, ozone (peroxone process), low oxidation transition metals such as Fe2+ from the Fenton process (von Sonntag, 2008), or even metal reactor walls and other non-metal surfaces (Hiroki and Laverne, 2005; Walter et al., 1955).   Furthermore, a *OH is capable of rapidly reacting with H2O2 to form another, less reactive radical, *HO2 (Equation 20). H2O2 + *OH  H2O + *HO2              (20) Due to the presence of transitional metals in sewage, it is in principal possible that some of the H2O2 is catalyzed by this mechanism.  Such catalysts operate via a homogeneous redox reaction, starting with the nucleophilic addition of H2O2 to a reduced metal and the transfer of one electron.  In addition to Fe(II/III), other transitional metals and their many complexes are also capable of acting as catalyst, for example: Cu (I/II), Ni(II), Co(II),  and Mn(II/III/IV) (Salem et al., 2000).  The reactivity of each metal depends on the pH, as well as their ligands, or lack thereof.  Generally speaking, complexed metals are more effective catalysts than their free ionic counterparts.  In the case of Fe, which has been widely studied and can be found in significant quantities in WAS, most of the necessary Fe2+ species becomes oxidized to Fe3+ when subjected to aerobic conditions, thereby preventing it from acting as a catalyst.  To exemplify this, a study found less than 0.2µM of Fe2+ per gram of organic matter from WAS, with most of it trapped in flocs and not available for reaction (Rasmussen and Per, 1996).  Furthermore, the Fenton process is acid catalyzed, requiring pHs of less than 3 to prevent the formation of Fe3+ precipitates.  Thus, at neutral pHs, *OHs are unlikely to be formed through the Fenton mechanism.  Interestingly, several researchers have reported that steel vessels and pipes can enhance H2O2 activation, resulting in a lower activation energy (Takagi and Ishigure, 1985; Zeronian and Inglesby, 1995), although the exact mechanism behind these observations has yet to be conclusively elucidated.  To summarize in the context of organic slurries, H2O2 may be activated to some degree from transitional metals and reactor walls, but its contribution to H2O2‟s overall reactivity is presumed to be minor in most cases, especially for substrates that utilize aerobic treatment processes. Alternatively, H2O2 can form *OHs without a catalysis by thermal degradation of its peroxide bond, as depicted in Equation 19 (Kazarnovskii, 1975; Takagi and Ishigure, 1985; Tessier and Forst, 1974).  H2O2 is fairly stable at room temperature and neutral pH, but thermally decomposes at higher temperatures.  22  Consequently, most H2O2 decomposition studies focus on vapour phase or supercritical conditions.  The handful of aqueous liquid phase decomposition studies are summarized below in Table 3. Table 3 – Compilation of kinetic data from aqueous H2O2 decomposition studies Ea (kJ/mol) A (1/s) T1/2 (s, at 90°C) Temperature Range (°C) Reactor Reference 71 6.4·105 1.8·103 100 – 280 Stirred tank a 54 6.4·105 60 Plug flow a 41 6.9 7.9·104 25 - 120 Pryex beaker b 49 1.6·103 9.0·102 150 - 350 Plug flow c a) Takagi and Ishigure (1985) b) Hiroki and Laverne (2005) c) Croiset et al. (1997)  The experimental data presented in Table 3 shows considerable discrepancies, particularly in regards to the pre-exponential factor (A).  As noted previously, the decomposition kinetics are dependent on the type of reactor due to the catalytic effect of surfaces, further complicating measurements.  In fact, early researchers were quick to observe how the reactor‟s material (metal or non-metal) and its preparation, impacted the observed activation energy (Walter et al., 1955).  Similarly, higher surface to volume ratios have been found to increase the rate of decomposition (Hiroki and Laverne, 2005), while Takagi and Ishigure's 1985 study concluded that their plug flow reactor yielded a lower Ea than their completely mixed batch reactor because of the former‟s greater surface area (see Table 3).  For future calculations, the recommended kinetic data that best models bulk thermal H2O2 decomposition would be A and Ea values of 6.4x105s-1 and 71kJ/mol, respectively.  The exact mechanism of H2O2 decomposition, particularly in the liquid phase, is often assumed to be a single step homolysis reaction that requires metal or surface catalysts at lower temperatures (i.e. <100°C).  However, a singular study by Kazarnovskii (1975), showed how H2O2 can decompose into a *OH at lower temperatures without a surface catalyst via its second dissociation product (Equation 21), which is in equilibrium with its first dissociation product (Equation 16). OH- + HO2-  O22- + H2O                  (21) This second dissociation product (O22-) subsequently reacts with H2O2 to produce a *OH (Equation 22). O22- + H2O2  O2- + OH- + *OH        (22) The final reaction shown in Equation 22 was considered to be the rate determining step.  In summary, the formation of *OHs from the MW/RF-H2O2 process can be divided into two temperature regimes: 23  1) Low Temperatures (<100°C):  decomposition occurs minimally via metal and surface catalysts, as well as from H2O2‟s second dissociation product 2) High Temperatures (>100°C): increased *OH formation from the thermally induced homolysis of H2O2‟s peroxide bond  Further support for the above division of mechanisms is provided by differential scanning calorimetry experiments (Wu et al., 2008), which reported that the thermal decomposition of H2O2 only starts after 67°C. In the absence of other carbon based compounds, the *OH will react with H2O2 to ultimately produce H2O and O2, as shown in Equations 23 and 24, respectively (Tessier and Forst, 1974).   *OH + H2O2  H2O + *HO2             (23) 2*HO2  H2O + O2                 (24) The reactions shown in Equations 20, 23, and 24, are likely to be outcompeted in the presence of organic wastes, especially for highly concentrated waste waters.  Because all subsequent reactions involving radicals are very rapid (see proceeding section) compared to the initial H2O2 decomposition step (Equation 19), the overall reactivity of H2O2 can be approximated, using the steady state assumption, to a first order mechanism, as shown in Equation 25.                                              (25) Exceptions to Equation 25 arise for very small and large H2O2 concentrations, where Equations 26 and 27 must be taken into consideration, respectively: *OH +*OH  H2O2                     (26) *OH + HO2-  OH- + HO2*                         (27)  2.2.1.3 Hydroxyl Radical Oxidation The formation of *OHs, and to a lesser extent other radicals, from H2O2 is desired as they are among the most powerful oxidizing species yet discovered.  They are capable of reacting with most organic compounds to form CO2, water, and inorganic salts.  Furthermore, they react non-selectively and rapidly with rate constants in the order of 106 to 1010M-1s-1 (Buxton et al., 1988).  As such, they have been identified for their potential to treat degradation resistant or non-biodegradable pollutants, including dyes, pesticides, aromatic compounds (ex. phenol), and pharmaceutical compounds, among many other 24  industrial contaminants (Wang and Xu, 2012).  In aqueous solutions, *OH reactions fall into three major categories (von Sonntag, 2006): 1) Addition to double bonds 2) H abstraction 3) Electron transfer Addition reactions (1) involve two or more reactants joining together to form new bonds.  In the context of *OHs, they can add themselves to compounds containing C=C, C=N, or S=O bonds, as exemplified in Equation 28. C2H4 + 2*OH  C2(OH)2H2     (28) Due to the electrophilic nature of *OHs, this type of reaction is regioselective, and the *OH will preferentially attack the more electron rich atom in a double bond.  The less favourable H abstraction reaction (2) involves the *OH bonding with a hydrogen atom and removing it from its parent molecule.  This leaves behind a single electron on the parent molecule, turning it into a radical (see Equation 29).   *OH + H-R  H2O + *R               (29) R can represent C, N, or S atoms.  For C-H bonds, abstractions preferentially occur in the following order: tertiary (CH) > secondary (CH2) > primary (CH3).  The newly created radical species can continue reacting in a similar fashion, creating a series of short lived radicals.  Referred to as radical chain reactions, this process essentially multiplies the degradation capacity of a *OH for a given substrate.  The last reaction, electron transfer (3), is believed to be an uncommon mechanism, and involves the direct transfer of an electron to the *OH, as exemplified in Equation 30. M + *OH  M+ + -OH                     (30) M is a metal complex that becomes oxidized after the loss of a valance electron.  The above three mechanisms allow *OHs to react with alcohol, amine, and amino acid functional groups to form acetic acids, amine oxides, and keto acids, respectively (Walter et al., 1955).   When *OHs are formed in biological systems, they are known to effectively deactivate DNA and degrade proteins through a variety of mechanisms, including the cleavage of peptide bonds (Stadtman and Levine, 2003).  EPS in WAS, as well as dairy manure, contain large fractions of structural polysaccharides that are difficult to degrade through biological treatment processes.  However, *OHs can depolymerize these 25  substances by disrupting their intra- and inter- hydrogen bonds, as has been shown with cellulose for example (Chirat and Lachenal, 1997; Wang and Gao, 2002).  2.2.2 Effect of Dielectric Heating on Reaction Rates Microwave ovens were first used by organic chemists to rapidly heat solutions and study organic reactions in 1986 (Kappe et al., 2012).  Since then, much literature has been published reporting enhanced reaction rates and skewed product distributions for reactions heated dielectrically, compared to those heated using conventional methods.  This led to much speculation of so called „specific‟, „athermal‟, or „non-thermal‟ affects, vague terms that encompassed several theoretical mechanisms that hypothesized how EMR may interact with atoms and molecules to influence their reactivity.  Such mechanisms were differentiated from those associated with thermal processes, and whose impacts on reactivity are well understood.  Following years of conflicting findings and heated debate in the literature, it is now widely accepted that most, if not all, rate enhancements observed for MW systems are solely due to thermal effects, and are not caused by any special interactions between EMR and the reactants (Baghurst and Mingos, 1992; Herrero et al., 2008; Kappe et al., 2012; Lidstrom et al., 2001; Moseley et al., 2007; Robinson et al., 2010; Shazman et al., 2007).  The source of the apparent discrepancies arise from the different ways that dielectric and conventional heaters pass kinetic energy to their substrates.  Dielectric heating, summarized in Section 2.1, selectively heats all polarizable molecules within its penetration depth.  Conversely, conventional heating relies on an external heat source that transfers kinetic energy through materials via conduction or convection.  The latter is ostensibly an inefficient method for delivering thermal energy to a substance as it requires large temperature gradients, as well as being dependent on the thermal conductivity of the substrate, solvent, and reaction vessel.  Consequently, performing comparison experiments between the two heating methods using equivalent temperatures and reactions is more difficult than was initially realized.  For conventional heating systems, the non-uniform temperature profile that arises within a substrate makes measuring a representative temperature near impossible, and often renders „average‟ temperatures meaningless.  For MW systems, the common placement of heat sensors have been shown to incorrectly measure the substrate‟s true bulk temperature (Kappe et al., 2012; Nüchter et al., 2004; Robinson et al., 2010), while inadequate stirring can disturb the electric field, resulting in temperature gradients (Moseley et al., 2007).  When extraordinary care is taken to control the temperature during a comparison study, previously observed differences disappear (Herrero et al., 2008; Kappe et al., 2012; Shazman et al., 2007).   Although the evidence for „athermal‟ or „non-thermal‟ effects have largely been disproved, the vastly different heating mechanisms inherent to dielectric ovens can be said to exhibit „specific‟ thermal effects.  26  To define more precisely, specific effects are thermal phenomena that are unique to dielectric heaters and impact chemical transformations relative to conventional heating processes.  It is these specific effects that are also partly responsible for the often observed differences between heating methods.  The major specific effects are described below: 1) Super Heating:  The ability of dielectric heaters to superheat solvents up to 40°C above their boiling temperatures under atmospheric conditions (Kappe et al., 2012).  Although still an area of active research, these observations are thought to arise from the rapid and uniform heating achieved by dielectric heaters, which inhibits the formation of nucleation sites necessary for boiling (Baghurst and Mingos, 1992).  2) Wall Effects:  Most conventional heating systems require the reactor vessel to be significantly hotter than the substrate to facilitate heat transfer.  In dielectric systems, the opposite is true, where the reactor is typically made of EMR transparent material and, therefore, of a lower temperature than the substrate. 3) Selective Heating:  Dielectric systems will preferentially heat molecules that are more polar (see Section 2.1), resulting in temperature differences between phases with different permittivity properties. 4) Rapid and Uniform Heating:  Properly designed dielectric heaters can more rapidly and uniformly heat substances compared to conventional methods, which will always require temperature gradients and be limited by their thermal conductivities.   2.3 Review of Previous Microwave and MW-H2O2 Research Much research has been conducted in the field of organic waste treatment using dielectric heaters.  To the author‟s best knowledge, the earliest such published account dates from 1977 (“Waste Water System Uses Microwaves,”), where a six step waste water treatment process used a microwave oven to disinfect dewatered solids for use as fertilizer.  At the time of this work, the use of dielectric heaters to treat a variety of organic wastes, often in combination with other technologies, is abundant in the literature.  The following review will summarize the areas of research and major conclusions in this field, with a focus on municipal sewage sludge and animal waste applications. 2.3.1 Treatment of Sludge from Municipal Waste Waters  The treatment of municipal sewage sludge being the most well studied substrate, several detailed reviews have been published that summarize the many benefits of the MW treatment process (Mudhoo and Sharma, 2011; Remya and Lin, 2011; Tyagi and Lo, 2013).  Reported benefits have been divided into two groups for the purpose of this review: digestion and nutrient recovery. 27  2.3.1.1 Benefits Pertaining to Digestion   The most common application of MWs for wastewater treatment involves the conditioning of sludge prior to an anaerobic digester (Eskicioglu et al., 2007a; Mehdizadeh et al., 2013; Tang et al., 2010).  An important goal of such a pretreatment is the production of high quality biosolids that can meet the US Environmental Protection Agency‟s (EPA) Class A designation, allowing them to be disposed of with fewer restrictions.  However, to meet EPA standards, the concentration of coliforms must be reduced to less than 1 000MPN/%TS (Walker et al., 1994).  Several studies have been published showing that MW ovens can disinfect municipal sludge to achieve such standards after they have been anaerobically digested (Hong, 2002; Hong et al., 2006; Pino-jelcic et al., 2002).  A significant finding is the selective targeting and deactivation of DNA by MWs, which often outperform conventional heating methods in terms of pathogen reduction (Kakita et al., 1995).  Many of the above sited studies have also reported other benefits from pretreating sludge prior to anaerobic digestion, including: digestion retention times shortened to that of several days; increased methane production (Kuglarz et al., 2013; Qiao et al., 2010); solubilization and hydrolysis of organics; and the enhancement of acidogenesis (Ahn et al., 2009).  Select results of these numerous benefits have been summarized below in Table 4.  These benefits generally improve at higher temperatures and with longer treatment times. Table 4 – Examples of beneficial results from MW pretreating sewage sludge prior to anaerobic digestion  Parameter Result Experimental Conditions Study Methane Production 50% CH4 increase Pretreatment at 80°C prior to mesophilic digestion Appels et al. (2013)  Solubilization of Organics 100% SCOD 80°C and 5.3%H2O2(v/v)/%TS Wong et al. (2006b) Solids Reduction 84% sludge volume reduction Anaerobic TWAS treated for 5min at 50W/g (~900°C) with 5wt% char Menendez et al. (2002) Acidogenesis Enhancement 24% of COD converted to acetic acid (CH3COOH) Two-step MW treatment at 120°C, pH 1.69, and 3.3%H2O2(v/v) Liao et al. (2007) Anaerobic Digester SRT Stable at SRTs as low as 5 days Continuous mesophilic digester pretreated at 175°C Toreci et al. (2009) Coliform Reduction 3 log CFU reduction in fecal coliforms (EPA Class A designation) Pretreatment at 85°C for <1min Hong et al. (2006)   Several studies have examined the effect of the MW treatment process on the dewaterability of municipal sludge (Wojciechowska, 2005).  Their results typically show that MW heating initially improves the sample‟s CST and other dewaterability indicators, but continual heating quickly reaches a threshold, after which the dewaterability deteriorates (Yu et al., 2009, 2010a).  Evidence has been presented that this may be due to the solubilization of loosely bound EPS at short treatment times, while additional heating causes tightly bound EPS to become loosely bound, thereby worsening dewaterability (Peng et al., 2013).  A 28  study by Eskicioglu et al. (2007b) best exemplifies the potential improvements to dewaterability, where a 40% improvement in CST was attained on anaerobic digested sludge after MW heating at 96°C. 2.3.1.2 Benefits Pertaining to Nutrient Recovery In addition to improving digestibility, many researchers have explored the recovery of nutrients from sewage sludge using MW based treatments, including: solubilization of phosphorus and release of orthophosphates (Danesh et al., 2008; Liao et al., 2005a); increasing struvite recovery (Wong et al. 2006a); solubilization of TKN and formation of ammonia (Chan et al., 2007); and the solubilization of organics as an internal source of carbon for denitrifying membrane bioreactors (Xu et al., 2015).  In terms of the MW-H2O2 process, it has been determined that temperature and H2O2 dosage are the most important factors for orthophosphate and ammonia release, respectively (Kenge et al., 2009).  Select studies and their results are summarized below in Table 5, to exemplify the proven benefits of MW treatment for nutrient recovery applications. Table 5 – Examples of beneficial results relating to the release and recovery of nutrients from sewage sludge Parameter Result Experimental Conditions Study Ammonia Formation 47% of TKN 120°C, 2.7%H2O2(v/v)/%TS, and acidic conditions Chan et al. (2007) P Solubilization 76% of TP 5min at 100°C Liao et al. (2005) Orthophosphate Formation 84% of TP 170°C for 5min with 2.5%H2O2(v/v)/%TS Liao et al. (2005a) 2.3.1.3 Combination of Microwave Treatment Process with Other Technologies The MW process has been coupled with several other technologies in an effort to further improve its capabilities.  For example, a MW-Fenton process was investigated, where Fe(II) was used to increase the formation of *OHs during a MW-H2O2 run.  However, the addition of Fe(II) was not found to improve  the solubilization of solids or nutrients when compared to a control experiment that only utilized H2O2 (Lo et al., 2008).  The combination of H2O2 and MW heating is used extensively in this work and research group; however, other oxidants and catalysts have been explored, including: ozone (Lo et al., 2015), ultrasound (Yeneneh et al., 2013), persulfate (S2O82-), UV-vis lamps, and various catalysts such as TiO2, as reviewed by Remya and Lin (2011).  Generally, the use of oxidants and catalysts improve the solubilization and destruction of organics, but it is uncertain if the additional costs they incur are cost effective.   Also widely researched is the effect of pH on the MW treatment process.  Acidic conditions, from the addition of H2SO4, were observed to increase the average particle size of the sludge and improve its dewaterability, while yielding comparable solubilization‟s of solids and nutrients (Chan et al., 2010).  Conversely, the addition of NaOH to raise the pH, was found to improve the disintegration of solids and 29  solubilization of nutrients, but at the cost of worsening the sludge‟s dewaterability (Chang et al., 2011; Chi et al., 2011; Doğan and Sanin, 2009).  For the MW-H2O2 process, it was suggested by Xiao et al. (2012) that alkali conditions catalyzed the decomposition of H2O2, thereby enhancing the degradation of organics. To the best of the author‟s knowledge, no study has yet to be published that uses RF EMR to treat municipal sludge.  Of interest, a singular study was found where RF heaters and a Fenton process were used to disinfect fecal coliforms in waste water effluent (Rodríguez-Chueca et al., 2014).  2.3.2 Treatment of Animal Wastes  Research into the MW treatment of animal wastes follows a similar line of investigation to that of sewage sludge, namely the use of MWs as a pretreatment process to either enhance the waste‟s digestibility or its resource recovery potential.  Although less studied than municipal sludge, the most common animal derived substrate discussed in the literature is dairy manure.   2.3.2.1 Benefits Pertaining to Digestibility  In relation to pretreating animal wastes for anaerobic digestion, research has been conducted on the solid (Lo and Liao, 2011) and liquid (Yu et al., 2010b) portions of mechanically separated DM.  These studies showed that higher temperatures, holding times, and oxidant dosages increased the solubilization of organic material and production of VFAs.  In the latter case, the fraction of SCOD to TCOD increased to 80% after treating at 90°C with a H2O2 dosage of 0.12%(v/v)/%TS under acidic conditions.  The biodegradability of DM, measured in terms of BOD5/COD, could also be substantially increased by a factor of 3.5 (Beszedes et al., 2012).  As such, the MW-H2O2 process has been shown to improve the manure‟s digestibility under anaerobic conditions (Jin et al., 2009), although only minimal gains were observed in another study by Chan (2013).  Interestingly, the methane potential from anaerobic digestion was not observe to increase, as would be expected (Chan et al., 2013). As was shown with sewage sludge, liquid animal wastes can be disinfected from bacteria, such as E. Coli (Niederwohrmeier et al., 1985), as well as viruses (Böhm et al., 1984) after heating to 65°C for less than a minute.  A singular study used RF dielectric heaters to successfully disinfect dairy, calf and swine wastewaters (Lagunas-Solar et al., 2005); the authors recognizing the potential for RF heaters to be used in farm applications where traditional treatment technologies would not be appropriate.  2.3.2.2 Benefits Pertaining to Nutrient Recovery The solubilization of nutrients for recovery from DM has been investigated for the following: orthophosphates, ammonia, K, Mg, and Ca (Qureshi et al., 2008).  For example, up to 80% of TP was 30  released as orthophosphate after treating at 170°C for five minutes (Pan et al., 2006).  Acidification of DM to pHs lower than 4 had a similar effect on increasing orthophosphate concentrations, typically making up most of the waste‟s TP (Lo et al., 2012).  Specifically, the use of oxalic acid not only released P, but greatly enhanced the settleability of DM and effectively removed free Ca ions that are known to inhibit the formation of recoverable struvite (Srinivasan et al., 2014b).  Building upon these studies, a successful pilot-scale system was able to increase orthophosphate levels by 50% after MW-H2O2-H+ treatment, 95% of which was recovered as struvite from a crystallizer (Zhang et al., 2015).  2.4 Design of Industrial Dielectric Heaters  Designing large throughput dielectric heaters that operate efficiently is an entire field of itself and outside the scope of this work.  For this reason, the purpose of the following section is to provide a brief overview of the technologies currently used in dielectric heaters and to highlight important considerations for full-scale applications of the MW/RF-H2O2 process.  Although theoretically the same, dielectrically heating materials in the MW or RF, range utilize different components and will be discussed separately.   2.4.1 Microwave Frequencies  Most MW frequencies, or bandwidths, are reserved for communication devices, leaving only 915, 2 450, and 5 800MHz for dielectric heaters, as agreed upon by the International Telecommunication Union (Venkatesh and Raghavan, 2004).  These frequencies correspond to wavelengths of 1 to 10cm in free space.  The 2.45GHz bandwidth is the most commonly used and the one found in household MWs, while 915MHz is typically used for larger scale applications.  The major uses for industrial MW heaters include food processing, disinfection, and heating plastics.  2.4.1.1 Magnetron There are two classes of devices for generating MW EMR, the klystron and the magnetron.  For industrial applications over 10kW, it is the magnetron that is exclusively used.  In short, a magnetron consists of a heated cathode inside a metal tube, the anode,  that has specially designed side cavities (Meredith, 1998).  The space between the anode and cathode is made to have a large voltage drop, an axially aligned magnetic field, and is kept under vacuum.  Electrons are emitted from the cathode and follow a circular path around the cavity due to the magnetic field, causing the electrons in the side cavities to resonate, thereby releasing EMR.  The size and shape of the cavities dictates the frequency of the released EMR.  The generated EMR is tapped by an antenna at one end of the magnetron that feeds into the connecting waveguide.  Magnetrons are highly efficient, typically converting 88% or more of their supplied electrical power into EMR (Meredith, 1998).  The lost energy is mainly dissipated as heat in the magnetron, hence the need for liquid cooling at industrial scales, although air cooling is also simultaneously used for other 31  components.  An estimate of the required coolant flow rate, Q, is given in L/min for water in Equation 31, based off of the magnetron‟s power rating, Prated, in kW (Schiffmann, 2006).                              (31) Consequently, magnetron failures are most often caused by the melting of its cathode, in which case, the magnetron can be refurbished for only 50 to 60% of its initial cost.  How the magnetron is operated will also impact its performance, where frequent stop/start cycling may reduce a normal lifespan by 30 to 50% (Meredith, 1998). 2.4.1.2 Waveguide A waveguide is simply a hollow metal tube, typically with a rectangular cross section, that is used to transfer EMR from a generator to the applicator.  Circular waveguides are not as common due to their greater complexity and often larger impedance mismatches (Meredith, 1998).  The EMR propagates in different „modes‟ in the waveguide, the most common of which is the TE10 (Metaxas and Meredith, 1983).  This means that the electric field‟s plane is transversal (TE) to the EMR propagating along the waveguide‟s length.  Furthermore, the waveguide‟s cross-sectional dimensions are such that only one half wavelength can develop (signified by the 10 index).  To accomplish a TE10 mode, the height is half that of its width, which in turn, is between 50 and 70% of the operating frequency‟s wavelength (Erickson, 1995).  For the 915MHz bandwidth, this corresponds to a standard cross-section of 0.248 by 0.124m (Meredith, 1998).  Energy transfer through a waveguide is highly efficient. 2.4.1.3 Applicator The applicator is essentially a modified waveguide that allows the generated EMR to irradiate into the load and transfer its power.  There are two major classes of applicators: multimode and single mode (Meredith, 1998).  Multimode uses a metal cage that houses the load and allows EMR emitting from a port to irradiate it.  Because the box‟s dimensions are several times larger than that of the operating wavelength, several modes and reflections are allowed to develop, creating standing waves.  Although simple and inexpensive to make, they often suffer from non-uniform heating distributions and lower energy transfer efficiencies. Single mode applicators, as their name suggests, are designed so that only one mode of the EMR‟s wave can develop.  They are available in several different configurations depending on the load‟s shape.  Axial travelling applicators have the load continuously moving along the propagating EMR‟s axis.  For liquid loads, the flow should be in the same direction as the EMR‟s wave to improve stability. Transverse traveling applicators have their substrates crossing the EMR‟s path, perpendicular to its axis.  The 32  simplest design for this configuration has a horizontal slit cut through the waveguide‟s shorter wall to allow the load to pass through.  The slit is located where the wave‟s electric field peaks to maximize energy transfer.  A variation of this would be the serpentine design, which has the load travelling through multiple slits in a single waveguide that doubles back on itself.  Lastly, a horn, or E field applicator is a waveguide that has its shorter dimension elongated.  This increases the electric field‟s cross section, while decreasing its power density; however, the uniformity of the electric field remains constant.  By using two opposing and synchronized horns, the system‟s efficiency can be improved by cancelling any standing waves.  Schematics of these applicators and their corresponding benefits and disadvantages are summarized in Table 6.   33  Table 6 – A comparison of common microwave applicator designs Type Schematica Advantagesb Disadvantagesb Axial TE10 Waveguide    Inexpensive; uniform transversal heating, good for loads with variable ε properties; can by highly efficient Load size limited by waveguide; requires loads with large tan(δ) values Transverse TE10 Waveguide    Simple; inexpensive Reflected EMR from load create inefficient standing waves; load size limited by waveguide Transverse Serpentine    Very high efficiencies are possible (>90%) Compounded reflected standing waves make tuning difficult Transverse Horn    Uniform heating of large loads; can have high efficiencies (<95%) with dielectrically constant loads Standing waves arise (can be cancelled using two synchronized horns); susceptible to changing ε properties that give rise to large reflected waves a) Black and red arrows show the pathway of the load and EMR, respectively  b) Information taken from Metaxas and Meredith (1983), and Meredith (1998)  Many other applicator designs exist, but the several presented here are the most appropriate for liquid loads.  Reviewing the above designs, it is recommended that future, large-scale MW-H2O2 systems 34  consider an axially travelling waveguide applicator, operating under a TE10 mode.  This configuration was chosen due to its low costs, high attainable efficiencies, and the compatibility between typical pipe diameters and standard 915MHz waveguide cross-sections (Metaxas and Meredith, 1983).  Even more importantly, these applicators are not as sensitive to changes in the load‟s dielectric properties and therefore, should suffer less from reflected power losses (Meredith, 1998).  Such losses were found to be a serious problem with the current 915MHz MW, which utilizes a horn type applicator (see Section 5.7.1).  Alternatively, the horn configuration would be an appropriate candidate for waste water treatment applications that require a modest temperature increase, thereby limiting the changes in the substrate‟s dielectric properties.   2.4.1.4 Tuning For low power MWs, multiscrew tuning rods are used to match the impedances of the load and generator (Meredith, 1998).  The screws, coming in sets of one to four, are typically made of brass and inserted into the waveguide or applicator at set wavelength intervals (1/2, 1/4, 1/8, etc.).  They can minimize the amount of reflected power by changing the MW‟s capacitance to compensate that of the load‟s (see Section 2.1.6 for theory).  Although inexpensive, their ability to tune the system may be physically limited and dependant on the experience of the operator.  As well, they require constant adjustment for loads with variable dielectric properties.  If they are inserted more than halfway through the waveguide, their electrical behaviour changes to that of an inductor, which can quickly cause arcing and mechanical resonating.  Larger scale systems would require expensive automatic tuning instruments to keep the impedances matched.  Because there will always be some reflected power, it is necessary to have a circulator that can divert the returning EMR into a water reservoir.  2.4.2 Radiofrequencies  RF is generally defined as occupying the bandwidth between 1 and 300MHz on the electromagnetic spectrum, corresponding to wavelengths of 1 to 300m in free space.  Due to the popular use of these frequencies for communication purposes, primarily as AM/FM radios, the following bands have been reserved by the International Telecommunication Union for non-communication purposes: 6.78, 13.56, 27.12, and 40.68MHz (Tang et al., 2005).  Industrial applications of RF heaters include: drying lumber and curing wood glue; drying of stock textiles; welding plastics; and in the food processing industry, for baking and post-bake drying of edible products (Schiffmann, 2006). 2.4.2.1 Generating the Radiofrequency Electromagnetic Radiation There are two common electronic designs used by industrial RF heaters (Tang et al., 2005; Taylor et al., 2003).  The older design has a transformer and rectifier that are used to increase the supplied power‟s 35  voltage and convert it into a direct current.  The high voltage current is then sent to a triode valve, which is part of an oscillator circuit that self-excites the current and makes it alternate at the desired RF.  Because the applicator is part of the generator‟s circuit, the power sent to the load can be controlled by simply adjusting the position of the applicator‟s electrodes (see Section 2.4.2.3).  This would change the strength of the electric field in the applicator, and in turn, affect the amount of dissipated power.  The second and more modern design utilizes a crystal oscillator and a triode valve to generate a current that alternates at the desired RF.  As this current is of low voltage, an amplifier is required to increase its power.  To connect the applicator to the triode, a co-axial cable is often employed to deliver the high power RF current (see following section).  The impedance of the current leaving the amplifier is permanently set at 50Ω and an impedance matching network on the applicator‟s circuit equates the load‟s impedance to this value.  In this case, the oscillator is permanently set at a desired frequency, as is the electrodes‟ position. A triode‟s construction is similar to that of a magnetron‟s, whereby a heated cathode is positioned inside a metal tube serving as the anode.  A wire grid is also placed between the two electrodes and the entire device is enclosed in a vacuum sealed shell.  When the cathode is heated, electrons travel to the anode at a rate that is dependent on the grid‟s applied voltage.  If the grid has a negative voltage, the escaping electrons will be repelled and the current through the anode reduced, allowing the power sent to the applicator to be controlled.  High power triodes are approximately 55 to 70% efficient, although newer solid-state amplifier models are up to 80% (Chindris and Sumper, 2012).   2.4.2.2 Co-axial Cable Since the RF heater used in this work is of the second design described in the preceding section, the co-axial cable will be discussed briefly here.  Unlike MWs, RF wavelength‟s are too long for waveguides, and therefore, typically use co-axial cables for transmission.  These cables consist of a metal wire, or bar, encapsulated inside a concurrent metal tube, with the annular space filled with a dielectric material, such as air.  Co-axial cables are generally not as efficient as waveguides (Metaxas and Meredith, 1983).    2.4.2.3 Applicator Electrodes Considering that RF heaters generate and deliver their energy as an alternating current, their applicators must be able to produce an electric field that can irradiate their intended loads.  This is accomplished by using a capacitor, comprised of two metal electrodes constructed such that they sandwich the substrate (Tang et al., 2005).  The „active‟ electrode is connected to the co-axial cable, while the second electrode is grounded.  In many cases, the outer shell of the RF oven functions as the ground electrode.  There are a 36  wide variety of capacitor applicators, but the types most appropriate for liquid loads are highlighted below.  The simplest type of applicator is the through-field electrode configuration, where two flat metal plates are set opposite one another.  Very easy and inexpensive to make, they also have the benefit of being able to heat large volumes of substrates with a variety of shapes, such as coiled piping.  If a substrate‟s surfaces are not parallel to the plates however, tuning and efficiency problems may arise.  Another variant of this design is the concentric through-field configuration that uses two cylindrical tubes as the electrodes, with the active electrode set inside the grounded one (Taylor et al., 2003).  This arrangement is especially suitable for liquid substrates that can flow through the annular space between the electrodes.   A second type of applicator is the staggered through-field configuration that consists of a series of off-set electrode rods, as depicted in Figure 6.  The substrate would pass between the upper and lower set of electrodes and through the elliptical shaped electric fields that would arise between them.  This type of applicator can produce large power densities and high heating rates for loads of an intermediate thickness (Tang et al., 2005), such as a rack of pipes.    Figure 6 – A staggered through-field applicator for RF heating systems When designing applicators, there are several rules of thumb to consider, as discussed by Wilson (1987).  Firstly, the electrode‟s dimensions should not exceed 10% of the operating wavelength to avoid a non-uniform voltage distribution over them and uneven heating of the load.  The electric field can be described as field lines that develop perpendicular to conductors, but bend around dielectric materials.  To ensure efficient energy transfer to the load, its surfaces should be positioned such that they are perpendicular or parallel to the electrodes and electric field lines.  The electrodes do not necessarily have to touch the substrate, as any air gaps would absorb negligible amounts of energy; however, an air gap may change the shape of the electric field around the load.  If this is found to suppress the heating rate, metal plates that directly touch the substrate can be used to focus the electric field through the load.  Lastly, there is a maximum electric field strength that can be generated between the electrodes, and 37  therefore, a maximum heating rate.  This limit is represented by the dielectric breakdown of an insulating material.  Once past this point, electrical resistance rapidly deteriorates, allowing a current to flow freely through the insulator.  Generally referred to as arcing, this phenomenon must be avoided as it is a safety hazard and can damage equipment.  The dielectric breakdown of air is 3x106V/m, although it may be lower for other substances found in the applicator (Schiffmann, 2006).  The presence of sharp corners or metallic dust in the applicator can also increase the risk of arcing by altering the shape of the electric field.  2.4.2.4 Tuning  When an applicator is first installed, its impedance is roughly matched to the generator through the use of stubs (Wilson, 1987).  Stubs are coiled strips of copper, or any other conducting material, with set dimensions that connect the electrodes.  Although this may appear to be short circuiting the system, due to the high frequencies involved, the stubs act as inductors and can match the applicator‟s impedance by affecting its reactance.  They can also be used to reduce the variation in voltage across an electrode by creating local voltage minimums, resulting in a more uniform heating rate.  To more exactly match the applicator‟s impedance, either a variable inductance device, or a capacitor with a variable electrode separation, is wired into its circuit.  In some cases, the applicator‟s electrodes are repositioned to change its capacitance directly.  For industrial scale systems, it is recommended to have an automatic matching network to maximize efficiency, especially for loads that have highly variable dielectric properties.   2.4.3 Comparing Microwave and Radiofrequency Dielectric Heaters It is difficult to construct a proper comparison between RF (27MHz) and MW (915MHz) heaters, as their efficiencies and costs depend on a variety of factors.  Additionally, economic studies are scarce and quickly become obsolete as technologies evolve.  A reasonably representative comparison of the two systems is presented in Table 7, using the most recent data available from the literature. Table 7 – Comparison of MW and RF components used for dielectric heating Parameter MW (915MHz) RF (27MHz) Generator Type Magnetron Power Triode Tube Maximum Power Output (kW)a 100 900 Maximum Power Densities (kW/m2)a 500 200 Lifespan (h)a 5 000 – 8 000 5 000 – 10 000 Approximate Capital Cost ($/kW)b 7 000 – 10 000 3 500 – 10 000 a) Data taken from Chindris and Sumper (2012)  b) Data taken from Schiffmann (2006)  38  Overall, it can be generally stated that RF generators are less expensive, commonly half the capital cost of their MW counterparts (Chindris and Sumper, 2012), and are capable of delivering much more power (Zhao, 2006).  Another benefit of a RF system is that power is not dissipated unless a load is present, thereby negating the need for a water reservoir to capture reflected EMR.  Both types of generators have comparable lifespans and their cost per rated kW decreases with higher power capacities.  Advantages of MW generators include their ability to generate greater power densities, and hence, higher heating rates, as well as their superior energy efficiencies.  However, overall system efficiencies at either frequency are comparable, where the superiority of one over the other often depends on the applicator‟s ability to effectively transfer energy into its load.  Hence, total energy efficiencies are typically quoted in the 50 to 70% range for both frequencies (Chindris and Sumper, 2012; Erickson, 1995; Marra, 2012).  As for benefits conferred by each frequency in and of itself, the significantly longer RF wavelength at 27MHz allows for greater volumes to be heated, while typically producing more spatially uniform heating rates (Chindris and Sumper, 2012).  This is balanced by a MW‟s shorter, but higher energy containing wavelengths, which are the reason for its ability to realize faster heating rates.  In summary, after weighing the pros and cons of each system, a RF dielectric heater would be the recommended choice for future projects, although both systems should be considered, on a case by case basis.     39  3.0 Methods and Materials The organic slurries investigated (Section 3.1), as well as the physical (Section 3.2) and chemical (Section 3.3) analyses performed on them, are described below.  For a detailed overview of the equipment used to carry out the MW/RF-H2O2 experiments and their standard operating procedures, refer to Section 4.0.  Analysis for the physical and chemical tests followed Standard Methods for the Examination of Water and Wastewater (1999), unless otherwise noted.  All tests were measured in triplicates using distinct samples.  A data point was dropped if the resulting average had a standard deviation higher than 10% and if it differed by 50% or more from the other two measurements.   3.1 Substrates Investigated 3.1.1 Waste Activated Sludge All WAS samples used in this work were taken from UBC‟s pilot WWTP (Department of Civil Engineering, Vancouver).  The plant is operated as an enhanced biological phosphorus removal (EBPR) process, combined with an aerobic membrane bioreactor (MBR), as depicted in Figure 7.  As such, the system recycles sludge between an anaerobic, anoxic, and aerobic zone and runs without a primary or secondary clarifier.  Two independent treatment trains operate with different solid retention times (SRT), but are otherwise identical.  Figure 7 – Process flow diagram of the pilot-scale WWTP that provided the WAS samples  Samples were taken from the same valve on the membrane‟s tank that is used to control the process‟s SRT by removing residual sludge.  The collected WAS, from either one or both trains, was well mixed and experimented on within 2h to preserve its characteristics.  The MBRs had SRTs that ranged from 20 to 80d over the course of this work, due to other ongoing research projects.  A dissolved oxygen content of 1ppm was maintained in the MBR and acetate was regularly added as a carbon source. 40  3.1.2 Dairy Manure DM samples were obtained from the UBC Dairy Education and Research Centre in Agassiz, British Columbia.  The research centre is a fully operating farm with 250 lactating cows and 250 calves, heifers, and dry cows.   Total wastewater flows are approximately 38m3/day, yielding 47 000L of wastewater per year, per lactating cow.  The animal‟s manure, as well as water used to wash the milk parlour, is collected and directed to a tank before passing through a liquid-solid separator unit (Daritech Inc, USA).  The separator is composed of two stages: a Sand Cannon that recovers sand used for bedding from the whole manure, and a solid-liquid separator (model DT360) that processes the remaining sand-free DM.  The final liquid and solid waste streams have expected TS values of 4 and 20%, respectively.  The liquid portion gets pumped to a well-mixed open top tank, before being pumped to larger holding tanks.  Samples of fresh liquid DM were collected straight from the effluent pipe entering the open tank.  Care was taken to minimize the collection of the aged bulk manure from the open tank.  Samples were stored at 4°C and well mixed, before conducting experiments. 3.1.3 Palm Oil Mill Effluent Samples of palm oil mill effluent (POME) were obtained from an active palm oil processing facility in Malaysia.  After receiving the samples, they were stored in sealed containers at 4°C.  Initial characterization revealed no chemical or physical differences between any of the samples provided. 3.2 Physical Analysis  3.2.1 Extraction of Soluble Fraction The soluble fraction of a sample was extracted to allow for additional analyses.  This was accomplished by first centrifuging the sample at 3 000 and 15 000rpm for WAS and DM, respectively.  The supernatant was then vacuum filtered using a 0.45µm nitrocellulose filter (09-719-555, Fisher Scientific) and the resulting effluent collected.   3.2.2 Total Solids Fractions (TS/TVS) Approximately 10mL of sample were added to pre-weighed aluminum cups and placed overnight in an oven set at 105°C.  The full cups before and after heating were weighed to determine the sample‟s total solids fraction (TS).  The total volatile solids (TVS) were measured by placing the samples in a muffler oven at 550°C for two hours and reweighing.  41  3.2.3 Suspended Solids Fractions (TSS/VSS) The 70mm diameter glass microfiber filters (1827-070, GE Healthcare Life Sciences) were washed by rinsing with three 10mL volumes of distilled water under vacuum.  The filters and their corresponding aluminum containers were placed in a muffler oven at 550°C for two hours and weighed.  To determine the total suspended solids (TSS), exactly 2 or 3mL of sample were vacuum filtered on the washed filters and rinsed with 10mL of distilled water, before heating at 105°C overnight and weighing.  Afterwards, the samples were placed in a muffler oven at 550°C for two hours and reweighed to find the fraction of volatile suspended solids (VSS). 3.2.4 Particle Size Distribution (PSD) A Mastersizer 2000 (Hydro2000S, Malvern Instruments) light scattering particle counter was used to fractionate the solids in a sample.  The instrument recorded particle sizes ranging from 0.010µm to 10 000µm and with an obscuration limit between 15 and 25%.  Each distinct sample was recorded as an average of three replicate measurements, and each experimental sample was then replicated three times.   3.2.5 Capillary Suction Time (CST) A Komline-Sanderson capillary suction time instrument was used with chromatography paper (17CHR, Whatman) and sample volumes of 5mL.  Replicates of three or more were taken as required by the quality of the data.  3.3  Chemical Analysis 3.3.1 Chemical Oxygen Demand (TCOD/SCOD) The closed reflux colorimetric COD method (5220-COD D) was followed using vials filled with a COD reagent made up of 2.8mL of 5.5gAg2SO4/L and 1.2mL of 61.2gK2Cr2O7/LH2SO4(99%).  Vials were given 2mL of sample, diluted such that the COD was within the range of 20 to 900mg/L, for both total (TCOD) and soluble (SCOD) fractions.  The vials were then sealed and digested at 150°C for 2h.  Each batch of reagent was calibrated with a five point set of standards using potassium hydrogen phthalate (85g/L for a 1gCOD/L solution).  Digested samples were measured at 600nm with a spectrophotometer (DR 2800, Hach).  Four measurements were made on each vial around its compass points to compile an average.  Samples were run in triplicates, using three distinct vials and sample volumes.  3.3.2 Total Phosphorus (TP/STP) To determine the total amount of phosphorus in  a sample, the appropriate amount of total (TP) or soluble (STP) sample (2.5 to 1000ugTP) was added to a vial containing 5mL of 134gK2SO4/H2SO4(20%v/v).  Samples were then digested for 180min at 135°C, followed by another 360min at 380°C.  After digestion, 42  vials were dosed with 30mL of distilled water, rotary shaken, and analyzed using a Flow Injection Analysis System (Quickchem 8000, Lachat Instruments) that followed the Ascorbic Acid Reduction colourimetric method (4500-P F).  If black particles were present after digestion, samples were passed through a coarse glass filter before analyzing.   3.3.3 Total Kjeldahl Nitrogen (TKN/STKN) Total (TKN) and soluble (STKN) samples were added to vials such that the total Kjeldahl nitrogen was between 5 and 2000µgTKN, before digesting as described in Section 3.3.2 (this digestion was combined with that of the TP analysis).  After digestion, vials were dosed with 30mL of distilled water, rotary shaken, and analyzed using a Flow Injection Analysis System (Quickchem 8000, Lachat Instruments), following method 4500-N D.  3.3.4 Ammonia (NH3) A sample‟s soluble fraction was diluted using distilled water such that its concentration of ammonia (NH3) was between 0.5 and 50mgNH3/L, and its total volume was 4mL.  One drop per sample of 5%(v/v)H2SO4 was added as a preservative.  Samples were analysed using a Flow Injection Analysis System (Quickchem 8000, Lachat Instruments) following method 4500-NH3 H. 3.3.5 Orthophosphates (OP) A sample‟s soluble fraction was diluted using distilled water such that its concentration of orthophosphate (OP) was between 0.25 and 25mgPO43-/L, and its total volume was 4mL.  One drop of 1g phenylmercuric acetate per litre of 20%(v/v) acetone was added as a preservative to each sample.  Samples were analysed using a Flow Injection Analysis System (Quickchem 8000, Lachat Instruments) following method 4500-P G. 3.3.6 Polyphosphates (PP) Soluble samples were diluted in 10mL vials using distilled water such that their final concentration ranged from 0.25 to 25mgPO43-/L.  To these, 1 drop of phenolphthalein indicator and 0.1mL of strong acid solution (3%(v/v)H2SO4 and 0.04%(v/v)HNO3) were added.  Partially sealed vials were digested (method 4500-P B) in an autoclave for 30min at a pressure within the range of 98 to 137kPa.  After cooling, 6N of NaOH was added dropwise until a permanent pink colour developed.  Lastly, samples were measured using the OP analysis method described in Section 3.3.5.  The polyphosphate (PP) fraction was calculated by subtracting the standard OP method‟s result from the one obtained here. 43  3.3.7 Volatile Fatty Acids (VFA) One drop of 50%(v/v)H3PO4 was added as a preservative to ~2mL of a sample‟s soluble fraction that contained between 2 and 200mg/L of acetic acid equivalents.  Samples were analysed using a gas chromatographic mass spectrometer (HP5580 Series 2, Hewlett Packard) using method 5560 D, for the following volatile fatty acids (VFA): acetic acid, propionic acid, iso-butyric acid, iso-valeric acid, and valeric acid.  VFA results were reported in units of mgCH3COOH/L by converting all measured VFAs into acetic acid based on their molar amounts. 3.3.8 Metals Analysis Total and soluble sample fractions were added to vials containing 2.5mL of HNO3 (67-70%(v/v)), 2.5mL HCl (36.5-38%(v/v)), and 20% of the sample‟s volume as 30%(w/w) H2O2.  Vials were digested for 2h at 140°C under closed reflux conditions.  If black particles were observed after digestion, samples were passed through a coarse glass filter before analyzing.  After digestion, samples were made up to 20 or 50mL with distilled water and analyzed using an inductively coupled plasma (ICP) atomic emission spectrometer (Optima 7300 DV, PerkinElmer), following method 3120 B.  The upper and lower detection limits for the analytes measured using this method (i.e. Na, K, P, Ca, and Mg) are shown in Table 8. Table 8 – Detection limits for the ICP metal analysis Analyte Lower Detection Limit (µg/L) Upper Detection Limit (mg/L) Na 1 500 K 1 P 10 Ca 0.1 Mg 0.1   44  4.0 The MW/RF-H2O2 Experimental Setup  4.1 2.45GHz Microwave Dielectric Heater 4.1.1 Description The 2.45GHz microwave oven (Ethos TC Digestion Labstation 5000, Milestone Inc., USA) is a 1kW batch reactor with 12 vessels, each with a capacity of 100mL (see Figure 8A).  The vessels are constructed to withstand temperatures and pressures of up to 220°C and 30 bar (Figure 8B).  This model has two magnetrons with a rotating diffuser to uniformly distribute the EMR.    Figure 8 – 2.45GHz MW oven: A) outside view, and B) view of vessels, where i) sample vessel, ii) sheath, iii) cap, iv) collar, v) spacers, and vi) vessel holder Magnetic stir bars are added to each vessel, allowing the contents to be completely mixed during an experiment.  One vessel accommodates a temperatures probe that is used to monitor the MW`s heating rate and is assumed to be representative of all the vessels.  To operate, the desired temperature and heating time is programmed into the controller. 4.1.2 Standard Operating Procedure The 2.45GHz MW was operated according to the following procedure for all experiments discussed in this work, unless otherwise noted: 1) Turn on the MW`s power switch 2) Add 30mL of sample to each vessel (Figure 8i) 3) Place vessel inside its corresponding sheath (Figure 8ii)  4) Place the cap (Figure 8iii), collar (Figure 8iv), and spacers (Figure 8v) on top of their corresponding vessel 5) Place the vessel assembly inside its corresponding holder (Figure 8vi) A) B) i) ii) iii) iv) v) vi) 45  6) Tighten the vessel holder‟s top screw with a ratchet until a clicking noise is heard, indicating that the correct torque has been applied and the vessel has been properly sealed  7) Align the vessels in the MW oven 8) Insert the temperature probe into vessel #1 and connect it to the port inside the MW 9) Add the top cap to vessels 10) Close the door and turn on the magnetic stirrer to the desired speed 11) On the controller‟s touch screen: i) After logging in, press „prep‟ ii) Press the „program‟ menu iii) Add the desired temperature/time steps iv) Click the rotation button v) Click „start‟ to begin the experiment 12) When the temperature regime has been completed and the sample cooled (~20min): i) Turn off the magnetic stirrer  ii) Click the rotation button to turn it off iii) Open MW door and remove vessels (reverse of steps #3 to #9) 4.2 915MHz Microwave Dielectric Heater 4.2.1 Original Design for Low Temperature Operation 4.2.1.1 915MHz MW Heater The continuous pilot-scale 915MHz MW heater (Sairem, France) was originally designed for temperatures below 100°C, as previously described in Srinivasan et al. (2014a), and shown in Figure 9.  The generator (Figure 9A) houses a magnetron capable of outputting 5kW of power through the aluminum waveguide (Figure 9B), which directs the EMR into the applicator (Figure 9C).  The applicator is of the horn variety and flanges vertically to spread the propagating EMR uniformly over twin silicone tubes (APSM series, C.P.E. Systems Inc.) that carry the substrate.  These tubes have inner diameters and lengths of 1.9cm and 1m, respectively, providing a total volume of 0.6L in the applicator (Figure 9D).  The waveguide and applicator were custom built for research purposes by RF Specialists Ltd (Canada).    46    Figure 9 – Model of the pilot-scale 915MHz MW heater, where: A) generator, B) waveguide, C) applicator, D) silicone tubing, E) tuning screws, F) water cooling system, G) isolator, H) holding tank, I) primary pump, J) H2O2 reservoir, and K) H2O2 pump Four brass screws (Figure 9E) serve as the tuning system and are spaced 1/3 of a wavelength apart (~11cm).  The tuning screws are used to match the impedance of the applicator to that of the substrate; see Section 2.1.6 for a theoretical background of impedance matching and Section 2.4.1.4 for the screws operation.  The MW is cooled by a fan and a water system (Figure 9F).  The influent cooling water must be between 17 and 22°C and have a minimum flow rate of 15L/min (50psi from municipal water supply).  The cooling water is used by the magnetron inside the generator, as well as the isolator (Figure 9G) to absorb the reflected EMR that has been redirected away from the magnetron.  Each of these components requires a minimum water flow rate of 6.5L/min.  The generator has a built-in electronic controller that allows the output power to be set and displays the amount of reflected power (RP).  Flow switches on the cooling water system are set to turn the generator off via the controller if not enough water is being delivered to either the magnetron or the isolator.    4.2.1.2 Auxiliary Systems To heat the substrate to the desired temperature, it must be recirculated through the applicator from its holding tank (Figure 9H) by a pump (Figure 9I).  The substrate is pumped upwards through the applicator to ensure that no gas bubbles become trapped, before returning to the tank.  H2O2 is pumped from a 1L reservoir (Figure 9J) into the substrate immediately before it enters the applicator to maximize the synergistic effect between the EMR and H2O2.  The H2O2‟s pump (Figure 9K) is a peristaltic variable frequency pump (CPT Series, Chem-Tech) with a maximum flow rate and pressure differential of A) B) C) D) E) F) G) H) I) J) K) 47  0.16L/min and 5.5bar.  A mixer is used in the holding tank to prevent the settling of solids and development of temperature gradients within the tank. 4.2.2 Modifications for High Temperature Operation To fulfill this work‟s objectives, the MW heater was modified to accommodate temperatures up to 130°C. The 915MHz MW itself required no changes, but rather all the auxiliary equipment needed to be upgraded in order to withstand the higher temperatures and pressures.  It also had to be a closed system in order to maintain the pressures generated by heating above 100°C.   4.2.2.1 Upgraded Auxiliary System The modified MW‟s overall hydraulic system is the same as that shown in Figure 9, except that more durable parts and extra instrumentation were required.  A complete process flow diagram for the upgraded MW system is given in Figure 10.  Figure 10 – Process flow diagram of the 915MHz MW modified for operation at temperatures above 100°C All process tubing has a diameter of 1.905cm (0.75in) and is constructed from 316 stainless steel braided hosing lined with a polytetrafluoroethylene, or PTFE, core (Swagelok, Burnaby BC).  The tubing‟s maximum pressure and temperature rating is 49.6bar at 203°C.  Tubing used for the H2O2 line has an outer diameter of 6mm and is also made of PTFE.  Seamless 316 stainless steel pipes (Swagelok, Burnaby 48  BC), with an outer diameter of 1.27cm (0.5in), are used for the vessel‟s venting system to allow the controlled discharge of any accumulated gases. The closed recirculation vessel is made of 1.905cm (0.75in) thick 316 stainless steel, with an approximate capacity of 43L (~25cm inner diameter and ~90cm height).  It is fed from its top and empties through a port at its bottom via connected tubing.  Although there is no mixing device inside the vessel, a 1.27cm (0.5in) outer diameter 316 stainless steel seamless pipe (Swagelok, Burnaby BC) is connected to the influent port and extends ~65cm inside the tank such that the incoming substrate is injected below the vessel‟s typical liquid level.  It was intended that this would help mix the vessel‟s contents and avoid the formation of significant temperature gradients or hydraulic short circuiting.  It is recommended that the new system treats between 10 and 20L of substrate per run, in order to minimize the formation of steam and attain reasonable heating rates.  Based on vapour-liquid equilibrium calculations, only 1.6 and 0.3%(w/w) of the substrate in the vessel would be turned into steam at 120°C for sample volumes of 10 and 20L, respectively.  Thus, the impact of steam on the treatment results is deemed negligible.  The H2O2 reservoir is a 1L beaker open to the atmosphere.  This means that H2O2 can only be added into the system at temperatures below 100°C. The primary pump (P-1 from Figure 10) used to recirculate the substrate is a two stage progressive cavity pump (Model NM021BY02S12B, Netzsch), capable of maintaining a flow rate of 10L/min at a maximum pressure differential of 6bar.  It has a variable frequency drive that is controlled by a separate variable frequency drive (VFD).  Pump curve measurements were taken to regress Equation 32 and allow the expected flow rate (Q), in L/min, to be calculated for a set frequency (f), in Hz (R2=0.986).                       (32)  The H2O2 pump (P-2 from Figure 10) is the same as that used in the original system and is described in Section 4.2.1.2.  A summary of all control valves and their function is provided in Table 9, where valve labels refer to those displayed in Figure 10.   49  Table 9 – Control valves used in the modified 915MHz MW system and their function Label* Valve Type Function V-1 Ball Allows substrate to be pumped from the reservoir to the vessel V-2 Globe Used to prime and drain the primary pump (P-1) V-3 Ball Isolates MW from primary pump (P-1) V-4 Ball Isolates H2O2 line and pump (P-2) V-5 Check Prevents backflow up the H2O2 line when V-4 is open V-6 Purge Allows air pockets to escape during start up V-7 Ball Used to manually relieve pressure in the vessel V-8 Pressure Relief Automatically opens when the vessel‟s pressure surpasses 3.6bar V-9 Plug Used to isolate vessel‟s level indicator in conjunction with V-10 V-10 Plug Used to isolate vessel‟s level indicator in conjunction with V-9 V-11 Ball Allows vessel to be drained or samples taken V-12 Ball Isolates recirculation of substrate from the vessel *Note: valve labels refer to those shown in Figure 10 4.2.2.2 Instrumentation  To measure the temperature of the substrate immediately exiting the MW (TI-1 from Figure 10), a fibre optic probe sits inside the tubing (Model T1S-03-WNO-PB05, Neoptix).  The temperature probe is connected to a signal conditioner (Reflex, Neoptix) that relays the data to a computer via a RS-232 cable.  A second thermocouple (Model KMQXL-125U-12, Omega), labeled as TI-2 in Figure 10, is used to measure the vessel‟s bulk temperature.  This thermocouple is connected to a data logging temperature meter (Model SDL200, Extech Instruments). The pressure of the vessel is measured at PI-1 (Figure 10) with an analogue pressure gauge (Model PGR-45LSS-60, Omega) fitted with a stainless steel diaphragm.  Pressure readings are recorded by hand as deemed necessary. A level indicator (LI-1 from Figure 10) is installed on the vessel to show how much substrate had been added.  It is constructed of a 0.953cm (3/8in) outer diameter semi-clear tube (Model 51805K73, McMaster-Carr) made of perfluoroalkoxy alkane.  The tube is positioned along the height of the vessel and connects to ports at its top and bottom.  Because the tube and fittings are not pressure rated, two plug valves (V-9 and V-10 on Figure 10) are used to isolate the level indicator at high temperatures.  A three phase multifunction power meter (Model Acuvim-CL-D-60, Accuenergy) is used to measure the real time energy and power demand of the MW‟s generator during an experiment.  To record its data, the power meter is connected to a computer via a USB cable.  The MW‟s power supply is plugged into the power meter, which, in turn, is plugged into the building‟s power grid. 50  4.2.3  Standard Operating Procedure  For all experiments using the 915MHz MW heater at temperatures above 100°C, the operating procedure detailed below was followed unless otherwise noted.  For experiments below 100°C, the same operating procedure was used except that the vessel‟s manual vent valve (V-7 from step 13) was kept open.  All valve and pump labels referenced in this procedure are shown in Figure 10Figure 10. 4.2.3.1 Start-up Procedure 1) Turn on cooling water system, ensuring that its within acceptable temperature and flow ranges 2) Plug in MW generator, pumps, and all instrumentation 3) Ensure V-6, V-7, V-9, and V-10 are open and all other valves are closed 4) Fill reservoir with well mixed substrate 5) Open V-1 and wait until substrate has flooded downstream tubing 6) Open V-12 and then V-11 7) Close V-11 and then V-12 after the flow rate through V-11 has reached a steady state (~10s) 8) Open V-2 fully and close after a steady flow is observed (~10s) 9) Open V-3, wait 10s, and close 10) Turn on the primary pump (P-1) at its maximum flow rate and open V-3 immediately afterwards 11) Close V-6 five seconds after starting pump P-1 (approximately when P-1 has reached full speed) 12) When the level in the vessel has reached the desired height, open V-12 then close V-1 13) Close V-7, V-9, and V-12 14) Set the pump to the desired flow rate via the VFD 15) Set the MW‟s „forward power‟ to 4.5kW and press „start‟ 16) When ready to inject H2O2: i) Open V-4 and then immediately start the H2O2 pump (P-2) ii) After dosing is complete, stop pump P-2 and immediately close V-4 4.2.3.2 Shut-down Procedure 1) Click the „Stop‟ button on the MW‟s control panel, unplug the generator, and turn off the cooling water after another 15min 2) Keeping P-1 running, wait until the temperature has decreased to 90°C or below 3) Open V-7 to release pressure in the vessel 4) Collect sample from V-11 or V-2 5) Turn off P-1 6) Open V-4 for five seconds, then close before disconnecting the H2O2 line (to relieve any accumulated pressure and prevent spraying H2O2) 51  7) Open all remaining valves to completely drain the system 4.2.4 Characterizing the Modified 915MHz Microwave System To characterize the modified MW‟s hydraulic behaviour, two tests with salt water (1gNaCl/L) were conducted at flow rates of 6 and 8.4L/min.  The applicator‟s effluent and the tank‟s bulk temperature for both runs are plotted in Figure 11, yielding overall heating rates of 1.1 and 1.2°C/min for the low and high flow rates, respectively.  Figure 11 – Temperature profiles of salt water heated with the modified 915MHz MW oven: A) 6L/min, and B) 8.4L/min Operating at the lower flow rate induced a larger temperature difference between the tank and applicator, particularly at the beginning of the run.  Additionally, when compared to the higher flow rate run, a lower flow rate resulted in a less consistent heating rate that exhibited greater fluctuations.  These observations suggest that the tank‟s contents are not being well mixed at lower flow rates, causing some of the bulk water to be short circuited.  Thus, it is recommended to operate the system at 8.4L/min or higher, to induce a greater mixing in the tank and attain a more uniform heating rate. For experiments using WAS and DM, different tuning screw arrangements were needed to maintain low RP values.  Recommended tuning screw positions are provided in Table 10 for both substrates, where tuning screw A is closest to the MW generator.  The WAS, in particular, required regular retuning throughout an experiment, due to its highly variable dielectric properties.    R² = 0.9555 0204060801001201401600 2000 4000 6000Temperature (°C) Time (s) A) Applicator EffluentTankR² = 0.9933 0204060801001201400 2000 4000Temperature (°C) Time (s) B) Applicator EffluentTank52  Table 10 – Height of tuning rods outside of the 915MHz MW’s waveguide for WAS and DM substrates Substrate Applicability  Tuning Screw Position (cm) A B C D WAS Low Temperature (20-70°C) 19.3 21.5 20 20.3 After Addition of H2O2 15.3 21.3 16.8 20.3 High Temperature (>100°C) <15.3 21.3 16.8 20.3 DM Entire Run (20-130°C) 19.0 15.5 16.5 16.5  It is important to note that the tuning rod positions provided in Table 10 are approximate, and ultimately depend on the exact characteristics of the substrate being heated.  These values, therefore, should be taken as a guide. 4.2.5 Recommendations for Future Modifications and Upgrades   After running the modified 915MHz MW system over the course of many experiments, several areas of improvement were identified.  These issues are described below and possible solutions are provided, to aid in future work.   The currently used vessel is oversized and inappropriate for the experiments typically undertaken.  Consequently, this has led to many difficulties, particularly in regards to the large volumes of substrate needed to run each experiment, which lowers the MW‟s maximum heating rate.  It is recommended that an alternative vessel be obtained with a smaller capacity of 10 to 15L.  This would also reduce the percent of substrate that is turned into steam.  A more important issue with the current vessel is that, due to its shape and size, it is highly probable that its contents are not completely mixed and therefore, do not have a uniform bulk temperature.  By designing the new vessel to have a smaller volume and a shape with a low width to height ratio, its characteristic flow pattern would exhibit a more uniform residence-time distribution.  In other words, having a vessel shaped like a pipe allows it to behave more akin to a truly continuous plug-flow system, thereby reducing the risk of undesirable temperature distributions developing within the vessel.  An alternative approach would be to install a mechanical mixer or diffuser in an effort to better mix the vessel‟s influent stream with its contents and promote a more uniform bulk temperature.  Considering the difficulty and costs associated with the installation of a mixer in the pressurized vessel, adding a diffuser directly to the vessel‟s influent pipe would likely be the less expensive and more practical option.  Incomplete mixing is the likely cause of another issue; the vessel‟s temperature probe fluctuates widely over the course of a run.  Although this could be due to EMR propagating down the tube and interfering with the thermocouple, the fluctuations however, typically disappear at higher temperatures, suggesting that large temperature gradients in the vessel are indeed the source of the inaccuracies.  Thus, improving the extent of mixing or using a more appropriately designed vessel, should also improve the temperature data‟s quality.  A potential solution to this specific problem 53  would be to change the position of the vessel‟s thermocouple and attempt to find more stable and representative temperature readings. The MW system was designed to withstand temperatures of up to 130°C.  If temperatures above that are desired, the vast majority of equipment would be safe to use except for select components that would require upgrading, as identified below.  Assuming a new operating temperature of 160°C, the fibre optic temperature probe (TI-1 from Figure 10) and its tube fitting would have to be replaced with a more durable substitute due to their limited pressure ratings.  In fact, starting at 130°C, temperature measurements were only intermittently recorded by the current probe.  Similarly, the fittings that connect the tubing inside the MW‟s applicator to the rest of the system would have to be replaced with pressure appropriate hardware.  Lastly, the pressure relief valve (V-8 from Figure 10) would have to be reset such that it releases at a pressure above 6.2bar.     It has been observed that the primary pump‟s flow rate (P-1 from Figure 10) using the substrates tested in this work, can deviate significantly from that expected when using water, thereby rendering the pump curve correlation presented in Equation 32 inaccurate.  Furthermore, it is common during start-up that not all of the air gets properly evacuated from the system.  When this occurs, the flow rate is vastly reduced.  For these reasons, it would be desirable to have the ability to measure and verify the system‟s flow rate during an experiment, either by installing an inline flow meter, or using an external portable device.   Lastly, it is recommended that a heat exchanger be used to quickly cool down the substrate after an experiment.  Currently, the system requires substantial cooling times for the substrate to reach a safe temperature that allows for its collection.  By holding the substrate at these high temperatures for such extended periods of time, it is probable that the experiment‟s treatment results would also be affected.  By rapidly force-cooling the substrate with a heat exchanger, this potential source of error can be negated.  4.3 The 27MHz Radiofrequency Dielectric Heater  4.3.1 Original Design for Batch Operation The pilot-scale 27MHz RF dielectric heater (Model KA64, Thermatron) was retrofitted for batch operations by RF Specialist Ltd. (Canada), as modeled in Figure 12. 54   Figure 12 – Model of the original batch mode 27MHz RF heater: A) generator, B) co-axial cable, C) outer shell and ground electrode, and D) active electrode A 13kW co-axial RF power triode (model ITL 5-1, Thales) is housed in the generator (Figure 12A).  It serves as an oscillator and amplifier by converting a DC electrical signal into an AC with a frequency of 27MHz, while increasing the output current‟s power to a maximum of 6kW.  The triode is air cooled by a fan and its power output is controlled by a silicon-controlled rectifier (SCR).  The RF AC current is then transferred through the co-axial cable (Figure 12B) to the oven (Figure 12C).   The transmission line terminates at a horizontally placed „active‟ electrode plate inside the oven (Figure 12D).  The walls of the oven act as the second ground electrode and samples are placed directly underneath the active electrode during operation.  The electrodes provide approximately 130L of space for samples; practically however, only 9L of sample can be heated at a time.  Two coiled copper pipes connecting the active electrode to the oven‟s walls are used as stubs.  The samples are held in EMR transparent containers with aluminum caps on their tops and bottoms.  The metal caps must be touching the sample to obtain a reasonable heating rate.  Because the vessels are not pressure rated, the RF heater was limited to temperatures below 100°C.  When the sample is ready inside the oven, a fibre optic temperature probe is inserted into its container through a hole in the oven‟s wall.  A major challenge encountered with this configuration was the electrode‟s highly irregular electric field that induced a variable heating rate, dependent on the size of the sample containers and their placement in the oven.  A series of tests using salt water allowed a contour map of the heating rate‟s spatial variations to be constructed using Origin® 8 software; see Figure 13.  55   Figure 13 – Contour map describing the change in heating rate relative to the centre of the batch RF heater’s active electrode Figure 13 shows the negative percent deviations in heating rate relative to the centre of the oven and were as high as 20% at the electrode plate‟s outer edges.  Several other disadvantages of the original batch design include:  High inefficiencies required low heating rates and small sample volumes  Samples were not mixed and could separate during an experiment  H2O2 had to be injected immediately prior to an experiment at room temperature, resulting in its partial deactivation by catalase and reduced reactivity for WAS (see Section 2.2.1)  Does not simulate a full-scale system, which would most likely be continuously operated  For the above reasons, it was decided to convert the batch heater into a continuously operated one. 4.3.2 Modifications for Continuous Operation 4.3.2.1 Changes to the RF Heater  RF Specialist Ltd. (Canada) converted the batch RF heater into a continuous mode by retrofitting the electrodes; all other components of the RF heater described in the previous section remained the same.  The new electrodes consist of three vertically positioned aluminum plates arranged parallel to one another in a sandwich-like formation.  This set of electrodes contain aligned holes that allow tubing to be coiled through them, as shown in Figure 14, where the middle plate is the active electrode that is connected to the co-axial cable and the two outer plates are grounded to the oven‟s walls. 56   Figure 14 – Upgraded electrode configuration for the continuously operated 27MHz RF heater EMR is emitted between the active electrode and its two ground counterparts, heating any fluid in the tubing contained within.  This electrode design was chosen so that the electric field is parallel to the tubing, to improve its uniformity.  Polyethylene supports, which are largely transparent to EMR, are used to hold the electrodes in place.  Polyethylene caps are also fitted into the electrodes‟ holes to prevent arcing.  One coiled copper stub was added to the bottom of the active electrode, connecting it to the oven‟s shell.  The tubing exits the oven through specially cut holes that are encased in copper piping vestibules that limit the leakage of EMR.    4.3.2.2 Auxiliary Equipment A tank and pump is used to recirculate the substrate through the electrodes until the desired temperature is attained, as summarized in the below process flow diagram (Figure 15).   Figure 15 – Process flow diagram of the continuous 27MHz RF heating system 57  The tubing used inside the RF oven was made of a poly braided platinum silicone fibre with an inner and outer diameter of 2.54cm (1in) and 3.569cm (1.405in), respectively (C.P.E. Systems Inc, Burnaby BC).  Approximately 14.3m of tubing is housed in the oven, providing a heating volume of 7.2L.  Tubing for the H2O2 line is made of PTFE with an outer diameter of 6mm.  All remaining tubing is 0.75in (inner diameter) high temperature resistant rubber (McMaster-Carr, USA).    Samples are pumped through the RF heater by P-1 (Figure 15), a progressive cavity pump with a maximum flow rate of 7.5L/min (Model 3315 3913315000, Moyno Inc.).  The pump‟s flow rate is controlled by a separate VFD.  H2O2 is injected into the tubing immediately before the sample enters the RF heater to maximize the synergistic effect between the EMR and H2O2.  A ball valve (V-1 from Figure 15), isolates the H2O2 line that draws from the 1L H2O2 reservoir.  The pump used to inject the H2O2 (P-2 from Figure 15) is the same one used by the 915MHz MW (see Section 4.2.1.2 for details).  Once the sample has passed through the heater, it goes back to the recirculation tank, which is a 0.91m (3ft) long section of 10.16cm (4in) NPT 316 stainless steel pipe, providing a holding capacity of 7.4L.  The total capacity of the RF heater, including the tubing and the tank, is 14.6L.   A recirculation tank was needed so that the considerable amount of tubing inside the oven could be filled, as well as providing additional substrate holding capacity to allow sampling throughout an experiment.  The reason for using a large pipe as the tank was to ensure that the system‟s flow regime simulated as closely as possible to that of a true continuous system.  A tall narrow tank will exhibit fluidic behaviour similar to a pipe, where a slug of fluid travels relatively together, while a wide tank will have lower flow velocities and likely suffer from spatially dependent resident times (i.e short circuiting).  Furthermore, the low velocities in a wide tank could allow larger particles to settle or the substrate to separate into different phases.  Even if a mixer was used in such a tank, the contents would have a bulk temperature made by the mixing of fluidic elements subjected to varying temperatures and irradiation times; it is therefore possible that this would yield different treatment results when compared to a fully continuous system.  For these reasons, it was decided that the tank should be as narrow as possible, while maintaining a holding capacity of ~7L within a reasonable length, hence the chosen tank described above.  Assuming the typical flow rate of 6.5L/min is used, the Reynolds number of the tank would range from 1 400 to 4 200 over the course of a treatment run (25 to 90°C for pure water).   Consequently, the flow regime is either in the transition or turbulent zone and can be regarded as a reasonably mixed environment.  A ball valve at the bottom of the recirculation tank (V-2 from Figure 15) is used to take samples throughout an experiment.  58  4.3.2.3 Instrumentation A fibre optic probe (TI-1 from Figure 15) and transmitter measures the temperature of the substrate as it exits the RF heater.  Both instruments are described previously in Section 4.2.2.2.  A second fibre optic probe (TI-2 from Figure 15) was added for later experiments to measure the applicator‟s influent temperature.  A power meter is connected to the RF‟s power supply and transmits data to a computer, also described in Section 4.2.2.2.  The RF generator has a SCR that controls the amount of power sent to the electrodes, as well as two dials that measure the „plate‟ and „grid‟ current in the triode.  The SCR functions by changing the negative voltage on the grid, displayed as a current on its corresponding dial.  By turning up the SCR knob, it decreases the grid‟s negative voltage, permitting a greater flow of current through the anode.  The anode‟s current on the „plate‟ dial is proportional to the power sent to the oven.  In practice, inefficiencies from the substrate‟s dielectric properties and impedance mismatches mean that this relationship is dependent on a variable efficiency factor, and therefore, is not a useful method for determining actual power output.  To avoid arcing in the oven or an overloading of the RF generator, the maximum current on the anode should not be exceeded.  If either of these situations occurs, the RF generator is wired to automatically shut down.  This maximum anode current, as measured by the „plate‟ dial, depends on the substrate as detailed in Section 4.3.4. 4.3.3  Standard Operating Procedure  The continuous RF heater was operated according to the below procedure for all experiments discussed in this work, unless otherwise noted.  All component labels refer to those displayed in Figure 15. 1) Plug in RF heater, pumps, and temperature probes  2) Ensure V-1, V-2, and V-3 are closed 3) Add ~7L of well mixed substrate to the recirculation tank 4) Open V-3 and close after a steady flow through the primary pump (P-1) is obtained to ensure that the stator is properly wetted 5) Turn on P-1 at a low flow rate 6) Turn off P-1 when the recirculation tank is nearly empty 7) The RF heater is now „charged‟ and up to another 7L of substrate can be added to the tank 8) Before starting an experiment, turn on P-1 to the desired flow rate via its VFD 9) Starting the RF heater:  i) Turn on the „power‟ breaker switch ii) Turn on the „RF‟ switch iii) Press black „start‟ button 59  iv) Slowly turn up the SCR power controller to the desired heating rate while keeping below grid and plate current maximums (see Table 11) 10) Adjust the tuning knob as desired to improve the heating rate (see Table 11) 11) Adding H2O2:  i) Open V-1  ii) Turn on H2O2 pump (P-2) immediately afterwards for the desired amount of time iii) Turn off P-2 iv) Close V-1 immediately  12) Samples can be taken from V-2 or V-3 at any point during an experiment  13) Ending an experiment: i) Press the red „stop‟ button ii) Turn down the SCR controller to zero iii) Turn off the „RF‟ switch iv) Turn off the „power‟ breaker switch v) Unplug the RF generator after 15min vi) Drain the system through V-2 and V-3 4.3.4 Characterizing the Continuous 27MHz Radiofrequency System The modified RF heater was tested with salt water (1gNaCl/L) at a flow rate of 6L/min. The resulting temperature profile was very consistent (R2 of 0.981) and gave an overall heating rate of 1.5°C/min for a plate current of 1A.  It is important to note that the heating rate is dictated by the plate current, and not the set power.  The power settings and tuning positions for the substrates tested in this work are summarized in Table 11. Table 11 – The continuous RF heater’s electronic settings for WAS and DM experiments Parameter Salt Water WAS Dairy Manure Tuning Position (1) 746 746 8125 Anode Current (A) 1 1 1.5 Grid Current (A) - 0.42 – 0.48 (start) 0.16 – 0.24 (end) 0.26 – 0.3 Power Range (%) 60 - 30 70 – 85 (start) 25 – 35 (end) 55 – 60   The anode current was maintained at the value indicated in Table 11, to ensure an even heating rate and that the RF‟s power limitations were not exceeded.  For WAS experiments, unlike those using DM, the anode current consistently increased with temperature, requiring the grid current to be reduced accordingly via the SCR power control knob.  60  4.3.5 Recommendations for Future Modifications and Upgrades    The primary issue with the continuous RF design is the substantial amount of foam that is produced after H2O2 is added to WAS.  The foam rises up the tank and often spills over the sides.  A cap with a tube leading to a separate bucket was used to control and capture the foam, but with only limited success.  It is recommended that the top of the tank is outfitted to either prevent the foam from leaving the tank, such as a properly fitted lid, or capture the foam as it leaves the tank.  Adjusting the substrate, by reducing its solids content or adding a foam suppressant, may also mitigate this issue.     61  5.0 Waste Activated Sludge Experiments This section will discuss the results from WAS experiments using MW and RF heaters.  Operating details of the various heaters can be found in Section 4.0, while the WAS samples and analyses are described in Section 3.0. 5.1 Effect of MW-H2O2 Process on Dewaterability  The primary objective of the following sets of experiments is to characterize the effect of the MW-H2O2 process on the dewaterabilility and digestibility of WAS at temperatures above 100°C, which had not been adequately investigated before.  The results were also used as a guide when choosing the temperature and H2O2 conditions for the 915MHz MW experiments (see Section 5.2).  The experiments presented in this section were conducted using the batch 2.45GHz MW.   5.1.1 Results A summary of the treatment conditions and select results for each run are provided in Table 12.  See Appendix A for the complete set of TS, TVS, TSS, VSS, TCOD, and CST data.   62  Table 12 – Experimental conditions and select results for experiments on WAS using the 2.45GHz batch MW Temperature (°C) H2O2 Dosage VFA (mg/L) SCOD (mg/L) CST (s) D10 (µm) D50 (µm) D90 (µm) %(v/v) %(v/v)/%TS Raw for Part 1 2.3±0.1 56±3 80±10 13.50±0.03 30.7±0.1 60±1 110 0.2 0.26 12±2 1 450±90 520±70 11.22±0.06 26.6±0.1 52.5±0.3 110 0.3 0.39 17±2 1 740±30 480±40 11.25±0.03 26.66±0.09 53.0±0.4 110 0.4 0.51 15±1 1 700±100 410±30 10.95±0.09 26.1±0.2 53±1 120 0.2 0.26 18±5 1 620±80 300±30 10.55±0.05 26.6±0.3 75.7±4 120 0.3 0.39 20±4 1 610±10 290±10 10.61±0.05 26.8±0.2 79±3 120 0.4 0.51 21±2 1 600±60 280±30 11.5±0.3 32±1 123±12 130 0.2 0.26 15±1 1 750±40 450±90 11.37±0.01 31.7±0.4 186±11 130 0.3 0.39 19±3 1 810±40 480±20 9.9±0.2 25.5±0.6 56±3 130 0.4 0.51 19±1 2 000±20 580±70 9.6±0.2 25.6±0.6 61±4 Raw for Part 2 4±2 68±4 80±10 11.3±0.2 29.3±0.4 64±2 150 0.2 0.21 16±3 2 750±70 620±30 6.9±0.2 24.8±0.7 190±70 150 0.8 0.83 24±2 3 150±20 780±20 4.3±0.2 19.4±0.4 100±10 130 0.8 0.83 36±1 3 800±100 440±30 3.5±0.1 17.82±0.04 100±20 110 0.8 0.83 31±1 2 600±20 720±90 5.8±0.2 24.3±0.2 61.4±0.6 5.1.2 Discussion on Digestion Indicators The formation of VFA (Table 12) is most strongly correlated with the H2O2 dosage (Pearson correlation coefficient, or PCC, of 0.84), rather than temperature (PCC 0.10). Thus, even above 100°C, higher H2O2 dosages are the primary driving force behind VFA production.  The likely explanation for this observation is that H2O2 is a source of otherwise limited oxygen that is needed to form the higher oxidative state VFAs.  TCOD does not significantly change with respect to temperature or H2O2 dosage (see Appendix A for data), suggesting minimal organic mineralization.  This is further validated by the TCOD‟s strong correlation with TS (PCC 0.93), as expected.  63  SCOD, shown in Table 12, is more affected by the H2O2 dosage (PCC 0.70) than temperature (PCC 0.57).  These correlations suggest that using very high temperatures for pilot-scale MW studies (see Section 5.2) is a less effective way of increasing SCOD than using higher H2O2 dosages.  Comparing the SCOD release to those obtained from later runs (see Section 5.2.1) shows that the impact of H2O2 on SCOD in these studies is lower than expected.  Although changes in the sludge‟s properties may be affecting this comparison, the fact that doubling the oxidant dosage resulted in only a couple of percent increases in SCOD (Table 12), suggests otherwise.  The cause of these differences is likely due to catalase, an enzyme that rapidly degrades H2O2 (see Section 2.2.1).  Catalase, however, is deactivated by 60°C (Wang et al., 2009), and if H2O2 is injected at this temperature or above, the SCOD is significantly improved.  For these 2.45GHz batch studies, H2O2 had to be added at the beginning of a run.  Therefore, a significant portion of the H2O2 was likely lost due to the catalase, and the importance of H2O2 to SCOD understated from the data presented in this section, as consistent with previous work (Chan et al., 2007).  5.1.3 Discussion on Dewaterability  A priority goal of these experiments was to determine if a temperature-H2O2 combination existed that improves dewaterability above 100°C.  Dewaterability is often approximated by the CST test (Vesilind, 1988), with results shown in Table 12.  Previous studies have consistently found that CST improves slightly after MW-H2O2 treatment, but then quickly deteriorates with longer treatment times, unless acid is added to the process (see literature review in Section 2.3.1).  Much literature is available showing the importance of pH on the dewaterability of WAS, where acidic conditions reduce the surface charge of cells and allow EPS to break away into solution (Neyens and Baeyens, 2003).  However, acidifying WAS would increase operating and capital costs, as well as requiring the sludge to be neutralized downstream.  As such, it would be preferable to improve dewaterability without adjusting the pH.  It was hypothesized that high temperature-H2O2 combinations had the potential to accomplish this.   Over the range of experimental conditions investigated, none improved the CST relative to the untreated sample, although an optimum temperature was observed at 120°C, minimizing the CST increase.  Any deviation from this temperature further increased CST, while H2O2 had no statistically significant impact at 120°C.  To see if much higher H2O2 dosages could improve CST at 110°C, a 0.83%(v/v)/%TS trial was conducted.  The resulting CST was much worse than at the optimum H2O2 dosage of 0.51%(v/v)/%TS (see Table 12).   Many of the remaining CST results were not statistically different from one another, due to their large standard deviations, making it difficult to accurately investigate their underlying mechanisms.  The minimal CST differences are likely due to H2O2 inactivation from catalase. Even higher H2O2 dosages, or a more effective H2O2 delivery, may yet achieve the desired CST improvements. 64  Previous studies have concluded that the greater the fraction of supracolloidal particles (1 to100µm), the harder the sludge was to dewater (Barber and Veenstra, 2015; Karr and Keinath, 2014).  This important factor is further investigated by using the sample‟s PSD data (see Table 12) to determine how it is affected by the MW-H2O2 process‟s operating conditions.  Over all these treatment runs, it is readily seen that the D10 and D50 values decreased with respect to their untreated samples.  This means that the frequency of smaller sized particles (less than 30µm) were increasing, supporting the conclusions drawn from the SCOD data.  D10 and D50 values were more dependent on H2O2 dosage (PCC -0.74 and -0.64, respectively) than with temperature (PCC -0.53 and -0.37, respectively).  In contrast, the D90 values either increased or decreased, compared to the raw control sample, depending on the treatment conditions.  For example, fewer larger sized particles were found for the treatment runs conducted at 110°C.   This was expected as the high temperatures and oxidative environment would work to degrade the solids into smaller constituents, thereby shifting the PSD function towards shorter diameters.  However, substantial increases in the number of large particles were found for the 130 and 150°C trials, with similar H2O2 dosages.  Thus, it appears that the MW-H2O2 process had a polarizing effect on PSDs, where mid-sized particles were either broken down or coalesced to form larger solids.  Examining the PCC of D90 to H2O2 reveals a weak negative correlation of -0.25.  Conversely, temperature had a stronger positive correlation to D90 (PCC 0.63), meaning that higher temperatures actually caused some particles to coalesce.  The opposing effects of temperature and H2O2 are illustrated using select PSD functions plotted in Figure 16; such trends have also been observed in previous MW-H2O2 studies conducted at lower temperatures (Chan et al., 2010).   Figure 16 – PSD functions for WAS experiments using the 2.45GHz batch MW 0123456789100.1 1 10 100 1000 10000Volume of Particles (%) Particle Diameter (µm) Raw110°C+0.8 H₂O₂%(v/v)/%TS 150°C+0.2 H₂O₂%(v/v)/%TS 65  From Figure 16, evidence of coalescence is found for the 150°C run with the rise of a second peak at ~500µm.  In summary, it is recommended that high temperature pilot-scale 915MHz MW studies be run at temperatures just over 100°C, with greater H2O2 dosages injected at 60°C, if improving the dewaterability of the MW-H2O2 process is desired.  5.2 Experiments on the 915MHz Microwave Heater at High Temperatures Three experiments were run on the modified 915MHz MW heater to investigate the effect of temperature and oxidant dosage above 100°C on the pilot-scale treatment of WAS.  The experimental conditions are summarized below in Table 13.   A second oxidant, ozone (O3), was also investigated in addition to H2O2.  An O3 generator (VMUS-4 AZCO Ind., Canada) was connected to an airflow meter and needle valve to control the effluent flow rate.  The airflow was set to 1L/min for all pertinent samples.  The produced O3 was directed through a diffuser secured to the bottom of a sealed reaction tank that held the substrate.  An electric mixer was installed in the tank to ensure the uniform ozonation of the WAS.       Table 13 – Experimental conditions for high temperature WAS runs using the 915MHz MW Heater Experimental Conditions MW Only MW-H2O2 O3+MW-H2O2 Volume (L) 20 Flow Rate (L/min) 6 Final Temperature (°C) 110 H2O2 Dosage %(v/v) - 0.6 0.6 %(v/v)/%TS - 0.68 0.77 H2O2 Addition (°C) - 60 60 O3 Dosage at 1L/min (min) - - 30 O3 Addition - - Prior to MW Overall Heating Rate (°C/min) 0.66* 1.39 1.51 Total Heating Time 2:15:20* 1:07:30 0:58:40 *Note: a temporary pressure leak caused the abnormally low heating rate and long run time  In short, the experiments were as follows: i) control MW run with no oxidant added, ii) typical MW-H2O2 run using a high H2O2 dosage (relative to previous 915MHz MW studies), and iii) dual oxidant run that pretreated the WAS with O3 before subjecting it to the MW-H2O2 process.  Select data is presented and discussed below; the full data set can be found in Appendix B. 5.2.1 Digestion Indicators The extent of COD solubilization is plotted in Figure 17, showing how increasing the oxidant dosage substantially raised the fraction of SCOD to a maximum of 87±3% for the O3+MW-H2O2 run.  In fact, the WAS appeared bleached in colour after this treatment.  A previous, oxidant-free experiment with the same MW heater treating WAS at 90°C resulted in a SCOD of 19±1% (Bailey, 2015).  Comparing this to 66  the MW only run presented in this section, only a ~10% SCOD gain was realized by raising the treatment temperature to 110°C.   Figure 17 – COD results from high temperature WAS experiments using the 915MHz MW heater, where numbers in brackets reflect the experiment’s respective raw sample When the MW only or MW-H2O2 treatment was utilized, the rising SCOD fraction was primarily caused by the breakdown of large organic molecules into smaller and more soluble products.  In contrast, by pretreating the WAS with O3, the resulting increase in the SCOD fraction was almost exclusively due to a decrease in TCOD, whereby organic matter is completely oxidized to CO2.  Other works have noted that O3 is effective at lysing cells and mineralizing SCOD, but does not attack nor solubilize  the sludge‟s floc structure (Zhang et al., 2009).  Furthermore, the raw sample subjected to only O3 reveals a synergistic effect between O3 and the MW-H2O2 process.  Combining these two oxidants consecutively resulted in an additional SCOD release of 27±3%, relative to the sum of their individual SCOD increases.  It is well known that the direct combination of O3 and H2O2 forms *OHs.  This occurs from O3 reacting with H2O2‟s dissociation product, further enhancing the solubilization of WAS (Wang and Xu, 2012).  However, because the O3 was used as a pretreatment for these experiments, *OH production from its reaction with H2O2 was likely minimal.   VFA production was also enhanced with higher oxidative treatment conditions.   Starting off at less than 4mg/L, total VFA increased to 21±2mg/L when only MW heating was used, but jumped to 245±9 and 298±6mg/L, for the MW-H2O2 and O3+MW-H2O2 runs, respectively.  The minimal increase in VFAs 020004000600080001000012000COD (mg/L) non-SCODSCOD67  gained from pretreating with O3 is reflected in the Raw+O3 sample‟s low VFA concentration of only 1.8±0.4mg/L.  These observations suggest that O3 has a limited ability to react with WAS to form VFAs, which agrees with its known reactivity and preference to mineralize organic matter. 5.2.2 Nutrient Release The speciation of STP is shown in Figure 18, where unaccounted for P is assumed to be organic.  TP was statistically equivalent for both raw samples and their initial STP levels were less than 5mg/L (data not shown, see Appendix B).   Figure 18 – STP speciation from high temperature WAS experiments using the 915MHz MW heater, where numbers in brackets reflect the experiment’s respective raw sample Adding an oxidant to the treatment process increased STP, although pretreating with O3 did not result in additional statistically significant gains over that of the MW-H2O2 process.  OP increased by 23 and 30% relative to the MW only run when H2O2 and O3+H2O2 were utilized, respectively.  PP was nearly stable over all MW treated samples, suggesting that the cells were completely lysed and all available PP was in solution at 110°C, without the need for H2O2 or O3.  An in-depth discussion of the possible relationships between the various forms of phosphorus can be found in Section 5.6, using the results from all WAS experiments. The solubilization of TKN followed the same trend as that of STP, with the two data sets yielding a PCC of 0.99.  The two MW runs utilizing oxidants achieved complete TKN solubilization.  Both raw samples had NH3 levels less than 0.5mg/L and required oxidative conditions to be released in significant amounts; the MW-H2O2 and O3+ MW-H2O2 runs produced 80.3±0.8 and 89±1mg/L of NH3, respectively.   Soluble Ca, K, and Mg levels are displayed in Figure 19 as a percent of their totals.  The solubilization of Mg followed a similar trend as P, where MW heating resulted in their partial release, but treatment runs, 050100150200250300350400450Raw+O₃ (2) MW (1) MW-H₂O₂ (2) O₃+MW-H₂O₂ (2) STP (mgP/L) OrganicOPPP68  utilizing oxidants, yielded near complete solubilization.  Ca was solely released with the use of oxidants, while K was largely solubilized during the MW only run.  The distinct trend for K can be explained by the fact that K+ is the primary intracellular cation in bacteria, where it is used as a co-factor for enzymes and to maintain the cell‟s osmotic pressure (Ballal et al., 2007; Sanin et al., 2006).  This means that, if the cell is lysed, a large flux of soluble K+ cations will be immediately released into solution.  From the SCOD and PP results discussed above, it is highly probable that most of the cells were lysed during the MW only run, hence the near complete solubilization of K without requiring oxidants.   Figure 19 – Solubilization of Ca, K, and Mg from high temperature WAS experiments using the 915MHz MW heater, where numbers in brackets reflect the experiment’s respective raw sample The relative molar ratios of Mg, NH3 and OP are shown in Table 14 in consideration of their importance to the formation of struvite as a possible method for recovering P.   Table 14 – Ratio of species required for struvite formation from WAS experiments using the 915MHz MW heater, where numbers in brackets reflect the experiment’s respective raw sample Run NH4+:Mg2+:PO43- (mol) MW (1) 1:2.5:2.6 MW-H2O2 (2) 3.5:1.6:1 O3+MW-H2O2 (2) 3.7:1.5:1  If only MW heating is utilized, NH3 is the limiting species.  But under oxidative treatment conditions, the limiting species becomes OP, as observed with previous bench scale batch studies (Wong et al., 2006b).  0102030405060708090100Solubilization (%) CaKMg69  Having the OP as the limiting species is preferable for two reasons: i) all problematic OP can be removed, and ii) no synthetic NH3 or Mg will have to be added, thereby reducing the operating costs of a struvite crystallizer.  5.2.3 Physical Properties Measurements pertaining to the physical properties of the WAS samples are presented in Table 15. Table 15 – Select physical sludge properties from high temperature WAS experiments using the 915MHz MW heater, where numbers in brackets reflect the experiment’s respective raw sample Sample TS (%) TVS (%) TSS (g/L) VSS (%) CST (s) Raw 1 0.877±0.005 80.4±0.1 8.60±0.03 80±1 117±4 MW (1) 0.857±0.003 80.5±0.4 5.8* 85* 1 250±40 Raw 2 0.775±0.005 81.2±0.3 7.43±0.08 79±3 165±5 Raw+O3 (2) 0.755±0.003 81.0±0.2 6.4±0.1 79±3 200±10 MW-H2O2 (2) 0.713±0.004 75.3±0.3 2.9±0.2 69±7 19±1 O3+MW-H2O2 (2) 0.525±0.003 74.2±0.5 1.4±0.1 43±9 12.9±0.4 *Note: only one measurement could be taken  The results in Table 15 follow identical trends as discussed in Section 5.2.1.  TS and TSS values decreased when subjected to greater oxidative conditions, with the lowest values observed for the O3+MW-H2O2 run.  The reduction of TS mirrors the change in TCOD (PCC 0.998), while TSS analogously followed SCOD trends (PCC 0.94).  Together, the TS and TSS data confirmed the previously observed synergetic effect between the O3 and the MW-H2O2 process. TVS and VSS (Table 15) fractions both decreased with greater oxidant treatment strengths, suggesting that the volatile organic portion of the WAS, which largely represents its biomass (Tchobanoglous et al., 2003), is preferentially attacked.  Previous studies have shown that starting at 145°C, the MW treatment of organics will form refractory compounds, such as humic acids and melanoidins (Shahriari et al., 2012).   The same study found that the MW-H2O2 process can increase humic acid concentrations at temperatures below 100°C.  As such, it was suggested that pretreating TWAS with the MW-H2O2 process produced less biodegradable soluble organics, as concluded from their suppressed ultimate methane yields and mesophilic digestion rates, at pretreatment temperatures as low as 60°C (Eskicioglu et al., 2008).  However, performing mass balance calculations with the TSS and VSS data for the MW runs that utilize oxidants, indicates that the preferential attack of volatile solids is the primary reason for this fraction‟s observed decline.  In light of the above cited studies though, it is also probable that the formation of 70  refractory substances is occurring at these treatment conditions.  For these two reasons, the remaining solids in the treated samples are likely to be more recalcitrant and less digestible.     CST (see Table 15) increased for the slow MW heating run, but decreased substantially relative to untreated samples, when subjected to the oxidant utilizing MW trials.  In contrast, the CSTs from the 2.45GHz batch MW experiments at 110°C (Section 5.1) were significantly worsened, most likely due to their lower H2O2 dosages and less effective delivery system.  The reason for the positive action of H2O2 and O3 in these experiments can be found from their PSD functions, presented in Figure 20, where particle size data for all samples is tabulated in Appendix B.  Figure 20 – Select PSD functions from high temperature WAS experiments using the 915MHz MW heater, where numbers in brackets reflect the experiment’s respective raw sample Both raw samples had similar PSDs, and thus, their average is plotted in Figure 20.  The Raw+O3 sample differed only slightly from the other raw samples, confirming that O3 primarily reacts with cells and dissolved organic matter, rather than flocs structures, as discussed in Section 5.2.1.  The two oxidant utilizing MW runs had nearly equivalent PSD functions, hence, only the MW-H2O2 run is shown in Figure 20.  Their dramatic PSD shifts resulted in 90% of their particles having a diameter less than 1.2µm.  This is the likely reason behind the observed CST improvements, where smaller particles are able to flow freely through the filter paper, without clogging, it and hold less bound water.  Reviewing the dewaterability studies cited in Section 5.1.3, these results fit entirely with past observations that particles in the supracolloidal range (1 to 100µm) inhibit dewaterability.  Furthermore, the MW only trial had a PSD that was still largely within this supracolloidal range, thereby preventing any CST improvements.  In terms of tightly and loosely bound EPS, the MW only run likely converted much of the former into the latter, resulting in its severe CST deterioration (see Section 2.3.1.1 for context).  Accordingly, the MW-05101520250.1 1 10 100 1000Volume (%) Size (µm) Average RawMW (1)MW-H₂O₂ (2) 71  H2O2 and O3+MW-H2O2 trials are presumed to completely solubilize both their tightly and loosely bound EPS.  5.2.4 Conclusions Pertaining to Commercial Applications  In summary, heating WAS to 110°C offered minimal gains in treatment efficacy compared to just below 100°C.  The notable benefits include the substantial formation of VFAs and enhanced COD solubilization at high oxidant dosages.  Considering the additional capital and operating costs required to run a process above 100°C and the recalcitrant nature of the treated substrate, such temperatures are not recommended unless high solids reduction or solubilization is desired.  Pretreating with O3 preferentially mineralized organics, but otherwise, had near negligible effects on the other treatment parameters.  The use of O3 with the MW-H2O2 process for commercial applications should be considered if very large sludge reductions are desired.  In such cases, the O3 should be injected concurrently with the H2O2 and immediately prior to the MW, in order to maximize their synergetic effects.  High H2O2 dosages are able to improve CSTs, as previously hypothesized, although the additional costs may not be justified by their benefits.   5.3 Experiments on the 27MHz Radiofrequency Dielectric Heater Using the 27MHz RF heater modified for continuous operation, four experiments were carried out to determine the effect of H2O2 on WAS.  A fifth experiment was then conducted to investigate the feasibility of operating the MW/RF-H2O2 system under a low temperature regime, with a particular focus on OP recovery.   5.3.1 Effect of Hydrogen Peroxide To investigate the effect of H2O2 with RF EMR, the following four experiments were conducted: RF heating only, low H2O2, medium H2O2, and high H2O2.  Details of each run‟s experimental conditions are summarized in Table 16.  Since the RF heater is operated below 100°C, samples were collected throughout an experiment to understand how process parameters changed with temperature. Table 16 – Experimental conditions for WAS experiments investigating the effect of H2O2 using the 27MHz RF heater  Conditions RF Only High H2O2 Medium H2O2 Low H2O2 Volume (L) 12 Flow Rate (L/min) 6.5 Final Temperature (°C) 90 H2O2 Dosage %(v/v) - 0.6 0.52 0.3 %(v/v)/%TS - 0.63 0.33 0.23 H2O2 Addition (°C) - 60 Heating Rate (°C/min) 1.50 1.47 1.46 1.47 Total Run Time (min:s) 49:30 49:40 51:20 49:30 72  5.3.1.1 Digestion Indicators The solubilization of COD, shown in Figure 21 as a percent of TCOD, was enhanced with the addition of H2O2 and increased with higher dosages.   Figure 21 – Percent of SCOD from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater  For all runs, the thermal hydrolysis of COD started at 45°C and plateaued after 60°C, unless H2O2 was added.  Furthermore, the SCODs were similar at 60°C, despite their initial TS varying from 0.96±0.01 to 1.579±0.003%.  Thus, thermal hydrolysis is independent of the substrate‟s solids content, and treating higher solids containing WAS would be more energy efficient.  After the addition of H2O2 at 60°C, the SCOD increased linearly, except for the highest H2O2 dosage tested, for which the SCOD plateaued after 75°C.  This suggests that a certain fraction of the solids is readily hydrolyzed, while the remaining non-dissolved organics require greater amounts of H2O2 per molecule to become soluble.  It can be approximated that this easily solubilized fraction of organics constituted ~45% of the TCOD for these experiments‟ substrates.    Comparing the TCOD between the end of a run at 90°C and its respective raw sample, no changes were observed for the RF only and low H2O2 trial, as confirmed by their constant TS measurements (see Table 18 in Section 5.3.1.3).  The medium H2O2 run yielded a similar decrease in its TCOD and TS of 9±5 and 13.8±0.3%, respectively, indicating that some organics were completely oxidized at this dosage.  For the high H2O2 trial, one would then expect a larger degree of mineralization; however, the results show a 9±1% decrease in TS and a 2.5±0.1% increase in TCOD.  A possible explanation for these contradictory results is that not all of this run‟s H2O2 may have fully reacted.  If such is the case, any residual H2O2 would interfere with the COD test by reacting with its reagent (K2Cr2O7), thereby falsely inflating its results (Lee et al., 2011).  Using the theoretical consumption of K2Cr2O7 by H2O2 and assuming that the 010203040506020 30 40 50 60 70 80 90SCOD (%) Temperature (°C) RF OnlyLow H₂O₂ Medium H₂O₂ High H₂O₂ 73  reduction in TS should be proportional to that of TCOD, the corrected TCOD is calculated to be 9 100±60mg/L.  The final SCODs at 90°C are plotted with respect to their H2O2 dosages in Figure 22 for all four experiments, along with the corrected high H2O2 run.    Figure 22 – Effect of H2O2 dosage on SCOD at 90°C from WAS experiments using the 27MHz RF heater The corrected high H2O2 run from Figure 22 still shows a decreased marginal return from the additional H2O2.  It also reveals how much H2O2 was likely left unconsumed, due to insufficient reaction time.  To delineate the above observations, the reactivity of H2O2 can be divided into three stages, following its addition to WAS (at 60°C):  1) Readily degradable organics (ex. loosely bound EPS) start to be solubilized 2) As degradation progresses, mineralization starts to be observed for sufficiently high H2O2 dosages 3) The rate of COD solubilization starts to decline as the readily degradable organics become completely solubilized, leaving behind less degradable and more refractory organics.  Some of these refractory substances may be produced by the MW/RF-H2O2 process itself, as discussed in Section 5.2.3.  After this threshold, the remaining organics require more H2O2 per mole to be solubilized or mineralized.  Depending on the heating rate, the above stages may be less distinct and overlap one another.  These claims are further supported by observing that the corrected SCOD released from the high H2O2 RF run (47±1%) was only 5±2% lower than that from the high temperature 915MHz MW-H2O2 experiment (Section 5.2.1), despite the RF run operating at a lower H2O2 dosage and temperature.  This again confirms the minimal returns from using higher temperature and H2O2 treatment conditions.  Alternatively, this may also hint at potential gains from using a RF dielectric heater, discussed later in Section 5.4.  One benefit of high temperature operation is that the H2O2 will react and decompose faster, 01020304050600 0.1 0.2 0.3 0.4 0.5 0.6 0.7SCOD at 90°C (%) H₂O₂ Dosage (%(v/v)/%TS) Measured RF RunsCorrected High H₂O₂ Run 74  thus, larger dosages can be completely consumed within the same treatment time.  This is supported by the nearly equivalent reduction in TCOD and TS for the high temperature 915MHz runs, indicating no residual H2O2.  It should be noted that variations in their substrates‟ physical, chemical, and biological characteristics may, instead, be responsible for the minimal SCOD differences observed between the two dielectric heaters.  The production of VFA, shown throughout each run in Figure 23, required H2O2 to continue increasing past 60°C, but yielded final concentrations that are independent of the dosage.  Figure 23 – VFA results from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater Instead, the final concentrations of VFAs for the three H2O2 utilizing runs are correlated to their substrates‟ initial TS values.  VFAs that are produced by thermal hydrolysis of WAS derive from long chain fatty acids, particularly those that are highly unsaturated, as well as from proteins and their amino acids (Wilson and Novak, 2009).  Higher solids containing WAS would contain greater amounts of lipids and proteins, explaining the results‟ dependence on initial TS values.  Interestingly, the 110°C 915MHz MW-H2O2 run (Section 5.2.1) had a final VFA concentration 2.7 times larger than that from the high H2O2 RF run.  Such a substantial increase shows that the production of VFAs is likely limited below 100°C. 5.3.1.2 Nutrient Release The release of OP throughout each run is plotted in Figure 24.  The final concentrations of OP at 90°C appear independent of the substrates‟ H2O2 dosage and TP.  The samples at 60°C though, do appear dependent on the H2O2 dosage, where lowering it resulted in a higher OP peak. 02040608010012020 30 40 50 60 70 80 90 100VFA (mg/L) Temperature (°C) RF OnlyLow H₂O₂ Medium H₂O₂ High H₂O₂ 75   Figure 24 – OP results from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater With respect to temperature, OP revealed a consistent relationship among all runs, whereby it rose rapidly at 45°C to a maximum at 60°C, before falling sharply thereafter.  OP levels then started to gradually recover after 75°C for oxidant utilizing treatment runs.  Similar OP trends have been observed for other MW treatment studies (Bailey, 2015; Chan et al., 2007; Kuglarz et al., 2013; Wong et al., 2006a).  The possible mechanisms behind these trends, as well as those observed from other WAS experiments, are discussed in detail in Section 5.6.  The speciation of TP into OP, PP, organic STP (i.e. the remainder of STP), and non-STP is shown in Figure 25, for all four runs at 90°C.  Figure 25 – TP speciation at 90°C from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater 010203040506020 40 60 80 100OP (mg/L) Temperature (°C) RF OnlyLow H₂O₂ Medium H₂O₂ High H₂O₂ 0100200300400500600700800RF Only Low H₂O₂ Medium H₂O₂ High H₂O₂ TP (mgP/L) Remaining TPOrganic STPPPOP76   From Figure 25, PP had no observable correlation with H2O2, but showed a greater release from substrate‟s with higher overall TPs.  A higher TP indicates a larger reservoir of PPs in the phosphorus accumulating organisms (PAO) found in the WAS, which was collected from an EBPR process.  Ostensibly, this allows a greater release of PP from the MW/RF-H2O2 process.  Considering temperature, PP was only partially released at 60°C, but by 75°C, it had nearly reached its maximum (see Appendix C).  This fits with the lysing of cells that commenced at 60°C during the MW-H2O2 process (Hong et al., 2006; Pino-jelcic et al., 2002), facilitating the release of the cell‟s internal PP reserves.  The amount of organic STP (Figure 25) was independent of H2O2 and TP, while the RF only and high H2O2 run alone showed a significant increase in organic STP with temperature (Appendix C).  Assumed to comprise proteins, DNA, membranes and other P containing intracellular components, organic STP evidently requires greater temperatures or oxidants to be converted into OP. Analysing the above TP data, measured using the colourimetric method, suggests that some samples may have under reported TP values.  In particular, the medium H2O2 run had a higher TS value (1.579±0.003%) than the low H2O2 run (1.33±0.01%), despite having statistically equivalent TP values.  It is well known that TP is correlated to TS in EBPR systems, suggesting that there remains undigested P in the total fractions.  The TKN results, which uses the same digestion procedure (see Section 3.3.3), also followed a similar trend (see Appendix C for data).  This further supports the proposition that the digestion was incomplete for these high TS runs.  A third piece of evidence is that the final PP values at 90°C were much higher for the medium H2O2 run, as would be expected, based on that run‟s larger TS result.  Comparing the TP and STP measurements made from the colourimetric method (Section 3.3.2) to those from the ICP method (Section 3.3.8), yield additional insights.  Although both sets of results followed a similar trend, major discrepancies wer observed when these two methods are compared to one another.  These discrepancies are plotted as the difference between the methods‟ STP and TP measurements in Figure 26.  A negative value in Figure 26 means that the TP or STP, as measured by the colourimetric method, is greater than that observed from the ICP method. 77   Figure 26 – Differences in TP and STP measurements between the colourimetric and ICP method for WAS experiments investigating the effect of H2O2 using the 27MHz RF heater Figure 26 shows that the RF only and high H2O2 runs‟ (lowest TS and TP) measurements were equivalent at all temperatures, while the medium and low H2O2 experiments had significant differences, particularly for their raw TP samples.  To summarize, it is clear that the ICP method requires a stronger digestion procedure for samples with a TS of ~1% or higher.  Similarly, for the colourimetric method, there were likely undigested TKN and TP for samples with a TS of 1.33% or greater.   The fraction of STKN increased linearly with temperature and the run‟s H2O2 dosage (see Appendix C).  The maximum STKN release was 70±10% for the high H2O2 run at 90°C.  The release of NH3 throughout all four runs is plotted in Figure 27.  Final NH3 concentrations at 90°C are also positively correlated to the H2O2 dosage, where no H2O2 actually resulted in a declining trend after ~60°C.  The thermal treatment of WAS produces NH3 via the hydrolysis and degradation of proteins (C. a. Wilson & Novak, 2009), particularly soluble proteins (Liu et al., 2009).  It follows that higher H2O2 dosages will enhance the release of soluble proteins.  Additionally, H2O2 is capable of reacting with certain proteins to directly produce NH3 (Walter et al., 1955), resulting in its observed trend.  -50050100150200250300350400RF Only High H₂O₂ Med H₂O₂ Low H₂O₂ DIfference in P Measurements (mgP/L) TP RawSTP 60°CSTP 75°CSTP 90°C78   Figure 27 – Release of NH3 from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater The solubilization of Ca, K, and Mg primarily occurred between 45 and 60°C.  Higher H2O2 dosages increased the final fraction of solubilized Ca and Mg, while K was nearly identical for all four runs.  The independence of K solubilization from H2O2 is because it is primarily found as an intracellular cation in WAS; see Section 5.2.2 for details.  Conversely, Ca is largely found outside the cell and is the primary ion that binds floc together, according to the divalent cation bridging theory (Cousin and Ganczarczyk, 1999; Neyens and Baeyens, 2003).  This means that large amounts of Ca are in the EPS, either as a binding cation, or trapped within the floc as one of its precipitates, such as CaCO3 (Peeters et al., 2011).  Thus, the enhanced destruction of floc caused by increasing the H2O2 dosage is presumed to release a higher fraction of Ca, either in its cation or precipitate form.  To a lesser extent, Mg is also found in EPS as a divalent bridging cation and precipitate (Bruus et al., 1992; Keiding and Nielsen, 1997), explaining its similar trend to Ca.  Additionally, Mg2+, and to a lesser extent Ca2+, act as counter ions to PPs in WAS and are known to be released in conjunction with each other (Jardin and Pöpel, 1996), further supporting the above observations (discussed in Section 5.6.3).  A benefit of releasing Mg is found in the changing molar ratio of struvite‟s constituents, presented in Table 17 for final samples taken at 90°C. Table 17 – Struvite molar ratios from WAS experiments investigating the effect of H2O2 at 90°C using the 27MHz RF heater  Run NH4+:Mg2+:PO43- (mol) RF Only 1:5.1:2.4 Low H2O2 2.8:3.7:1 Medium H2O2 1.7:3.4:1 High H2O2 1:4.4:1.1 051015202530350 20 40 60 80 100NH₃ (mg/L) Temperature (°C) RF OnlyLow H₂O₂ Medium H₂O₂ High H₂O₂ 79   Struvite ratios behaved similarly to those observed for the high temperature 915MHz MW experiments (see Section 5.2.2), whereas with RF heating only, the limiting species was NH3, but with the use of H2O2, OP becomes limiting.  Once again, the use of H2O2 creates favourable conditions for the formation of struvite.  5.3.1.3 Physical Properties  Select physical properties of the WAS samples are summarized in Table 18.  For the complete set of data, refer to Appendix C. Table 18 – Select physical properties from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater  Run Sample TS (%) TSS (g/L) VSS (%) CST (s) RF Only Raw 0.99±0.01 9.9±0.2 80.0±0.4 160±50 90°C 0.96±0.01 6.9±0.3 88±2 830±10 Low H2O2 Raw 1.33±0.01 12.7±0.3 81.1±0.8 540±20 90°C 1.320±0.001 8.3±0. 3 86±2 1200±100 Medium H2O2 Raw 1.579±0.003 15.9±0.5 82.7±0.3 590±70 90°C 1.361±0.004 7.3±0.1 82±2 700±100 High H2O2 Raw 0.96±0.01 9.17±0.07 83±2 180±50 90°C 0.870±0.003 3.60±0.09 84±1 220±70  TS, as discussed in Section 5.3.1.1, decreased slightly with higher H2O2 dosages due to the complete oxidation of organic compounds.  It is strongly correlated with TCOD, except in the case where residual H2O2 persists.  TVS changed negligibly with H2O2, unlike the decreases observed with the higher temperature and oxidant utilizing 915MHz MW runs (Section 5.2.3).  As expected, TSS was strongly correlated to SCOD, with PCC values greater than -0.96 for all treatment runs.  The VSS results were observed to increase at 90°C for the RF only and low H2O2 runs, while remaining unchanged with higher H2O2 dosages.  The high temperature oxidant free 915MHz trial also exhibited VSS increases.  From these two sets of experiments, it appears that the volatile fraction of WAS increased when only dielectric heating or low H2O2 dosages were used, but then decreased at higher dosages.  It can be postulated that the first by-products created from the MW/RF-H2O2 process are volatile compounds that are then consumed or out produced by more refractory substances, if sufficient amounts of H2O2 are provided.  Together, the TSS and TVS suggest that, at the temperature-H2O2 combinations tested with the RF heater, the treated WAS was not substantially less recalcitrant.  From Table 12, the raw samples‟ CST values increased with higher TS, which has been pointed out by other researchers as a drawback of using the CST test as an indicator for dewaterability (Vesilind, 1988).  80  Treating with the RF-H2O2 process did not improve the CST compared to the initial raw value, for any of the runs.  However, the relative increases in CST after treatment lessened with higher H2O2 dosages (PCC -0.85).  In fact, at the highest H2O2 dosage tested, the increase in CST was not statistically significant.  With even higher H2O2 concentrations, the CST can drop to below that of the initial raw sample, as is seen from the high temperature 915MHz MW runs (Section 5.2.3).  Therefore, the studies presented thus far demonstrated that a threshold exists with respect to H2O2, whereby increasing the amount of H2O2 initially worsens a substrate‟s dewaterability, until a maximum CST is reached.  As the H2O2 dosage is increased further, CST starts to decline until it is below its initial value.  To better understand the changes in CST, select particle size functions are plotted in Figure 28.  Since the RF only and low H2O2 runs had nearly identical curves to their respective raw samples‟, they are omitted from Figure 28.  As well, the raw samples‟ PSD functions were very similar between the medium and high H2O2 runs; thus, their average is plotted in Figure 28 for clarity.    Figure 28 – Select PSD functions at 90°C from WAS experiments investigating the effect of H2O2 using the 27MHz RF heater The above plot shows how higher H2O2 dosages resulted in broader and shorter PSD peaks that are shifted towards lower particle sizes, particularly to sizes less than 1µm.  Looking at the impact of each particle size category on CST, it was determined that more particles in the range of 10 to 50µm are correlated to higher CSTs.  This information can be used as a guide in future studies to estimate the impact of a MW/RF-H2O2 treatment on a substrate‟s dewaterability.  The quoted particle size range also fits with previous studies, which found that supracolloidal particles (1 to 100µm) inhibit a sludge‟s dewaterability (Barber and Veenstra, 2015; Karr and Keinath, 2014).  According to the results presented here, to 01234567890.1 1 10 100 1000Volume (%) Size (µm) Average RawMedium H₂O₂ High H₂O₂ 81  improve the CST of a WAS sample, the total volume percent of particles within the size range of 10 to 50µm must be reduced to less than 38%. 5.3.2 Low Temperature Study From the RF studies presented in the preceding section, OP was observed to consistently peak at 60°C.  As this is also the optimal temperature to inject H2O2, it was decided to attempt a low temperature, low H2O2 continuous RF experiment.  The goal was to investigate the possibility of maximizing OP release at more economical treatment conditions.  The chosen experimental conditions are summarized in Table 19.  Select results are presented in the following sections; for the complete data set, see Appendix D. Table 19 – Experimental conditions for a low temperature WAS run using the 27MHz RF heater  Condition Value Volume (L) 12 Flow Rate (L/min) 6.5 Final Temperature (°C) 60 H2O2 Dosage %(v/v) 0.55 %(v/v)/%TS 0.41 H2O2 Addition (°C) 60 Heating Rate (°C/min)* 2.01 Total Run Time (min:s) 81:30 Holding Time (min:s) 60 *Note: does not include holding time  Samples were taken before treatment and 15, 30, 45, and 60 minutes after the addition of H2O2 at 60°C.  A sample at 60°C, immediately prior to the injection of H2O2, was to be taken at a subsequent experiment using the same WAS, but technical difficulties prevented the experiment from occurring.  However, another 60°C sample was taken 13 days later from a RF experiment using WAS at similar conditions as those listed in Table 19 (see the low H2O2 experiment in Section 5.3.1).  As such, this sample is used in the proceeding discussion to approximate the aforementioned missing data point.  A comparison of the raw samples from each respective experiment is provided in Table 20, showing minimal differences for the majority of parameters.  The validity of using this later experiment‟s 60°C sample, referred to henceforth as the „borrowed‟ sample, is addressed individually for each treatment parameter.     82  Table 20 – Comparison of raw samples from the low temperature and low H2O experiments using the 27MHz RF heater on WAS Parameter Raw Sample Low Temperature RF Study Low H2O2 RF Study (Section 5.3.1) TS (%) 1.352±0.001 1.33±0.01 TVS (%) 80.12±0.01 80.01±0.06 TSS (g/L) 13.7±0.2 12.7±0.3 VSS (mg/L) 84±2 81.1±0.8 TCOD (g/L) 14.0±0.2 13.44±0.09 SCOD (mg/L) 143±7 131±5 CST (s) 260±40 540±20 TP (mg/L) 640±20 710±20 STP (mg/L) 11.7±0.2 0.3±0.5 OP (mg/L) 5.2±0.8 0.111±0.002 TKN (mg/L) 1070±20 1050±30 STKN (mg/L) 14±1 8.5±0.7 NH3 (mg/L) 5.61±0.02 0.11±0.04 VFA (mg/L) 40±10 5.4±0.6 5.3.2.1 Digestion Indicators  TCOD was equivalent across all samples, indicating that minimal mineralization occurred and that the H2O2 was largely consumed, as confirmed by near constant TS values (Table 21).  The fact that no residual H2O2 was observed is noteworthy, as the H2O2 dosage used is near the consumption threshold found for the previous RF studies at 90°C.  After 60°C and the addition of H2O2, the SCOD increased substantially, but did not change further over the course of the holding time.  Table 21 – Select digestion parameters for the low temperature WAS experiment using the 27MHz RF heater Sample TCOD (g/L) SCOD (%) TS (%) VFA (mg/L) Raw 14.0±0.2 1.03±0.05 1.352±0.001  48±2 60°C (Borrowed) 13.27±0.08 11.0±0.2 1.291±0.005 43±3 15min 14±1 25±2 1.326±0.003 52±1 30min 14.0±0.2 27±2 1.311±0.002 51±2 45min 14.0±0.5 26±1 1.310±0.003 50±2 60min 14.0±0.3 26.0±0.8 1.325±0.001 19±3  By considering the borrowed 60°C sample, it is apparent that the majority of the SCOD increase was due to H2O2, with thermal degradation being responsible for the remaining SCOD gains.  It is appropriate to use the borrowed data point, as the raw samples from each experiment had TS, TVS, TSS, VSS, TCOD, and initial SCOD measurements that differed by less than 8% (although the two samples‟ properties are significantly different at the 95% confidence level).  To determine if H2O2 was used more effectively at higher treatment temperatures, as expected from a review of the literature (Section 2.2.1), the previously 83  presented RF experiments at 90°C are used as a comparison.  By extrapolating their SCOD results to this run‟s H2O2 dosage of 0.41%(v/v)/%TS, and using the RF only run as a baseline, the contribution from thermal degradation and H2O2 to the solubilization of COD is estimated in Figure 29 at 60 and 90°C.  The error bars displayed in Figure 29 are calculated by propagating the standard deviation of all pertinent measurements and do not include other possible sources of uncertainty.  Figure 29 – Contributions to SCOD gains from the low temperature RF run and an estimated run at 90°C for WAS Figure 29 reveals an impressive synergistic effect between H2O2 and temperature, where using the same H2O2 dosage (0.41%(v/v)/%TS) at 90°C, yielded a considerable SCOD gain, beyond that produced from the additional thermal degradation.  In consequence, H2O2 was estimated to be ~2 times more effective for a given dosage, when used at a treatment temperature of 90°C instead of 60°C.  This synergy can be partly explained by the formation of the more destructive *OHs at higher temperatures (see Section 2.2.1.2).   The total VFAs were constant at 60°C for holding times up to 45 minutes, but quickly declined thereafter (Table 21).  Interestingly, the VFAs hardly increased above the raw value, which is at odds with previous results.  It is possible this is due to the raw sample‟s abnormally high initial VFA concentration.   5.3.2.2 Nutrient Release The release of OP during the experiment is plotted in Figure 30.  Considering the borrowed data point, OP peaked at 60°C, as is expected according to previous studies, but then dropped to a steady state level after 15 minutes.  Possible mechanisms to explain the OP‟s trend are detailed in Section 5.6.2. 0102030405060RF Run at 60°C RF Run at 90°CSCOD (%) Contribution from H₂O₂ Thermal DegradationContribution from 60 to 90°CThermal DegradationContribution from 20 to 60°C84   Figure 30 – OP release from the low temperature WAS experiment using the 27MHz RF heater  PP was stable with respect to holding time at approximately 195mg/L.  It was also largely released at 60°C, again matching with previous observations.  Analysing the nutrient data for its struvite recovery potential, the molar ratio of NH3:Mg2+:PO43- changes from having a limiting species of NH3 at 60°C, to that of OP, after 15 minutes of holding time.  This change is due to rising NH3 and declining OP concentrations after the 60°C sample.  The borrowed 60°C data point is appropriate to include in this analysis because the raw samples from both experiments had TP and TKN levels that differ by 10% or less.   5.3.3 Conclusions Pertaining to Commercial Applications Summarizing the RF experiments presented in this section, 27MHz RF dielectric heaters were comparable to MWs at 915MHz, in terms of their treatment capabilities.  Regarding the effective use of H2O2 to maximize the solubilization of COD, the following points bear consideration:   1) The rate of H2O2 consumption varies with the substrate and treatment temperature.  When using high dosages, sufficient reaction time must be available to ensure that the H2O2 is fully utilized 2) Substrates have a H2O2 dosage threshold, where the marginal SCOD return from a higher dosage starts to decline due to the consumption of readily solubilized organics 3) The impact of H2O2 on SCOD is greater when used at higher temperatures, due to a synergetic effect  Investigating the feasibility of a low temperature MW/RF-H2O2 treatment regime reveals that OP cannot be maximized, as its initial release is inherently unstable and declines within 15 minutes.  Furthermore, holding the substrate at 60°C does not improve any of its other treatment parameters.  Achieving large SCOD fractions would, therefore, be challenging as higher H2O2 dosages would have to be used; 020406080100120140Raw 60°C 15min 30min 45min 60minOP (mg/L) 85  however, such dosages would not be consumed as efficiently or as rapidly at the lower treatment temperature.  In terms of operating costs, treating at 60°C would reduce energy expenses, but the additional H2O2 that would be required may offset these savings.  To conclude, the use of a low temperature MW/RF-H2O2 treatment process is not recommended for WAS.    5.4 Comparison Study The results from the previous RF experiments suggest that, although a RF dielectric heater was comparable to that of a MW heater, significant differences may exist.  Thus, the goal of this study was to determine if any such differences are measurable.  To accomplish this, a comparison experiment was undertaken that required both dielectric heating systems to treat the same sample of WAS, using the same experimental conditions.  The chosen conditions and any differences between the two runs are noted in Table 22.   Table 22 – Experimental conditions for the dielectric heater comparison study using WAS Conditions 915MHz 27MHz Substrate WAS (TS of 1.579±0.003%) Volume (L) 20 12 H2O2 Dosage %(v/v) 0.52 %(v/v)/%TS 0.33 H2O2 Addition (°C) 60 Heating Rate (°C/min) 1.49 1.46 Flow Rate (L/min) 6 5.4.1 Summary of Results Each set of results are compared using the t-test and evaluated at the 95% confidence level, to discern between significant differences and those that fall within the range of probable errors.  The statistical analysis is summarized in Table 23, with non-significant differences reported as „Equal‟ and significant changes indicated by their percent difference.  Positive percent differences signify that the MW‟s result is larger than that of the RF heaters, and vice-versa.  The RF run used in this comparison study is the same as the medium H2O2 experiment presented in Section 5.3.1, while the MW run‟s complete data set is tabulated in Appendix E.   86  Table 23 – Statistical test results from the dielectric heater comparison experiment using WAS Parameter 60°C (%) 75°C (%) 90°C (%) TS 0.5 -5.7 9.8 TVS -0.2 Equal 1.2 TSS Equal Equal 18.0 VSS Equal Equal Equal TCOD Equal Equal 7.2 SCOD  -8.3 Equal Equal OP  -36.6 14.3 Equal TP Equal Equal Equal STP  -9.5 Equal 16.3 PP  Equal Equal 15.3 NH3  -15.3 15.7 Equal TKN  -12.6 -17.6 Equal STKN  4.3 Equal 13.9 VFA  -59.4 Equal Equal Sol Ca Equal Equal 4.2 Sol K Equal Equal Equal Sol Mg -7.6 5.6 Equal CST - - 29.9 D10 - - 40.9 D50 - - Equal D90 - - -16.8 Note: positive percent differences indicate that a higher value was measured for the 915MHz MW 5.4.2 Discussion on Digestion Indicators  From Table 23, the TS and TCOD are nearly equal over both runs at 60°C, but by 90°C, the RF run is significantly lower for both parameters.  It appears that the RF heater is able to start mineralizing organics at these treatment conditions, while the MW heater‟s results are not statistically different from their initial raw values.  This surprising result could be explained by the differences between the dielectric oven‟s heating mechanisms, as explained in Section 5.4.5.  Regarding the soluble solids content, the SCOD at 60°C had a significant increase in favour of the RF heater, suggesting that the heating mechanisms in the RF range are more effective at thermally degrading biomass.  Although the absolute SCOD is higher for the RF heater at 90°C (by ~8%), it is not a statistically significant gain.  However, the RF run‟s greater TSS reductions at 90°C are statistically real, implying that the SCOD difference was also a true increase and that the larger errors associated with the COD analysis prevent it from passing the t-test.  Taking these trends into account explains how the final fraction of solubilized COD is significantly higher, by ~16%, for the RF run.  VFA production is not significantly different between the two heaters except at 87  60°C, where the RF heater is over 50% higher. This apparent advantage at 60°C matches the SCOD observations, further supporting the possibility that RF heating may enhance thermal degradation.  5.4.3 Discussion on Nutrient Release Considering the STP and STKN data at 90°C (Table 22), the MW heater shows a larger release, implying that a greater amount of proteins are being solubilized and degraded.  However, the additional solubilization of nitrogen is not supported by the release of NH3, which is equal at 90°C.  It is important to note that the TP analysis from the colourimetric and ICP method use digestion procedures that are likely inadequate for this study‟s high solids containing WAS (see Section 5.3.1.2 for discussion).  This means that the STP and TP measurements may be lower than their actual values and that the observed difference between the two heaters is less certain.  Since the STKN data relies on the same digestion method as the colourimetric TP analysis, the above uncertainties also apply to its results.   PP had a larger release for the MW heater, indicating that its cells were lysed to a greater extent.  OP though, did not show a clear trend, as the RF heater‟s substantial gain at 60°C disappeared by the end of the run.  This enhancement at 60°C matches with the many other solubilization improvements observed with the RF heater.  Since the nutrients required for struvite (NH3, Mg2+, and PO43-) are found to be statistically equivalent between both frequencies, neither one can be said to offer an advantage in terms of recoverable struvite. Reviewing the above nutrient observations, there appears to be an advantage with MW heaters in terms of solubilizing TP, TKN, and PP, but issues with the analytical digestion method cast an unquantifiable uncertainty on the former two parameters.  Before H2O2 is added though, the RF heater is able to release more NH3 and OP. 5.4.4 Physical Properties TS and TSS are not equal at 90°C (Table 23), as supported by the COD data discussed above.  The TVS reveals minimal differences, although they were significant at the 95% confidence level, while VSS was equal at all temperatures.  It can be demised from the TVS and VSS that the refractory nature of the WAS was not substantially different between the two runs.   PSD curves are plotted in Figure 31 for the samples treated at 90°C, showing that both heaters shifted the PSD profile towards smaller particles, as regularly seen with the MW/RF-H2O2 process. 88   Figure 31 – PSD functions at 90°C from the dielectric heater comparison experiment using WAS The RF heater though, has a greater PSD shift that enhances the formation of smaller and larger particles. This observation is confirmed by the statistically different cumulative D values (Table 23), where the D10 results are larger and the D90 results are smaller for the MW heater.  This additional breakdown of particles by the RF heater collaborates with the SCOD and TSS results discussed above. A substantial difference of nearly 30% is observed between the CST results, with the RF heater yielding the lower value.  In fact, the CST for the RF heater was not significantly different from that of the raw initial sample, albeit large standard deviations (22%) are found.  The superior CST performance can be explained by the smaller volume fraction of particles with sizes between 10 to 60µm for the RF heater (59%), compared to the MW heater (72%).  The importance of this particle size range on a WAS‟s dewaterability is discussed in Section 5.3.1.3.  This correlation means that the mechanisms causing greater solids degradation for the RF heater are also likely responsible for indirectly lowering the CST. 5.4.5 Theoretical Explanation of Observations The majority of the observed differences between the two heaters pertain to the hydrolysis and general degradation of organics, where changes in TSS, TCOD, SCOD (fraction), D10, D50, and D90 values suggest a superior destruction of solids by the RF heater.  These differences can be attributed to the distinct heating mechanisms that dominate at each respective EMR frequency range (see Section 2.1.3 for detailed explanations of the various mechanisms).  Furthermore, the impacts of the various dielectric loss mechanisms on biological systems are summarized in Section 2.1.4, where conclusions made regarding the effectiveness of each loss phenomena appear to be validated in this study.  Specifically, surface polarizations that manifest at RFs allow charges to accumulate along cell membranes.  This could induce the dielectric breakdown of the cell and enhance the frequency of cell lysing.  Similarly, the dominance of ionic conductance at 27MHz, known to be especially important when heating TWAS (shown in Section 01234567890.1 1 10 100 1000Volume (%) Size (µm) RawMW Heater at 90°CRF Heater at 90°C89  2.1.4.2), allow the ionic rich intracellular environment and cytoplasm to be selectively heated above that of the surrounding bulk water.  As such, these mechanisms together target cells and their membranes, theoretically resulting in their enhanced breakdown.  More cell lysing would result in a greater release of their internal components, explaining the observed shift in PSD and TSS reductions.  A secondary consequence is that the pieces of cells and their once internally held organic constituents would be more available for attack by H2O2.  Normally, H2O2 would have to diffuse through the membrane before reacting with such substances, thereby offering a certain degree of protection.  The TCOD and TS reductions could be the result of this effect.  WAS contains large concentrations of extracellular ions and so the ionic conductance mechanism is clearly well suited to heating the bulk water as well.  These two dielectric loss phenomena therefore, are plausible candidates for the RF heater‟s positive performance, particularly at 60°C before the addition of H2O2. Conversely, for the MW frequency range, one would expect that the dominant dipolar rotation mechanism would have a significant impact on the WAS‟s high water content.  However, much of that water is bound by floc and EPS, resulting in a reduced ability to be dielectrically heated.  Interestingly, after the addition of H2O2, several of the solid degradation indicators show a temporary rebound in favour of the MW heater (Table 23), notably for TS and SCOD.  This could be due to the more effective heating of the highly polarizable H2O2 molecule (Maroulis, 1992), which would focus thermal energy on it and improve its reactivity and decomposition into *OHs.  If this is true, such effects are still outperformed by the RF heater‟s mechanisms before the end of the run.  Similarly, the greater solubilization of STP and STKN, if indeed a replicable effect, could be explained by the dipolar rotation heating mechanism in the MW range.  Many proteins are known to have large dipoles due to their helical structure and charge carrying regions (Felder et al., 2007).  Thus, their potentially enhanced breakdown, when exposed to MW EMR, could result in higher STP and STKN levels. CST results are also observed to improve with the RF heater.  The enhanced solids degradation explained above is likely to be at least partially responsible for this.  Another potential reason why RF heating has such a profound effect on CST could be due to its ionic conductance mechanism targeting floc bridging cations.   By exciting these cations and focusing thermal energy on the flocs, the EPS could be more effectively destroyed and broken down into smaller particles, supporting the above conclusions. 5.4.6 Major Conclusions  To answer this study‟s original question, there were significant differences between a 27MHz RF and a 915MHz MW heater.  Overall, the RF heater outperformed the MW, in terms of solubilizing and degrading organics, most likely due to its unique dielectric loss mechanisms (i.e. ionic conductance and 90  surface polarization).  These effects also resulted in a superior CST.  MW heating may offer advantages with respect to the thermal activation of H2O2 and solubilization of proteins via its dipolar rotation mechanism, although it is recommended to further investigate these claims to obtain more certain results.  For the design of commercial scale processes, a RF heater appears to offer real benefits over a MW and should be given full consideration. 5.5 Effect of Heating Rate on Pilot-scale Systems Commercial designs of the MW/RF-H2O2 process would likely use high-powered, dielectric heaters to rapidly treat a substrate continuously.  Thus, it is important to understand the effect of higher heating rates on treatment efficacy.  Previous experiments using batch MW‟s found that a faster heating rate improved the SCOD release, until a threshold was reached (Lo et al., 2010).  However, these gains are likely due to the fact that a faster heating rate limited the exposure time of H2O2 to catalase, until temperatures above 60°C inactivated the enzyme.  Since the construction of pilot-scale, dielectric heaters, that allow H2O2 to be injected at any temperature, the effect of heating rate has not been re-examined until now.  To investigate the impact of the heating rate, the 915MHz MW experiment presented in the previous section is used as a control study.  A second experiment was then undertaken using the same heater and operating conditions, except for the heating rate, which was increased by a factor of 2.5.  The faster heating rate was accomplished by reducing the volume of treated WAS.  A comparison of the treatment conditions for each run is provided in Table 24.  The results for the slow heating rate control study are discussed in full in Section 5.4, and the complete data set for the fast heating rate run is available in Appendix F. Table 24 – Experimental conditions for the heating rate comparison study using a 915MHz MW heater on WAS  Parameter Fast Heating Rate Slow Heating Rate Volume (L) 4.5 20 H2O2 Dosage %(v/v) 0.52 0.52 %(v/v)/%TS 0.36 0.33 H2O2 Addition (°C) 60 Heating Rate (°C/min) 3.91 1.49 Flow Rate (L/min) 6 5.5.1 Digestion Indicators COD measurements for the raw and 90°C samples are compared in Table 25, along with t-test results used to determine if the differences between the two runs are statistically significant (evaluated at the 95% confidence level).  If the two data points are statistically equivalent, the t-test result is reported as „Equal‟.  Otherwise, the percent difference is reported, where a positive value indicates that the higher result is from the slower heating rate run. 91  Table 25 – COD results from the heating rate comparison study using a 915MHz MW heater on WAS Sample COD Parameter Slow Heating Rate Fast Heating Rate T-test Results Raw TCOD (g/L) 16.1±0.6 18.6±0.2 -16% SCOD (g/L) 0.142±0.003 0.145±.004 Equal SCOD (%) 0.88±0.04 0.78±0.02 11% 90°C TCOD (g/L) 14.78±0.07 15.8±0.1 6.2% SCOD (g/L) 5.7±0.2 6.1±0.3 Equal SCOD (%) 39±1 39±2 Equal Note: positive percent differences indicate that a higher value was measured for the slow heating rate  The fact that the SCOD does not differ for either of the samples confirms that solids solubilization is governed by the H2O2 dosage and final temperature, and not the heating rate.  The exception to this claim would be for high H2O2 dosages that no longer have enough time to completely react due to an increased heating rate.  In such a case, the SCOD fraction can be expected to decline.  At the conditions tested here, there is no evidence to suggest residual H2O2 in either of the runs.  The VFA results, shown in Figure 32, are nearly three times higher by 90°C under a slower heating rate.  This difference accelerates after 60°C, matching previous observations that VFA production is markedly dependent on H2O2, although it is not necessarily influenced by the dosage (Section 5.3.1.1).    Figure 32 – VFA results from the heating rate comparison study using a 915MHz MW heater on WAS 5.5.2 Nutrient Release Phosphorus and its dependency on the heating rate are briefly summarized below; for a complete discussion on the chemistry and mechanisms behind the observed behaviour of P, refer to Section 5.6.  Starting with OP, the effect of heating rate on its release is clearly shown in Figure 33.  Both runs peak at 60°C, but the OP from the faster heating rate run does not recover from its post maximum decline, as it 02040608010012020 30 40 50 60 70 80 90 100VFA (mg/L) Temperature (°C) Slow Heating RateFast Heating Rate92  does for a slower heating rate.  This confirms that the peaking behaviour of OP is strictly governed by temperature and its subsequent recovery is primarily governed by the heating rate.    Figure 33 – OP results from the heating rate comparison study using a 915MHz MW heater on WAS Interestingly, the maximum peak is higher for the faster heating rate run, although it is not certain whether this is due to the heating rate or differences in the substrates‟ initial characteristics.  To determine if the consistently observed OP trend is specific to dielectric heating, WAS was heated conventionally on a hot plate until 90°C for a sample with a H2O2 dosage and heating rate of 0.31%(v/v)/%TS and 6.3°C/min (see Figure 33).  The standalone sample demonstrated that the overall trend is inherent to the WAS, but that the OP release and recovery kinetics may be influenced by the heating method.  The speciation of STP into OP, PP, and organic fractions is plotted in Figure 34 for the 90°C samples.  Both experiments‟ raw samples have statistically equivalent TP concentrations.    0204060801001200 20 40 60 80 100OP (mg/L) Temperature (°C) Slow Heating RateFast Heating RateFast Heating Rate (Hot Plate)93   Figure 34 – STP speciation from the heating rate comparison study at 90°C using a 915MHz MW heater on WAS From Figure 34, the advantageous effect of a lower heating rate on the release of PP, and to a lesser extent, organic STP, is shown.  The observed gains are presumably due to the additional time spent at elevated temperatures that enhance the solubilization of P containing organics, such as DNA and phospholipids, and the release of PP from cell fragments.  Likewise for nitrogen, a faster heating rate results in a lower final STKN and NH3 release, as shown in Figure 35.  The TKN of the two substrates‟ raw samples were statistically equivalent.    Figure 35 – STKN and NH3 results from the heating rate comparison study using a 915MHz MW heater on WAS 050100150200250300350400450Fast Heating Rate Slow Heating RateSTP (mgP/L) OrganicPPOP051015202530010020030040050060070080010 30 50 70 90NH₃ (mgN/L) STKN (mgN/L) Temperature (°C) Slow Heating Rate (STKN)Fast Heating Rate (STKN)Slow Heating Rate (NH₃) Fast Heating Rate (NH₃) 94  Both nitrogen parameters have similar release rates that increase sharply after 60°C, ostensibly due to the addition of H2O2 and its well-known ability to produce NH3 directly from proteins (Walter et al., 1955).  Noting that SCOD was equivalent between the two runs, it is somewhat surprising that STKN differed to the degree observed in Figure 35.  A possible explanation for this discrepancy is found from low temperature (60-90°C) thermal hydrolysis studies on WAS, which suggest that longer heating times and higher temperatures substantially favour the solubilization of proteins over carbohydrates (Xue et al., 2015).  In relation to the results presented here, this means that the slower heating rate solubilized more proteins than carbohydrates, giving the former greater exposure to H2O2.  Thus, the overall SCOD would not necessarily change significantly, but the distribution of macromolecules available for reaction with H2O2 would favour proteins.  Fitting with the other nutrient data, the final solubilizations of Ca, K, and Mg for the lower heating rate run were 40±10, 19±4, and 46±7% higher than for a faster heating rate, respectively.  It is worth repeating that the high solids content of the WAS used in these experiments may not have been completely digested by the ICP method (see Section 5.3.1.2 for details).  Consequently, these results should be treated with caution.  Despite the vast differences between each run‟s nutrient releases, the molar ratios for struvite recovery indicate that OP is the limiting species in both cases, although the slower heating rate has over twice the struvite potential.   5.5.3 Physical Properties TS results (Table 26) correspond with their respective TCOD values.  Both parameters were nearly constant for the slower heating rate run, but decreased over the course of the faster heating rate experiment.  This indicates a greater degree of mineralization for the faster heating rate run, despite both runs having equivalent final SCOD fractions.  The affinity for H2O2 to mineralize organics, versus solubilizing them, is clearly a function of the substrate‟s inherent and dynamic characteristics.  Such changes in the properties of WAS are difficult to quantify, as both run‟s initial TVS (~80.7%) and VSS (~82%) were nearly identical.  For large-scale applications, any MW/RF-H2O2 system must be flexible in its operation to accommodate shifting substrate properties.  As seen with previous experiments, TSS also decreased proportionally with SCOD (Table 26).   95  Table 26 – TS and TSS results from the heating rate comparison study using a 915MHz MW heater on WAS  Temperature (°C) TS (%) TSS (g/L) Fast Heating Rate Slow Heating Rate Fast Heating Rate Slow Heating Rate 20 1.454±0.002 1.579±0.003 15.2±0.4 15.9±0.5 60 1.305±0.004 1.535±0.004 11.4±0.3 13.6±0.4 75 - 1.438±0.004 - 10.6±0.2 90 1.19±0.01 1.508±0.001 7.7±0.4 8.9±0.2  Treated samples have statistically equivalent CST values of ~925s.  However, the slower heating rate run‟s raw sample had a higher CST (590±70s) than that of the fast heating rate run (400±20s), indicating a slight advantage with the former.  Knowing that CST and PSD are highly related, it accords well that their initial and final D values are nearly the same (see Appendix E and F).   5.5.4 Conclusions Pertaining to Commercial Applications Rapid heating rates, typically generated by large-scale dielectric heaters, will significantly inhibit the release of VFA, PP, OP, and NH3.  Consequently, the potential for struvite recovery would also be reduced.  SCOD, though, will not be impacted, unless the increased heating rate no longer provides sufficient time for the H2O2 to fully react.  Thus, the reactivity between H2O2 and the substrate needs to be carefully characterized in order to maximize the process‟s efficiency.   If maintaining high levels of VFA, PP, OP or NH3 is desirable, the MW/RF-H2O2 process can be designed such that the substrate is exposed to elevated temperatures for longer periods of time.  The most obvious method to accomplish this would be to use dielectric heaters that generate low energy densities, and therefore, slower heating rates.  This would require, either very large applicators to make up for the reduced energy densities, or a higher number of low energy density dielectric heaters if the required applicator size is not practical.  However, capital and maintenance costs would increase, as well as the system‟s overall footprint.  A second option is to have a fraction of the substrate recirculate through the dielectric heater to increase the substrate‟s overall residence time in the applicator.  Similarly, though, this would increase the size and capacity of the heaters and pumps, thereby substantially raising the system‟s costs.  Instead, the recommended solution is to use an insulated tank following the MW/RF-H2O2 process, to hold the substrate near its final treatment temperature.  The size of the tank, and hence its retention time, can be chosen to achieve the desirable treatment objectives.  If heat losses from the tank are significant, a small fraction of the substrate can be recirculated through the heater to maintain the necessary holding temperature.  Since the substrate would not be exposed to EMR in the tank, a potential disadvantage is that the release of nutrients and reactivity of H2O2 is reduced relative to a slow heating 96  dielectric oven.  Future studies should investigate this proposed treatment regime, to confirm its feasibility.   5.6 Behaviour of Phosphorus in the MW/RF-H2O2 Process  The behaviour of phosphorus and its various species observed throughout a treatment run is clearly caused by several overlapping and likely competing processes. This is in addition to a substrate that is non-constant and heterogeneous in its characteristics and likely interacting with phosphorus.  All these factors combined make understanding the exact chemistry and underlying mechanisms of the phosphorus extremely difficult to elucidate, with any reasonable degree of confidence.  The explanations provided below are intended as a review of the potential processes that may be occurring and attempt at ranking them according to their probability.  Future experiments that could shed more light on MW/RF-H2O2 phosphorus chemistry will also be recommended where appropriate. 5.6.1 Distribution of Phosphorus in Waste Activated Sludge  For many biological wastes, including WAS, phosphorus can be classified as one of the following (Janssen et al., 2002): 1) Orthophosphate (OP) 2) Polyphosphates or condensed phosphorus (PP) 3) Organic phosphorus (soluble or non-soluble) 5.6.1.1 Orthophosphates in Waste Activated Sludge OP is a chemistry term that refers to the PO43- group and can be further classified as either physically or chemically bound phosphate (XPO4), or as reactive phosphate that is found in free solution (HnPO4n-3, n=0,1,2 or 3).  In neutral waters, reactive phosphate is primarily in the HPO42- and H2PO4- form.  The chemically bound species are most often metal complexes that form precipitates, while the physically bound form is when a PO43- group becomes adsorbed to the surface of precipitates, such as metal hydroxides, or organic solids.  Due to these properties, OP can be used to soften water by forming insoluble precipitates with Ca, Mg, and Fe (Griffith, 1995).  OP can be measured using the method described in Section 3.3.5 without the use of an acid digestion.  This analytical method is widely assumed to only measure reactive phosphates; however, it is not known to what degree chemically or physically bound phosphates are also measured (Neethling et al., 2009).  To add to the confusion, many of the chemically and physically bound phosphate particulates will pass through a 0.45µm filter and thus be included in STP measurements. 97  5.6.1.2 Polyphosphates in Waste Activated Sludge PPs are chains of OPs linked together with P-O-P bonds.  The formation of PP from OP is referred to as condensation and can create linear (Figure 36A), cyclic or metaphosphates (Figure 36B), and branched (Figure 36C) structures.  Branched and cyclic PPs greater than four monomers are not as stable in solution and thus, linear compounds are more often found in biological systems (Corbridge, 1990).    Figure 36 – Classifications of PPs, where: A) linear (triphosphoric acid), B) cyclic (trimetaphosphate), and C) branched PPs are stored in PAO‟s, such as those found in EBPR processes, in the form of granules that are composed of 100 to 200 OP monomers (Kuroda et al., 2002).  Studies have shown that these granules are located inside the cells, but their exact relationship to intracellular structures have not been conclusively delineated (Seviour et al., 2003).  Research suggests that the PP may be associated with either the cytoplasm, intracellular membranes, or complexed with DNA, RNA, or proteins.  In terms of their chemistry, PP`s mid-chain OH groups are vastly more acidic than their terminal ones (Corbridge, 1990). This allows them to form complexes with all metal ions, although alkali metal complexes are not stable (Rashchi and Finch, 2000).  In cells, it appears that the most important metals for complexing PPs are Mg and K, but not Ca (Seviour et al., 2003).  Measuring PP requires a mild acid digestion, such as the one described in Section 3.3.6, to hydrolyze the condensed phosphates into individual reactive OPs.  The solution is then analyzed using any standard OP method and the returned value has the sample‟s initial OP concentration subtracted from it to determine the PP fraction.  It is understood that a small fraction of PP will be hydrolyzed by the OP analysis (i.e. without any digestion step), but this is assumed to be minimal.  Additionally, it is possible that chemically and physically bound phosphates can be converted into OP by the PP acid digestion, especially if they are present in the soluble phase (Neethling et al., 2009).  Realizing these inaccuracies, PP measured using this analytical procedure is more properly called the acid-hydrolyzable phosphate fraction.   5.6.1.3 Organic Phosphorus in Waste Activated Sludge Organic P is bound in WAS mainly in the form of DNA (nucleotides), membranes (phospholipids), ATP, sugar phosphates, and phosphoamides.  These organic structures are typically the most difficult to liberate A) B) C) Note: diagrams generated using Chemdoodle© (http://web.chemdoodle.com/demos/sketcher) 98  P from and require strong acid solutions digested at high temperatures.  This fraction is determined by subtracting the OP and PP fractions from a TP measurement (see Section 3.3.2 or 3.3.8).  Thus, a TP analysis performed on a properly digested sample will have all its P converted into OP, including physically and chemically bound phosphate, as well as PP.  Some organic P will be found in the soluble phase, but the majority resides in the total non-soluble fraction.  Lastly, it is expected that a certain amount of organic P is converted into OP using the PP method.  It bears repeating that the analytical procedures described in this work for P speciation do not necessarily match theory.   5.6.2 Behaviour of Orthophosphates The behaviour of OP in WAS throughout a MW/RF-H2O2 treatment run can be described in three distinct phases, according to previous studies: 1) Initial and rapid increase from 45 to 60°C 2) Rapid decrease from 60 to 75°C 3) A gradual recovery after ~75°C These three phases and their potential governing mechanisms are discussed in the proceeding sections. 5.6.2.1 Initial Rise At the beginning of a MW/RF-H2O2 run, OP rapidly increased, starting at 45°C and peaked abruptly at 60°C, regardless of the H2O2 dosage (see Section 5.3.1.2) or heating rate (see Section 5.5.2).  Three processes have been identified that could contribute to its initial rise and temperature dependent peak: 1) Metal phosphate complexes released from EPSs 2) Dephosphorylation of organically bound phosphates (P-O-C bonds)  3) Release of phosphates physically bound to organic solids or metal precipitates As described above, phosphates can form metal complexes with Al, Fe, K, Mg, Na, and Ca, as well as create various mixed metal phosphate complexes, such as struvite.  These complexes are of the form MH2PO4, M2HPO4, and M3PO4 for alkali metals (ions shed one valence electron), and M(H2PO4)2, MHPO4, and M3(PO4)2 for alkaline earth metals (ions readily shed two valence electrons), where M is a representative metal.  Although such complexes have solubility limits that are typically above the concentrations of their respective metals in WAS (Corbridge, 1990), the metal complexes can be adsorbed to organic solids or trapped in EPSs.  Upon heating to 60°C, EPSs start to breakdown and could release the bound complexes into free solution.  However, as discussed in Section 5.6.1.1, it is not known if such metal phosphate complexes would be measured by the OP method used throughout this work. 99  A more likely explanation for the rapid rise in OP is dephosphorylation.  Chemically, this refers to the replacement of a –OPO(OH)2 functional group on a carbon macromolecule with a hydroxyl (OH) species (Corbridge, 1990).  Many proteins, particularly enzymes, are regulated by binding with OP.  When OP binds with such proteins, called phosphorylation, they become active or inactive due to changes in their tertiary and secondary structure via interfering with hydrogen bonds.  Furthermore, proteins are well known to denature at temperatures of 50°C or higher, which could precisely explain why OP spikes at 60°C, but is hardly measurable at 45°C (Corbridge, 1990).  Thus, as proteins located inside or outside the cells are degraded, dephosphorylation could occur, where their bonded OP is released into free solution.  In the context of biochemical processes, other studies have shown that phosphorylation is completely depressed by 50°C (Ladd and Semenovich, 1967) and dephosphorylation is enhanced as temperatures are increased to 42°C for proteins (Rokka et al., 2000).   Other phosphate groups that are physically bound to organic solids or metal precipitates could similarly be released into solution at these temperatures, further contributing to the OP‟s rise.  These phosphates are typically held in place by van der Wall interactions with their respective solids.  It is therefore reasonable to postulate that the increased molecular vibrations from higher temperatures would help liberate physically bound P.  One study by Choi et al. (2009) attempted to fractionate P in activated sludge samples taken from the aerobic tank of an MBR.  Their results indicate that ~60% of the substrate‟s TP is physically bound to Fe and Al oxide and hydroxide precipitates, although this fraction is also expected to contain an unknown amount of P chemically bound to Fe and Al (i.e. metal phosphates).  Another ~30% was extracted from a fraction associated with „organically bound‟ P, but the exact amount of this P that is physically bound to organic matter is unknown.   It is plausible that all three mechanisms are occurring in tandem to various degrees.  However, it is deemed by this author that dephosphorylation is the dominant mechanism responsible for the measured OP release, partly due to its better understood temperature dependency.  5.6.2.2 Crash at Medium Temperatures A sharp decrease in OP after 60°C was consistently observed, again independent of the H2O2 dosage or heating rate.  The decline quickly occurs even if the substrate is held at 60°C, as done in Section 5.3.2.  Since P is conserved in a sample, the OP must be transforming into another species or becomes insoluble.  One possibility is that the OPs are condensing into PPs, which also increase after 60°C.  To evaluate this mechanism, the decrease in OP was compared to the increase in PP from all of this work‟s applicable experiments (Table 27).  The t-test results displayed in Table 27 report statistically equivalent changes at the 95% confidence level as „Equal‟ and non-equivalent changes as their absolute differences.   100  Table 27 – Comparison of the changes in OP and PP between 60 and 75°C from WAS experiments presented in this work  Experiment OP Decrease from 60 to 75°C (mg/L) PP Increase from 60 to 75°C (mg/L) T-test Results (mg/L) 27MHz Heating Only 78±1 83±4 Equal 27MHz Low H2O2 76.2±0.8 64±3 |12|±3 27MHz Medium H2O2 64±2 1±26 |60|±30 27MHz High H2O2 85±1 100±10 |20|±10 915MHz Slow Heating Rate 22±1 -30±20 |50|±20  From the above data, the changes between the two phosphorous species are not statistically equivalent except in one case.  Furthermore, and more importantly, the decrease in OP is larger than the increase in PP for three of the remaining cases, confirming that OP turning into PP cannot be a major mechanism.  In fact, the condensation of OPs typically require acidic or basic pH‟s and temperatures of several 100°Cs (Rashchi and Finch, 2000; Wazer, 1958), conditions that were not investigated in these MW/RF-H2O2 treatment runs.   A more likely reason for the OP decrease is the presence of metal cations in the WAS that have the potential to complex with OP, as previously described.  If such complexes are indeed forming, they could prevent OP from being measured by either precipitating out of solution or resisting conversion into OP, during the reactive phosphate analysis.  In particular, many alkali metals, including Na+ and K+, have dissociation constants with OP that are between 102 and 107 times greater than for the alkaline earth metals, Ca2+ and Mg2+ (Wazer, 1958), highlighting OP‟s preference to complex with the latter.  Another way of quantifying the formation of metal complexes is the stability constant, K, which is an equilibrium constant that represents the ratio of the metal complex to its dissociation products in aqueous solutions (Equation 33).                     (33) Where M is a metal, L is the phosphate ligand, and ML is the metal complex that would form.   Data available from the literature reveals K values much greater than 1 for several metal phosphate complexes, indicating that metals will bind with OPs if available (Saha et al., 1996; Smith and Martell, 1976).   A selection of stability constants are presented in Table 28 for various metal complexes.   101  Table 28 – Stability constants and thermodynamic data for select aqueous metal phosphate complexes  Complex logKa ΔH (kcal/mol)b ΔS (cal/mol/°C)b [NaHPO4]- 0.60 8 30 [KHPO4]- 0.49 6 22 [MgHPO4] 1.88 3 23 [CaHPO4] 1.70 3 23 Note: data taken from Smith and Martell (1976),see paper for discussion on accuracy a)  Values measured at 25°C and with an ionic strength of 0.2 b)  Values measured at 25°C and with an ionic strength of 0  K constants generally increase with lower ionic strength solutions and when less hydrogen‟s participate in the complexion.  The positive change in entropy (ΔS) shows that the formation of the complexes is spontaneous.  With respect to temperature, the positive enthalpy of formation values (ΔH) mean that K will increase at higher temperatures.  Additional evidence for this mechanism is found from the observed solubilization of Ca and Mg that primarily occurred by 60°C, during a treatment run (see Section 5.3.1.2).   Combining the above information, the proposed mechanism for the rapid decline of OP is delineated as follows: the release of OP via dephosphorylation and the solubilization of metal cations largely occurred by 60°C, after which, the newly freed reactive OP began to complex with the solubilized metals, thereby making it unavailable for the reactive P analysis.  In the context of treating WAS with the MW/RF-H2O2 process, the WAS‟s high ion content will suppress complexion, but the formation of the complexes will be increasingly favoured, as the temperature rises throughout a run.   5.6.2.3 Recovery After approximately 75°C, OP started to gradually increase linearly as the treatment temperature continued to rise.  The reappearance of OP is likely due to the hydrolysis of PP into single OP units.  When the hydrolysis occurs at the terminal PP groups, OP increases rapidly, as has been observed under neutral and basic conditions in WAS (Corbridge, 1990; Rashchi and Finch, 2000; Schrodter et al., 2012).  A second reaction pathway is the breakage of P-O-P bonds, or scissoring, at random places along a PP chain.  This mechanism is acid catalyzed and unlikely to be a dominate mechanism.  Interestingly, longer chain PPs that have more than 10 OP units show an increased stability with each additional phosphate group (Rashchi and Finch, 2000), suggesting that the hydrolysis rate may increase with time as the PP chains become shorter.  A third, not as well understood mechanism, is the degradation and subsequent formation of metaphosphates from PPs larger than four phosphate units (Griffith, 1995).  Such a reaction is also unlikely to be dominant due, to the observed release of reactive OP. At room temperatures and neutral pHs, P-O-P  bonds are unstable in water but their hydrolysis occurs at very slow rates, with half-life‟s in the order of years (Griffith, 1995; Kulaev et al., 2004; Rashchi and 102  Finch, 2000; Schrodter et al., 2012; Wazer, 1958).  Their propensity to hydrolyze is because the P-O-P bonds are thermodynamically unstable when compared to orthophosphoric acid, having an enthalpy of -42kJ/mol (Rashchi and Finch, 2000).  However, the hydrolysis increases by several orders of magnitude with higher temperatures and lower or higher pHs (Wazer, 1958).  The presence of cations can either inhibit or promote hydrolysis, if the reaction is taking place in acidic or basic media, respectively (Schrodter et al., 2012).  With respect to modeling the hydrolysis of PP, the reaction rate can be generally described as first order (Griffith, 1995; Omelon and Grynpas, 2008).  Due to the slow reaction rates, only a few studies have looked at PP‟s hydrolysis at neutral pHs and temperatures below 100°C, as presented below. A general summary of PP degradation from Wazer (1958), shows that a 5% conversion is achievable within a time range of minutes to one hour depending on the pH.    A more recent study by Rulliere et al. (2012), reports that ~5% of PPs were converted into OP and trimetaphosphate after 10 minutes at 100°C with a slightly acidic pH of 5.6.  Their results lend additional support to the three hydrolysis reaction pathways discussed above.  Also noteworthy is their observation that Ca ions enhanced the degradation of PPs with more than four OP units to trimetaphosphates, although the final formation of OP remained the same.  Trimetaphosphates are very stable and required Ca and temperatures of 120°C before they started to break down.  If they are being formed by the MW/RF-H2O2 process, it may inhibit OP struvite recovery efforts (see Section 5.6.4).   Another study that heated WAS conventionally, reported that PP hydrolyzed to OP starting at 70°C.  At 90°C, a 60% conversion was obtained after holding for two hours (Kuroda et al., 2002), which fits well with the above PP heat treatment study.  Kuroda et al. (2002) also measured a significant amount of trimetaphosphate that increased with treatment time, further suggesting that only a limited fraction of PP can be converted into the desired reactive OP.   Going back to this work‟s WAS treatment studies (Section 5.3.1.2 and 5.5.2) and assuming that all OP released from 75 to 90°C is produced via PP degradation, conversion rates ranged from approximately 8 to 13%.  This appears higher than expected according to the above quoted literature; however, the addition of H2O2 and the solubilized metal ions may be catalyzing the hydrolysis.  The role of H2O2 is supported by an absence of an OP recovery for the 27MHz RF only run (Section 5.3.1.2) and for an earlier 915MHz MW only experiment (Bailey, 2015).  As well, the high temperature 915MHz MW only run resulted in a 30% lower OP recovery (Section 5.2.2).  Correlating the H2O2 dosage to the estimated PP conversion yielded a PCC of 0.81 for the experiments presented in this work. 103  5.6.3 Release of Polyphosphates Generally, soluble PP is first measured at 60°C, rapidly increased until 75°C, and then plateaued by 90°C.  Since the formation of PP from soluble OPs is unlikely (Section 5.6.2.2), PP is released from insoluble organics where it is entrapped.  Although the majority of PP is stored inside the cell as granules, it has been shown that a portion is also adsorbed to the cell‟s outer membrane (Choi et al., 2009).  Thus, the initial release of PP likely comes from these extracellular deposits.  As the cells become increasingly lysed at higher temperatures, the intracellular granules are released into solution and the soluble PP continues to rise.  The study by Kuroda et al. (2002) reported the release of PP from an EBPR process‟s WAS after heating it to 70°C for one hour, but found no observable damage to cell membranes when samples were examined under a microscope; this confirmed that initial PP release are indeed from extracellular deposits.  As discussed in Section 5.6.1.2, PPs are often complexed with divalent metal cations, particularly Mg.  This is partially confirmed by analysing the combined data from WAS experiments conducted at 90°C, revealing that the release of PP is most strongly correlated with the release of Mg (PCC 0.97), followed by Ca (PCC 0.59) and K (PCC 0.45).   A secondary mechanism that may aid the solubilization of PP at higher temperatures (>75°C) is their hydrolysis into OP, as discussed in Section 5.6.2.3.  As PP chains are shortened, either by terminal phosphate group losses, mid-chain P-O-P breakages, or the formation of cyclic phosphates, they become more soluble.  Shorter, more soluble, PP chains will obviously be found in the sample‟s filtered fraction while longer less soluble PPs have a greater propensity to adsorb to and be trapped by organic matter.  Work done on human blood for example, found that PPs with less than 50 phosphate units were soluble and those of longer chain lengths insoluble (Corbridge, 1990).   5.6.4 Maximizing Phosphorus Recovery 5.6.4.1 The Orthophosphate Strategy Phosphorus recovery efforts often focus on reactive OP due to the development of several methods that are capable of extracting them (Guo et al., 2014; Morse et al., 1998).  The major chemical OP recovery processes include precipitation, using Ca(OH)2 or FeCl3, and the formation of struvite via a fluidized bed reactor.  Thus, one strategy to maximize the amount of recoverable P from the MW/RF-H2O2 process is to promote the availability of reactive OP from the dephosphorylation, de-adsorption, and PP hydrolysis mechanisms.  Such a strategy can be realized by taking advantage of OP‟s peak release that occurs at 60°C.  According to previous results, the dephosphorylation of proteins and the de-adsorption of physically bound OPs at 60°C is enhanced with faster heating rates and when no H2O2 is used.  Although this treatment regime would be very energy and cost efficient, if an EBPR process is used, the majority of 104  the P would still be in the form of PP, and therefore, not available for recovery.  Furthermore, the OP released at 60°C is not stable due to the rapid formation of metal-OP complexes (see Section 5.6.2.2).  Consequently, the recovery of OP would have to occur immediately after treatment, or metal chelaters would have to be used to prevent the formation of such complexes.  For example, ethylenediaminetetraacetic acid (EDTA) is a widely used chelator that has been shown to effectively release OP from calcium precipitates found in DM (Zhang et al., 2010).  Alternatively, the formation of OP could be maximized by enhancing the hydrolysis of PP that occurs at the end of a MW/RF-H2O2 run.  According to previous experiments and a review of the literature (Section 5.6.2.3), this method would benefit from the following treatment conditions: high temperatures, particularly 120°C or above (Rulliere et al., 2012); long holding times, unless very high temperatures are used (ex. 10mins at 120°C); high H2O2 dosages; and lower pHs, as has been investigated  by Chan et al., (2007).  Such a treatment regime would clearly require more energy and higher operating costs, but may be economically viable due to the increased recovery of P, especially for EBPR processes.   5.6.4.2 The Polyphosphate Strategy A second strategy is to extract the PP from the WAS without converting it into reactive OP.  The recovered PP can then be separately processed into OP or used directly as a fertilizer.  In this case, the goal of the MW/RF-H2O2 process would be to maximize the solubilization of PP while minimizing its subsequent hydrolysis.  To balance these two mechanisms, mild treatment conditions are recommended in order to inhibit the hydrolysis of PP: temperatures below 100°C; low H2O2 dosages; and short holding times.  PPs can be extracted by precipitating them with CaCl2, among other salts, as was demonstrated by Kuroda et al. (2002), who recovered 65% of the soluble PPs after 30 minutes at room temperature.  Regarding the direct use of PP as a fertilizer, it is often incorrectly pointed out that it is a „slow release‟ nutrient since H2PO4- and HPO42- are considered to be the primary forms of P that are biologically available to plants.  However, controlled field experiments that compare the use of PP to traditional „plant available‟ P fertilizers have found no difference in crop yields or P uptake (Dick and Tabatabai, 1987; Schield, 1974; Torres-Dorante et al., 2006).  It has been suggested that naturally occurring microorganisms that are widely prevalent in the soil are capable of hydrolyzing PP enzymatically, thereby negating any observable differences in P uptake.  Additionally, PP fertilizers have other benefits, including the ability to co-deliver micronutrients (Mg, Ca, etc.) and be produced in highly concentrated solutions.  In summary, this P recovery strategy has the potential to economically yield a large supply of useable fertilizer by avoiding PP‟s otherwise necessary conversion into OP.   105  5.6.4.3 Improving Phosphorus Recovery with the WASSTRIP Process The waste activated sludge stripping to remove internal phosphorus (WASSTRIP) process is a recent, but proven method, to biochemically convert the PAO‟s internal PP reserves into OP (Schauer, 2012).  The WASSTRIP process accomplishes this by holding WAS from an EBPR plant in an anaerobic tank with a short-retention time (U.S. Patent No. 7,604,740, 2009).  To stimulate the release of OP, the WAS is held for at least 0.5h and a source of readily degradable carbon is added, such as VFA rich primary sludge.  After the holding tank, the WAS is put through a separator, where the collected solids are sent to an anaerobic digester and the supernatant, enriched with P and Mg, is pumped directly to a struvite crystallizer.  This process has several benefits; in particular, it is an energy efficient method to increase the production of struvite by preventing its unrecoverable formation in the digester.  Although pretreating the WAS with the MW/RF-H2O2 process would provide a source of VFAs to the WASSTRIP‟s anaerobic tank, the two processes are unlikely to be compatible because the harsh pretreatment conditions would lyse much of the microfauna and solubilize their PP reserves.  Consequently, the microorganisms would have to first rebound and reuptake the PP, before they are able to convert it into OP, and as a result, the anaerobic holding tank would require a much longer retention time.  In essence, this turns the holding tank into another digester and the issues originally solved by the WASSTRIP process reappear.  However, if the appropriate treatment conditions are chosen, a MW/RF-H2O2 system could fulfill the function of the WASSTRIP process itself.  For example, pretreating the WAS at high temperatures and H2O2 dosages would thermochemically release OP and Mg and improve its dewaterability.  As done in WASSTRIP, a separator could then be effectively used to send the P and Mg rich supernatant directly to a nutrient recovery unit.  Although the MW/RF-H2O2 process would be much more expensive than that of the WASSTRIP, the many other benefits associated with a MW/RF-H2O2 pretreatment make it a more attractive option. 5.6.4.4 Summary  Selecting a P recovery strategy would depend on the substrate‟s characteristics, financial constraints, and other treatment objectives.  To clarify the options presented in this section, recommended treatment conditions that are best suited for each strategy are summarized in Table 29.  If it is desirable to maximize the release of OP by promoting the hydrolysis of PP (see Table 29), it is also recommended to use a separator on the pretreated WAS in order to send its supernatant directly to a nutrient recovery unit.   106  Table 29 – Recommended MW/RF-H2O2 treatment conditions for various P recovery strategies  Parameter Maximizing OP at its Initial Peak Maximizing OP via PP Hydrolysis Maximizing the Release of PP Temperature 60°C High (≥90°C) 90°C Heating Rate Fast Slow Slow Holding Time N/R Long (depends on temperature) 10mins pH Adjustment N/R Low N/R H2O2 Dosage N/R High ~0.3%(v/v)/%TS Note: N/R stands for Not Recommended  5.7 Energy Efficiency of Dielectrically Heating Waste Activated Sludge  The power and temperature data for the 915MHz MW and 27MHz RF heaters are discussed separately due to their different modes of operation.  For a description of the dielectric heaters‟ design and operating procedures, refer to Section 4.0. 5.7.1 915MHz Microwave Heater Temperature and power data collected during the 110°C MW-H2O2 experiment (see Section 5.2) is discussed below, to showcase the energy consumption patterns of the 915MHz MW heater operating at temperatures above 100°C.  This run can be considered representative of the other high temperature 915MHz WAS experiments.  The temperature profile of the applicator‟s effluent and the holding tank is plotted in Figure 37 for this run.   Figure 37 – Temperature profile of the 110°C MW-H2O2 run On average, the substrate‟s temperature, after passing through the applicator was 11°C higher than that of the tank.  At the beginning of the run, the temperature difference was as high as 19°C, but this gap narrowed to a few degrees by the end.  This indicates that the tank may not be completely mixed, particularly at lower temperatures (see Section 4.2.4 for full discussion).  The larger fluctuations observed 0204060801001200 1000 2000 3000 4000 5000Temperature (°C) Time (s) Applicator EffluentTank107  for the tank‟s temperature measurements further supports this theory.  Compared to the salt run in Figure 11, heating WAS appears qualitatively similar, albeit with greater temperature fluctuations, as would be expected for the more viscous and heterogeneous substrate.   The power meter measures the instantaneous total power consumed by the MW generator.  Using the MW‟s controller, the forward power (FP), or the power sent from the magnetron to the applicator, is set by the user.  The controller also reports the amount of reflected power (RP) from the applicator due to impedance mismatches between it and the substrate.  This allows the net FP to be calculated, representing the energy that is absorbed by the substrate and not reflected back.  Lastly, the actual power absorbed by the substrate can be calculated from its overall temperature rise during an experiment.  Using the data described above, the efficiencies of the MW‟s various components are reported in Table 30 for the 110°C MW-H2O2 experiment.  Table 30 – Representative efficiency values for the 915MHz MW heater treating WAS at high temperatures   Efficiency Formula Value (%) Generator FP/Total Power 58 Applicator RP/FP 81 Heat Loss Actual Power/Net FP 55 Total Actual Power /Total Power  25  Once the FP is set, typically at 4.5kW, the total power consumption stays constant, according to the generator efficiency value displayed in Table 30.  The missing energy is consumed by the generator‟s control systems and air cooling fan, as well as from the magnetron in the form of heat loss, due to the imperfect conversion of electrical energy into EMR.  The applicator‟s efficiency (Table 30) shows how much power is reflected, and ultimately lost to the MW‟s isolator.  This efficiency factor varies widely over the course of a run, but can be partially managed by tuning the MW‟s applicator (see Section 2.1.6).  Heat losses (Table 30) from the entire system are estimated using the ratio of the net FP to the theoretical power required to heat the substrate.  A fraction of this lost power would also include the power that becomes stored (Ps) in the substrate due to polarization phenomena induced by EMR.  However, using Equation 5 and data collected by Bobowski et al. (2012), Ps is estimated to be negligible (i.e. <<1%).  Combined, the above sources of inefficiencies result in the poor, overall conversion of electricity into thermal energy. In the context of full-scale MW systems, the generator‟s efficiency is largely fixed and is expected to be similar to that found for the pilot-scale heater.  In contrast, the high heat losses observed here could be drastically reduced for a commercial system by using insulation, a design that does not require a holding tank, and faster heating rates.  As for the applicator‟s efficiency, it can also be minimized in commercial 108  applications by using an appropriate tuning regime and process design.  To better characterize and understand this inefficiency, the RP throughout the 110°C MW-H2O2 and 90°C MW-H2O2 (Section 5.4) runs are plotted in Figure 38.    Figure 38 – The RP throughout select 915MHz MW experiments using WAS, where: A) addition of H2O2, B) low temperature tune, C) post-H2O2 tune, and D) high temperature tune These runs demonstrate several key features of WAS and its changing impedance.  Firstly, the addition of H2O2 increased the RP by 10 to 20% (Figure 38A).  This is primarily due to the formation of CO2 from readily oxidized organics and partly from the introduction of a H2O2 phase, which has a different temperature and set of dielectric properties.  Both reasons essentially result in sharp dielectric phase boundaries that disturb the spatial uniformity of the applicator‟s electric field and the substrate‟s impedance.  Immediately after the addition, RP fluctuated widely for the above reasons, before stabilizing shortly thereafter at a higher value; this allowed the MW to be tuned and the impedance rematched.   This stability is achieved because the initial flux of CO2 reduces to a lower rate and the H2O2 becomes heated to the substrate‟s temperature and completely mixed within it.  The second noteworthy observation is the RP‟s temperature dependency, which showed similarities between all the 915MHz MW experiments using WAS.  Using Figure 38, a three step tuning regime is prescribed to maximize efficiency: 1) Low Temperature Tune:  a single tune can largely match the impedance between 20 and 70°C, as depicted in the 90°C run (Figure 38B).  For the 110°C experiment, the MW was pre-tuned for the low temperature range, resulting in a lower initial RP that declines until the addition of H2O2. 2) Post-H2O2 Tune:  after the RP has stabilized from the injection of H2O2, the system can be returned to greatly reduce the impedance mismatch (Figure 38C)   010203040506020 40 60 80 100 120Reflected Power  (%) Applicator's Effluent Temperature (°C) 90°C MW-H₂O₂ 110°C MW-H₂O₂ A) B) C) D) 109  3) High Temperature Tune:  as the temperature in the applicator surpasses ~100°C, the RP rapidly starts to increase again, requiring a third tuning, as seen with the 110°C run (Figure 38D).   It is important to note that the characteristics of the WAS and the treatment conditions appear to impact the RP.  This is observed in Figure 38, where the RP declined at 100°C for the 90°C run, but rapidly increased for the 110°C run, at the same temperature.  Ultimately, WAS that will be treated by the MW-H2O2 process must have its dielectric and impedance properties measured, when designing full-scale systems.    5.7.2 27MHz Radiofrequency Heater Temperature and power data collected from the RF low H2O2 experiment (Section 5.3.1) is used as a representative run to showcase the energy consumption characteristics of the 27MHz RF heater.  The temperature of the WAS entering and exiting the applicator is plotted in Figure 39.  Figure 39 – Representative temperature profiles from the 27MHz RF low H2O2 experiment using WAS The temperature rise was fairly constant (R2=0.98) and similar to the salt water test run (see Section 4.3.4), indicating that 1g/L of NaCl can be used to simulate WAS in the RF range.  However, the heating rate does decrease marginally for the second half of the run, suggesting that a tuning may be useful at higher temperatures.  The total power consumed, as measured by the power meter, and the actual power absorbed by the WAS as it passes through the RF heater, are plotted simultaneously in Figure 40, for the same low H2O2 experiment. 01020304050607080901000 500 1000 1500 2000 2500 3000 3500Temperature (°C) Time (s) Applicator's EffluentApplicator's Influent110   Figure 40 – Representative power data from the 27MHz RF low H2O2 experiment using WAS Throughout the run, the set power had to be continually turned down in order to maintain a stable plate current, and therefore, a constant heating rate.   If such corrective action is not taken, the plate current and total power consumption would rise, potentially overloading the generator.  Adjustments made to the power setting are visible in Figure 40 wherever the total power function shows a vertical drop.  Examining the effect of adding H2O2 at 60°C shows that the total and actual power consumptions fluctuate briefly before stabilizing, after which, the plate current and total power increased at a greater rate.  The reason for this behaviour, as discussed above with the 915MHz MW heater, is due to changes in the substrate‟s overall dielectric properties, caused by the H2O2 producing CO2 and increasing the heterogeneity of the substrate.  Consequently, more frequent and drastic corrective action, via the power setting, is required during this phase.   The ratio of the two power functions constitutes the RF heater‟s total efficiency, where the unaccounted for energy includes the following: circuits‟ resistances, control system, air cooling fan, triode inefficiencies, and heat losses.  Over the entire run, the average efficiency is 16%, with a range of 10 to 35%.  The majority of the extreme efficiency values were observed immediately after the addition of H2O2.  It is important to reiterate that impedance mismatches in a RF system will lower the substrate‟s heating rate, but will not necessarily reduce its energy efficiency.  This is because the capacitor type applicator cannot withdraw more energy from its power supply than can be absorbed by its load.  In other words, if the substrate has a poorly matched impedance, it will yield a slower heating rate, but the total power consumed by the generator will be proportionally lower.  This means that no energy is actually lost due to mismatched impedances, as occurs in the 915MHz MW heater.  Furthermore, the increasing plate current with temperature confirmed that the dielectric properties of the substrate gradually improved at 01234567890 20 40 60 80 100Power (kW) Temperature (°C) TotalActual111  higher temperatures.  Thus, the RF system only requires one tuning to achieve a reasonable heating rate over the entire run.    5.7.3 Comparing Dielectric Heaters for the Treatment of Waste Activated Sludge By comparing the total energy efficiencies of the two dielectric heaters treating WAS, it can be concluded that the 915MHz MW is 1.6 times more efficient than that of the 27MHz RF heater.  However, it should be noted that the RF heater used in this work is several decades older than the MW system, preventing an unbiased comparison.  Furthermore, both systems could have their efficiencies drastically improved by installing better designed applicators and insulation to reduce heat losses.  According to the literature review covered in Section 2.4.3, industrial scale dielectric heaters, using either frequency, typically have similar energy efficiencies.  Therefore, assuming that both dielectric heaters can be modified to attain near equivalent efficiencies, the RF heater is the recommended choice for treating WAS for the following reasons: 1) In the RF range, the WAS‟s dielectric properties and impedance are more stable when the temperature is increased and H2O2 is added  2) WAS becomes easier to heat dielectrically as the temperature is increased, due to its favourably changing dielectric properties in the RF range  3) The RF heater only requires one tuning for temperatures up to 100°C, while the MW heater needs at least two  4) It is operationally easier to tune any impedance mismatches with the RF heater 5) Mismatched impedances do not result in wasted energy, as is the case for a MW heater   112  6.0 Dairy Manure Experiments The liquid fraction of DM samples was treated with the MW/RF-H2O2 process using the 27MHz RF and 915MHz MW dielectric heaters.  Details pertaining to the DM and its origins can be found in Section 3.1.2.  Descriptions of the 27 and 915MHz heaters can be found in Sections 4.3 and 4.2, respectively. 6.1 Preparing Dairy Manure for Treatment Following previous work, the DM was acidified prior to the MW/RF-H2O2 process to increase the solubilization of OP (Lo and Liao, 2011; Yu et al., 2010b).  It was found that lowering the pH to below 4 imparted no additional changes in terms of SCOD or nutrient releases (Lo et al., 2012); as such, all DM experiments using the RF heater (Section 6.2) and 915MHz MW heater (Section 6.3) had their pH adjusted to 4 or below with H2SO4.  The pH and other preparation procedures are summarized for all untreated DM samples in Table 31. Table 31 – Raw samples from all DM experiments and their method of preparation Label pH H2SO4 Dosage (mL/L) Acidification Time (d) Sieved 1a 6 0 0 No 1b 2.63 5.35 1 No 2a 6.46 0 0 No 2b 3.42 5.59 1 No 2c 3.35 5.59 3 No 2d 3.35 5.59 3 Yes  Accurately acidifying DM is difficult, where some samples had less acid added but yielded a lower pH than others (see Table 31).  The pH can shift with time, as observed for sample 2c after being acidified for an additional two days.  Another operational difficulty encountered was the severe foaming that occurs immediately after the addition of H2SO4.  To mitigate this, the acid was gently mixed in small volumes over the course of an hour.  Acidification was also done the day before an experiment, to allow the foam to settle overnight.  Full-scale, MW/RF-H2O2 systems would have to take this into consideration by using a large mixing tank with a sufficient residence time and foam capacity, during acidification.  If foam is allowed to enter the dielectric heater, large impedance mismatches and inefficiencies are likely to occur.  The effects of sample preparation on select DM properties are shown in Table 32. Table 32 – Select chemical and physical properties of all untreated DM samples Parameter 1a 1b 2a 2b 2c 2d TS (%) 3.51±0.02 4.09±0.01 3.77±0.02 4.15±0.08 3.42±0.01 3.35±0.01 TCOD (g/L) 38±1 40±5 44±3 40±1 34±2 34±1 SCOD (g/L) 8.7±3 6.55±0.07 10.3±1 8.3±0.2 8.3±0.2 8.1±0.2 113  OP (mg/L) 4.6±0.4 191±2 4.0±0.1 205±4 178±3 179±4 STP (mg/L) 22±4 240±20 20.2±0.4 237±8 216±6 221±6 Sol Ca (mg/L) 184±7 770±30 220±10 860±20 800±200 800±300 Sol K (g/L) 1.70±0.02 1.69±0.03 1.7±0.1 1.66±0.03 1.27±0.06 1.21±0.01 Sol Mg (mg/L) 164±5 267±5 160±8 262±5 270±10 270±4  The most significant change, as noted at the beginning of this section, was the release of OP after acidification (samples 1a to 1b and 2a to 2b in Table 32).  Similarly, Ca and Mg were greatly solubilized, while K was largely solubilized to begin with.  These trends can be explained by reviewing the inorganic chemistry of manures (Sommer et al., 2013).  The divalent nutrients, Ca and Mg, are primarily found either adsorbed to negatively charged organic particles, or as precipitates with phosphate and carbonate.  This is confirmed by the low initial concentrations of OP, Ca, and Mg in the pre-acidified samples.  When the pH is lowered, these nutrients are released from their physically and chemically bound forms into free solution.  K though, like other monovalent nutrients, has a reduced affinity to complex and adsorb to negatively charged surfaces, due to its lower electrostatic interactions.  For this reason, K is mainly found in the soluble phase.  Holding the acidified DM for up to one day (samples 1b and 2b) does not significantly change their available nutrients, showing that the release process is fairly rapid.  Acidified DM left for three full days though (sample 2c), showed a slight drop of approximately 10% for all solubilized nutrients.   A secondary effect of acidification was the subsequent increase in TS (samples 1b and 2b).  This is likely due to the formation of salts from the sulphuric acid‟s conjugate base.  This would also explain why the TCOD did not change accordingly.  Conversely, SCOD decreased upon acidification.  It was proposed in previous work that this drop is due to the coalescence of colloidal particles caused by lowering the pH  (Lo and Liao, 2011).  Most organic particles are negatively charged due to the loss of a proton from carboxylic, ammonium, and thiol functional groups located on their surfaces (Sommer et al., 2013).  Thus, a low pH would reverse the protons‟ dissociation and the particle‟s negative surface charges, allowing the particles to coagulate.   During the 915MHz MW experiments, the high solids containing DM was observed to strain the primary pump and reduce its flow rate.  To overcome pumping difficulties without lowering the solids content through dilution, the DM was passed through a screen with mesh sizes of 1.981mm (#10 Taylor Series) for the second MW run.  By removing the largest sized particles, particularly the hairs and partially digested fibres, the pumps are able to operate at their expected flow rates.  As shown by sample 2d (Table 32), passing the DM through the screen results in a 0.07±0.01% decrease in TS, while no significant change in TCOD was observed.  It is recommended to screen DM for all future MW/RF-H2O2 projects to 114  reduce the wear on pumps.  However, prescreening also decreases the DM‟s total nutrient concentrations by 10 to 20% for P, Ca, Mg, and K (Figure 41).    Figure 41 – The total concentrations of select nutrients in acidified DM before and after screening  Taking into account the minimum reduction in TS, the proportion of nutrients removed by the screen is quite high.  These nutrients are likely organically bound, as confirmed by the constant solubilized nutrient levels that are unaffected by the screen (Table 32).  Recovering the separated nutrients would be difficult and likely require an additional acid hydrolysis processing step (a note on the P values reported in Figure 41 and throughout the following sections: TP measurements from the ICP analysis are used instead of those from the colourimetric method, due to large errors observed in the latter; see Appendixes G and H for full data set). 6.2 27MHz Radiofrequency Heater  Two 27MHz RF experiments were undertaken at different H2O2 injection temperatures, 25 and 60°C, to determine its effect on the treatment process‟s efficacy.  The injection temperature is known to be important for WAS (Bailey, 2015), but it has not yet been shown if this is also true for DM.  A literature search for the presence of catalase in DM, the enzyme believed to be responsible for the reduced H2O2 reactivity at lower temperatures in WAS, did not return any results.  It is likely, however, that catalase is not found in significant quantities in liquid DM, due to its microbiology primarily consisting of anaerobic bacteria (Sommer et al., 2013).  In contrast, WAS is largely comprised of catalase producing aerobic microorganisms.  Adding H2O2 at higher temperatures may still be beneficial because the decomposition of H2O2 into the more effective *OHs would be favoured over the direct reaction between H2O2 and organic matter (see Section 2.2.1).  For the above reasoning, it was deemed worthwhile to conclusively 020040060080010001200140016001800P Ca K MgTotal Nutrients (mg/L) Before Screening (Sample 2a)After Screening (Sample 2d)115  show the effect of H2O2‟s injection temperature in the context of DM.  An equally important objective was to validate the RF-H2O2 treatment process for DM in a 27MHz RF heater, which has not been reported in the literature to date.  Table 33 summarizes the treatment conditions for the two experiments.  Sample 1b from Table 31 represents the raw untreated sample for both runs.  Select results that pertain to the goals of this study are presented below; see Appendix G for the complete data set.  Table 33 – Experimental conditions for DM runs using the 27MHz RF heater Conditions Low Temperature Addition High Temperature Addition Substrate Raw sample 1b (see Table 31) Temperature (°C) 90 Volume (L) 12 H2O2 Dosage %(v/v) 1.40 %(v/v)/%TS 0.40 H2O2 Addition (°C) 25 60 Heating Rate (°C/min) 2.62 2.57 Flow Rate (L/min) 6 Holding Time (min) 0 10 6.2.1 Digestion Indicators  SCOD increased with respect to temperature and is shown in Figure 42 for both runs.  By the end of the experiments at 90°C, the SCOD was 8±3% higher for the high temperature H2O2 addition run.    Figure 42 – SCOD from DM experiments using the 27MHz RF heater As hypothesized above, the likely explanation for this apparent increase is the enhanced reactivity of H2O2. At higher temperatures, the formation of the extremely reactive *OHs is possible, while lower temperatures only permit H2O2 to directly attack organic substances (see Section 2.2.1).  Thus, by adding H2O2 at 60°C instead of 25°C, a larger fraction is able to decompose into *OHs instead of being 02468101214160 20 40 60 80 100SCOD (g/L) Temperature (°C) Low Temperature AdditionHigh Temperature Addition116  consumed by other less effective reactions, thereby resulting in the improved SCOD fraction.  Although the SCOD enhancement is minimal, these results show that the H2O2 injection temperature is important, even for non-catalase containing substrates.  It should be noted that the highest measured SCOD release (34±2%) represents only a modest gain over the raw un-acidified DM sample (23±1%).  The TCOD was statistically equivalent over all samples at ~40g/L (see Appendix G).  This suggests that the H2O2 was fully consumed during the experiments and that limited amounts of organics were mineralized.  For the high temperature injection run, an additional sample was taken after holding the DM for 10 minutes at 90°C.  Its SCOD though,wa not significantly different than that found for the sample taken immediately at 90°C, indicating that the thermal degradation of DM occurred rapidly for a given temperature.   Similar results were obtained for the production of VFA; an 11% increase was observed for the high temperature H2O2 addition run, compared to its low temperature counterpart.  However, the high temperature run was not statistically different from the acidified raw sample (2500±30mg/L).  This suggests that the MW/RF-H2O2 process is oxidizing VFAs to CO2 at a rate faster or equal to their production. 6.2.2 Nutrient Release Adding H2O2 at a higher temperature did not have a significant effect on the solubilization of most nutrients, as shown in Table 34.  The differences between the high temperature injection run, relative to the low temperature run, are evaluated in Table 34, using the t-test at the 95% confidence level.  Non-significant changes are reported as „Equal‟ and changes that pass the statistical test are reported as their percent difference.  The significantly different OP and STKN results yielded only minor increases in favour of the higher temperature H2O2 injection run.   Table 34 – Comparison of nutrient releases from DM experiments using the 27MHz RF heater Nutrient H2O2 Addition at 25°C H2O2 Addition at 60°C (%)* OP (mg/L) 175±2 6 STP (mg/L) 268±9 Equal NH3 (mg/L) 810±20 Equal STKN (mg/L) 1110±20 12 Sol Ca (mg/L) 850±30 Equal Sol K (mg/L) 1600±40 Equal Sol Mg (mg/L) 254±7 Equal *Note: positive percent differences indicate an increase over the 25°C H2O2 addition run  Comparing the treated samples to the acidified raw sample (sample 1b in Table 32), it is clear that the MW/RF-H2O2 treatment process had a minimal effect on nutrient release.  For all the nutrients presented in Table 34, changes were less than 10% with no clear positive or negative trend.  The only exception was 117  found for STKN and NH3, which were 14% lower for the 60°C injection run, and 13% higher for the 25°C injection run, respectively.  Overall, acidifying DM was the main driving force behind the solubilization of nutrients.  The potential for struvite recovery was substantial due to the large amounts of released OP and the fact that it is the limiting species.  The molar ratios of the nutrients required to form struvite are nearly constant between the acidified raw sample and the two treated runs (90°C) at 26:5.5:1 for NH4+:Mg2+:PO43-, respectively.  Crystallizing struvite though, will be hindered by high levels of Ca, which is found in a molar ratio of two with Mg and will preferentially complex with OP (Huchzermeier and Tao, 2012; Le Corre et al., 2005).  To overcome this drawback, oxalic acid, metal chelators or more Mg will have to be added to lower this ratio to 0.25 or below, according to previous work (Srinivasan et al., 2014b).   6.2.3 Comparison to Previous Experiments Comparing the results discussed above to previous experiments reveals the importance of the DM‟s initial characteristics and how it‟s prepared.  Studies conducted a year prior to this work on the 915MHz MW at 90°C, using nearly equivalent H2O2 dosages (0.41%(v/v)/%TS), yielded comparable results in terms of increasing SCOD and releasing nutrients (Bailey, 2015).  The major differences are related to the DM‟s initial concentrations of NH3 and VFA, which were 37% and 28% higher for the previous study, respectively.  The solubilization of COD was also higher at 40±3%, compared to only 31±2% for the 27MHz RF heater with a low temperature H2O2 addition.  However, examining the initial un-acidified SCOD fractions revealed similar improvements between the two sets of experiments.  It is not possible to conclude whether or not treatment efficacy differences exist between either frequency, with the currently available data.   Earlier work, using a continuous 2.45GHz MW by Yu et al. (2010b), shows that even larger SCOD gains of 80% are possible at 90°C, despite using a lower H2O2 dosage (0.12%(v/v)/%TS).  This particular study used diluted DM with a TS of only 0.83% and a smaller sized screen for prefiltering (1mm sieve size).  Furthermore, the process used to separate the DM at the farm has since changed with the addition of a viable sand removal unit.  Cleary, how the DM is separated and prepared before an experiment can drastically influence the treatment efficacy of the MW/RF-H2O2 process. 6.2.4 Conclusions  Adding H2O2 at 60°C confers a minor advantage over 25°C, in terms of SCOD, OP and STKN.  The improvements are likely due to the enhanced reactivity of H2O2, whereby the formation of *OHs are favoured at higher temperatures.  The proof of concept for using the 27MHz RF heater to treat DM as part of the RF-H2O2 process has been established, yielding comparable results to previous work with the 118  915MHz MW heater.  However, it is clear from this study and past results that the vast majority of released nutrients are achieved through acidification.  Additionally, only modest gains in SCOD are attained at the temperatures and H2O2 dosages experimented thus far, suggesting that future work should focus on more intensive treatment conditions, to realize further SCOD improvements.   6.3 915MHz Microwave Heater The primary objective of this set of experiments was to investigate the effect of high H2O2 dosages and treatment temperatures above 100°C, using the 915MHz MW heater.  This follows from previous work using the 27MHz RF heater (see preceding section), that yielded poor SCOD gains at less severe treatment conditions.  Two experiments were run at 110 and 130°C, with the intent to compare their results to earlier work done at 90°C.  The ultimate goal was to determine if it is worth operating the MW/RF-H2O2 process above 100°C for the treatment of DM.  Based on previous results, the H2O2 was added at a lower temperature, due to the minimal effect on SCOD, and to avoid the sudden formation of CO2 in the applicator.  Gas bubbles in the applicator induce large RP fluctuations when treating WAS and it was decided that the slight reduction in SCOD was an acceptable loss to avoid such issues.  The use of a pilot-scale, MW heater to treat DM at temperatures above 100°C has not been previously reported in the literature.  A summary of the experimental conditions is shown in Table 35.  Substrate preparation is detailed in Table 31 for samples 2b and 2d, which represent the raw samples for the 110 and 130°C runs, respectively.  Select results that pertain to the goals of this study are presented below; see Appendix H for the complete data set. Table 35 – Experimental conditions for DM runs using the 915MHz MW heater Conditions Low Temperature Run High Temperature Run Substrate Raw Sample 2b (Table 31) Raw Sample 2d (Table 31) Temperature (°C) 110 130 Volume (L) 20 H2O2 Dosage %(v/v) 2.11 %(v/v)/%TS 0.56 0.63 H2O2 Addition (°C) 35 Heating Rate (°C/min) 2.07 2.19 Flow Rate (L/min) 6 6.3.1 Digestion Indicators  The SCOD and TCOD after treatment at 110 and 130°C are compared in Figure 43, along with the previous section‟s RF run at 90°C, with a low temperature H2O2 addition. 119   Figure 43 – COD from DM experiments at final treatment temperatures, where the code in brackets reflect the experiment’s respective raw sample From Figure 43, the effect of higher temperatures and H2O2 dosages is clear, where TCOD decreased substantially with increasing temperature due to the enhanced mineralization of organics.  The absolute SCOD increased from the 90 to the 110°C run, but actually decreased at 130°C.  However, the larger reductions in TCOD at 130°C result in near equivalent SCOD fractions between the two high temperature runs.  This suggests that higher temperatures and H2O2 dosages preferentially mineralize organics, opposed to solubilizing them.   This can be explained by understanding that DM is highly recalcitrant, due to the animal‟s low protein and high cellulose diet, resulting in large fractions of non-biodegradable carbohydrate fibres and lignin (Sommer et al., 2013).  Thus, the solubilized organics, particularly VFAs and proteins, are quickly mineralized to CO2, while the larger fibres and partially digested plant matter hydrolyze more slowly.  Consequently, it is likely that the remaining DM is highly refractory, as supported by the TVS results decreasing with higher treatment temperatures (see Appendix H for data).  If it is desirable to solubilize this degradation resistant fraction of DM, higher temperatures, greater oxidative conditions, and longer holding times would have to be considered alongside the additional costs such treatment conditions would incur.   Complementing the previous RF results, VFA concentrations were marginally higher after treatment with the 915MHz MW by ~9% for either run.  Although these results would suggest that harsher treatment conditions are beneficial to the production of VFA, the molar ratio of acetic acid to its longer chain counterparts decreases from ~13 before treatment, to ~3 after either run.  This shows that the VFA increases are actually due to the conversion of longer chain VFAs into acetic acid.  In fact, harsher 05101520253035404550Raw 1b 90°C (1b) Raw 2b 110°C (2b) Raw 2d 130°C (2d)COD (g/L) non-SCODSCOD120  treatment conditions may enhance the mineralization of acetic acid and reduce the total concentration of VFAs. 6.3.2 Nutrient Release Nutrient measurements for the 110 and 130°C runs are summarized in Table 36.  Comparing the two runs‟ absolute values shows that they are equivalent to each other for all nutrients except soluble K.  Taking into account that their untreated raw samples had different initial characteristics, the 130°C run actually yielded a superior release of OP.  This is shown in Table 36 where the differences between the treated and raw sample are evaluated at the 95% confidence level.  Non-significant changes are reported as „Equal‟ and those that are statistically significant are reported as the percent difference relative to the raw sample.   Table 36 – Nutrient results from DM experiments using the 915MHz MW heater Nutrient Low Temperature Run High Temperature Run Value (mg/L) Difference from Raw Sample 2b* (%) Value (mg/L) Difference from Raw Sample 2d* (%) STP 278±5 17 283±8 28 OP 229±5 11 240±4 34 STKN 1370±30 30 1410±70 30 NH3 865±9 Equal 870±60 Equal Sol Ca 570±30 -34 532±7 Equal Sol K 1290±60 -23 1110±30 -8 Sol Mg 220±40 Equal 250±10 Equal *Note: positive percent values indicate that the treated sample is larger than its raw sample  The increasing OP with higher treatment temperatures is likely from the degradation of organic P.  A study by He et al. (2004) found that phytic acid is the predominant form of hydrolysable organic P, relative to DNA and monoester derived sources.  Phytic acid is used by plants and cereals for P storage and cannot be degraded by the cattle‟s digestive system; but researchers have developed a method to hydrolysis it into OP using MW heating and various acids (March et al., 1998).  It is proposed that the higher temperatures and longer heating rates used in this set of MW-H2O2 experiments is able to likewise degrade phytic acid, resulting in the observed OP increase. The other improved nutrient release is STKN and it increased similarly for both runs.  As concluded with the RF experiments, the release of NH3, Ca, and Mg was primarily achieved after acidification of the DM.  In terms of struvite recovery, molar ratios of the pertinent nutrients were near constant between raw and treated samples at ~21:5:1 for NH4+:Mg2+:PO43-, respectively.  A slight benefit was observed at higher treatment temperatures, where the inhibiting Ca2+:Mg2+ ratio declined from ~2 to 1.3 at 130°C.   121  Due to changes in the DM‟s characteristics and farm management practices, it is difficult to compare the experimental results in Table 36 to previous studies.  For example, earlier work by (Srinivasan et al., 2014b) on liquid DM using similar treatment conditions resulted in substantially lower STKN and SCOD fractions, but a higher OP release.  It is therefore more appropriate to use the 90°C RF experiments presented in the previous section (Table 34) to assess this study‟s results.  Such an analysis shows that the only distinct advantage of treating above 100°C, in terms of solubilizing nutrients, is with OP and STKN, which were 31% and 24% higher at 110°C, respectively, compared to the RF run (25°C H2O2 addition).   6.3.3 Conclusions Treating DM at temperatures above 100°C, using a pilot-scale 915MHz MW heater, demonstrated that its SCOD can be appreciably improved, primarily through the mineralization of soluble organics.  Potential concerns with the use of high temperatures and H2O2 dosages include the less digestible nature of the treated DM and the potential mineralization of its VFAs, especially acetic acid.  Regarding the release of nutrients, benefits were observed with OP and STKN for temperatures above 100°C.   It is worth reiterating that acidification is still responsible for the majority of the solubilized nutrients.  In summary, it is recommended that a high temperature MW/RF-H2O2 system be considered for commercial applications, due to its ability to enhance the reduction of solids and recovery of struvite. 6.4 Energy Efficiency of Dielectrically Heating Dairy Manure The power and temperature data for the 915MHz MW and 27MHz RF heaters are discussed separately, due to their different modes of operation.  For a description of the dielectric heaters‟ design and operating procedures, refer to Section 4.0. 6.4.1 915MHz Microwave Heater The temperature profile for the 110 and 130°C experiments are plotted in Figure 44, yielding overall heating rates of 2.1 and 2.2°C/min, respectively. 122   Figure 44 – Temperature profiles from DM experiments using the 915MHz MW heater Each run‟s temperature profile shows a consistent heating rate that is over 25% faster than those recorded for the WAS experiments, despite both sets of runs using the same volume of substrate (20L) and set FP (4.5kW).  The reason for this improved heating rate is clearly shown in the DM runs‟ RP data (Figure 45).  Figure 45 – RP fraction throughout DM experiments using the 915MHz MW heater The 110°C DM experiment was conducted first and required one tuning when the RP suddenly spiked at ~80°C (see Figure 45).  The new tuning arrangement adequately matched the system‟s impedance for the second DM experiment at all temperatures leading up to 130°C and reduced the average RP to below 10%, for both runs.  Furthermore, the overall efficiency of the 110°C run was 43%, nearly 50% higher than that of a typical WAS run.  The DM‟s greater heating efficiency stems from its more favourable 0204060801001201400 500 1000 1500 2000 2500 3000 3500Applicator Exit Temperautre (°C) Time(s) 110°C130°C02468101214161820 40 60 80 100 120 140Reflected Power (%) Applicator Exit Temperature (°C) 110°C130°C123  dielectric properties that allow its impedance to be better matched and presumably give it a larger ε” value.  The superior dielectric properties are perhaps due to the DM‟s higher concentration of soluble ions, which would increase the effect of the ionic conductance heating mechanism (Section 2.1.3.2).  Additionally, this heating mechanism could be enhanced by the lack of EPS and flocs in the DM, resulting in a less structured substance that grants ions a greater mobility when irradiated with EMR.  Another important factor for dielectrically heating materials in the MW range is the amount of bound water that would inhibit dipolar rotations (Section 2.1.3.1).  Without an active EPS matrix, it is expected that this fraction of bound water is less than that of WAS for an equivalent TS content.   This may also be the reason why the dielectric properties appear constant with respect to temperature and the addition of H2O2.  To summarize in the context of full-scale applications: DM was efficiently heated with the 915MHz MW, required only one tuning arrangement up to 130°C, and its dielectric properties were not significantly affected by temperature or H2O2. 6.4.2 27MHz Radiofrequency Heater The RF DM experiment, with H2O2 added at 60°C, is used as a representative run for this section.  Its temperature profile is given in Figure 46, which is nearly constant and without significant fluctuations.  Figure 46 – Representative temperature profile from the DM experiment using the 27MHz RF heater with H2O2 added at 60°C The average heating rate was 2.6°C/min, or ~75% faster than that found for WAS.  As seen with the 915MHz MW, after retuning the RF heater for DM, no further tuning is required and only minor power adjustments are needed to maintain the plate current at 1.5A.  Consequently, the total and actual instantaneous power consumption remained constant over the entire run (Figure 47). 01020304050607080901000 500 1000 1500 2000Temperature (°C) Time (s) Applicator's InfluentApplicator's Effluent124   Figure 47 – Representative power data from the DM experiment using the 27MHz RF heater with H2O2 added at 60°C The overall energy efficiency for both runs is 20%, again higher than that found for WAS.  DM is more efficiently heated due to favourable dielectric properties, particularly those relating to ionic conductance, as discussed in the preceding section.  Of equal importance is that the DM‟s dielectric properties did not appear to change with temperature or the addition of H2O2 in the RF range. 6.4.3 Comparison of Dielectric Heaters for the Treatment of Dairy Manure DM clearly exhibits favourable dielectric properties for heating in either the MW or RF EMR range.  As a result, substantially higher heating rates and energy efficiencies for both dielectric heaters were realized.  The MW heater is over twice as efficient as the RF heater, although that does not necessarily apply for properly designed full-scale systems (as discussed in Section 5.7.3).  Neither of the system‟s efficiencies are affected by temperature or compromised after the addition of H2O2.  With the data currently available, there is no clear advantage for either frequency in terms of treating DM, although a perceived benefit of using the RF system resides in its simpler mode of operation.  It is recommended that both types of heaters be considered when designing commercial scale systems.    02468101220 40 60 80 100Power (kW) Applicator Influent Temperature (°C) TotalActual125  7.0 Palm Oil Mill Effluent Experiments An exploratory set of experiments were undertaken on palm oil mill effluent samples (POME), described in Section 3.1.3, to evaluate the MW/RF-H2O2 process as a viable treatment option.  Three experiments were heated using the batch 2.45GHz MW (see Section 4.1) according to the treatment conditions shown in Table 37.   Table 37 – Experimental conditions for POME runs using the 2.45GHz batch MW Conditions Run 1 Run 2 Run 3 Substrate POME (TS of 5.08±0.01%) Temperature (°C) 120 150 150 Volume (mL) 120 120 120 H2O2 Dosage %(v/v) 2.5 2.5 5 %(v/v)/%TS 0.49 0.49 0.98 Heating Rate  (°C/min) 20 20 20  To determine if the MW/RF-H2O2 process is appropriate for POME, experimental conditions were chosen to investigate the effect of temperature and H2O2.  High temperatures and H2O2 dosages were picked with the intent that any promising results would be followed by additional experiments investigating more economic treatment conditions.  In other words, if the above temperature-H2O2 combinations did not yield positive results, it is unlikely that the MW/RF-H2O2 process would be economically used to treat POME.  Select results are presented in the proceeding sections; see Appendix I for the complete data set. 7.1 Digestion Indicators The TCOD and SCOD results are shown in Figure 48; higher temperatures and H2O2 dosages affected larger reductions in TCOD, with up to 63% of the TCOD mineralized for the highest dosage tested (run 3). 126   Figure 48 – COD from POME experiments using the batch 2.45GHz MW heater SCOD increased when subjected to the mildest treatment conditions (run 1), but then decreased below that of even the raw sample for the more severe treatment runs.  This shows that the rate of mineralization is greater than that of solubilization for temperature-H2O2 combinations above those used by run 1.  Looking at the difference between runs 1 and 2, it appears that increasing the temperature preferentially mineralized SCOD.  Likewise, the difference between runs 2 and 3 revealed that increasing the H2O2 dosage preferentially mineralizes the non-dissolved COD.  The substantial mineralization rates observed with the higher H2O2 dosage is expected because forming CO2 from organic matter requires the availability of molecular oxygen, which is provided by H2O2 itself.  Furthermore, POME is an organic colloid suspension with emulsified oils and greases comprising up to 0.7% of its total mass (Wu et al., 2009).  These constituents are particularly vulnerable to attack and subsequent mineralization by H2O2 due to their large surface areas and easily oxidized aliphatic molecular structures.   The high cellulose and protein contents of POME, as outlined in Section 1.1.3, will also become hydrolyzed through H2O2‟s various reaction mechanisms (reviewed in Section 2.2.1). TSS decreased with harsher treatment conditions relative to the raw sample, and was strongly correlated with TCOD (PCC of 0.95).  VSS decreased similarly, although not to the same extent as TSS; starting at 91.6±0.2% in the raw sample, it declined significantly at the 95% confidence level to 88.6±0.8, 84±1, and 72±4% for runs 1 to 3, respectively.  This trend suggests that the remaining organics in treated POME are potentially more recalcitrant and less degradable.  Since lignin is a major carbohydrate found in POME and is resistant to oxidative degradation, its proportion is likely increasing as the treatment process advances. 01020304050607080Raw Run 1 Run 2 Run 3COD (g/L) non-SCODSCOD127  Considerable amounts of VFA were produced by the MW-H2O2 process, as plotted in Figure 49.  VFA increased the most for run 3, by a factor of nearly 4.5, relative to the untreated sample.    Figure 49 – VFA results from POME experiments using the batch 2.45GHz MW heater Examining the differences between the three experiments, it appears that increasing the temperature by 30°C was more effective for raising VFA levels than doubling the H2O2 dosage, although large errors lower the certainty of this observation.  This trend may be due to H2O2‟s strong preference for mineralizing organics, which would include VFAs.  POME is known to contain high concentrations of long chain fatty acids, especially palmitic and oleic acid (Wu et al., 2009).  These are likely being broken down into acetic acid by the MW-H2O2 process, potentially explaining the sizable VFA gains. 7.2 Nutrient Release TP fractions are plotted in Figure 50, showing the change in OP and STP with treatment.  The STP fraction was 86% or higher for all three runs and only significantly increased at the highest H2O2 dosage tested (run 3). 0123456Raw Run 1 Run 2 Run 3VFA (g/L) 128   Figure 50 – TP fractions from POME experiments using the batch 2.45GHz MW heater Considering that TP is largely solubilized in the raw sample, it is likely that the majority of P is in the form of phosphates that are physically adsorbed to negatively charged colloidal particles or chemically complexed with minerals, such as Mg2+ and Ca2+.  Although PPs were not measured, they are unlikely to be present, since plants store their P in the form of phytic acid.  Organic P appeared to be minimal, judging by the limited amount of undissolved P.  This insoluble organic P, and an unknown quantity of soluble organic P, is likely composed of phytic acid, DNA, and phospholipids from cell membranes.  OP increased with temperature, but then decreased after the H2O2 dosage was doubled.  This could be due to its complexion with metal ions that become liberated by the additional H2O2.  At its greatest release, OP accounted for 36% of STP in run 2, although an abnormaly low TP measurement could be inflating this value.  P results from the colourimetric analytical method were in general agreement with those from the ICP method (data not shown, see Appendix I). STKN behaved similarly to STP, where it increased with higher temperatures, but appeared to decrease from a higher H2O2 dosage, as shown in Figure 51. 020406080100120140160180200Raw Run 1 Run 2 Run 3TP (mgP/L) non-STP Remaining STP OP129   Figure 51 – Fraction of STKN from POME experiments using the batch 2.45GHz MW heater Treatment increased NH3 by 63% compared to the untreated raw sample (24±2mg/L), but remained unchanged over all three runs.  Together, the minimal STKN and NH3 gains suggest that the degradation of proteins is limited at all treatment conditions, despite the fact that they represent over 10% of POME‟s dry matter (Wu et al., 2009).  A potential explanation for this unexpected trend comes from previous work by Ho et al. (1984), who reported that most proteins in POME are tightly bound to the insoluble particulate fraction.  Furthermore, it has also been found that POME protiens are less degistable (Wu et al., 2009).  Thus, the MW-H2O2 process could be preferentially hydrolyzing other more available and reactive organic components in the POME, such as oils and lipids, leaving the insoluble proteins relatively untouched.     The soluble fractions for Ca, K, and Mg are summarized in Figure 52, which have total concentrations of 380±10, 3 170±70, and 500±10mg/L, respectively.   0102030405060Raw Run 1 Run 2 Run 3STKN (%) 130   Figure 52 – Soluble Ca, K, and Mg fractions from POME experiments using the batch 2.45GHz MW heater Both K and Mg are completely solubilized in the raw sample, and therefore, are not affected by the MW-H2O2 process.  In stark contrast, Ca decreased substantially with harsher treatment conditions, relative to the untreated sample.  A possible explanation of this is the complexion of Ca by organic ligands or OP, as mentioned above, thereby removing it from the soluble phase.  For struvite recovery, this is beneficial because Ca will preferentially complex with OPs over Mg and hinder the formation of struvite (Huchzermeier and Tao, 2012; Le Corre et al., 2005).  To evaluate the potential for struvite recovery from POME, the molar ratios of its constituents are presented in Table 38. Table 38 – Important struvite ratios from POME experiments using the batch 2.45GHz MW heater Experiment NH4+:Mg2+:PO43- (mol) Ca2+:Mg2+ (mol) Run 1 2:18:1 0.21 Run 2 1:12:1 0.06 Run 3 1:19:1 0.12  The struvite ratios show that, for all treatment scenarios, OP is the limiting species and has the potential to be removed without the need for chemical additions.  Furthermore, it has been previously shown that the Ca to Mg ratio must be kept below 0.25 to avoid inhibiting struvite crystallization (Srinivasan et al., 2014b), a criterion that was also met by all three runs (Table 38). 7.3 Recommendations for Treatment Strategies and Future Experiments It should be noted that details regarding the source of the POME samples were not provided.  Furthermore, a review of the literature suggests that the POME samples have characteristics that are more closely aligned with sludge derived from the crude oil separator used in the palm oil extraction process (Wu et al., 2010).  Typically, POME is a mixture of three waste streams produced by separate steps in the 020406080100120140Raw Run 1 Run 2 Run 3Soluble Metal Fraction (%) Ca K Mg131  extraction process, including the separator sludge (outlined in Section 1.1.3).  This means that the POME samples may not be representative of an average mill‟s combined effluent, which should be taken into consideration for future studies. The POME experiments presented establish that the MW/RF-H2O2 process can effect significant improvements in this waste‟s characteristics, particularly in regards to the reduction of solids and the formation of VFAs.  Since the POME leaves the mill with a temperature in the range of 80 to 90°C, only treatment temperatures above 100°C should be considered for future studies.  Although this increases the expected capital costs of a full-scale, treatment system, the preheated nature of the waste stream will reduce the energy load.  Experiments also indicate a potential case for the recovery of struvite.  Several hurdles that must first be overcome, include maximizing the release of OP and improving the POME‟s settleability.  Before additional experiments can be recommended, general treatment strategies need to be delineated to guide future investigations.  A review of the literature allows several such strategies to be outlined. One strategy would be to maximize the mineralization of solids to reduce the required capacities of downstream treatment processes.  The current results show that higher H2O2 dosages appear to be more effective at reducing TCOD, although temperature-H2O2 interactions may also be important.  If the solids content is reduced enough, a digestive treatment step may not be needed.  In this case, a physical or chemical solids removal unit could, instead, be used to polish the POME and allow its discharge.  For example, many researchers have proposed the use of flocculation and coagulation, air flotation, and adsorption technologies to force the settling of POME solids (Gobi and Vadivelu, 2013; Wu et al., 2010).  The major downside of this strategy is the high costs associated with harsher treatment conditions.  As well, if a digestion step is still required, treatment conditions that result in large solids reductions appear to deteriorate the POME‟s digestibility, as suggested by the observed decline in VSS and likely larger lignin fraction.  Similarly, a second strategy is to focus efforts on increasing the digestibility of the POME to enhance downstream aerobic or anaerobic digesters (Vijayaraghavan et al., 2007; Yoochatchaval et al., 2011), or facultative pond systems currently in use.  The results obtained so far indicate that treatment conditions near those used by run 1 will increase the fraction of SCOD and production of VFA, while avoiding the previously mentioned negative effects on digestibility.  Studies though, would have to confirm that the treated samples are indeed more digestible by employing batch digesters, BOD tests, or methane potential assays. 132  Lastly, the MW/RF-H2O2 process can be used to treat the POME by converting it directly into a desirable resource.  For example, the positive benefits of applying raw POME on crop fields has already been shown; however, the release of nitrogen and phosphorus from it occurs slowly in the soils, making it a difficult to use fertilizer (Wu et al., 2009).  The MW/RF-H2O2 process has the potential to improve these properties, thereby reducing the reliance on synthetic fertilizers used by oil palm farmers.  Recent studies have also investigated the possibility of using POME as a carbon feedstock for biological processes, particularly for polyhydroxyalkanoate (PHA) producing microorganisms (Lee et al., 2014).  PHA can be separated from the microorganisms and used to produce biodegradable plastics, but requires expensive feed stocks containing high concentrations of VFAs.  The substantial formation of VFAs observed in the above experiments makes this a promising application.  Accordingly, it is recommended to investigate high temperatures and low H2O2 dosages to maximize VFAs. Ultimately, the adoption of one strategy will be dictated by its cost-effectiveness.  From the limited data obtained so far, it is provisionally recommended to focus future research efforts on the conversion of POME into fertilizer.  This strategy allows the POME to be completely utilized within the locality of a mill, while featuring the most promising financial case due to the expected reductions in operating costs.    133  8.0 Design of a Full-scale RF-H2O2 Treatment System The purpose of this section is to recommend a system design that maximizes the benefits of the MW/RF-H2O2 process for the full-scale treatment of WAS from a municipal WWTP.  The proposed design can then be further optimized based on a specific WAS‟s characteristics.  It is the author‟s intention that the MW/RF-H2O2 sizing method, developed over the proceeding sections, can be used to systematically complete future feasibility studies to aid in the evaluation of potential applications.  Based on the culmination of previous work, WAS was picked to showcase the MW/RF-H2O2 process due to its promising results. 8.1 Design Criteria A full-scale process is proposed for the treatment of TWAS from a municipal WWTP serving 1 million people.  The design criteria for the MW/RF-H2O2 process are presented in Table 39.  The treatment of primary sludge and is not considered in this analysis (see Appendix J for a complete overview of all assumptions and design parameters, as well as the sizing methodology developed for a MW/RF-H2O2 process). Table 39 – Design criteria for the proposed full-scale RF-H2O2 system treating TWAS Parameter Value Reasoning Capita 1 000 000 - Wastewater Produced (L/capita/d) 460 Tchobanoglous et al. (2003) WAS Produced (kg/m3 wastewater) 0.08 Tchobanoglous et al. (2003) TS of TWAS (%) 8 Assumed to be dewatered mechanically Tchobanoglous et al. (2003) TSS (g/L) 75 Based off of TS  Operation (h/d) 20 Allowing down time for maintenance purposes Initial Temperature (°C) 15 Average of winter and summer temperatures 8.2 Process Overview The proposed treatment system uses 27MHz RF dielectric heaters and a H2O2 dosage of 0.35%(v/v)/%TS to treat the TWAS at 95°C, as schematically shown in Figure 53.  Equipment labels displayed in Figure 53 are referenced throughout the remainder of Section 8.0. 134    Figure 53 – Process flow diagram for the proposed full-scale RF-H2O2 system treating TWAS The WAS is thickened before the RF-H2O2 process to improve its energy efficiency (Table 39).  To summarize the system‟s major components, a heat exchanger (E-1) recovers thermal energy from the treated effluent to preheat the influent and reduce energy costs.  The H2O2 is pumped from its reservoir (V-1) and conventionally preheated (E-2), before being mixed with the TWAS (E-3).  The combined stream is then immediately heated by three RF ovens connected in parallel.  An optional holding tank (V-2) follows the RF heaters if it is desirable to achieve certain treatment objectives outlined in Section 8.3.4.  The treated effluent is intended for a thermophilic, two phase acid/gas anaerobic digester (V-3 and V-4).  Characteristics of each stream are summarized in Table 40, where stream labels refer to those displayed in Figure 53.   Table 40 – Summary of stream characteristics for the proposed full-scale RF-H2O2 system treating TWAS Stream Flow Rate (L/min) Temperature (°C) TS (%) TSS (g/L) A 383 15.0 8 75 B 383 50.0 8 75 C 15 15.0 0 0 D 15 50.0 0 0 E 133 50.0 7.7 72 F 398 95.0 7.6 37 G 398 94.5 7.6 37 H 398 56.3 7.6 37 8.3 Discussion of Proposed Design  This section explains the proposed treatment system in greater detail and describes each component individually.  Design reasoning and important assumptions made are also discussed. 135  8.3.1 Heat Exchangers Thermal energy is recovered from the RF heater‟s effluent to preheat the untreated TWAS, using a spiral plate heat exchanger (E-1).  This model is chosen due to its high efficiencies and ability to handle two slurry streams without major fouling (Perry and Green, 1999).  E-1 is estimated to have a thermal efficiency and required plate area of 85% and 25m2, respectively.  Although expensive, the heat exchanger greatly reduces the energy load of the process by 44%.  The heat exchanger also allows the treated TWAS to be sent directly to the digester without the need for a separate cooling system. A vertical cylindrical furnace (E-2) is used to preheat the H2O2 using biogas recovered from the anaerobic digester.  This furnace type is chosen, despite its low thermal efficiencies, because it is more appropriate for the low heat duty demanded by the H2O2 stream: 60.5kW (Perry and Green, 1999).  The total biogas demand for the furnace is estimated to be 71 000m3/yr.  The reason for preheating the H2O2 is to ensure that it attains the final treatment temperature in the RF heater, thereby maximizing its reactivity.  The static mixer, shown as E-3, is used to thoroughly homogenize the two streams.  This will aid the H2O2‟s reactivity and increase the uniformity of the substrate‟s dielectric properties.  Preheating both streams prior to the RF heater is also beneficial for the system‟s energy efficiency.  By lowering the necessary temperature increase in the heater‟s applicator, the substrate‟s dielectric properties will vary less, allowing its impedance to be better matched to the RF system. 8.3.2 Dielectric Heaters A 27MHz RF type heater is chosen over a 915MHz MW due to the conclusions reached from the literature review (Section 2.4.3), namely: lower costs per installed kW; fewer required units due to larger power capacities; greater penetration depth; and simpler designs.  As well, other researchers have found that the dielectric properties of TWAS and related materials are more favourable for heating in the RF range (Section 2.1).  This was confirmed by the pilot-scale studies presented in this work, which also concluded that heating WAS at RFs resulted in less fluctuations from changing temperatures and the addition of H2O2 (Section 5.7.3).  Lastly, a set of experiments that directly compares the two frequencies indicates that the RF heater yields superior treatment results, particularly in regards to the degradation of organics (Section 5.4). Three RF heaters, each with a power capacity of 900kW, are required to heat the combined stream of TWAS and 70% pure H2O2.  The heaters are set to run at 83% capacity under normal operating conditions, to accommodate fluctuations in the TWAS‟s flow rate and temperature.  They are installed in parallel so that the failure of one heater does not halt the entire process.  To size the applicator, the dielectric properties of TWAS reported in the literature are used and their changes with temperature 136  estimated.  The maximum power density and capacity available from an industrial RF heater are picked, due to the scale of the treatment process.  Sizing calculations returned an electrode separation of 8cm and a total electrode plate area of 2.5m2.  The separation distance is optimized with respect to the electrode‟s required area and the applicator‟s total volume (see Appendix J for details).  Although calculations assume a flat parallel plate electrode configuration, the reported dimensions can be used as an approximation for other electrode designs, as reviewed in Section 2.4.2.3.  The system‟s overall heating rate is expected to be 30°C/min. 8.3.3 Operating Conditions and Expected Treatment Results A final treatment temperature of 95°C was picked to maximize the solubilization and mineralization of organics, as well as the release of PP and VFAs.  Although these parameters have been shown to benefit from even higher temperatures, it was decided not to treat above 100°C for the following reasons: substantially greater equipment costs; additional operating difficulties, particularly with start-up; additional safety hazards for operators, and a lower marginal return rate in terms of treatment efficacy (see Section 5.2).  The major disadvantage of operating at 95°C is the deterioration of the substrate‟s dewaterability.  Recent work has shown that this can be largely mitigated by using higher H2O2 dosages; however, such dosages are not feasible at a treatment temperature of 95°C, as discussed below.  Overall, the impaired dewaterability is not considered to be overly inhibiting, as the WAS is set to be thickened before the RF-H2O2 process and sent to a digester immediately afterwards.    The chosen H2O2 dosage is based on previous pilot-scale RF experiments (Section 5.3.1), which indicate that above a dosage of 0.35%(v/v)/%TS, a decreasing marginal return rate is observed for TSS reductions.  It is hypothesized that this is due to an insufficient time-temperature combination that results in the incomplete consumption of H2O2.  This in turn is partly caused by the increasing refractory nature of the substrate as the readily oxidized organics are completely hydrolyzed.  Thus, it is likely that the most cost effect H2O2 dosage occurs at this point.  Using kinetic data from the literature to estimate the thermal decomposition of H2O2 into *OHs, a conversion of only 0.3% was calculated for the process‟s heating rate (see Appendix J).  Although these calculations should be used to approximate the conversion‟s order of magnitude, it highlights the effect of the heating rate on H2O2‟s reactivity.  It is recommended that future studies attempt to better characterize H2O2‟s reactivity with WAS, particularly with respect to heating rates in the range of 30°C/min.  Given the chosen H2O2 dosage and operating temperature, TS and TSS are expected to decrease by 1 and 48%, respectively (see Appendix J).  To ensure an adequate supply of H2O2, it is recommended that one week‟s worth of H2O2 is stored onsite, requiring its reservoir tank (V-1) to have a capacity of 150m3. 137  8.3.4 Conditional Modifications  A holding tank (V-2) with a retention time of 3h and volume of 85m3 is recommended if certain treatment objectives are desired.  These include the enhanced degradation of solids, improved dewaterability, or maximizing P recovery.  For the first two cases, higher H2O2 dosages would be required to realize them.  To improve the TWAS‟s dewaterability, a required dosage in the range of 0.8%(v/v)/%TS is estimated from  previous work (Section 5.2.3).  Such dosages though, are unlikely to be fully consumed in the applicator, as discussed in the preceding section.  Installing an insulated holding tank after the RF heater offers a potential solution, where any residual H2O2 in the heater‟s effluent has the opportunity to fully react.  Although the substrate is not being irradiated in the tank, holding it at 95°C would maintain the H2O2‟s reactivity.  The estimated heat losses from the tank indicate that the effluent temperature would only drop by 0.5°C (see Appendix J). Thus, the thermal conversion of H2O2 into *OHs is estimated to be 54% in the tank, a substantial improvement over that of the applicator. The holding tank is also appropriate in the last case where maximizing the recovery of P is desired.  This is a particularly attractive objective if the TWAS contains a significant quantity of PP, as observed from EBPR processes.  PP must first be converted into OP before it can be recovered by a struvite crystallizer.  From the literature, it is estimated that approximately 90% of the PP can be thermally hydrolyzed at the tank‟s chosen retention time (Appendix J).  If the treated substrate‟s settleability is adequate, the crystallizer could be placed after the holding tank and a clarifier.  Alternatively, struvite could be recovered from the digesters supernatant.  Another modification can be recommended if foaming or impedance mismatches in the applicator are found to be suppressing the heating rate.  By changing the RF heaters‟ arrangement so that they are connected in series, these issues can be partly mitigated.  If connected in series, each heater becomes responsible for only a 15°C temperature rise.  This allows them to be more effectively tuned, and therefore, any impedance mismatches minimized over their respective temperature ranges.  Holding tanks with short retention times, in the order of a few minutes, can also be added between the heaters to control foaming.  The tanks would allow the foam to settle or be collected, thereby preventing bubbles from entering the applicator.   8.3.5 Anaerobic Digester It is recommended that a thermophilic anaerobic digester, operating at ~54.3°C, is used to maximize the reduction of VSS by taking advantage of the substrate‟s residual heat.  Thermophilic microorganisms can induce a greater solids reduction using a smaller reactor volume compared to their mesophilic counterparts.  Furthermore, the digester should be operated using a two phase, acid/gas configuration.  138  With this arrangement, the substrate first passes through a small tank where it primarily undergoes hydrolysis and acidogenesis (V-3).  The substrate then enters the larger methanogenic tank where most of the biogas is generated (V-4).  This style of digester is well suited for the MW/RF-H2O2 process, which thermochemically advances the TWAS part way through the hydrolysis and acidogenesis phase.  Thus, the acid phase tank would require a very short SRT, likely in the order of 1 day or less.  Actual SRT‟s typically range from 1 to 2 days and their pH is allowed to drop to 6 or below (Sieger et al., 2004; Tchobanoglous et al., 2003).  The second methanogenic tank is maintained at neutral pHs and a longer SRT of 10 days or more.  Again, due to the substrate being „pre-digested‟ by the MW/RF-H2O2 process, it is expected that a shorter SRT can be achieved for equivalent VSS reductions, as suggested by previous studies (see Section 2.3.1).  Thermophilic acid/gas digesters have several other benefits, including: higher VSS reductions of 50 to 60% due to the optimization of each biochemical phase; enhanced foaming control; increased biogas production; and the production of Class A biosolids (Tchobanoglous et al., 2003; Turovskiy and Mathai, 2006).  A significant disadvantage is the substantial odour potential arising from the acid phase tank‟s off gas, caused by the high concentrations of VFAs (Willis and Schafer, 2006).  The offensive gas can be treated separately, such as by scrubbing it through the methanogenic tank to recapture the VFAs.  With this recommended digester, its effluent TSS is estimated to be 19g/L, compared to 42g/L if the TWAS is not pretreated with the RF-H2O2 process. 8.3.6 Process Control Considerations TWAS treated using a RF heater is prone to thermal runaway.  This occurs when the TWAS‟s temperature increases rapidly as its ability to absorb EMR improves at higher temperatures, creating an uncontrollable feedback loop.  To prevent such a situation, it is important that the substrate does not stay in the applicator for longer than its intended residence time.  This can be ensured via automatic shut-off controls that cut the heater‟s power supply if its effluent temperature rises above 100°C, or if its flow rate drops below a preset threshold.  Additionally, an automatic impedance matching system should be installed to improve the heater‟s efficiency, as it is likely that the TWAS‟s dielectric properties are daily and seasonally dependent.   To prevent a buildup of gas in the applicator that could reduce its flow rate or disturb its impedance, an automatic air release valve should be installed at the exit of each heater.  As vapour and CO2 is produced by the RF-H2O2 process, the valve will continuously cycle open to expel any accumulating gases. The applicator itself should also be installed at an upwards angle to facilitate the rapid expulsion of any bubbles formed in it.   139  Maintaining a consistent influent temperature to the digester is essential to prevent thermal shock to its microorganisms, especially for thermophilic operations (Tchobanoglous et al., 2003).  In fact, it is recommended that the digester‟s temperature does not vary by more than 0.5°C/day (Turovskiy and Mathai, 2006).  To ensure this is achievable without relying extensively on additional heat exchangers, the power output of the RF heater should be automatically adjusted in response to its effluent temperature.  However, auxiliary heaters would still be required to maintain the digester‟s temperature when the RF heaters are not in operation and to offset their heat losses.  It is recommended that these heaters are furnaces that are fuelled using the digester‟s biogas.  This is recommended partly due to the convenient availability of the biogas, but more importantly, because furnaces can be turned on and ramped up rapidly, in the event that the RF heaters unexpectedly go offline. 8.3.7 Concluding Remarks   Since WAS‟s characteristics can vary widely, the most cost effective treatment conditions need to be determined beforehand for the substrate under consideration.  The retention time of the holding tank, or whether it is needed at all, also needs to be resolved based on the chosen H2O2 dosage, economic viability of recovering PP as struvite, and availability of space.  If P recovery is desired, the placement of the struvite crystallizer and any necessary auxiliary equipment would depend on the dewaterability of the treated substrate.  Also of importance are the substrate‟s dielectric properties and their relationship with temperature. These must be determined beforehand to optimize the heater‟s design and improve its efficiency.  The proposed RF-H2O2 system presented in this work is intended to provide a framework from which future studies and projects can build upon. 140  9.0 Summary of Work This section summarizes key findings and recommendations from literature reviews and original experimental work. 9.1 Chemistry of the MW/RF-H2O2 Process  H2O2 mineralizes and solubilizes organics by either reacting directly with them, or by forming *OHs via metal catalysts, surfaces, equilibrium products, or thermal decomposition  No conclusive evidence for athermal or non-thermal MW effects  Evidence for a MW‟s „specific thermal‟ effects that confer advantages over conventional heating  Specific thermal effects include: selective heating of polarizable molecules; volumetric heating mechanisms; and higher power densities  Can improve the effectiveness of H2O2‟s reactivity by promoting the formation of *OHs  9.2 Experiments with Waste Activated Sludge 9.2.1 Effect of the MW-H2O2 Process on Physical Properties and Solid’s Reduction  VFA, SCOD, and TCOD improvements are more dependent on H2O2 than high temperatures  Higher temperatures cause larger particles to coalesce, while higher H2O2 dosages shift the PSD towards smaller particles  CST improvements require higher H2O2 dosages rather than higher temperatures 9.2.2 Effect of High Temperatures and Oxidant Dosages Using a 915MHz Microwave Heater  Substantial VFA production requires H2O2 but not O3   Confirmed the use of high oxidant dosages to improve CST without acidification  Major benefit of combining O3 with the MW-H2O2 process is the substantial reduction in TS  Heating at 110°C (no H2O2) offers minimal gains in treatment efficacy compared to 90°C  Major advantage of treating above 100°C is the ability to use higher H2O2 dosages 9.2.3 Effect of the H2O2 Dosage Using a 27MHz Radiofrequency Heater  Large H2O2 dosages require more time or higher temperatures to completely react   Approximately 50% of TCOD is readily solubilized with H2O2  MW/RF-H2O2 process is more energy efficient when treating substrates with a higher TS   VFA production requires H2O2 but is independent of its dosage 141   Release of NH3 and STKN are dependent on the H2O2 dosage, while OP and STP are not  Data supports the conversion of PP into OP and the formation of OP metal complexes   Confirmed CST threshold with respect to increasing H2O2 dosages  9.2.4 Feasibility of a Low Temperature Treatment Using a 27MHz Radiofrequency Heater  No treatment parameter improves when substrate is held at 60°C  H2O2 has significant synergistic effects with temperature at 90°C  VFA levels decline after 45min when substrate is held at 60°C   Initial OP peak at 60°C is inherently unstable and declines within 15min  Low temperature treatment processes are not recommended for full-scale systems  9.2.5 Comparing the Treatment Efficacy of a 915 and 27MHz Dielectric Heater   RF heater yields a 17% higher SCOD fraction and a 30% lower CST   It is recommended that full-scale systems consider RF heaters 9.2.6 Effect of Heating Rate for Full-scale Applications  SCOD is dictated by a treatment‟s final temperature and H2O2 dosage, as long as there is sufficient time for the H2O2 to fully react  Initial OP peak only occurs at 60°C, but its subsequent recovery is dependent on the heating rate  A slower heating rate enhances the release of VFA, PP, STKN and NH3   The PSD and CST are not affected by the heating rate 9.2.7 Behaviour of Phosphorus in the MW/RF-H2O2 Treatment Process  Initial OP release is likely due to dephosphorylation and de-adsorption processes  OP‟s steep decline after 60°C is likely due to the formation of metal complexes  Gradual OP recovery after its crash are likely due to the hydrolysis of PPs   PP is rapidly released starting at 60°C, but plateaus by 75°C  Two promising strategies for maximizing the recovery of P are: i) Enhance the release of PP for direct recovery by treating at 90°C for 10min with a H2O2 dosage of 0.3%(v/v)/%TS ii) Maximize the conversion of PP into OP by using high temperatures (>90°C), large H2O2 dosages, long holding times, and/or low pHs  142  9.2.8 Energy Efficiency of the MW/RF-H2O2 Treatment Process  915MHz heater has a superior overall efficiency (25%) than that of the 27MHz heater (16%)  Advantages of 27MHz heater: requires one tuning regime; more stable efficiency; and is operationally easier to tune  915MHz heater requires three tuning regimes: low temperature (20 to 70°C); post H2O2 injection; and high temperature (>~100°C) 9.3 Experiments with Liquid Dairy Manure 9.3.1 Effect of Dairy Manure Preparation  Acidification is primarily responsible for the release of VFAs, OP, Mg, and Ca  Prescreening with a 1.981mm sized mesh allows DM to be pumped more efficiently  Prescreening removes 0.07%TS and reduces the total amount of P, Mg, Ca, and K by 10 to 20%  9.3.2 Effect of Hydrogen Peroxide’s Injection Temperature Using a 27MHz Radiofrequency Heater  Solubilization of COD is 8±3% higher when H2O2 is added at 60°C instead of 25°C  Slight OP and STKN improvements for the higher H2O2 injection temperature  Treatment improvements may be due to the enhanced formation of *OHs from H2O2 9.3.3 Effect of High Temperatures and Hydrogen Peroxide Dosages Using a 915MHz Microwave Heater  High H2O2 dosages at temperatures above 100°C vastly improve SCOD, OP, and STKN results   A higher H2O2 dosage is primarily responsible for SCOD gains, opposed to heating above 100°C   Only clear benefit of heating to 130 over 110°C is a minor increase in OP   DM treated at high H2O2 dosages and above 100°C is likely to be less digestible 9.3.4 The Energy Efficiency of Treating Liquid Dairy Manure  DM yields faster heating rates,  higher energy efficiencies, and less impedance fluctuations compared to WAS  DM requires only one tuning regime for either dielectric heater, up to temperatures of 130°C  Improved DM energy efficiencies and more favourable dielectric properties are likely due to its higher more mobile ion concentrations   915MHz heater is more than twice as efficient compared to the 27MHz heater 143  9.4 Exploratory Experiments with Palm Oil Mill Effluent   MW-H2O2 process preferentially mineralizes up to 63% of the COD   H2O2 appears to be more important for TCOD reductions than temperature  Nutrient ratios are favourable for struvite recovery, but the substrate‟s dewaterability requires improvement    Higher temperatures improve VFA production over higher H2O2 dosages  Release of other nutrients only improved minimally after treatment  Most promising treatment strategy is to prepare the POME for its direct use as a fertilizer 9.5 Conclusions The original work presented in this thesis describes the design and construction of a pilot-scale 915MHz high temperature MW and a continuous 27MHz RF dielectric heater for the treatment of WAS, liquid DM, and POME.  Experiments on these waste substrates using the MW/RF-H2O2 process, confirmed that treatment results obtained in previous, bench-scale, studies are reproducible at larger continuous scale applications.  Thus, the MW/RF-H2O2 process is shown to be a viable method for improving the digestibility and nutrient recovery potential of several organic waste waters.  Results and an extensive literature review also suggest that using the RF heater offers advantages relating to its operation, cost, and treatment efficacy.  Conclusions from this work were used to design a full-scale RF-H2O2 system for treating TWAS at a municipal WWTP serving one million people.   9.6 Recommendations for Future Work 9.6.1 Specific Areas Requiring Additional Research  Characterize the reaction kinetics of H2O2, particularly with respect to heating rate  Characterize the dielectric properties of substrates with respect to frequency and temperature  Future WAS experiments should focus on treatment temperatures just below 100°C  Future DM experiments should focus on treatment temperatures above 100°C  Continue to investigate the feasibility of treating POME for use as a fertilizer at treatment temperatures above 100°C  Upgrade the current 915MHz MW heater such that it better simulates plug flow behaviour    Implement a foam suppression or diversion device for the 27MHz RF heater 9.6.2 Selecting an Industrial Dielectric Heater  Future 915MHz MW heaters should consider an axial traveling wave applicator  144   Benefits of a 915MHz MW heater are their more efficient generators that are capable of higher power densities and heating rates   The benefits of a 27MHz RF heater are: cheaper generators; higher total power capacities; and larger penetration depths  The overall energy efficiency of either dielectric heater are expected to be comparable for full-scale systems 9.6.3 Design of a Full-scale RF-H2O2 Treatment System for Thickened Waste Activated Sludge  A method for sizing MW/RF-H2O2 systems is developed for feasibility studies  TWAS from 1 000 000 people could be treated with three 900kW 27MHz RF heaters  Recommended treatment conditions are 95°C and 0.35%H2O2(v/v)/%TS  A heat exchanger can be used to preheat the substrate and reduce energy costs  A post heating holding tank is recommended if higher H2O2 dosages or P recovery is desirable  RF heaters are estimated to have an overall energy efficiency of 58%, and an electrode area and separation of 2.5m2 and 8cm, respectively    A two phase acid/gas thermophilic digester is suggested for the treated TWAS   The case study estimates that TSS could be reduced from 75 to 19g/L after digestion    145  Bibliography Ahn, J.H., Shin, S. G., and Hwang, S. (2009). Effect of microwave irradiation on the disintegration and acidogenesis of municipal secondary sludge. Chemical Engineering Journal, 153(1-3), 145–150. doi:10.1016/j.cej.2009.06.032 American Public Health Association, Water Environment Federation, and American Water Works Association. (1999). Standard Methods for the Examination of Water and Wastewater. (L. Clesceri, A. Greenberg, and A. Eaton, Eds.) (20th ed.). American Public Health Associaiton. Appels, L., Baeyens, J., Degrève, J., and Dewil, R. (2008). Principles and potential of the anaerobic digestion of waste-activated sludge. Progress in Energy and Combustion Science, 34(6), 755–781. doi:10.1016/j.pecs.2008.06.002 Appels, L., Houtmeyers, S., Degreve, J., Impe, J. Van, and Dewil, R. (2013). Influence of microwave pre-treatment on sludge solubilization and pilot scale semi-continuous anaerobic digestion. Bioresource Technology, 128, 598–603. Baghurst, D. R., and Mingos, M. P. (1992). Superheating effects associated with microwave dielectric heating. Journal of the Chemical Society, Chemical Communications, (674), 674–677. Bailey, S. (2015). Pilot-scale Microwave Treatment of Wastewater Slurries: Assessment of a 915MHz Microwave Generator and Custom Applicator. University of British Columbia. Ballal, A., Basu, B., and Apte, S. K. (2007). The Kdp-ATPase system and its regulation. Journal of Biosciences, 32(3), 559–568. doi:10.1007/s12038-007-0055-7 Barba, A. A., and D‟Amore, M. (2002). Relevance of Dielectric Properties in Microwave Assisted Processes. In S. Costanzo (Ed.), Microwave Materials Characterization (pp. 91–118). InTech. Barber, J. B., and Veenstra, J. N. (2015). Evaluation of biological sludge properties influencing volume reduction. Journal of the Water Pollution Control Federation, 58(2), 149–156. Baur, R. (2009). U.S. Patent No. 7,604,740. Retrieved from ww.google.com/patents/US6099288w Beaulieu, M. (2004). Manure Management in Canada. In Farm Environment Management (p. 52). Ottawa: Statistics Canada. Retrieved from http://publications.gc.ca/Collection/Statcan/21-021-M/21-021-MIE2004001.pdf Beszedes, S., Szabo, G., and Geczi, G. (2012). Application of thermal and microwave pre-treatments for dairy wastewater sludge. Annals of Faculty Engineering Hunedoara - International Journal of Engineering, 3, 231–236. Bobowski, J., Johnson, T., and Eskicioglu, C. (2012). Permittivity of waste-activated sludge by an open-ended coaxial line. Progress in Electromagnetic Research Letters, 29(January), 139–149. Böhm, R., Kuhlmann, R. H., and Strauch, D. (1984). The effect of microwave treatment on viruses in liquid manure. Agricultural Wastes, 9(2), 147–154. doi:10.1016/0141-4607(84)90069-6 146  Brandjes, P. J., Wit, J. De, Meer, H. G. Van Der, and Keulen, H. Van. (1996). Environmental Impact of Animal Manure Management. Wageningen: International Agriculture Centre. Retrieved from http://www.fao.org/ag/againfo/programmes/en/lead/toolbox/Refer/IACman.PDF Bruus, J., Nielsen, P., and Keiding, K. (1992). On the stability of activated sludge flocs with implications to dewatering. Water Research, 26(12), 1597–1604. doi:10.1016/0043-1354(92)90159-2 Buxton, G. V., Greenstock, C. L., Helman, W. P., and Ross, A. B. (1988). Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals. Journal of Physical and Chemical Reference Data, 17(2), 513. doi:10.1063/1.555805 Canadian Council of Ministers of the Environment. (2010). A Review of the Current Canadian Legislative Framework for Wastewater Biosolids. Winnipeg. Retrieved from http://www.ccme.ca/files/Resources/waste/biosolids/pn_1446_biosolids_leg_review_eng.pdf Canadian Council of Ministers of the Environment. (2012). Guidance Document for the Beneficial Use of Municipal Biosolids, Municipal Sludge and Treated Septage. Winnipeg. Retrieved from http://www.ccme.ca/files/Resources/waste/biosolids/pn_1473_biosolids_guidance_eng_1.0.pdf Canadian Water and Wastewater Association. (2001). National Survey of Wastewater Treatment Plants National Survey of Wastewater Treatment Plants. Ottawa. Retrieved from http://www.cwwa.ca/pdf_files/freepub_wastewaterreport.PDF Chan, I. C. Y. (2013). Anaerobic Digestion of Dairy Manure Using the Microwave Hydrogen Peroxide Advanced Oxidation Process. The University of British Columbia. Chan, I., Srinivasan, A., Liao, P. H., Lo, K. V, Mavinic, D. S., Atwater, J., and Thompson, J. R. (2013). The effects of microwave pretreatment of dairy manure on methane production. Natural Resources, (4), 246–256. Chan, W. I., Liao, P. H., and Lo, K. V. (2010). Effects of irradiation intensity and pH on nutrients telease and solids destruction of waste activated sludge using the microwave-enhanced advanced oxidation process. Water Environment Research, 82(11), 2229–2238. doi:10.2175/106143010X12681059116419 Chan, W. I., Wong, W. T., Liao, P. H., and Lo, K. V. (2007). Sewage sludge nutrient solubilization using a single-stage microwave treatment. Journal of Environmental Science and Health, Part A, 42(1), 59–63. doi:10.1080/10934520601015735 Chang, C. J., Tyagi, V. K., and Lo, S. L. (2011). Effects of microwave and alkali induced pretreatment on sludge solubilization and subsequent aerobic digestion. Bioresource Technology, 102(17), 7633–7640. doi:10.1016/j.biortech.2011.05.031 Chi, Y., Li, Y., Fei, X., Wang, S., and Yuan, H. (2011). Enhancement of thermophilic anaerobic digestion of thickened waste activated sludge by combined microwave and alkaline pretreatment. Journal of Environmental Sciences, 23(8), 1257–1265. doi:10.1016/S1001-0742(10)60561-X Chindris, M., and Sumper, A. (2012). Industrial Heating Processes. In Electrical Energy Efficiency: Technologies and Applications (pp. 295–334). Chichester: John Wiley and Sons, Ltd. 147  Chirat, C., and Lachenal, D. (1997). Effect of hydroxyl radicals on cellulose and pulp and their occurrence during ozone bleaching. Holzforschung, 51(2), 147–154. Choi, H. J., Choi, C. H., and Lee, S. M. (2009). Analyses of phosphorus in sewage by fraction method. Journal of Hazardous Materials, 167(1-3), 345–350. doi:10.1016/j.jhazmat.2008.12.133 Corbridge, D. (1990). Phosphorus: An outline of its chemistry, biochemistry, and technology (4th ed.). Amsterdam: Elsevier. Cousin, C. P., and Ganczarczyk, J. J. (1999). Effect of calcium ion concentration on the structure of activated sludge flocs. Environmental Technology, 20(11), 1129–1138. doi:10.1080/09593332008616910 Croiset, E., Rice, S. F., and Hanush, R. G. (1997). Hydrogen peroxide decomposition in supercritical water. American Institute of Chemical Engineers, 43(9), 2343–2352. Danesh, P., Hong, S. M., Moon, K. W., and Park, J. K. (2008). Phosphorus and heavy metal extraction from wastewater treatment plant sludges using microwaves for generation of exceptional quality biosolids. Water Environment Research, 80(9), 784–795. doi:10.2175/106143008X276714 Dick, R., and Tabatabai, M. (1987). Polyphosphates as sources of phosphorus for plants. Fertilizer Research, 118, 107–118. Doğan, I., and Sanin, F. D. (2009). Alkaline solubilization and microwave irradiation as a combined sludge disintegration and minimization method. Water Research, 43(8), 2139–2148. doi:10.1016/j.watres.2009.02.023 Erickson, J. (1995). Dielectric Heating. In Handbook of Electrical Heating for Industry (pp. 102–118). New York: Institute of Electrical and Electronics Engineers, Inc. Eskicioglu, C., Droste, R. L., and Kennedy, K. J. (2007a). Performance of anaerobic waste activated sludge digesters after microwave pretreatment. Water Environment Research, 79(11), 2265–2273. doi:10.2175/106143007X176004 Eskicioglu, C., Kennedy, K. J., and Droste, R. L. (2007b). Enhancement of batch waste activated sludge digestion by microwave pretreatment. Water Environment Research, 79(11), 2304–2317. doi:10.2175/106143007X184069 Eskicioglu, C., Prorot, A., Marin, J., Droste, R. L., and Kennedy, K. J. (2008). Synergetic pretreatment of sewage sludge by microwave irradiation in presence of H2O2 for enhanced anaerobic digestion. Water Research, 42(18), 4674–4682. doi:10.1016/j.watres.2008.08.010 Felder, C. E., Prilusky, J., Silman, I., and Sussman, J. L. (2007). A server and database for dipole moments of proteins. Nucleic Acids Research, 35(SUPPL.2), 512–521. doi:10.1093/nar/gkm307 FMC Corporation. (2003). Hydrogen Peroxide - Standard Grade. Philadelphia. 148  Gobi, K., and Vadivelu, V. M. (2013). By-products of palm oil mill effluent treatment plant - A step towards sustainability. Renewable and Sustainable Energy Reviews, 28, 788–803. doi:10.1016/j.rser.2013.08.049 Griffith, E. (1995). General Phosphate Chemistry (As Applied to Fibers). In Phosphate Fibers (pp. 51–90). New York: Springer Science+Business Media. Guo, W.Q., Luo, H.C., Yang, S.S., Wu, Q.L., and Peng, S.M. (2014). Emerging technologies for phosphorus removal and recovery: A review. Applied Mechanics and Materials, 507, 702–706. doi:10.4028/www.scientific.net/AMM.507.702 He, Z., Griffin, T. S., and Honeycutt, C. W. (2004). Phosphorus distribution in dairy manures. Journal of Environmental Quality, 33(4), 1528–1534. doi:10.2134/jeq2004.1528 Herrero, M. A., Kremsner, J. M., and Kappe, C. O. (2008). Nonthermal microwave effects revisited: on the importance of internal temperature monitoring and agitation in microwave chemistry. The Journal of Organic Chemistry, 73(1), 36–47. doi:10.1021/jo7022697 Hill, J. M., and Jennings, M. J. (1993). Formulation of model equations for heating by microwave radiation. Applied Mathematical Modelling, 17, 369–379. Hill, J. M., and Pincombe, A. H. (1992). Some similarity temperature profiles for the microwave heating of a half-space. Journal of the Australian Mathematical Society, 33(January 1991), 290–320. Hiroki, A., and Laverne, J. A. (2005). Decomposition of hydrogen peroxide at water-ceramic oxide interfaces. Journal of Physical Chemistry B, (109), 3364–3370. doi:10.1021/jp046405d Ho, C. C., Tan, Y. K., and Wang, C. W. (1984). The distribution of chemical constituents between the soluble and the particulate fractions of palm oil mill effluent and its significance on its utilisation/treatment. Agricultural Wastes, 11(1), 61–71. doi:10.1016/0141-4607(84)90055-6 Hong, S. M. (2002). Enhancement of pathogen destruction and anaerobic digestibility using microwaves. University of Wisconsin-Madison. Hong, S. M., Park, J. K., and Lee, Y. O. (2004). Mechanisms of microwave irradiation involved in the destruction of fecal coliforms from biosolids. Water Research, 38(6), 1615–1625. doi:10.1016/j.watres.2003.12.011 Hong, S. M., Park, J. K., Teeradej, N., Lee, Y. O., Cho, Y. K., and Park, C. H. (2006). Pretreatment of sludge with microwaves for pathogen destruction and improved anaerobic digestion performance. Water Environment Research, 78(1), 76–83. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/16553169 Huchzermeier, M. P., and Tao, W. (2012). Overcoming challenges to struvite recovery from anaerobically digested dairy manure. Water Environment Research, 84(1), 34–41. doi:10.2175/106143011X13183708018887 Janssen, P., Meinema, K., and van der Roest, H. (2002). Background of the Bio P Process. In Biological Phosphorus Removal: Manual for Design and Operation (pp. 15–33). London: IWA Publishing. 149  Jardin, N., and Pöpel, H. J. (1996). Behavior of waste activated sludge from enhanced biological phosphorus removal during sludge treatment. Water Environment Federation, 68(6), 965–973. Jin, Y., Hu, Z., and Wen, Z. (2009). Enhancing anaerobic digestibility and phosphorus recovery of dairy manure through microwave-based thermochemical pretreatment. Water Research, 43(14), 3493–3502. doi:10.1016/j.watres.2009.05.017 Jones, C. (1999). Applications of Hydrogen Peroxide and Derivatives. (J. Clark, Ed.). Royal Society of Chemistry. doi:10.1039/9781847550132 Kakita, Y., Kashige, N., Murata, K., Kuroiwa, A., Funatsu, M., and Watanabe, K. (1995). Inactivation of Lactobacillus bacteriophage PL-1 by microwave irradiation. Microbiology and Immunology, 39(8), 571–576. Kappe, O., Stadler, A., and Dallinger, D. (2012). Methods and Principles in Medicinal Chemistry: Microwaves in Organic and Medicinal Chemistry (2nd ed.). Hoboken: Wiley-VCH. Karr, P. R., and Keinath, T. M. (2014). Influence of particle size on sludge dewaterability. Journal of the Water Pollution Control Federation, 50(8), 1911–1930. Kazarnovskii, I. (1975). Mechanism of spontaneous decomposition of hydrogen peroxide in aqueous solutions. Proceedings of the USSR Academy of Sciences, 221(2), 353–356. Keiding, K., and Nielsen, P. H. (1997). Desorption of organic macromolecules from activated sludge: Effect of ionic composition. Water Research, 31(7), 1665–1672. doi:10.1016/S0043-1354(97)00011-0 Kenge, A. a, Liao, P. H., and Lo, K. V. (2009). Factors affecting microwave-enhanced advanced oxidation process for sewage sludge treatment. Journal of Environmental Science and Health, Part A, 44(11), 1069–1076. doi:10.1080/10934520903005020 Kremer, F., and Schonhals, A. (Eds.). (2003). Broadband Dielectric Spectroscopy. New York: Springer-Verlag Berlin Heidelberg. doi:10.1007/978-3-642-56120-7 Kuang, W., and Nelson, S. O. (1998). Low-frequency dielectric properties of biological tissues: A review with some new insights. American Society of Agricultural Engineers, 41(1), 173–184. Kuglarz, M., Karakashev, D., and Angelidaki, I. (2013). Microwave and thermal pretreatment as methods for increasing the biogas potential of secondary sludge from municipal wastewater treatment plants. Bioresource Technology, 134, 290–297. doi:10.1016/j.biortech.2013.02.001 Kulaev, I., Vagabov, V., and Kulakovskaya, T. (2004). The Biochemistry of Inorganic Polyphosphates (2nd ed.). Chichester: John Wiley and Sons, Ltd. Kumar, R. R., Park, B. J., and Cho, J. Y. (2013). Application and environmental risks of livestock manure. Journal of the Korean Society for Applied Biological Chemistry, 56(5), 497–503. doi:10.1007/s13765-013-3184-8 150  Kuroda, A., Takiguchi, N., Gotanda, T., Nomura, K., Kato, J., Ikeda, T., and Ohtake, H. (2002). A simple method to release polyphosphate from activated sludge for phosphorus reuse and recycling. Biotechnology and Bioengineering, 75(3), 2–7. doi:10.1002/bit.10205 Ladd, C., and Semenovich, A. (1967). The role of cellular reactions in adaptation of multicellular organisms to environmental temperature. In A. Troshin (Ed.), The cell and environmental temperature (p. 462). New York: Pergamon Press. Lagunas-Solar, M. C., Cullor, J. S., Zeng, N. X., Truong, T. D., Essert, T. K., Smith, W. L., and Piña, C. (2005). Disinfection of dairy and animal farm wastewater with radiofrequency power. Journal of Dairy Science, 88(11), 4120–4131. doi:10.3168/jds.S0022-0302(05)73096-4 Le Corre, K. S., Valsami-Jones, E., Hobbs, P., and Parsons, S. a. (2005). Impact of calcium on struvite crystal size, shape and purity. Journal of Crystal Growth, 283(3-4), 514–522. doi:10.1016/j.jcrysgro.2005.06.012 Lee, E., Lee, H., Kim, Y. K., Sohn, K., and Lee, K. (2011). Hydrogen peroxide interference in chemical oxygen demand during ozone based advanced oxidation of anaerobically digested livestock wastewater. International Journal of Environmental Science and Technology, 8(2), 381–388. Lee, W. S., Chua, A. S. M., Yeoh, H. K., and Ngoh, G. C. (2014). Influence of temperature on the bioconversion of palm oil mill effluent into volatile fatty acids as precursor to the production of polyhydroxyalkanoates. Journal of Chemical Technology and Biotechnology, 89(7), 1038–1043. doi:10.1002/jctb.4197 Liao, P. H., Lo, K. V., Chan, W. I., and Wong, W. T. (2007). Sludge reduction and volatile fatty acid recovery using microwave advanced oxidation process. Journal of Environmental Science and Health. Part A, 42(5), 633–639. doi:10.1080/10934520701244417 Liao, P. H., Wong, W., and Lo, K. V. (2005a). Advanced oxidation process using hydrogen peroxide/microwave system for solubilization of phosphate. Journal of Environmental Science and Health, Part A, 40(9), 1753–1761. doi:10.1081/ESE-200068038 Liao, P. H., Wong, W., and Lo, K. V. (2005b). Release of phosphorus from sewage sludge using microwave technology. Journal of Environmental Engineering and Science, 4, 77–81. doi:10.1139/S04-056 Lidstrom, P., Tierney, J., Wathey, B., and Westman, J. (2001). Microwave assisted organic synthesis - a review. Tetrahedron, 57(45), 9225–9283. Liu, X. L., Liu, H., Du, G. C., and Chen, J. (2009). Improved bioconversion of volatile fatty acids from waste activated sludge by pretreatment. Water Environment Research, 81(1), 13–20. doi:10.2175/106143008X304640 Lo, K. V, Chan, W. I., Lo, I. W., and Liao, P. H. (2010). The effects of irradiation intensity on the microwave-enhanced advanced oxidation process. Journal of Environmental Science and Health, Part A, 45(2), 257–262. doi:10.1080/10934520903430087 151  Lo, K. V, Chan, W. W. I., Yawson, S. K., and Liao, P. H. (2012). Microwave enhanced advanced oxidation process for treating dairy manure at low pH. Journal of Environmental Science and Health, Part B, 47, 362–367. doi:10.1080/03601234.2012.646175 Lo, K. V, and Liao, P. H. (2011). Microwave Enhanced Advanced Oxidation in the Treatment of Dairy Manure. In U. Chandra (Ed.), Microwave Heating (pp. 91–106). Rijeka. Lo, K. V, Liao, P. H., and Yin, G. Q. (2008). Sewage sludge treatment using microwave-enhanced advanced oxidation processes with and without ferrous sulfate addition. Journal of Chemical Technology and Biotechnology, 83, 1370–1374. doi:10.1002/jctb Lo, K. V, Srinivasan, A., Liao, P. H., and Bailey, S. (2015). Microwave oxidation treatment of sewage sludge. Journal of Environmental Science and Health, Part A, 50(8), 882–889. doi:10.1080/10934529.2015.1019811 Madaki, Y. S., and Seng, L. (2013). Palm oil mill effluent (POME) from Malaysia palm oil mills: Waste or resource. International Journal of Science, Environment and Technology, 2(6), 1138–1155. March, J. G., Grases, F., and Salvador, A. (1998). Hydrolysis of phytic acid by microwave treatment: Application to phytic acid analysis in pharmaceutical preparations. Microchemical Journal, 59(3), 413–416. doi:10.1006/mchj.1998.1604 Markx, G. H., and Davey, C. L. (1999). The dielectric properties of biological cells at radiofrequencies: Applications in biotechnology. Enzyme and Microbial Technology, 25, 161–171. Maroulis, G. (1992). A complete description of the electric dipole moment, polarizability, and hyperpolarizability tensors of hydrogen peroxide. Journal of Chemical Physics, 96(8), 6048–6052. Marra, F. (2012). Microwave and Radio-Frequency Heating Processes for Food. In J. Ahmed and M. Rahman (Eds.), Handbook of Food Process Design: Volume II (1st ed., pp. 1101–1127). Blackwell Publishing Ltd. Mehdizadeh, S. N., Eskicioglu, C., Bobowski, J., and Johnson, T. (2013). Conductive heating and microwave hydrolysis under identical heating profiles for advanced anaerobic digestion of municipal sludge. Water Research, 47(14), 5040–5051. doi:10.1016/j.watres.2013.05.055 Menendez, J., Inguanzo, M., and Pis, J. (2002). Microwave-induced pyrolysis of sewage sludge. Water Research, 36, 3261–3264. Meredith, R. (1998). Engineer’s Handbook of Industrial Microwave Heating. London: The Institution of Electrical Engineers. Metaxas, A., and Meredith, R. (1983). Industrial Microwave Heating. London: Peter Peregrinus Ltd. Mikutta, R., Kleber, M., Kaiser, K., and Jahn, R. (2005). Review: Organic matter removal from soils using hydrogen peroxide, sodium hypochlorite, and disodium peroxodisulfate. Soil Science Society of America, 65, 120–135. 152  Morse, G. K., Brett, S. W., Guy, J. a., and Lester, J. N. (1998). Review: Phosphorus removal and recovery technologies. Science of the Total Environment, 212(1), 69–81. doi:10.1016/S0048-9697(97)00332-X Moseley, J. D., Lenden, P., Thomson, A. D., and Gilday, J. P. (2007). The importance of agitation and fill volume in small scale scientific microwave reactors. Tetrahedron Letters, 48(35), 6084–6087. doi:10.1016/j.tetlet.2007.06.147 Mudhoo, A., and Sharma, S. K. (2011). Microwave Irradiation Technology In Waste Sludge And Wastewater Treatment Research. Critical Reviews in Environmental Science and Technology, 41(11), 999–1066. doi:10.1080/10643380903392767 Neethling, J., Smith, S., Eleuterio, L., and Bingham, A. (2009). Measuring Low Phosphorus Concentrations. Alexandria. Retrieved from https://www.werf.org/c/KnowledgeAreas/NutrientRemoval/compendium/08-Measuring_Low_Phosphorus.aspx Neyens, E., and Baeyens, J. (2003). A review of thermal sludge pre-treatment processes to improve dewaterability. Journal of Hazardous Materials, 98(1-3), 51–67. doi:10.1016/S0304-3894(02)00320-5 Niederwohrmeier, B., Bohm, R., and Strauch, D. (1985). Microwave Treatment as an Alternative Pasteurisation Process for the Disinfection of Sewage Sludge: Experiments with the Treatment of Liquid Manure. In Inactivation of Microorganisms in Sewage Sludge by Stabilisation Processes (pp. 135–147). New York: Elsevier Science Publishing Co. Nüchter, M., Ondruschka, B., Bonrath, W., and Gum, A. (2004). Microwave assisted synthesis – a critical technology overview. Green Chemistry, 6, 128–141. doi:10.1039/b310502d Omelon, S. J., and Grynpas, M. D. (2008). Relationships between polyphosphate chemistry, biochemistry and apatite biomineralization. Chemical Reviews, 108(11), 4694–4715. doi:10.1021/cr0782527 Orfanidis, S. J. (2014). Electromagnetic Waves and Antennas. New Brunswick: Rutgers University. doi:10.1016/B978-075064947-6/50011-3 Pajoum-Shariati, F., Sarrafzadeh, M.H., Mehrnia, M.R., Sarzana, G., Ghommidh, C., Grasmick, A., and Heran, M. (2013). Dielectric monitoring and respirometric activity of a high cell density activated sludge. Environmental Technology, 35(4), 425–31. doi:10.1080/09593330.2013.831459 Pan, S.H., Lo, K. V., Liao, P. H., and Schreier, H. (2006). Microwave pretreatment for enhancement of phosphorus release from dairy manure. Journal of Environmental Science and Health, Part A, 41(4), 451–458. doi:10.1080/03601230600619068 Peeters, B., Dewil, R., Lechat, D., and Smets, I. Y. (2011). Quantification of the exchangeable calcium in activated sludge flocs and its implication to sludge settleability. Separation and Purification Technology, 83(1), 1–8. doi:10.1016/j.seppur.2011.04.008 153  Peng, G., Ye, F., and Ye, Y. (2013). Effects of microwave irradiation on dewaterability and extracellular polymeric substances of waste activated sludge. Water Environment Research, 85(3), 278–285. doi:10.2175/106143012X13461650921130 Perry, R., and Green, D. (1999). Perry’s Chemical Engineers' Handbook (7th ed.). McGraw-Hill Companies. Retrieved from http://pubs.acs.org/doi/pdf/10.1021/ed027p533.1\nhttp://www.lavoisier.fr/livre/notice.asp?ouvrage=1037804 Pethig, R. (1984). Dielectric properties of biological materials: Biophysical and medical applications. Transactions of Electrical Insulation, 19(5), 453–474. Pino-jelcic, S. A., Hong, S. M., and Park, J. K. (2002). Enhanced anaerobic biodegradability and inactivation of fecal coliforms and salmonella spp . in wastewater sludge by using microwaves. Water Environment Research, 78(2), 209–216. doi:10.2175/106143005X90498 Piyasena, P., Ramaswamy, H. S., Awuah, G. B., and Defelice, C. (2003). Dielectric properties of starch solutions as influenced by temperature, concentration, frequency and salt. Journal of Food Process Engineering, 26(1), 93–119. doi:10.1111/j.1745-4530.2003.tb00591.x Qiao, W., Wang, W., Zhu, C., and Zhang, Z. (2010). Biogas recovery from microwave heated sludge by anaerobic digestion. Science in China Series E: Technological Sciences, 53(1), 144–149. doi:10.1007/s11431-010-0025-4 Qureshi, A., Lo, K. V., and Liao, P. H. (2008). Microwave treatment and struvite recovery potential of dairy manure. Journal of Environmental Science and Health, Part B, 43(4), 350–357. doi:10.1080/03601230801941709 Rashchi, F., and Finch, J. (2000). Polyphosphates: a review their chemistry and application with particular reference to mineral processing. Minerals Engineering, 13(10), 1019–1035. Rasmussen, H., and Per, N. (1996). Iron reduction in activated sludge measured with different extraction techniques. Water Research, 30(3), 551–558. Remya, N., and Lin, J.G. (2011). Current status of microwave application in wastewater treatment - a review. Chemical Engineering Journal, 166(3), 797–813. doi:10.1016/j.cej.2010.11.100 Robins Environmental. (2013). Canadian Biogas Study Technical Document. Retrieved from http://www.biogasassociation.ca/bioExp/images/uploads/documents/2014/biogas_study/Canadian_Biogas_Study_Technical_Document_Dec_2013.pdf Robinson, J., Kingman, S., Irvine, D., Licence, P., Smith, A., Dimitrakis, G., … Kappe, C. O. (2010). Understanding microwave heating effects in single mode type cavities - theory and experiment. Physical Chemistry Chemical Physics, 12(18), 4750–4758. doi:10.1039/b922797k Rodríguez-Chueca, J., Mediano, A., Ormad, M. P., Mosteo, R., and Ovelleiro, J. L. (2014). Disinfection of wastewater effluents with the Fenton-like process induced by electromagnetic fields. Water Research, 60, 250–258. doi:10.1016/j.watres.2014.04.040 154  Rokka, A., Aro, E. M., Herrmann, R., Andersson, B., and Vener, A. (2000). Dephosphorylation of photosystem II reaction center proteins in plant photosynthetic membranes as an immediate response to abrupt elevation of temperature. Plant Physiology2, 123(4), 1525–1535. Rulliere, C., Perenes, L., Senocq, D., Dodi, A., and Marchesseau, S. (2012). Heat treatment effect on polyphosphate chain length in aqueous and calcium solutions. Food Chemistry, 134(2), 712–716. doi:10.1016/j.foodchem.2012.02.164 Saha, A., Saha, N., Zhao, J., Gregán, F., Sajadi, S. A. A., Song, B., and Sigel, H. (1996). Stability of metal ion complexes formed with methyl phosphate and hydrogen phosphate. Journal of Biological and Inorganic Chemistry, 1, 231–238. Salem, I. A., El-maazawi, M., and Zaki, A. B. (2000). Kinetics and mechanisms of decomposition reaction of hydrogen peroxide in presence of metal complexes. International Journal of Chemical Kinetics, 32, 643–666. doi:10.1002/1097-4601(2000)32:11<643::AID-KIN1>3.0.CO;2-C Sanin, F., Vatansever, A., Turtin, I., Kara, F., Durmaz, B., and Sesay, M. L. (2006). Operational conditions of activated sludge: influence on flocculation and dewaterability. Drying Technology, 24(10), 1297–1306. doi:10.1080/07373930600838009 Schauer, P. (2012). Recover and reuse. Water and Wastes Digest, (May), 30–31. Retrieved from www.wwdmag.com/sites/default/files/30_AA2_Ostara.pdf Schield, S. (1974). Comparative performance of polyphosphate fertilizers for three row crops. Kansas State University. Schiffmann, R. F. (2006). Microwave and Dielectric Drying. In Handbook of Industrial Drying (3rd ed.). CRC/Taylor and Francis. Schrodter, K., Bettermann, G., Staffel, T., Wahl, F., Klein, T., and Hofmann, T. (2012). Phosphoric Acid and Phosphates. In Ullmann’s Encyclopedia of Industrial Chemistry. Wiley-VCH. doi:10.1002/14356007.a19 Seviour, R. J., Mino, T., and Onuki, M. (2003). The microbiology of biological phosphorus removal in activated sludge systems. FEMS Microbiology Reviews, 27(1), 99–127. doi:10.1016/S0168-6445(03)00021-4 Shahriari, H., Warith, M., Hamoda, M., and Kennedy, K. J. (2012). Anaerobic digestion of organic fraction of municipal solid waste combining two pretreatment modalities, high temperature microwave and hydrogen peroxide. Waste Management, 32(1), 41–52. doi:10.1016/j.wasman.2011.08.012 Shazman, A., Mizrahi, S., Cogan, U., and Shimoni, E. (2007). Examining for possible non-thermal effects during heating in a microwave oven. Food Chemistry, 103(2), 444–453. doi:10.1016/j.foodchem.2006.08.024 Sieger, R., Brady, P., Donovan, J., and Shea, T. (2004). High Performance Anaerobic Digestion. Retrieved from wef.org/WorkArea/DownloadAsset.aspx?id=2250 155  Smith, R., and Martell, A. (1976). Critical Stability Constants Volume 4: Inorganic Complexes (Vol. 4). New York: Plenum Press. doi:10.1007/978-1-4757-5506-0 Sommer, S., Christensen, M., and Schmidt, T. (2013). Animal Manure Recycling: Treatment and Management. Somerset: John Wiley and Sons. von Sonntag, C. (2008). Advanced oxidation processes: mechanistic aspects. Water Science and Technology, 58(5), 1015–1021. doi:10.2166/wst.2008.467 von Sonntag, C. (2006). The Hydroxyl Radical. In Free-Radical-Induced DNA Damage and Its Repair (pp. 48–75). Heidelberg: Springer. Sorrentino, R., and Bianchi, G. (2010). Impedance Matching. In Microwave and RF Engineering (pp. 163–185). Hoboken: John Wiley and Sons, Ltd. Spiro, B. M., and Griffith, W. P. (1997). The mechanism of hydrogen peroxide bleaching. Textile Chemist and Colorist, 29(11), 12–13. Srinivasan, A., Bailey, S., MacSween, J., Victor, K., Koch, F., and Liao, P. H. (2014a). A continuous-flow 915 MHz microwave treatment of dairy manure. Journal of Environmental Engineering and Science, 9(3), 155–157. Srinivasan, A., Nkansah-Boadu, F., Liao, P. H., and Lo, K. V. (2014b). Effects of acidifying reagents on microwave treatment of dairy manure. Journal of Environmental Science and Health, Part B, 49(7), 532–539. doi:10.1080/03601234.2014.896681 Stadtman, E. R., and Levine, R. L. (2003). Free radical-mediated oxidation of free amino acids and amino acid residues in proteins. Amino Acids, 25(3-4), 207–218. doi:10.1007/s00726-003-0011-2 Statistics Canada. (2008). A geographical profile of livestock manure production in Canada. EnviroStats, 2(2), 12–17. Retrieved from http://www.statcan.gc.ca/pub/16-002-x/16-002-x2008004-eng.pdf Statistics Canada. (2014). Number of Farms, Dairy Cows and Heifers. Canadian Dairy Information Centre. Retrieved from http://www.dairyinfo.gc.ca/index_e.php?s1=dff-fcilands2=farm-fermeands3=nb Takagi, J., and Ishigure, K. (1985). Thermal Decomposition of Hydrogen Peroxide and Its Effect on Reactor Water Monitoring of Boiling Water Reactors. Nuclear Science and Engineering, 89, 177–186. Tang, B., Yu, L., Huang, S., Luo, J., and Zhuo, Y. (2010). Energy efficiency of pre-treating excess sewage sludge with microwave irradiation. Bioresource Technology, 101(14), 5092–5097. doi:10.1016/j.biortech.2010.01.132 Tang, J., Wang, Y., Chan, T. V. C. T., and Introduction, I. (2005). Radio-Frequency Heating in Food Processing. In Novel Food Processing Technologies (pp. 501–523). Washington: CRC Press. Taylor, S., Jones, P., Nakai, S., and Jennings, J. (2003). Electroheat and Materials Processing. In Electrical Engineer’s Reference Book (16th ed.). Elsevier. 156  Tchobanoglous, G., Burton, F., and Stensel, D. (2003). Wastewater Engineering (4th ed.). Metcalf and Eddy Inc. Tessier, A., and Forst, W. (1974). Mechanism of Hydrogen Peroxide Pyrolysis. Canadian Journal of Chemistry, 52(i), 794–797. Toreci, I., Kennedy, K. J., and Droste, R. L. (2009). Evaluation of continuous mesophilic anaerobic sludge digestion after high temperature microwave pretreatment. Water Research, 43(5), 1273–1284. doi:http://dx.doi.org/10.1016/j.watres.2008.12.022 Torres-Dorante, L. O., Claassen, N., Steingrobe, B., and Olfs, H.W. (2006). Fertilizer-use efficiency of different inorganic polyphosphate sources: effects on soil P availability and plant P acquisition during early growth of corn. Journal of Plant Nutrition and Soil Science, 169(4), 509–515. doi:10.1002/jpln.200520584 Towler, G. P., and Sinnott, R. K. (2008). Chemical Engineering Design: Principles, Practice and Economics of Plant and Process Design. Elsevier. Turovskiy, I. S., and Mathai, P. K. (2006). Wastewater Sludge Processing. Hoboken: John Wiley and Sons, Inc. Tyagi, V. K., and Lo, S. L. (2013). Microwave irradiation: A sustainable way for sludge treatment and resource recovery. Renewable and Sustainable Energy Reviews, 18(71), 288–305. doi:10.1016/j.rser.2012.10.032 Venkatesh, M. S., and Raghavan, G. S. V. (2004). An overview of microwave processing and dielectric properties of agri-food materials. Biosystems Engineering, 88(1), 1–18. doi:10.1016/j.biosystemseng.2004.01.007 Vesilind, P. (1988). Capillary Suction Time as a Fundamental Measure of Sludge Dewaterability. Journal of the Water Pollution Control Federation, 60(2), 215–220. Vijayaraghavan, K., Ahmad, D., and Ezani Bin Abdul Aziz, M. (2007). Aerobic treatment of palm oil mill effluent. Journal of Environmental Management, 82(1), 24–31. doi:10.1016/j.jenvman.2005.11.016 Walker, J., Knight, L., and Stein, L. (1994). A Plain English Guide to the EPA Part 503 Biosolids Rule. Washington. Retrieved from http://water.epa.gov/scitech/wastetech/biosolids/503pe_index.cfm Walter, S., Satterfield, C., and Wentworth, R. (1955). Hydrogen Peroxide. New York: Reinhold Publishing Corporation. Wang, J. L., and Xu, L. J. (2012). Advanced oxidation processes for wastewater treatment: formation of hydroxyl radical and application. Critical Reviews in Environmental Science and Technology, 42(3), 251–325. doi:10.1080/10643389.2010.507698 Wang, W., and Gao, P. (2002). A peptide-mediated and hydroxyl radical HO-involved oxidative degradation of cellulose by brown-rot fungi. Biodegradation, 12, 383–394. 157  Wang, Y., Wei, Y., and Liu, J. (2009). Effect of H2O2 dosing strategy on sludge pretreatment by microwave-H2O2 advanced oxidation process. Journal of Hazardous Materials, 169(1-3), 680–684. doi:10.1016/j.jhazmat.2009.04.001 Waste Water System Uses Microwaves. (1977). Chemical Engineering, 84(3), 77–78. Wazer, J. (1958). Phosphorus and its Compounds. New York: Interscience Publishers. Willis, J., and Schafer, P. (2006). Advances in Thermophilic Anaerobic Digestion. In Water Environment Federation’s Annual Technical Exhibition and Conference (pp. 5378–5392). Water Environment Foundation. Wilson, C. a., and Novak, J. T. (2009). Hydrolysis of macromolecular components of primary and secondary wastewater sludge by thermal hydrolytic pretreatment. Water Research, 43(18), 4489–4498. doi:10.1016/j.watres.2009.07.022 Wilson, T. (1987). Radio-Frequency Dielectric Heating in Industry. Palo Olta. Wojciechowska, E. (2005). Application of microwaves for sewage sludge conditioning. Water Research, 39(19), 4749–4754. doi:10.1016/j.watres.2005.09.032 Wong, W. T., Chan, W. I., Liao, P. H., Lo, K. V., and Mavinic, D. (2006a). Exploring the role of hydrogen peroxide in the microwave advanced oxidation process: Solubilization of ammonia and phosphates. Journal of Environmental Engineering and Science, 5, 459–465. doi:10.1139/S06-020 Wong, W. T., Chan, W. I., Liao, P. H., and Lo, K. V. (2006b). A hydrogen peroxide/ microwave advanced oxidation process for sewage sludge treatment. Journal of Environmental Science and Health, Part A, 41(11), 2623–33. doi:10.1080/10934520600928086 Wu, L., Chen, K., Cheng, S., Lee, B., and Shu, C. (2008). Thermal decomposition of hydrogen peroxide in the presence of sulfuric acid. Journal of Thermal Analysis and Calorimetry, 93(1), 115–120. Wu, T. Y., Mohammad, A. W., Jahim, J. M., and Anuar, N. (2009). A holistic approach to managing palm oil mill effluent (POME): Biotechnological advances in the sustainable reuse of POME. Biotechnology Advances, 27(1), 40–52. doi:10.1016/j.biotechadv.2008.08.005 Wu, T. Y., Mohammad, A. W., Jahim, J. M., and Anuar, N. (2010). Pollution control technologies for the treatment of palm oil mill effluent (POME) through end-of-pipe processes. Journal of Environmental Management, 91(7), 1467–1490. doi:10.1016/j.jenvman.2010.02.008 Xiao, Q., Yan, H., Wei, Y., Wang, Y., Zeng, F., and Zheng, X. (2012). Optimization of H2O2 dosage in microwave-H2O2 process for sludge pretreatment with uniform design method. Journal of Environmental Sciences, 24(12), 2060–2067. doi:10.1016/S1001-0742(11)60998-4 Xu, R., Zhang, Q., Tong, J., Wei, Y., and Fan, Y. (2015). Internal carbon source from sludge pretreated by microwave-H2O2 for nutrient removal in A(2)/O-membrane bioreactors. Environmental Technology, 36(7), 827–836. doi:10.1080/09593330.2014.963694 158  Xue, Y., Liu, H., Chen, S., Dichtl, N., Dai, X., and Li, N. (2015). Effects of thermal hydrolysis on organic matter solubilization and anaerobic digestion of high solid sludge. Chemical Engineering Journal, 264, 174–180. doi:10.1016/j.cej.2014.11.005 Yeneneh, A. M., Chong, S., Sen, T. K., Ang, H. M., and Kayaalp, A. (2013). Effect of ultrasonic, microwave and combined microwave - ultrasonic pretreatment of municipal sludge on anaerobic digester performance. Water, Air, and Soil Pollution, 224, 1559. doi:10.1007/s11270-013-1559-4 Yoochatchaval, W., Kumakura, S., Tanikawa, D., Yamaguchi, T., Yunus, M. F. M., Chen, S. S., … Syutsubo, K. (2011). Anaerobic degradation of palm oil mill effluent (POME). Water Science and Technology, 64(10), 2001–2008. doi:10.2166/wst.2011.782 Yu, Q., Lei, H., Li, Z., Li, H., Chen, K., Zhang, X., and Liang, R. (2010a). Physical and chemical properties of waste-activated sludge after microwave treatment. Water Research, 44(9), 2841–2849. doi:10.1016/j.watres.2009.11.057 Yu, Q., Lei, H., Yu, G., Feng, X., Li, Z., and Wu, Z. (2009). Influence of microwave irradiation on sludge dewaterability. Chemical Engineering Journal, 155, 88–93. doi:10.1016/j.cej.2009.07.010 Yu, Y., Lo, I. W., Liao, P. H., and Lo, K. V. (2010b). Treatment of dairy manure using the microwave enhanced advanced oxidation process under a continuous mode operation. Journal of Environmental Science and Health, Part B, 45(8), 804–809. doi:10.1080/03601234.2010.515474 Zeronian, S. H., and Inglesby, M. K. (1995). Bleaching of cellulose by hydrogen peroxide. Cellulose, 2, 265–272. Zhang, G., Yang, J., Liu, H., and Zhang, J. (2009). Sludge ozonation: disintegration, supernatant changes and mechanisms. Bioresource Technology, 100, 1505–1509. doi:10.1016/j.biortech.2008.08.041 Zhang, H., Lo, V. K., Thompson, J. R., Koch, F. a, Liao, P. H., Lobanov, S., … Atwater, J. W. (2015). Recovery of phosphorus from dairy manure: a pilot-scale study. Environmental Technology, 36(11), 1398–1404. doi:10.1080/09593330.2014.991354 Zhang, T., Bowers, K. E., Harrison, J. H., and Chen, S. (2010). Releasing phosphorus from calcium for struvite fertilizer production from anaerobically digested dairy effluent. Water Environment Research, 82(1), 34–42. doi:10.2175/106143009X425924 Zhao, Y. (2006). Radio Frequency Dielectric Heating. In D.W. Sun (Ed.), Thermal Food Processing: New Technologies and Quality Issues (pp. 469–492). Paris: Taylor and Francis Group.   159  APPENDIX A – Remaining Data for the 2.45GHz Microwave Experiments on Waste Activated Sludge Table 41 – Complete TS, TVS, TSS, VSS, and TCOD data from experiments on WAS using the 2.45GHz batch MW Sample TS (%) TVS (%) TSS (g/L) VSS (%) TCOD (g/L) Part 1 Raw 1 0.779±0.003 79.4±0.3 8.0±0.1 84±3 8.2±0.1 110°C+0.2% 0.795±0.003 79.5±0.1 6.1±0.2 90±2 8.6±0.08 110°C+0.3% 0.801±0.001 79.6±0.1 5.93±0.03 91±1 8.9±0.3 110+0.4% 0.796±0.004 79.53±0.04 5.8±0.1 91±2 8.42±0.04 120°C+0.2% 0.783±0.005 79.6±0.1 6.1±0.2 89±1 9.1±0.5 120°C+0.3% 0.79±0.01 79.4±0.2 5.71±0.05 90±2 8.7±0.2 120°C+0.4% 0.775±0.004 79.2±0.2 5.7±0.1 88±2 8.6±0.08 130°C+0.2% 0.818±0.007 80.0±0.2 6.0±0.1 88±1 9.1±0.2 130°C+0.3% 0.817±0.004 80.3±0.2 5.89±0.08 90.0±0.4 8.4±0.2 130°C+0.4% 0.813±0.004 80.8±0.1 5.9±0.2 89±1 8.6±0.3 Part 2 Raw 2 0.96±0.01 78.8±0.4 9.1±0.1 84.3±0.7 10.9±0.3 150°C + 0.2% 1.02±0.01 77.5±0.8 6.8±0.3 87±1 12±1 150°C + 0.8% 0.96±0.05 80±3 6.2±0.3 85.2±0.7 10.5±0.4 130°C + 0.8% 0.99±0.02 77.3±0.5 5.2±0.2 87.1±0.4 9.9±0.4 110°C + 0.8% 0.968±0.002 79.1±0.1 6.1±0.3 87±1 10.5±0.2     160  APPENDIX B – Complete Data for the High Temperature 915MHz Microwave Experiments on Waste Activated Sludge Table 42 – Complete data set from high temperature WAS experiments using the 915MHz MW heater Sample Raw 1 MW Only Raw 2 Raw 2 + O3 H2O2 H2O2 + O3 TS (%) 0.877±0.005 0.857±0.003 0.775±0.005 0.755±0.003 0.713±0.004 0.525±0.003 TVS (%) 80.4±0.1 80.5±0.4 81.2±0.3 81.0±0.2 75.3±0.3 74.2±0.5 TSS (g/L) 8.60±0.03 5.8* 7.4±0.08 6.4±0.1 2.9±0.2 1.4±0.1 VSS (%) 80±1 85* 79±3 79±3 69±7 43±9 TCOD (g/L) 9.88±0.07 9.9±0.2 8.5±0.2 8.3±0.2 7.7±0.2 5.0±0.1 SCOD (g/L) 0.108±0.003 2.982±0.008 0.12±0.01 0.79±0.04 4.00±0.08 4.34±0.07 SCOD (%) 1.10±0.03 30.0±0.5 1.4±0.1 9.5±0.6 52±2 87±3 OP (mg/L) 0.04±0.02 103±1 1.12±0.01 48±1 127±1 134.2±0.6 PP (mg/L) - 57±2 - - 58±1 66±1 Non Sol. PP (mg/L) - 48±2 - - 40±1 0±2 TP (mg/L) 230±30 250±10 260±20 264±4 290±20 236±1 STP (mg/L) 2.1±0.1 120±10 4.2±0.1 50±1 230±10 250±10 STP (%) 0.9±0.1 48±5 1.6±0.1 18.9±0.5 77±6 104±5 NH3 (mg/L) 0.18±0.02 7.21±0.03 0.405±0.005 3.16±0.02 80.3±0.8 89±1 TKN (mg/L) 410±60 430±70 410±30 430±10 440±40 400±20 STKN (mg/L) BDL 140±40 1±1 40±3 380±30 450±40 STKN (%) BDL 30±10 0.3±0.2 9.3±0.8 80±10 110±10 VFA (mg/L) 3±1 21±2 0.63±0.04 1.8±0.4 245±9 298±6 CST (s) 117±4 1 250±40 165±5 200±10 19±1 12.9±0.4 D10 (µm) 11.4±0.2 5.4±0.1 10.0±0.3 9.48±0.03 0.461±0.002 0.462±0.001 D50 (µm) 29.1±0.2 22.9±0.1 27.6±0.5 26.8±0.1 0.599±0.03 0.606±0.001 D90 (µm) 66.6±0.6 71±3 66±1 64.9±0.9 1.118±0.004 1.172±0.001 Total Ca (mg/L) 130±20 - 77.1±0.7 - - - Sol Ca (mg/L) 25±6 22±2 13.1±0.2 16.6±0.9 67±4 58.6±0.4    161  Sample Raw 1 MW Only Raw 2 Raw 2 + O3 H2O2 H2O2 + O3 Sol Ca (%) 20±6 18±4 17.0±0.3 21±1 87±5 75.9±0.9 Total K (mg/L) 75±1 - 65.9±0.3 - - - Sol K (mg/L) 7.2±0.5 68±1 7.8±0.3 28.4±0.3 60±2 62.8±0.4 Sol K (%) 10±1 90±2 11.8±0.4 43.1±0.5 91±2 95.3±0.7 Total Mg (mg/L) 66.8±0.4 - 58.1±0.6 - - - Sol Mg (mg/L) 1.6±0.1 25.9±0.1 1.98±0.05 10.7±0.2 51.4±0.6 51.1±0.5 Sol Mg (%) 2.5±0.2 38.8±0.3 3.4±0.1 18.4±0.4 88±1 88±1 ICP TP (mg/L) 264±2 - 241±2 - - - ICP STP (mg/L) BDL 153±1 1.7±0.2 47.8±0.9 196±2 199±3 ICP STP (%) BDL 58±1 0.71±0.08 19.8±0.4 81±1 82±2 pH - - 6.66 6.48 3.84 3.78 Note: BDL refers to Below Detection Limit *Only one sample was analyzed   162  APPENDIX C – Complete Data for the 27MHz Radiofrequency Experiments Investigating Hydrogen Peroxide on Waste Activated Sludge Table 43 – Complete data set from a no H2O2 experiment on WAS using the 27MHz RF heater Sample Raw 45°C 60°C 75°C 90°C TS (%) 0.99±0.01 0.93±0.01 0.92±0.01 0.94 ±0.01 0.96±0.01 TVS (%) 81.4±0.5 79.9±0.6 80.9±0.3 81.7±0.5 81.9 ±0.7 TSS (g/L) 9.9±0.2 8.9±0.1 7.4±0.1 7.23±0 6.9±0.3 VSS (%) 80.0±0.4 82±2 86±1 88±2 88±2 TCOD (g/L) 9.92±0.07 9.6±0.2 9.7±0.1 10.1±0.1 9.8±0.2 SCOD (g/L) 0.04±0.01 0.20±0.01 1.40±0.01 1.87±0.06 2.02±0.01 SCOD (%) 0.4±0.1 2.1±0.1 14.5±0.2 18.6±0.6 20.6±0.4 OP (mg/L) 0.96±0.01 15.2±0.1 165±1 86.7±0.6 75.5±0.00 TP (mg/L 320±10 295±8 300±10 312±8 310±20 STP (mg/L) 0.6±0.1 15.4±0.3 168±3 207±7 216±2 STP (%) 0.18±0.04 5.2±0.2 55±2 66±3 70±4 PP (mg/L) - - 36±4 119±1 133±1 PP (%) - - 21±2 57±2 61.6±0.7 NH3 (mg/L) 0.23±0.01 11.2±0.6 12.7±0.2 7.3±0.3 5.57±0.09 TKN (mg/L) 680±50 610±20 650±80 640±30 630±40 STKN (mg/L) 1.8±0.6 10.3±0.3 104±3 194±5 230±10 STKN (%) 0.27±0.09 1.7±0.1 16±2 30±2 36±3 VFA (mg/L) 1.9±0.1 2.3±0.3 15±1 10±6 17±1 CST (s) 160±50 - - - 830±10 D10 (µm) 11.1±0.4 - - - 8.7±0.2 D50 (µm) 29.3±0.5 - - - 26.8±0.1 D90 (µm) 63±2 - - - 62±1 Total Ca (mg/L) 92±4 - - - - Sol Ca (mg/L) 10.6±0.1 12±1 22.4±0.5 32.7±0.9 30.2±0.4 Sol Ca (%) 12±1 14±1 24±1 35±2 33±2 Total K (mg/L) 101±2 - - - - Sol K (mg/L) 17.2±0.3 27.0±0.5 75±2 81±2 83±2 Sol K (%) 17.0±0.5 26.7±0.7 74±2 80±2 82±3 Total Mg (mg/L) 73±3 - - - - Sol Mg (mg/L) BDL 0.6±0.3 33.9±0.4 37.7±0.9 41.0±0.7 Sol Mg (%) BDL 0.9±0.3 47±2 52±2 56±2 ICP TP (mg/L) 330±10 - - - - ICP STP (mg/L) BDL 12.5±0.4 180±2 202±5 216±4 ICP STP (%) BDL 3.8±0.2 54±2 61±3 65±3 Note: BDL refers to Below Detection Limit   163  Table 44 – Complete data set from a high H2O2 dosage experiment on WAS using the 27MHz RF heater Sample Raw 45°C 60°C 75°C 90°C TS (%) 0.96±0.01 0.88±0.01 0.879±0.002 0.86±0.01 0.870±0.003 TVS (%) 80.8±0.3 82±2 82.2±0.2 81.4±0.4 81.0±0.5 TSS (g/L) 9.17±0.07 8.41±0.05 6.87±0.09 5.2±0.3 3.60±0.09 VSS (%) 83±2 83±2 86.6±0.5 85±2 84±1 TCOD (g/L) 10.00±0.09 9.4±0.2 9.4±0.2 10.2±0.04 10.25±0.06 SCOD (g/L) 0.078±0.008 0.182±0.003 1.240±0.005 4.8±0.1 5.4±0.1 SCOD (%) 0.8±0.1 1.93±0.04 13.1±0.3 47±1 52±1 OP (mg/L) 0.97±0.02 14.2±0.2 109.8±0.6 33.6±0.5 54.3±0.6 TP (mg/L 310±20 250±20 290±10 280±20 260±30 STP (mg/L) 0.3±0.3 13.9±0.9 123±4 164±2 202±7 STP (%) 0.1±0.1 5.5±0.5 42±2 58±3 80±10 PP (mg/L) - - 45.5±0.8 110±3 136±1 PP (%) - - 37±1 67±2 67±2 NH3 (mg/L) 0.04±0.03 9.3±0.1 7.7±0.2 20.8±0.2 27.5±0.7 TKN (mg/L) 640±70 500±60 630±60 600±50 550±90 STKN (mg/L) 0.7±0.3 14±2 86±5 280±40 400±30 STKN (%) 0.11±0.05 2.7±0.5 14±1 46 ±7 70±10 VFA (mg/L) 1.7±0.5 2.3±0.2 15±4 70±10 92±2 CST (s) 180±50 - - - 220±70 D10 (µm) 10.2±0.1 - - - 1.25±0.05 D50 (µm) 28.1±0.2 - - - 14.7±0.2 D90 (µm) 64±2 - - - 70±20 Total Ca (mg/L) 93.1±0.9 - - - - Sol Ca (mg/L) 13.0±0.5 14.1±0.2 18.3±0.4 39±7 54±2 Sol Ca (%) 13.9±0.5 15.1±0.3 19.6±0.4 42±8 57±2 Total K (mg/L) 98.6±0.9 - - - - Sol K (mg/L) 15.7±0.2 24.7±0.4 70±1 75±2 85±1 Sol K (%) 15.9±0.3 25.0±0.5 71±1 77±2 86±1 Total Mg (mg/L) 68.0±0.5 - - - - Sol Mg (mg/L) BDL 0.62±0.05 25.8±0.3 34.2±0.7 51±2 Sol Mg (%) BDL 0.91±0.07 38.0±0.5 50±1 75 ±3 ICP TP (mg/L) 311±3 - - - - ICP SP (mg/L) BDL 11.4±0.2 144±6 162±8 212±7 ICP SP (%) BDL 3.7±0.1 46±2 52±3 68±2 Note: BDL refers to Below Detection Limit   164  Table 45 – Complete data set from a medium H2O2 dosage experiment on WAS using the 27MHz RF heater Sample Raw 60°C 75°C 90°C TS (%) 1.579±0.003 1.528±0.001 1.52±0.01 1.361±0.004 TVS (%) 80.88±0.08 80.71±0.06 79.9±0.5 79.3±0.4 TSS (g/L) 15.9±0.5 13.4±0.2 11.5±0.5 7.3±0.1 VSS (%) 82.7±0.3 85.5±0.2 84.0±0.3 82±2 TCOD (g/L) 16.1±0.6 16.3±0.1 16.0±0.6 14.6±0.4 SCOD (g/L) 0.142±0.003 2.55±0.04 4.7±0.1 6.6±0.3 SCOD (%) 0.88±0.04 15.6±0.3 30±1 45±2 OP (mg/L) 0.323±0.006 120±2 56.0±0.5 104.7±0.8 TP (mg/L 690±20 680±3 646±9 650±30 STP (mg/L) 2.8±0.2 309±1 260±10 340±10 STP (%) 0.40±0.03 45.4±0.3 40±2 53±3 PP (mg/L) - 252±8 250±20 357±3 NH3 (mg/L) 1.20±0.02 7.2±0.3 17.9±0.1 32.5±0.5 TKN (mg/L) 1 110±40 1 180±60 1 140±50 1 100±100 STKN (mg/L) 8±2 215±4 390±50 570±20 STKN (%) 0.7±0.2 18±1 34±4 50±6 VFA (mg/L) 4±1 36±7 64±2 104±4 CST (s) 590±70 - - 700±100 D10 (µm) 9.2±0.1 - - 3.2±0.1 D50 (µm) 23.46±0.09 - - 19±1 D90 (µm) 54.5±0.1 - - 68±2 Total Ca (mg/L) 142±4 - - - Sol Ca (mg/L) 20±1 44±2 44±3 68±1 Sol Ca (%) 14±1 31±2 31±2 48±2 Total K (mg/L) 193±6 - - - Sol K (mg/L) 32±2 136±4 146.1±0.9 160±3 Sol K (%) 17±1 70±3 76±2 83±3 Total Mg (mg/L) 140±2 - - - Sol Mg (mg/L) 4.7±0.2 55.5±0.9 56.2±0.3 92±2 Sol Mg (%) 3.3±0.2 39.6±0.9 40.1±0.7 65±2 ICP TP (mg/L) 568±8 - - - ICP STP (mg/L) 4.3±0.2 274±6 244±3 351±6 ICP STP (%) 0.76±0.04 48±1 43.0±0.8 62±1    165  Table 46 – Complete data set from a low H2O2 dosage experiment on WAS using the 27MHz RF heater Sample Raw 60°C 75°C 90°C TS (%) 1.33±0.01 1.291±0.005 1.298±0.001 1.320±0.001 TVS (%) 80.01±0.06 79.80±0.01 79.4±0.3 79.7±0.1 TSS (g/L) 12.7±0.3 10.7±0.2 9.5±0.5 8.3±0.3 VSS (%) 81.1±0.8 89±2 88±1 86±2 TCOD (g/L) 13.44±0.09 13.27±0.08 14.0±0.4 12.9±0.3 SCOD (g/L) 0.131±0.005 1.46±0.02 2.9±0.2 4.07±0.02 SCOD (%) 0.97±0.04 11.0±0.2 21±2 31.5±0.7 OP (mg/L) 0.111±0.002 133±1 48.0±0.3 68.2±0.6 TP (mg/L 710±20 670±20 680±20 690±60 STP (mg/L) 0.3±0.5 370±10 410±40 440±20 STP (%) BDL 55±2 61±6 64±6 PP (mg/L) - 113±9 217±6 238±2 NH3 (mg/L) 0.11±0.04 6.3±0.4 7.41±0.07 11.4±0 TKN (mg/L) 1 050±30 1 010±20 1 010±10 1 000±30 STKN (mg/L) 8.5±0.7 147±5 250±10 380±20 STKN (%) 0.82±0.07 15±1 25±1 38±2 VFA (mg/L) 5.4±0.6 43±3 80±10 106±3 CST (s) 540±20 640±30 - 1 200±100 D10 (µm) 8.9±0.1 9.07±0.05 - 7.4±0.4 D50 (µm) 23.2±0.3 23.31±0.07 - 22.6±0.7 D90 (µm) 56±2 55.3±0.2 - 54±1 Total Ca (mg/L) 180±10 - - - Sol Ca (mg/L) 26±3 69.0±0.6 65±1 85±1 Sol Ca (%) 15±2 39±3 37±3 49±4 Total K (mg/L) 289±9 - - - Sol K (mg/L) 38±3 240±10 236±2 257±5 Sol K (%) 13±1 82±4 82±3 89±3 Total Mg (mg/L) 109±3 - - - Sol Mg (mg/L) 16.1±0.3 57±1 56.4±0.3 72±2 Sol Mg (%) 14.7±0.5 53±2 52±2 66±2 ICP TP (mg/L) 399±10 - - - ICP STP (mg/L) 25.3±0.7 221±5 212±3 270±10 ICP STP (%) 6.3±0.3 56±2 53±2 68±4 Note: BDL refers to Below Detection Limit    166  APPENDIX D – Complete Data for the Low Temperature 27MHz Radiofrequency Experiment on Waste Activated Sludge Table 47 – Complete data set from a low temperature experiment on WAS using the 27MHz RF heater Sample Raw 15min 30min 45min 60min TS (%) 1.352±0.001 1.326±0.003 1.311±0.002 1.310±0.003 1.325±0.001 TVS (%) 80.12±0.01 77.88±0.06 78.3±0.1 78.6±0.1 77.7±0.2 TSS (g/L) 13.7±0.2 10.3±0.4 9.6±0.3 9.3±0.3 11.0±0.2 VSS (%) 84±2 82±2 81±2 81±1 82±1 TCOD (g/L) 14.0±0.2 14±1 14.0±0.2 14.0±0.5 14.0±0.3 SCOD (g/L) 0.143±0.007 3.4±0.1 3.9±0.2 3.7±0.1 3.64±0.09 SCOD (%) 1.03±0.05 25±2 27±2 26±1 26.0±0.8 OP (mg/L) 5.8±0.2 52±2 50±1 51.3±0.8 54.3±0.6 PP (mg/L) - 194±5 195.3±0.6 191±1 194 ±1 TP (mg/L 640±20 650±20 650±40 630±10 640±30 STP (mg/L) 11.7±0.2 318±7 317±5 326±8 322±3 STP (%) 1.8±0.1 49±2 49±3 52±2 51±2 NH3 (mg/L) 5.61±0.02 16.5±0.1 17.47±0.06 18.23±0.06 19.9±0.1 TKN (mg/L) 1 070±20 1 130±80 1 080±80 1 080±30 1 140±60 STKN (mg/L) 14±1 317±1 346±7 370±10 393±1 STKN (%) 1.3±0.1 28±2 32±3 34±2 35±2 VFA (mg/L) 48±2 52±1 51±2 50±2 19±3 CST (s) 260±40 1 300±100 - - 1 100±100 D10 (µm) 0.5277±0.0006 0.454 ±0 0.4403±0.0006 0.424±0.001 0.4067±0.0006 D50 (µm) 0.6757±0.0006 0.584±0 0.573±0 0.564±0.002 0.548±0.003 D90 (µm) 1.182±0.001 0.984±0.001 0.9707±0.0006 0.963±0.006 0.940±0.01 Total Ca (mg/L) 140±20 - - - - Sol Ca (mg/L) 15±8 36.5±0.5 36.0±0.5 36±3 35.7±0.8 Sol Ca (%) 11±6 26±3 26±3 26±4 26±3 Total K (mg/L) 161±2 - - - - Sol K (mg/L) 50±1 138±2 137±2 138.1±0.9 136±2 Sol K (%) 30.9±0.8 86±2 85±2 86±1 84±2 Total Mg (mg/L) 110±2 - - - - Sol Mg (mg/L) BDL 43.6±0.8 43.3±0.7 43±1 43.9±0.2 Sol Mg (%) BDL 40±1 39±1 39±1 39.9±0.8 ICP TP (mg/L) 492±9 - - - - ICP STP (mg/L) BDL 243±1.63 241±5 238±2 236±1 ICP STP (%) BDL 49±1 49±1 48±1 48.0±0.9 Note: BDL refers to Below Detection Limit   167  APPENDIX E – Complete Data for the Low Temperature 915MHz Microwave Experiment on Waste Activated Sludge Table 48 – Complete data set from a 90°C WAS experiment using the 915MHz MW heater Sample Raw 60°C 75°C 90°C TS (%) 1.579±0.003 1.535±0.005 1.438±0.004 1.508±0.001 TVS (%) 80.88±0.08 80.58±0.03 79.3±0.3 80.2±0.1 TSS (g/L) 15.9±0.5 13.6±0.4 10.6±0.2 8.9±0.2 VSS (%) 82.7±0.3 84±2 82±2 85±1 TCOD (g/L) 16.1±0.6 16.7±0.7 15.5±0.2 15.8±0.1 SCOD (g/L) 0.142±0.003 2.35±0.04 5.2±0.3 6.1±0.3 SCOD (%) 0.88±0.04 14.1±0.7 34±2 39±2 OP (mg/L) 0.323±0.006 87.8±0.8 65.3±0.6 103.2±0.6 TP (mg/L 690±20 690±50 600±30 650±30 STP (mg/L) 2.8±0.2 282±2 272±3 406±3 STP (%) 0.40±0.03 41±3 46±2 63±3 PP (mg/L) - 260±10 230±20 420±30 NH3 (mg/L) 1.20±0.02 6.3±0.3 21.2±0.4 32.9±0.2 TKN (mg/L) 1 110±40 1 047±1 970±70 1 100±100 STKN (mg/L) 8±2 220±3 400±40 660±20 STKN (%) 0.7±0.2 21.4±0.3 42±5 61±6 VFA (mg/L) 4±1 23±3 62±5 100±10 CST (s) 590±70 - - 950±50 D10 (µm) 9.2±0.1 - - 5.4±0.4 D50 (µm) 23.46±0.09 - - 21.3±0.7 D90 (µm) 54.5±0.1 - - 58±4 Total Ca (mg/L) 142±4 - - - Sol Ca (mg/L) 20±1 40.5±0.9 47±5 71±1 Sol Ca (%) 14±1 28±1 33±3 50±2 Total K (mg/L) 193±6 - - - Sol K (mg/L) 32±2 139±2 147±2 161±1 Sol K (%) 17±1 72±2 76±2 83±3 Total Mg (mg/L) 140±2 - - - Sol Mg (mg/L) 4.7±0.2 52±1 59.6±0.3 95±1 Sol Mg (%) 3.3±0.2 36.8±0.9 42.4±0.7 68±1 ICP TP (mg/L) 568±8 - - - ICP STP (mg/L) 4.3±0.2 254±6 247±7 361±5 ICP STP (%) 0.76±0.04 45±1 43±1 64±1     168  APPENDIX F – Complete Data for the Fast Heating Rate 915MHz Microwave Experiment on Waste Activated Sludge  Table 49 – Complete data set from a fast heating rate WAS experiment using the 915MHz MW heater Sample  Raw 60°C 90°C TS (%) 1.454±0.002 1.305±0.004 1.19±0.01 TVS (%) 80.52±0.06 80.05±0.06 79.0±0.1 TSS (g/L) 15.2±0.4 11.4±0.3 7.7±0.4 VSS (%) 81.4±0.8 86±1 85±4 TCOD (g/L) 18.6±0.2 - 14.78±0.07 SCOD (g/L) 0.145±0.004 - 5.7±0.2 SCOD (%) 0.78±0.02 - 39±1 OP (mg/L) 0.55±0.01 113.5±0.5 45.4±0.3 PP (mg/L) - 178±1 204.6±0.3 TP (mg/L 680±10 600±20 557±4 STP (mg/L) 2.9±0.4 320±5 290±10 STP (%) 0.4±0.1 54±2 53±2 NH3 (mg/L) 0.29±0.05 7.8±0.1 21.7±0.2 TKN (mg/L) 1 100±30 1 000±10 890±30 STKN (mg/L) 4.8±0.5 200±4 370±10 STKN (%) 0.4±0.1 20.1±0.5 42±2 VFA (mg/L) 2.0±0.1 10±1 36±3 CST (s) 400±20 - 900±100 D10 (µm) 9.28±0.07 - 5.50±0.07 D50 (µm) 23.6±0.1 - 22.0±0.2 D90 (µm) 55.5±0.5 - 60.0±0.2 Total Ca (mg/L) 147±5 - - Sol Ca (mg/L) 22±2 42.4±0.5 52±5 Sol Ca (%) 15±1 29±1 35±3 Total K (mg/L) 191±3 - - Sol K (mg/L) 30.79±0.03 126±1 135±5 Sol K (%) 16.1±0.3 66±1 71±3 Total Mg (mg/L) 141±3 - - Sol Mg (mg/L) 4.8±0.1 56.6±0.7 65±3 Sol Mg (%) 3.4±0.1 40±1 46±2 ICP TP (mg/L) 580±10 - - ICP STP (mg/L) 4.6±0.7 276.±0.6 260±10 ICP STP (%) 0.8±0.1 47.5±0.8 45±2     169  APPENDIX G – Complete Data for the 27MHz Radiofrequency Experiments on Dairy Manure Table 50 – Complete data set from a 25°C H2O2 addition experiment on DM using the 27MHz RF heater Sample Raw Raw + Acid (Day of) Raw + Acid (Day Before) 60°C 90°C TS (%) 3.51±0.02 4.09±0.01 4.19±0.03 4.00±0.05 4.12±0.03 TVS (%) 72.2±0.1 68.8±0.1 69.1±0.2 63.8±0.2 64.4±0.4 TSS (g/L) 22±3 25.4±0.5 25±1 22.4±0.1 21±2 VSS (%) 76.9±0.5 85±2 83±1 81±2 82±2 TCOD (g/L) 38±1 39.5±0.5 40±5 41±3 41±3 SCOD (g/L) 8.7±0.3 6.2±0.3 6.55±0.07 10.7±0.4 12.7±0.3 SCOD (%) 23±1 15.7±0.9 16±2 26±2 31±2 OP (mg/L) 4.6±0.4 191±2 191±2 - 175±2 TP (mg/L 360±20 380±30 420±10 - 380±60 STP (mg/L) 22±4 245±6 240±20 - 268±9 STP (%) 6±1 65±5 58±6 - 70±10 NH3 (mg/L) 670±50 910±40 920±10 - 810±20 TKN (mg/L) 1 900±200 2 100±200 2 150±40 - 2 010±40 STKN (mg/L) 400±200 1 085±9 1 070±30 - 1 110±20 STKN (%) 20±10 53±4 50±2 - 55±1 VFA (mg/L) 2 470±30 2 470±20 2 500±30 - 2 270±50 Total Ca (mg/L) 1 140±20 - - - - Sol Ca (mg/L) 184±7 760±10 770±30 - 850±30 Sol Ca (%) 16.2±0.7 67±2 68±3 - 75±3 Total K (mg/L) 1 790±50 - - - - Sol K (mg/L) 1 700±20 1 670±50 1 690±30 - 1 600±40 Sol K (%) 95±3 93±4 94±3 - 89±3 Total Mg (mg/L) 309±5 - - - - Sol Mg (mg/L) 164±5 257±4 267±5 - 254±7 Sol Mg (%) 53±2 83±2 86±2 - 82±3 ICP TP (mg/L) 277±8 - - - - ICP STP (mg/L) 70±2 191±2 192±1 - 211±7 ICP STP (%) 25.3±0.9 69±2 69±2 - 76±3    170   Table 51 – Complete data set from a 60°C H2O2 addition experiment on DM using the 27MHz RF heater Sample Raw 60°C 90°C 90°C + 10min TS (%) 3.51±0.02 4.1±0.3 4.39±0.01 4.4±0.1 TVS (%) 72.2±0.1 72±6 65.9±0.2 65.0±0.4 TSS (g/L) 22±3 21±1 26±5 20±1 VSS (%) 76.9±0.5 83±1 83±3 86±3 TCOD (g/L) 38±1 39±2 40±2 43±1 SCOD (g/L) 8.7±0.3 7.79±0.03 13.8±0.2 13.7±0.3 SCOD (%) 23±1 20±1 34±2 32±1 OP (mg/L) 4.6±0.4 - 183±3 189±3 TP (mg/L 360±20 - 390±20 380±30 STP (mg/L) 22±4 - 274±7 280±10 STP (%) 6±1 - 71±3 74±7 NH3 (mg/L) 670±50 - 860±30 840±30 TKN (mg/L) 1 900±200 - 2 120±90 2 030±40 STKN (mg/L) 400±200 - 1 240±50 1 250±90 STKN (%) 20±10 - 59±3 62±4 VFA (mg/L) 2 470±30 - 2 530±40 2 570±10 Total Ca (mg/L) 1 140±20 - - - Sol Ca (mg/L) 184±7 - 840±40 850±20 Sol Ca (%) 16.2±0.7 - 74±4 75±2 Total K (mg/L) 1 790±50 - - - Sol K (mg/L) 1 700±20 - 1 640±50 1 650±10 Sol K (%) 95±3 - 91±4 92±3 Total Mg (mg/L) 309±5 - - - Sol Mg (mg/L) 164±5 - 261±8 263±4 Sol Mg (%) 53±2 - 85±3 85±2 ICP TP (mg/L) 277±8 - - - ICP STP (mg/L) 70±2 - 215±2 222±1 ICP STP (%) 25.3±0.9 - 78±2 80±2     171  APPENDIX H – Complete Data for the 915MHz Microwave Experiments on Dairy Manure  Table 52 – Complete data set from a 110°C DM experiment using the 915MHz MW heater Sample Raw Raw + Acid  (Day of) Raw + Acid  (Day Before) Run TS (%) 3.77±0.02 4.21±0.08 4.15±0.05 3.5±0.6 TVS (%) 72.8±0.2 70±1 68±1 63±2 TCOD (g/L) 44±3 40.8±0.2 40±1 28.5±0.5 SCOD (g/L) 10.3±0.1 9.3±0.6 8.3±0.2 16±1 SCOD (%) 23±2 23±2 20.8±0.8 55±5  OP (mg/L) 4.0±0.1 204±4 205±4 229±5 TP (mg/L 430±30 350±30 357±3 330±40 STP (mg/L) 20.2±0.4 223±8 237±8 278±5 STP (%) 4.7±0.3 63±6 66±2 84±10 NH3 (mg/L) 760±30 840±10 870±50 865±9 TKN (mg/L) 2 400±200 1 900±200 1 890±60 1 800±300 STKN (mg/L) 880±50 940±90 1 060±40 1 370±30 STKN (%) 36±4 49±7 56±3 80±10 VFA (mg/L) 3 220±20 2 930±20 2 920±20 3 060±90 Total Ca (mg/L) 1 050±80 - - - Sol Ca (mg/L) 220±10 880±20 860±20 570±30 Sol Ca (%) 20±2 83±6 81±6 54±5 Total K (mg/L) 1 600±70 - - - Sol K (mg/L) 1 700±100 1 660±40 1 660±30 1 290±60 Sol K (%) 109±8 104±5 104±5 81±5 Total Mg (mg/L) 400±20 - - - Sol Mg (mg/L) 160±8 263±8 262±5 220±40 Sol Mg (%) 40±3 66±4 65±3 50±10 ICP TP (mg/L) 336±8 - - - ICP STP (mg/L) 77±1 198±7 196±5 194±9 ICP STP (%) 22.9±0.6 59±3 58±2 58±3    172  Table 53 – Complete data set from a 130°C DM experiment using the 915MHz MW heater Sample Raw Raw + Acid (Before Sieve) Raw + Acid (After Sieve) Run TS (%) 3.77±0.02 3.42±0.01 3.35±0.01 2.41±0.02 TVS (%) 72.8±0.2 68.2±0.1 67.4±0.1 57.2±0.7 TCOD (g/L) 44±3 34±2 34±1 19±1 SCOD (g/L) 10.3±0.1 8.3±0.2 8.1±0.2 10.1±0.5 SCOD (%) 23±2 25±1 24±1 54±4 OP (mg/L) 4.0±0.1 178±3 179±4 240±4 TP (mg/L 430±30 338±2 400±10 295±8 STP (mg/L) 20.2±0.4 216±6 221±6 283±8 STP (%) 4.7±0.3 64±2 50±8 96±4 NH3 (mg/L) 760±30 770±20 750±60 870±60 TKN (mg/L) 2 400±200 1 800±80 2 500±500 1 500±100 STKN (mg/L) 880±50 1 030±60 1 080±30 1 410±70 STKN (%) 36±4 57±4 43±8 97±9 VFA (mg/L) 3 220±20 2 570±60 2 670±30 2 930±80 Total Ca (mg/L) (% of Before Sieve) 1 050±80 - 880±20 (84±6) - Sol Ca (mg/L) 220±10 800±200 800±300 532±7 Sol Ca (%) 20±2 90±20 90±30 60±1 Total K (mg/L) (% of Before Sieve) 1 600±70 - 1420±60 (89±6) - Sol K (mg/L) 1 700±100 1270±60 1 210±10 1110±30 Sol K (%) 109±8 89±6 85±4 78±4 Total Mg (mg/L) (% of Before Sieve) 400±20 - 346±5 (86±4) - Sol Mg (mg/L) 160±8 270±10 270±4 250±10 Sol Mg (%) 40±3 79±3 78±2 73±3 ICP TP (mg/L) (% of Before Sieve) 336±8 - 278±9 (83±3) - ICP STP (mg/L) 77±1 149±2 147±2 212±5 ICP STP (%) 22.9±0.6 54±2 53±2 76±3    173  APPENDIX I – Complete Data for the 2.45GHz Microwave Experiments on Palm Oil Mill Effluent Table 54 – Complete data set from POME experiments using the 2.45GHz batch MW heater Sample Raw Raw + 2.5% 120°C + 2.5% 150°C + 2.5% 150°C + 5% TS (%) 5.08±0.01 - - - - TVS (%) 82.4±0.1 - - - - TSS (g/L) 25.7±0.3 23* 17±2 15.0±0.6 7±1 VSS (%) 91.6±0.2 91* 88.6±0.8 84±1 73±4 TCOD (g/L) 75±1 82* 67.4±0.9 56±4 28±4 SCOD (g/L) 35.9±0.4 61* 42.4±0.9 25.7±0.9 14.4±0.4 SCOD (%) 48±1 73.9* 50±10 46±4 52±8 OP (mg/L) 93* 75* 104±21 147±2 98±24 NH3 (mg/L) 24±2 69 39* 39±2 40±10 TP (mg/L 180±10 160* 174±7 152±5 179±3 STP (mg/L) 159 ±4 135* 160±4 135±9 176±5 STP (%) 86±6 84* 92±4 89±6 98±3 TKN (mg/L) 1 330±90 1 140* 1 240±50 1 180±50 1 280±40 STKN (mg/L) 530±10 502* 590±30 650±8 642±2 STKN (%) 39±3 44* 48±3 55±2 50±2 VFA (mg/L) 1 174* 1 429* 3 400±300 4 700±200 5 240±90 Total Ca (mg/L) 380±10 - - - - Sol Ca (mg/L) 325* 274* 170±70 92±6 40±20 Sol Ca (%) 86* 73* 40±20 25±2 12±5 Total K (mg/L) 3 170±70 - - - - Sol K (mg/L) 3 910* 2750* 3 400±500 3 600±300 3 000±200 Sol K (%) 123* 87* 110±20 112±8 95±8 Total Mg (mg/L) 500±10 - - - - Sol Mg (mg/L) 510* 430* 480±10 470±10 440±10 Sol Mg (%) 102* 86* 95±3 93±3 87±4 ICP TP (mg/L) 197±4 - - - - ICP STP (mg/L) 164* 139* 167±5 171±5 137±8 ICP STP (%) 83* 71* 85±3 87±3 70±4 *Note: only one sample was taken   174  APPENDIX J – Summary of Assumptions and Calculations Used to Design the Full-scale RF-H2O2 System  A complete summary of all assumptions and estimated values are shown in Table 55.  These values are used in the full-scale RF-H2O2 treatment process sizing method, detailed below the table.   Table 55 – Summary of assumptions and parameters used to design a full-scale RF-H2O2 system Parameter Value Reasoning Fluid Properties TWAS Density (kg/m3) 992.2 Based on water at 40°C TWAS Heat Capacity (J/kg/K) 4 178.5 Based on water at 40°C 70% H2O2 Density (kg/m3) 1 294 Overall density at 20°C from manufacturer‟s data (FMC Corporation, 2003) 70% H2O2 Heat Capacity  (J/kg/K) 4 178.5 Conservatively based on water at 40°C Anaerobic Digester VSS Reduction (%) 60 Tchobanoglous et al. (2003) Energy Content of Biogas (kJ/m3) 22 400 Tchobanoglous et al. (2003) Heat Exchangers E-1 Heat Transfer Coefficient (W/m2/°C) 1 000 Low range of data from Perry and Green (1999) E-1 Efficiency (%) 85 Due to heat loss E-1 Heat Duty (kW) 1 066 Calculated from above data E-1 Heat Transfer Area (m2) 25 Calculated from above data E-2 Efficiency (%) 80 Lower range taken from Towler and Sinnott (2008) due to the use of a vertical cylinder furnace.  This furnace type is suitable for low heat loads but are more inefficient (Perry and Green, 1999). E-2 Heat Duty (kW) 60.5 Calculated from H2O2 dosage and above efficiency Biogas Required for E-2 (m3/h) 9.7 Calculated from above values Dielectric Properties ε' (F/m) 7.53·10-10 From Bobowski et al. (2012), assumed to be independent of temperature according to Piyasena et al. (2003) ε" at 60°C (F/m) 3.78·10-9 From Bobowski et al. (2012), corrected using correlations from Piyasena et al. (2003) Penetration Depth at 60°C (m) 0.07 Calculated from above values ε" at 95°C (F/m) 5.37·10-9 From Bobowski et al. (2012), corrected using correlations from Piyasena et al. (2003) Penetration Depth at 95°C (m)  0.05 Calculated from above values RF Heater Power Rating per Heater (kW) 900 Largest capacities as quoted by Chindris and Sumper (2012), chosen due to improved economics Maximum Energy Density (kW/m2) 200 High range of  data from Chindris and Sumper (2012) Maximum Electric Field (V/m) 2 688 Based on above values at 95°C Circuitry Efficiency (%) 92 Schiffmann (2006)  175  Parameter Value Reasoning RF Heater Triode Efficiency (%) 70 High range of data from Chindris and Sumper (2012)  Applicator Efficiency (%) 90 Schiffmann (2006) Total Efficiency (%) 58 Calculated from above values Additional Heating Capacity (%) 15 To account for fluctuations in flow and temperature Depth of Applicator (cm) 8 After optimization (see below) Cross Section (m2) 2.5 Calculated from above values Holding Tank Retention Time (h) 3 To achieve desired PP conversion and H2O2 decomposition Extra Capacity Factor (%) 15 For fluctuations in flow rate Total Volume (L) 85 000 Based on above design values and rounded Width (m) 5.5 Square footprint for superior mixing  Height (m) 2.8 Assumed 2:1 aspect ratio Heat Loss from Floor (W/m2/°C) 2.3 300mm concrete on a semi-moist floor, taken from Tchobanoglous et al. (2003) Heat Loss from Top (W/m2/°C) 1.4 100mm concrete cover with insulation, taken from Tchobanoglous et al. (2003) Heat Loss from Sides (W/m2/°C) 0.7 300mm concrete above ground walls with insulation, taken from Tchobanoglous et al. (2003) Total Heat Loss (kW) 14.9 Calculated from above values Reaction Kinetics  TSS Reduction (%) 48 Regressed from RF results on WAS (Section 5.3.1) Decomposition of H2O2 into *OH in Applicator (%) 0.3 Corrected for 50 to 95°C using kinetic data from Takagi and Ishigure (1985) for a completely mixed batch reactor (see Table 3) Decomposition of H2O2 into *OH in Tank (%) 54 Corrected for 95°C using kinetic data from Takagi and Ishigure (1985) for a completely mixed batch reactor (see Table 3) PP Conversion to OP (%/min) 0.5 For 90 to 100°C based on Rulliere et al. (2012) and Kuroda et al. (2002) Conversion of PP in Tank (%) 90 Calculated from above data  To size the full-scale RF-H2O2 treatment process, the dielectric properties of the TWAS and their dependence on temperature must first be determined.  For this analysis, TWAS data is used from Bobowski et al. (2012).  The heat loss factor (ε”), is corrected for changes with temperature using correlations from Piyasena et al. (2003), developed for salt and starch solutions.  From this study, a salt solution is picked that matches the initial dielectric properties of TWAS.  The following correlation is used as shown in Equation 34 (R2=0.998).    (         )                   (34) ε” is in V/m and T is the temperature in kelvin.  The ε‟ is assumed to be constant as the data from Piyasena et al. (2003) suggests that it should not change by more than 6%.  Next, the maximum electric 176  field that is generated in the applicator should be chosen from a manufacturer or picked within industry standard guidelines.  Due to the large heating requirements, the maximum energy intensity, Ip, from a commercially available RF heater is used (Chindris and Sumper, 2012).   Equation 35 allows the resulting maximum electric field to be calculated:  |  |      √   √              (35) ηf is the minimum overdesign factor which takes into account fluctuations in flow rate and influent temperature, chosen as 1.15 for this analysis.  The maximum electric field is calculated at both temperature extremes and the lower value taken.  The flow rate of H2O2 and TWAS must be determined, as well as the desired temperature difference across the RF heater.  The H2O2 dosage should be chosen based on the desired TSS reduction and can be estimated using a linear model, such as the one shown in Equation 36 (R2=0.92), that is regressed from pilot-scale RF heater experiments (Section 5.3.1).                     (36) TSSred is the reduction in TSS (%) and d is the H2O2 dosage in %(v/v)/%TS.  With the total flow rate passing through the RF heater known, the total power (PT) required by the heaters can be calculated from the stream‟s change in enthalpy using Equation 37:        (     )                  (37) q is the flow rate, ρ is the density, Cp is the heat capacity, Tf is the effluent temperature, T0 is the influent temperature, and ηRF is the efficiency of the RF heater.  In this analysis, ηRF is chosen as 58% as explained in Table 55.  The number of RF heaters is calculated after selecting a power rating for each individual heater.  Because the capital cost per kW and the lifespan both improve for higher power rated RF heaters, their maximum available size of 900kW is chosen (Chindris and Sumper, 2012).  This requires three heaters to meet the process‟s necessary Pt.  Combining Equations 6, 7, and 8 and integrating with respect to the electrode separation distance (z), Equation 38 is derived for the dissipated power:               |  |    (     ⁄   )               (38) A is the cross section of the electrode plates.  Remembering that dp and ε” change with temperature, the dimensions of the applicator are analytically determined in MS Excel® using rectangle numerical integration.  This is accomplished by performing an energy balance between the substrate passing through a single heater and the power dissipated over a single time step.  A time step of 0.1s is used as a smaller 177  time step offered negligible improvements in accuracy.  By advancing the number of time steps (i.e. adding more rows) for a given applicator depth (z), the electrode area required to raise the TWAS to the desired temperature can be determined.  This process is reiterated with various penetration depths to create electrode area and applicator volume curves, as shown in Figure 54.  Figure 54 – Possible electrode dimensions that are capable of heating the required TWAS flow rate From Figure 54, it is clear that increasing the electrode‟s separation will require larger applicator volumes but smaller electrode areas, which decreases rapidly for small separations.  The optimal electrode separation for this analysis is chosen at 8cm in order to take advantage of the electrode area‟s exponential decay, while minimizing the applicator‟s overall volume (see Figure 54).  Lastly, the decomposition of H2O2 into *OH is estimated using its rate determining step (Equation 23) and kinetic data from Takagi and Ishigure (1985).  For the holding tank, the conversion (X), or fraction of H2O2 that undergoes decomposition, is calculated using Equation 39.    (       ⁄ )    (       ⁄ )           (39) T is the temperature of the holding tank, assumed to be nearly constant, and η is the chosen retention time.  Calculating the conversion in the applicator requires a system of ordinary differential equations to be solved due to the changing temperature.  By modeling the applicator as a plug flow reactor, the conversion is related to its volume, v, in Equation 40.  The temperature is also related to the applicator‟s volume in Equation 41 via its heating rate (q), which can be determined from the above data.          (   )    (    )⁄               (40) 00.050.10.150.20.250.30.35024681012140 5 10 15 20Applicator Volume (m3) Electrode Area (m2) Distance Between Electrodes (cm) Electrode AreaApplicator Volume178                     (41) Equations 40 and 41 are solved in Matlab® using appropriate initial conditions and the applicator‟s previously determined volume as the limit for v.  If a greater conversion is desired, either a larger tank can be utilized or the entire process can be reiterated with a different RF power rating to achieve an improved conversion.  All other auxiliary equipment are sized according to design manuals using the factors and references detailed in Table 55.  The information calculated in the above method can be used to work out preliminary costs for future feasibility studies.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0216471/manifest

Comment

Related Items