UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effect of mixing on peroxymonosulfate generation Shaharuzzaman, Mohammad 1998

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

Item Metadata

Download

Media
831-ubc_1999-0609.pdf [ 5.58MB ]
Metadata
JSON: 831-1.0058948.json
JSON-LD: 831-1.0058948-ld.json
RDF/XML (Pretty): 831-1.0058948-rdf.xml
RDF/JSON: 831-1.0058948-rdf.json
Turtle: 831-1.0058948-turtle.txt
N-Triples: 831-1.0058948-rdf-ntriples.txt
Original Record: 831-1.0058948-source.json
Full Text
831-1.0058948-fulltext.txt
Citation
831-1.0058948.ris

Full Text

Effect of Mixing on Peroxymonosulfate Generation by  Mohammad Shahamzzaman Ph.D. (Organic Chemistry), University of Missouri, USA, 1996  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF CHEMICAL ENGINEERING  We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA October, 1998  In  presenting  this  degree at the  thesis  in  partial fulfilment  of  University of  British Columbia,  I agree  freely available for reference copying  of  department publication  this or of  and study.  this  his  or  her  requirements that the  I further agree  thesis for scholarly purposes by  the  may be  representatives.  It  thesis for financial gain shall not  is  that  CA^JLV^CAJ  % BlOTeSO^YtX.  granted  allowed  The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  01-2-8- «W  S^u^*  1  advanced  permission for extensive  permission.  Department of  an  Library shall make it  by the  understood be  for  ^  that without  head  of  my  copying  or  my written  ABSTRACT  Efficient generation of pulping and bleaching chemicals is essential to economic production of pulp. In some cases the success of a proposed process will depend on whether the key chemical can be generated economically at an industrial scale. Peroxymonosulfate (PMS, Na S0 ) has shown to be an effective and selective TCF 2  5  bleaching agent for both delignification and brightness development. The commercial production of PMS is achieved by reacting concentrated  H2SO4  with 70%  H2O2.  Conversion is limited to 70% by equilibrium, and production costs are high for commercial bleaching applications. With the increase of oxygen delignification throughout the world, a cheaper method to generate alkaline PMS is desired. This is because an acidic PMS stage, if placed between two alkaline oxygen treatments, would require additional alkali to reach the required bleaching pH. The additional cost of the added alkali would significantly increase the total cost associated with use of acidic PMS. PMS can also be generated by the oxidation of sodium sulfite with oxygen or air in alkali media. This shows promise as an economic alternative for basic PMS generation. Past generation of PMS in the laboratory was limited to 20% yield and 3.8 g/L NasSOs. However, the reaction is mixing-sensitive and optimization of mixing and reaction conditions allow both the yield and chemical concentration to be increased. The effects of key parameters (feed-time, energy dissipation rate, sodium sulfite concentration, reactant volume ratio, oxygen pressure and temperature) were verified in a number of different batch reactors (medium-intensity, PHT, high-intensity, rotor-stator). The addition time of sodium sulfite (feed time) was found to affect both the PMS yield and concentration. For very short feed times (<60 sec), there is a point where the yield of PMS is a maximum. Experiments indicate that for each set of reaction conditions an optimal feed time must be found to achieve the maximum PMS yield. An increase in energy dissipation (0.33 W/kg - 1000 W/kg) rate was always found to increase PMS generation efficiency. A weaker sodium sulfite solution was useful to generate PMS at higher yield and lower concentration. By manipulating mixing parameters (0.5M Na^SC^, ii  4°C, 950 W/kg,  Voxygenated water/V dium sulfite SO  soiution=25),  PMS yield was increased to 54.4%  at 1.6 g/L. The concentration (2.0M Na S0 , 4°C, 770 W/kg, 2  solution  3  V  o x y g  e n a t e d W V o d i u m sulfite S  -  25) was increased to 9.8 g/L, but the PMS yield dropped to 33%. Mathematical modelling shows that by adjusting mixing, both the yield and concentration of PMS can be increased. Generation limits are expected to be about 80% yield at 10.0 g/L PMS at 5000 W/kg. Oxygen solubility was found to be a limiting factor as observed experimentally in the P T reactor by performing experiments at increased H  pressure. The same conclusion was made from mass transfer calculations. Both indicate that oxygen depletion can be a major barrier in running experiments at higher (1.5 2.0M) sodium sulfite concentration. Sulfite solutions could not be prepared at concentrations higher than 2M because of the solubility limitation of sodium sulfite. PMS yield was decreased by almost 20% when reactions were carried out at room temperature (20 - 25°C) as opposed to 4°C. An engulfment model (E-model) predicts that a high yield and concentration of PMS is possible. However, our experimental results still lag behind the model prediction. This is possibly due to limitation in achieving maximum average energy dissipation (950 W/kg) in a medium-intensity reactor. A target of 50% yield at 10.0 g/L of PMS might be possible in a reactor that takes into account of all the mixing parameters together.  iii  TABLE OF CONTENTS  Abstract  ii  List of Tables  vii  List of Appendix Tables  viii  List of Figures  xii  Acknowledgement  xiii  1  INTRODUCTION 1.1  1  BLEACHING TECHNOLOGY  1  1.1.1  Category I.  5  1.1.2  Category II.  5  1.1.3  Category III  6  1.2  D I M E T H Y L DIOXIRANE ( D M D )  11  1.3  PEROXYMONOSULFATE ( P M S )  14  1.3.1  Generation of Acidic Peroxymonosulfate (PMS)  23  1.3.2  Generation of Basic Peroxymonosulfate (PMS)  25  1.4  C H E M I S T R Y A N D K I N E T I C S OF SULFITE O X I D A T I O N  33  1.5  M I X I N G SENSITIVE C H E M I C A L R E A C T I O N S  35  2  OBJECTIVES  39  3  EXPERIMENTAL  41  3.1  MIXERS USED  3.1.1  41  Experiments in a beaker  42  iv  3.1.2  Medium-Intensity Mixer  3.1.3  P  3.1.4  High-Shear Mixer  3.1.5  Dynamic Rotor-Stator Mixer  44  Reactor  m  47 50 53  3.2  CHEMICALS USED  56  3.3  QUANTIFICATION M E T H O D  56  3.4  M A S S B A L A N C E OF S 0 " , S 0 " A N D S 0 " IONS USING F T I R A N D I O D O M E T R I C TITRATION 2  5  R E S U L T S  4  A N D  2  4  2  3  D I S C U S S I O N S  57  5 8  4.1  TEST REPRODUCIBILITY  58  4.2  C O N F I R M A T I O N OF P R E V I O U S T E S T S IN B E A K E R  58  4.3  T E S T S IN M E D I U M - I N T E N S I T Y M I X E R  59  4.3.1  Effect of Feed Time  61  4.3.2  Effect of Energy Dissipation  63  4.3.3  Effect of Sodium Sulfite Concentration  4.3.4  Effect of Volume Ratio of Oxygenated water to Sodium Sulfite  4.3.5  Effect of Temperature  7 7  4.3.6  Other Tests  7 2  70  4.4  P  4.5  HIGH-SHEAR MIXER  74  4.6  ROTOR-STATOR M I X E R  75  H  T  REACTOR  68  73  5  S U M M A R Y  6  F U T U R E  7  N O M E N C L A T U R E  7 8  8  R E F E R E N C E S  8 0  W  A N D  O  R  C O N C L U S I O N S  7 6  K  7 7  V  9  APPENDICES  85  A. Procedures for iodometric titration for PMS  86  B. Calculation of 0.8 ppm C u  88  2+  C. Preliminary estimation of the ratio of moles of O2 to the moles of Na2SC>3 available 88 D. Error analysis  91  E. List of Tables  92  F. FTIR spectra of PMS  106  vi  LIST OF TABLES  Table 1: Total Pulp Capacity (1997)  1  Table 2: Bleaching of Chemical Pulps Over the Past 50 Years (Chirat et al., 1997)  3  Table 3: Classification of Bleaching Chemicals (Lachenal et al., 1993)  5  Table 4: Ultraviolet spectra of PMS anions (Marsh et al, 1989)  17  Table 5: Reduction Potentials of Oxidants (Anderson, 1997)  19  Table 6: Peroxyacid Use in the Bleaching Sequence (Anderson et al., 1995)  20  Table 7: Generation of Peroxymonosulfate Using Sodium Sulfite (Springer etal. 1988) 28 Table 8: Peroxymonosulfate Generation from Sulfite Air Oxidation Catalyzed by Various Transition Metal Ions (Chen et al., 1996) Table 9: Effect of Water Source on Na S0 Generation (Chen et al., 1996) 2  5  30 31  Table 10: Chemical Analysis of Cu -Catalyzed Sulfite Oxidation Liquor (Chen et al., 2+  1996)...; Table 11: Different Mixers Used  32 :  41  Table 12: Dimensions of Medium-Intensity Mixers  47  Table 13: Chemicals used  56  vii  LIST OF APPENDIX TABLES Table E-I: Effect of feed time in the medium intensity mixer A (N=166 rpm; £ ve 0.33 W/kg; Reactor temperature = 4°C; Volume of water = 1.0 L; Volume of NaOH = 25 ml; Volume of CuS0 = 20 ml; 1.5 M Na S0 = 45 ml; Total volume = 1.09 L) 93 =  a  4  2  3  Table E-II: Effect of feed time in the medium intensity mixer (N=560 rpm (save 2.67 W/kg); Reactor temperature = 4°C; Volume of water =1.0 L; Volume of NaOH = 25 ml; Volume of CuS0 = 20 ml; 1.5 M ; Na S0 = 45 ml; Total volume = 1.09 L 93 =  :  4  2  3  Table E-III: Effect of feed time in the medium intensity mixer A (N=2323 rpm (s 33.4 W/kg); Reactor temperature = 4°C; Volume of water = 1.0 L; Volume of NaOH = 25 ml; Volume of CuS0 = 20 ml; 1.5 M Na S0 = 45 ml; Total volume = 1.09 L 94 =  ave  4  2  3  Table E-IV: Effect of feed time in the medium intensity mixer B (Save-720 W/kg; N=23 lOrpm; Reactor temperature = 4°C; Volume of water = 0.5 L; Volume of NaOH = 12 ml; Volume of CuS0 = 10 ml; 0.5 M Na S0 = 20 ml; Total volume = 0.542 L 94 4  2  3  Table E-V: Effect of energy dissipation in the medium intensity mixer A (Reactor temperature = 4°C; Volume of water = 1.0 L; Volume of NaOH = 25 ml; Volume of CuS0 - 20 ml; 0.5 M Na S0 = 45 ml; Total volume = 1.09 L 95 4  2  3  Table E-VI: Effect of energy dissipation in the medium intensity mixer B (Reactor temperature = 4°C;Volume of water = 0.5 L; Volume of NaOH = 12 ml; Volume of CuS0 = 10 ml; 0.5 M Na S0 = 20 ml; Total volume, V = 0.542 L; Feed time, t d = 12.3 s 95 4  T  2  3  fce  Table E-VII: Effect of energy dissipation in the medium intensity mixer B (Reactor temperature = 4°C; Volume of water = 0.5 L; Volume of NaOH = 12 ml; Volume of CuS0 = 10 ml; 1.5 M Na S0 = 20 ml; Total volume, V = 0.542 L; Feed time, t d = 11.25 s 96 4  T  2  3  fce  Table E-VIII: Effect of Na S0 cone, in the medium intensity mixer B (s 900 W/g, N=2300rpm; Reactor temperature = 4°C; Volume of water = 0.5 L; Volume of NaOH = 12 ml; Volume of CuS0 = 13 ml; (0.5,1.5, 2.0 M) Na S0 = 20 ml; Total volume, V = 0.545 L; Feed time, tfced = 12 s 96 ==  2  3  ave  4  2  3  T  Table E-IX: Effect of Na S0 cone, in the medium intensity mixer B (s ve = 610 W/kg, N=2300rpm; Reactor temperature = 4°C; Volume of water = 0.4 L; Volume of NaOH = 10 ml; Volume of CuS0 = 8 ml; (0.5,1.5, 2.0 M) Na S0 = 20 ml; Total volume, V = 0.438 L; Feed time, t d = 11.8 s 96 2  3  a  4  2  3  T  fee  Table E-X: The effect of energy dissipation (medium-intensity mixer A) on PMS yield (Reactor temperature = 4°C; Volume of water = 2.0 L; IN  viii  NaOH = 50 ml; CuS0 (0.1 g/1) = 40 ml; 2 M Na S0 = 45 ml; pH 12.6; Total volume, V = 2.135 L) 97 4  2  3  T  (  Table E-XI: The effect of energy dissipation (medium-intensity mixer A) on the yield of PMS (Reactor temperature = 4°C; Volume of water = 2.0 L; IN NaOH = 50ml; CuS0 (0.1 g/L) = 40 ml; 0.5 M Na S0 = 45 ml; pH = 12.65; Total volume, V = 2.135 L) 97 4  2  3  T  Table E-XII: The effect of feed time (medium-intensity mixer B) on PMS generation (Volume of water = 0.5 L; IN NaOH = 12 ml; CuS0 (0.1 g/L) = 10 ml; 0.5 M Na S0 = 20 ml; Total volume, V = 0.542 L; pH = 12.55; Reaction temperature = 4°C; s e = 905 W/kg, N=2400rpm) 97 4  2  3  T  av  Table E-XIII: The effect of volume ratio (medium-intensity mixer B) on PMS yield (Volume of water = 0.5 L; IN NaOH = 12 ml; CuS0 (0.1 g/L) = 13 ml; pH = 12.53; N=2400rpm) 98 4  Table E-XIV: The effects of initial volume (medium-intensity mixer B) on PMS yield (20mL 2.0 M Na S0 , pH=12.5, N=2300rpm) 98 2  3  Table E-XV: The effects of initial volume (medium-intensity mixer B) on PMS yield (20mL 1.5 M Na S0 , pH=12.5, N=2300rpm) 98 2  3  Table E-XVI: The effects of initial volume (medium-intensity mixer B) on PMS yield (20mL 0.5 M Na S0 , pH=12.5, N=2300rpm) 99 2  3  Table E-XVII: The effect of changing volume and feed time in mediumintensity mixer B (Volume of water = 0.45 L; IN NaOH = 9.8 ml; CuS0 (0.1 g/L) = 12 ml; 2.0 M Na S0 = 45 or 50 ml; Total volume, V = 0.5168 or 0.5218 L; pH = 12.50; Reaction temperature = 4°C; e = 770 W/kg; N=2323rpm) 99 4  2  3  T  ave  Table E-XVIII: The effect of temperature on the yield of PMS in mediumintensity mixer B 100 Table E-XIX: Oxygen Solubility in Water at 23°C calculated from "Handbook for Engineers" 100 Table E-XX: Comparison of PMS yields  100  Table E-XXI: P T Mixer (Reaction temperature = 23°C; Volume of water = 1.5 L; IN NaOH = 36 ml; CuS0 (0.1 g/L) = 40 ml; 1.5 M Na S0 = 60 ml; Total volume, V = 1.636 L; Feed time =11.5 sec; e = 240 W/kg; N=2000rpm) 101 H  4  2  T  3  ave  Table E-XXII: The effect of feed time in P T Mixer (Reaction temperature = 23°C; Volume of water =1.5 L; IN NaOH = 36 ml; CuS0 (0.1 g/L) = 40 ml; 1.5 M Na S0 = 60 ml; Total volume, V = 1.636 L; Pressure = 100 psi; £ = 240 W/kg, N=2000rpm) 101 H  4  2  3  T  ave  Table E-XXIII: PHT Mixer (Reaction temperature = 23°C; Volume of water = 1.5 L; IN NaOH = 36 ml; CuS0 (0.1 g/L) = 40 ml; 1.5 M Na S0 = 60 4  ix  2  3  ml; Total volume, VT = 1.636 L; Pressure = 20 psi; Feed time = 20 sec) 101 Table E-XXIV: The effect of feed time in high-intensity mixer (Reaction temperature = 23°C; Volume of water = 2.5 L; IN NaOH = 65 ml; CuS0 [0.1 g/L] = 80 ml; 2.0 M Na S0 = 50 ml; Total volume, V = 2.695 L; s = 263 W/kg, N=2000rpm) 102 4  2  3  T  ave  Table E-XXV: Effect of energy dissipation in high-intensity mixer on PMS yields (Reaction temperature = 23°C; Volume of water = 2.5 L; IN NaOH = 65 ml; CuS0 [0.1 g/L] = 80 ml; 2.0 M Na S0 = 50 ml; Total volume, V = 2.695 L; s = 263 W/kg, N=2000rpm, tf=22s) 102 4  T  2  3  ave  Table E-XXVI: The effect of feed time in High intensity mixer (Reaction temperature = 23°C; Volume of water = 2.5 L; IN NaOH = 65 ml; CuS0 (0.1 g/L) = 80 ml; 2.0 M Na S0 = 50 ml; Total volume, V = 2.695 L; s = 2080 W/kg, N=4000rpm) 102 4  2  3  T  ave  Table E-XXVII: Model predictions at high-energy dissipation rate (V /V =44.5; T=4°C; pH=12.5; V=545mL;k]/k =44; C =2.0M) A  B  2  B0  103  Table E-XXVIII: Effect of feed time in PHT reactor (Volume of water = 1.8 L; IN NaOH = 50 ml; CuS0 (0.1 g/L) = 50 ml; 2.0 M Na S0 = 60 ml; Total volume, V = 1.96 L; e = 240 W/kg, N=2000rpm) 103 4  2  T  3  ave  Table E-XXIX: Effect of pressure in PHT reactor on PMS yield (Volume of water = 1.8 L; IN NaOH = 50 ml; CuS0 (0.1 g/L) = 50 ml; 2.0M Na S0 = 60 ml; Total volume, V = 1.96 L; s = 240 W/kg; N=2000rpm; V / V = 31.67) 103 4  2  3  T  A  ave  B  Table E-XXX: Effect of pressure in PHT reactor on PMS yield (Volume of water = 1.8 L; IN NaOH = 50 ml; CuS0 (0.1 g/L) = 50 ml; 2.0 M Na S0 = 60 ml; Total volume, V = 1.96 L; s = 240 W/kg; N=2000rpm; V / V = 31.67) 104 4  2  3  T  A  ave  B  Table E-XXXI: Effect of pressure in PHT reactor on PMS yield (Volume of water = 1.8 L; IN NaOH = 50 ml; CuS0 (0.1 g/L) = 50 ml; 0.5 M Na S0 = 60 ml; Total volume, V = 1.96 L; £ = 240 W/kg; N=2000rpm; V / V = 31.67) 104 4  2  3  T  A  ave  B  Table E-XXXII: Oxygen solubility measured by oxygen meter in P T Reactor 104 H  Table E-XXXIII: Mass balance of S0 ", S0 ", and S0 " ions using FTIR and Iodometric Titration (P T mixer, N=2000rpm, pH = 12.5, T = 15°C) 105 2  3  2  5  2  4  H  Table E-XXXIV: Dynamic rotor-stator mixer (Reaction temperature = 23°C; pH =12.5; Mixer setting # 7) 105 Table E-XXXV: Effect of feed time in dynamic mixer (Reaction temperature = 23°C; Volume of water = 0.5 L; IN NaOH = 12 ml; CuS0 (0.1 g/L) = 4  x  14 ml; 2.0 M Na S0 = 20 ml; Total volume, V = 0.546 L; s e= 3380 W/kg; Mixer setting #7) 105 2  3  T  av  Table E-XXXVI: Reproducibility of yield and concentration of PMS (Medium-Intensity reactor config. A; Conditions: Reactor temperature = 4°C; Volume of water = 2.0 L; IN NaOH = 50ml; CuS0 (0.1 g/L) = 40 ml; 0.5 M Na S0 = 45 ml; pH = 12.65; Total volume, V = 2.135 L; tf=55-60s;8ave=219W/kg; N=2050rpm) 106. 4  2  3  T  xi  LIST O F  F I G U R E S  Figure 1.1: Formation of Dimethyldioxirane  12  Figure 1.2: Catalytic Nature of Dimethyldioxirane  12  Figure 2.1: PMS Application for Different Initial Pulp Strengths  40  Figure 3.1: Experimental Setup for Graduated Cylinder or Beaker, showing use of stirrers .43 Figure 3.2: Photograph of Medium-Intensity Mixer Showing operation setup 45 Figure 3.3: Schematic of Medium-Intensity Mixer Showing Feed Point Location  46  Figure 3.4: Photograph of PHT Mixer Showing controls and rotor  48  Figure 3.5: Photograph of High-Shear mixer showing rotor-stator configuration used Figure 3.6: Schematic of feed point location and dimensions of the highintensity mixer  52  Figure 3.7: Photograph of Dynamic Rotor-Stator Mixer  54  Figure 3.8: Schematic of Feed Point Location of Rotor-Stator Mixer  55  51  Figure 4.1: Effect of feed time on PMS yield at different mixing intensities..62 Figure 4.2: The Effect of Energy Dissipation on PMS Yield and Concentration  63  Figure 4.3: Effect of Energy Dissipation on PMS Yield  64  Figure 4.4: Effect of Energy Dissipation on PMS Yield (T=4°C, pH=12.5, V =0.542L, 0.5M Na S0 ) 65 T  2  3  Figure 4.5: Comparison of E-model prediction and experimental values for the effect of energy dissipation on the Yield of PMS (mediumintensity mixer of config.B; V /V =44; T=4°C; pH=12.5; V =545mL; tf=41-60s; Cu =0.8ppm) 67 A  B  2+  T  Figure 4.6: Effect of Sodium Sulfite Concentration on PMS Yield  ..69  Figure 4.7: Effect of Feed Time on PMS Yield (Medium-Intensity Mixer config.B; 0.5M Na SO =20mL; T=4°C; pH=12.55; V =542mL; Save=905W/kg; Cu =0.8ppm) 69 2  3  T  2+  Figure 4.8: Effect of Sodium Sulfite Concentration on PMS Yield  71  Figure 4.9: Effect of Temperature on PMS Generation  72  xii  ACKNOWLEDGMENT I would like to take the opportunity to express my gratitude to those who have assisted me to complete this work. First and foremost, I would like to thank the Almighty, Omnipotent, Omniscient, Omnipresent Allah Ta'Ala Whose mercy we all need at every moment. I proceed towards Him seeking an end to my difficulties or to fulfil my needs at every stage because He says: "If you walk towards me, I run towards you". I like to thank my supervisor, Dr Chad Bennington for giving me the opportunity to perform this research. His encouragement and stimulating guidance, as well as financial support, were invaluable during the past couple of years. This work wouldn't have been possible without countless help of Joseph Mmbaga, a real friend in my laboratory. His continuous support has been a source of strength. Thanks God to bring me closer to him. I hope and believe that God will reward him for his noble deeds. I would also like to thank Dr. Vilas Rewatkar for extending his help especially in the final phases of this work. Also, the suggestions received from  Dr. Jianxin Chen and  Professor Colin Oloman are greatly appreciated. The friendship of the former and present members of the Mixing Group is acknowledged. Special thanks to Joseph, Vilas and Min-Wong for the lively discussions about world politics, economics and other things. They have been enjoyable to work with and be around. The help of the technical staff of the Pulp and Paper Center workshop, Peter, Tim, Ken and Rita, was essential to the successful completion of the present study. Last, but not the least, to my caring parents, wife and daughter, not only do I dedicate this dissertation to, but also my life and career, for without them, it could not mean near enough to me. This accomplishment of mine is also theirs. I love all of you more than words can say.  Xlll  1 INTRODUCTION 1.1  BLEACHING  TECHNOLOGY  Bleaching plays an important role in the economics of the pulp and paper industry. Table 1 represents a total pulp capacity by Canada and world. A large fraction of total pulp requires bleaching. According to the FAO (Food and Agricultural Organization), paper production will grow from 240 million tons per year in 1991 to 444 million tons per year in 2010. Consequently this will increase the demand for bleaching agents. This increased demand in paper is also very good news for the forest industry.  Table 1: Total Pulp Capacity (1997)  Millions of Tonnes  CANADA  WORLD  Chemical Pulp  9.4  37.6  Bleached Softwood kraft  7.2  18.5  Bleached Hardwood Kraft  1,8  15.7  Sulfite  0.3  1.7  High Yield  2.1  3.5  Newsprint  9.7  38.7  Market Pulp  11.4  41.1  Bleaching is a chemical process applied to cellulosic materials to increase their brightness. The pulp properties depend on its principal components- cellulose, hemicellulose, and lignin. Cellulose is a linear polymer of anhydroglucose and constitutes 40-50% of most wood species. It is essentially colorless and does not require bleaching. But cellulose degradation and lower pulp yield can occur during bleaching of pulp fibers 1  in order to decolorize other components. Hemicelluloses are non-uniformly linked carbohydrate polymers and usually comprise 20-30% of the wood. They are also colorless and relatively stable and require no bleaching. Hemicelluloses are extremely important for bonding in paper sheets. Lignin is structurally different from cellulose and hemicelluloses and comprises 26-32% of softwoods and 20-28% of hardwoods. The absorbance of visible light by wood pulp fibers is caused mainly by the presence of lignin. Pulping process removes most of the lignin present in wood leaving approximately 3-5% lignin in a pulp entering the bleach plant. Lignin is a randomly linked polymer of phenyl propane building blocks. Lignin derivatives form most of the color-producing chromophores, which darken the pulp. The aim of bleaching is to remove the lignin derivative chromophores after cooking to reach a target brightness level. Bleaching of pulp is done to achieve a number of objectives. The most important of these is to increase the brightness of the pulp so that it can be used in paper products such as printing grades and tissue papers. For chemical pulps an important benefit is the reduction of fiber bundles and shives as well as the removal of bark fragments. This improves the cleanliness of the pulp. Bleaching also eliminates the problem of yellowing of paper in light, as it removes the residual lignin in the unbleached pulp. Resin and other extractives present in unbleached chemical pulps are also removed during bleaching, and this improves the water absorbency, which is an important property for tissue paper grades. The papermaking properties of chemical pulps are changed after bleaching. Removal of the residual lignin in the pulp increases fiber flexibility and strength. On the other hand, a lowered hemicellulose content results in a lower swelling potential of the fibers and a reduced bonding ability of the fiber surfaces. In the manufacture of bleached chemical pulps, the residual lignin is oxidatively fragmented to a point where dissolution takes place. Thus one should be aware of the bleaching chemistry. Whatever bleaching agent is used, high brightness of the pulp must be achieved without compromising the strength of the final product. Cellulose  2  degradation during bleaching can significantly reduce the strength o f the final product. Bleaching reactions can also lead to significant dissolution of substance from the pulp, decreasing the yield o f final product and affecting production cost. The consideration o f bleaching chemicals in context o f health and safety is also very important.  Some  bleaching chemicals are highly reactive and thus proper equipment design, operation management, materials of construction are required. Over the past 10 years, the volume and range o f chemicals used to bleach pulp has changed significantly. Yesterday's highest performing chemicals have been dropped, establishing progress for more environmentally friendly alternatives and a completely different outlook on the bleaching chemical perspective. This has caused a radical change in chemical pulp bleaching sequences, primarily due to environmental pressures (Table 2). Tabic 2: Bleaching of Chemical Pulps Over the Past 50 Years (Chirat et al., 1997)  Year  1 9 5 0 - 1975  1975 - 1990  >.l 990  C  C  0  D  P  Stages  E  E  (Eo)  (DC)  (PO)  used in  H  (DC)  (E )  C  Q  the  D  (EP)  0  z  bleaching  (E )  PA  sequence  (E p)  Px  0 P  0  0  PXA  X  One outcome o f this has been a substantial increase in the variety o f bleaching sequences, which are employed in today's pulp industry. Market opportunities and local regulations drive the implementation of such bleaching sequences. In addition, modern  3  bleaching chemicals made it more complex to choose a bleaching sequence. A bleaching sequence for a chemical pulp consists of a number of stages, each of which has a specific function. The early part of a sequence is designed to remove the major portion of the residual lignin in the pulp. Unless this is done high brightness cannot be reached. The later stages in the sequence are the so-called brightening stages, in which chromophores in the pulp are eliminated and the brightness increased to a high level. It is not only important to consider the general trends in bleaching, but also to understand that all the bleaching  processes  are  not  equal.  Unbleached  pulp  appears  brown  because  chromophores absorb blue light. Bleached pulp is whiter because it reflects more blue light than it did before bleaching. Bleaching success is thus measured by the improvement in reflectance that is indicated by the reduced concentration of chromophores in the pulp sheet. But even in aiming at removing the residual lignin, there is the question of whether this target level should be very high (>90% ISO) or only moderately high (-85% ISO). A good knowledge of the chemical reactions of each bleaching agent on pulp components is required to establish the right combination of bleaching stages. It is important to fully understand how each bleaching agent affects lignin (which needs to be removed), carbohydrates (the degradation of which has to be prevented) and inorganic material (metal ions which can interfere or contribute to the bleaching process). Most bleaching chemicals are oxidizing agents that generate acidic groups in the residual lignin (Nelson, 1998). If a bleaching stage is done under acidic conditions, it is followed by an alkaline extraction to remove the water-insoluble acidic lignin products. The fundamental chemistry of various common bleaching agents has been simply categorized by Lachenal and Nguyen-Thi (1993). It has been suggested that the main bleaching chemicals can be divided into three categories based on their reactivity. This concept is summarized in Table 3.  4  1.1.1  Category I  Chlorine (Cb) is capable of reacting with' all types of aromatic structures in residual lignin. This makes chlorine the most efficient bleaching chemical available so far. Each chlorine-containing chemical has an equivalent oxygen-based chemical. Ozone (O3) and chlorine are placed in the same category because they react with aromatic rings of both etherified and non-etherified phenolic structures in lignin as well as with double bonds. Ozone is less selective than chlorine, as it also attacks the carbohydrates in pulp. These chemicals are well studied for use in the first part of a sequence, as they are very efficient at degrading lignin. Table 3: Classification of Bleaching Chemicals (Lachenal et al., 1993) Category  Reaction Sites in Lignin  I  II  III  Aromatic rings  Free phenolic groups  Carbonyl groups  Double bonds  Double bonds  Cl  C10  Bleaching Chemicals  0  1.1.2  2  0  3  2  2  NaOCl H 0 2  2  Category II  Chlorine dioxide  (CIO2)  and oxygen ( O 2 ) are grouped together because they both  react primarily with free phenolic groups (phenolic groups not attached with other functional groups) in lignin, leaving the other phenolic groups almost intact, unless the ambient conditions are made more extreme (Chirat et al., 1997). They are not as effective as chlorine and ozone in degrading lignin. Chlorine dioxide is used extensively in the early stages of bleaching sequences as a replacement for chlorine even though it is 5  slightly less effective. The classification is somewhat simplistic in this respect and, moreover, does not take into account that chlorine dioxide is reduced to hypochlorous acid and that oxygen is reduced to hydrogen peroxide during the bleaching reactions. Chlorine dioxide is also used as a brightening agent in the later part of a sequence. In oxygen bleaching, the substrate is activated by providing alkaline conditions to ionize free phenolic hydroxyl groups in the residual lignin. The resulting anionic sites are electron-rich and therefore vulnerable to attack by oxygen. A n alternate pathway for initiation of the radical chain reaction is abstraction of a hydrogen atom from an unionized phenolic group or other functional group to give the corresponding organic radical (McDonough, 1996). Thus oxygen also reacts exclusively with free phenolic groups, which makes it similar to chlorine dioxide as far as delignification is concerned. 1.1.3  Category III Sodium hypochlorite (NaOCl) and hydrogen peroxide (H2O7) react almost  exclusively with carbonyl groups under normal conditions. Hydrogen peroxide itself is not very efficient when used at moderate temperatures (<70°C). It can react only on carbonyl-containing structures (quinone-type compounds) and its use is limited to color removal. The formation of radicals such as HO* and O2* seems to be needed to achieve a substantial delignification and raising the temperature can do this. The role of radicals as intermediates during most bleaching stages in delignification is still a matter of debate. One drawback of the occurrence of radical species is that they also react readily with cellulose and lead to degradation of the fiber. The selection of a suitable bleaching sequence based on this classification has been discussed by Lachenal and Nguyen-Thi (1993). A n efficient bleaching sequence should contain at least one chemical from each category. A chlorine-based sequence such as C E H E D and a TCF sequence as OZEP are a couple of examples of this principle. Metal ions in pulp must be controlled. Some metal ions, such as iron and copper, favor the formation of radical species during hydrogen peroxide, oxygen or ozone 6  treatments, leading to subsequent cellulose degradation. They can also be a source of chemical loss. For example, manganese, iron and copper cations catalytically decompose hydrogen peroxide. Magnesium on the other hand, is beneficial to oxygen and peroxide stages. Metal ions can be removed and/or controlled by acid or chelation stages. Chlorine was the most efficient and universally used bleaching chemical. But its use is being eliminated throughout the world as a result of environmental pressure. The main effect of eliminating chlorine has been a large decrease in the formation of chlorinated organics and the reduction or elimination of chlorinated dioxins in the bleaching effluents and in the bleached papers. These have both fallen to undetectable levels. Most mills have opted for the partial or total replacement of chlorine by chlorine dioxide. The net result is an increase in chemical costs of 20-30%. Effluents from chlorine dioxide treatments are not usually recycled to the kraft recovery system and these effluents are generally not harmful, especially after they have been subjected to secondary treatment. However, chlorine dioxide delignification normally generates a small amount of adsorbable organic halogens (AOX). This has resulted in an increased demand for pulps bleached without chlorine dioxide and all other chlorine-containing compounds (Cockram, 1991). Some mills have taken the opportunity to install oxygen delignification ahead of the bleaching line, which makes it possible to maintain the chemical cost at about the same level, but this also represents a $30 (Canadian) million investment in most cases. These drawbacks are the direct result of poorer delignification power of chlorine dioxide compared to chlorine. Selectivity may be defined as the amount of lignin removed from pulp (measured as the reduction in kappa number) for a given amount of carbohydrate degradation (measured as the decrease in viscosity). Oxygen is not a selective bleaching chemical as compared with chlorine or chlorine dioxide (or a combination of both). The latter chemicals can remove about 90% of the lignin in a kraft pulp, measured after alkaline extraction. Usually, oxygen can remove only about 50% of the lignin before degradation of the carbohydrates becomes extensive. Today, oxygen is the cheapest bleaching chemical, but capital cost is still the major drawback of oxygen delignification. 7  The problem is not only the level of investment involved, but also the fact that the money is spent to achieve a rather poor result («40-50% delignification) because of its limited selectivity (Munro, 1987). Despite this, efforts have been made to use oxygen much more extensively. More and more mills are installing a second oxygen stage - with or without intermediate washing. The last four to five years have also seen the appearance of hydrogen peroxide stages under oxygen pressure (PO) and at high temperature (90-110°C) used either at the beginning of a bleaching sequence or at the end. If (PO) is used as a modified O stage, then delignification increases by about 20% (Chirat et al., 1997). One advantage of using (PO) instead of a P stage is that the process is more efficient, requiring reduced retention times and chemicals. The use of hydrogen peroxide at high temperature and oxygen pressure makes it very important to have proper control of the metal ions with a chelating stage (Q). Hydrogen peroxide is not an effective delignifying agent either under alkaline or acidic conditions (Fossum et al., 1980). Ozone is known to be a very effective delignifying agent (Gratzl, 1991, Liebergott et al., 1984; 1992), but it's less selective towards lignin reactions due to aggressive radical reactions, which causes undesirable cellulose degradation and results in lower pulp viscosity and strength in totally chlorine-free (TCF) bleaching sequences. TCF bleaching uses chemicals that do not contain chlorine, such as oxygen, ozone, hydrogen peroxide, and peroxyacids. Elemental chlorine-free (ECF) bleaching involves replacement of all the molecular chlorine in a bleaching sequence with chlorine dioxide. The term ECF bleaching is usually interpreted to mean bleaching with chlorine dioxide as the only chlorine-containing bleaching chemical. In ECF and TCF bleaching sequences, the use of hydrogen peroxide can either . predominantly delignify or brighten depending on its position within the sequence. Alkaline hydrogen peroxide under normal bleaching conditions does not provide effective delignification; it is rather a good brightening agent. The reactivity of hydrogen peroxide and its bleaching effectiveness have increased considerably in recent years by chemical 8  conversions to more reactive peracids (Troughton et al., 1994), by using peroxide bleaching activators (Hammer et al., 1991; Chen, 1998), or by using high bleaching temperatures (>100°C) under pressure (Roy et al., 1995). In any case, the effectiveness of chlorine dioxide as a bleaching agent still surpasses that of the above bleaching agents. Much controversy has recently been stirred up over the TCF- or ECF-pulp bleaching question. This is typical when the pulp industry transfers to major new technologies that create technology gaps between producers. A similar debate occurred in the 1950s when chlorine dioxide was introduced. TCF pulp does not require a chlorine dioxide generator and mills with a closed-water system do not need a chloride-removal system. Croon (1997) has shown that investment and operating costs will be very similar for TCF and ECF pulp mills and lower than for recently built mills. The choice of technology for new mills and major rebuilds, then, will be guided by economic and product quality considerations. The superiority of TCF over ECF bleaching in terms of environmental impact is questionable. But there are some significant disadvantages in TCF bleaching, which explains the lack of interest still being expressed by most of the pulp producers. The most important of these is that bleaching a kraft pulp to high brightness (90% ISO) is not possible without sacrificing some strength properties. The problem of cellulose degradation during TCF bleaching has been extensively studied. Taking an OPZ(EO)P sequence, for example, it was shown that each stage might contribute to some cellulose depolymerization. One critical factor is the amount of ozone introduced in the sequence. For charges higher than 5-6 kg/tonne, the cellulose may be slightly depolymerized and oxidized. This last effect makes the pulp sensitive to any alkaline environment such as (EO)P which leads to further chain cleavage (Chirat et al., 1997). Consequently, despite the fact that such a sequence is close to optimum efficiency in terms of delignification and bleaching power, it is penalized by the occurrence of several degradation mechanisms taking place on cellulose.  9  One possible solution to the problem of cellulose degradation during TCF bleaching to 90% ISO, is to limit the charge of ozone and to introduce some nondegrading bleaching agents in the sequence. The only reagents that demonstrate this property so far, are the peroxyacids (peroxyacetic, peroxymonosulfuric acids). Peroxyacids are proving particularly interesting as it is very lignin specific and likely to be highly popular if cost and safety issues can be overcome (Beal, 1997). But more research is needed to reduce the chemical cost to acceptable levels when peroxyacids are used. Pulp and paper producers realize that environmental regulations that are in force or that will be introduced in the coming years will require less polluting processes as well as the placement of tighter controls on waste management. One way of reducing the environmental impact of a bleached kraft mill is to reduce bleaching chemical requirements. The use of chlorine to bleach chemical pulp to high brightness results in undesirable organochlorine compounds (dioxins and furans). Using non-chlorinebleaching agents can reduce the discharge of these compounds in the mill effluent or by reducing the amount of chlorine required to reach the desired brightness. As a result the implementation of new bleaching technologies which do not use chlorine base chemicals continue to be developed in the pulp and paper industry worldwide to meet stricter environmental requirements and a shifting consumer expectation. Development work on oxygen based bleaching agents that use oxygen, ozone, peroxide and peroxyacid has been proceeding in laboratories, pilot facilities and commercial installations for many years. Chlorine-free bleaching systems also potentially allow recycling of bleaching effluents into the mill as the corrosive chloride ion is removed. Spent effluent from these bleaching stages may be recycled to the black liquor and combusted in the recovery furnace rather than being released into the environment. These bleaching agents may also, in conventional bleaching sequences, be used to decrease or eliminate chlorine, chlorine dioxide and the level of AOX in the effluent.  10  The problem of using oxygen based chemicals is to find compounds that do not adversely affect pulp strength. Dimethyldioxirane (DMD), peroxymonosulfuric acid and its salts are such promising bleaching agents.  1.2 Dimethyl dioxirane (DMD) DMD is highly electrophilic and known to easily oxidize carbon-carbon double bonds in aliphatic and aromatic structures, which are abundant in residual lignin. DMD is an organic peroxide that can be made by oxidizing acetone with the potassium salt of Caro's acid (peroxymonosulfuric acid). The reaction inserts an extra oxygen atom into the carbonyl group of acetone, forming a three-membered ring structure that contain two oxygen atoms and one carbon atom with two methyl substituents as shown below in Figure l . l . The strained ring structure and peroxidic character make DMD a powerful and reactive oxidizing agent. Dioxiranes are a class of powerful electrophilic oxidants, efficient in oxygen transfer and selective in reactivity. They are also catalytic in that the ketone is not consumed, as illustrated in Figure 1.2. Recently, it has been shown that dimethyldioxirane (DMD) can be used as an effective and selective TCF bleaching agent for both softwood and hardwood kraft pulps (Lee et al., 1993, 1994). Effective bleaching of kraft pulp with either isolated or in-situ generated DMD has also been confirmed (Ragauskas, 1993; McDonough et al., 1994). The reactions of DMD with lignin model compounds have been studied and have shown that DMD may oxidize the aromatic rings of both etherified and non-etherified lignin model compounds (Argyropoulos et al., 1994). DMD is therefore superior to other bleaching agents such as  CIO2, O2,  and  H2O2,  phenolic hydroxyl groups.  11  which act only on structures having free  o R  HOOSO3-  R  acetone  HO  peroxymonosulfuric acid  HO-  O  ^SO,  H 0 2  O—O  X  D  R  +  S0 24  KO^O  "SO3-  R  DMD  Figure 1.1: Formation of Dimethyldioxirane  Figure 1.2: Catalytic Nature in Dimethyldioxirane Reaction  12  The potential of DMD for pulp-bleaching applications was recognized by Lee and co-workers (Lee et al., 1993). DMD treatment followed by an extraction stage also delignified a softwood kraft pulp to 90% ISO brightness with a severe viscosity loss (Lee et al., 1994). They showed that the kappa number of a hardwood kraft pulp was reduced from 16.4 to 3.4 (79% delignification) by 2.5% isolated DMD (generated by adding potassium peroxymonosulfate to an aqueous acetone solution of sodium bicarbonate and vacuum distilling the DMD in acetone), and that of a softwood kraft pulp was reduced from 31.5 to 5.4 (83% delignification) by 12.5% in situ generated DMD followed by 3.3% NaOH. Ragauskas showed that the kappa number of a softwood kraft pulp was reduced from 39.5 to 12.5 (68% delignification) by 5% isolated DMD. McDonough et al., 1994 also verified the effectiveness of DMD as a delignifying and bleaching agent by evaluating DMD in a short sequence with oxygen and peroxide stages. The sequences evaluated were OAQP [Oxygen (O), DMD (A), EDTA (Q), hydrogen peroxide (P)], in which the A stage was primarily a delignifying stage and OQPA, in which DMD played a brightening role. It was concluded that DMD was both effective and selective as a delignifying agent but was neither effective nor selective as a brightening agent. In a TCF bleaching sequence, OAQP holds considerable promise because it achieved brightness of mid 80's with modest viscosity loss that did not result in decreased pulp strength. Such selectivity obtained by DMD treatment and alkaline extraction was equivalent to that of conventional chlorination and extraction when implemented on a pilot scale (Lee et al., 1994a; 1994b) bleaching trial at the PAPRICAN pilot bleach plant and was considered to be technically and commercially viable. Peroxymonosulfuric acid, an oxygen donor for DMD generation was generated on-site from 98% H 2 S O 4 and 70% H2O2  using an industry-proven technology from DuPont. H2SO4  + H 0 ^ 2  2  H2SO5  + H 0 2  (1)  Since the reaction is exothermic, the mixture was cooled to maintain the temperature within the range of 15-45°C. Temperatures higher than 45°C were avoided to minimize decomposition of peroxymonosulfuric (also called Caro's acid). The formation 13  of  H2SO5  70%  H2O2  is favoured by minimizing the amount of water in the reaction mass and thus is recommended for this process. Lee and co-workers (1994) proposed that  acetone could be recovered by a distillation system similar to stripper systems used by Kraft mills to treat foul condensates. The inorganic chemicals, Na2CC>3 and  H2SO4,  could  be regenerated by using a green liquor carbonation process and a wet sulfuric acid production method. Despite the promise shown by DMD, the operation of commercial bleach plants with acetone and its required recovery pose significant problems for the implementation in industry. The capital investment and the complexity of this reaction are major barriers in its implication at industrial scale. The practical application of this technology will depend on the extraction and recovery of acetone from the bleaching effluent. Therefore, commercialization of DMD to produce a high brightness pulp with consistently good strength is unlikely and has yet to be proven. Search for other environmentally friendly bleaching agents still continued. Peroxymonosulfuric acid and its salts show promise as such bleaching agents.  1.3 Peroxymonosulfate (PMS) Peroxyacids, also called peracids, are a class of compounds that have been considered for use in pulp bleaching and evaluated to replace or to augment the more traditional bleaching chemicals (Anderson, 1997). A peroxyacid is an acid that contains a perhydroxyl group (-OOH) in place of hydroxyl group of its parent acid. So sulfuric acid is the parent acid of peroxymonosulfuric acid, one peroxyacid of greatest recent interest. Peroxymonosulfuric acid is generated by the action of sulfuric acid on hydrogen peroxide as shown in equation 2. The hydroxyl group of sulfuric acid and the proton of hydrogen peroxide that form water in the reaction are typed in bold face.  14  0 II 0=S—OH 1 OH  + HOOH  =  oII  0=S—O-O-H I OH  + H 0 7 2  Caro's acid, or peroxymonosulfuric acid (2)  In solution, percarboxylic acids are more volatile than carboxylic acids. Percarboxylic acids are also weaker acids than their parent acids. Generally, the pKa of a percarboxylic acid is 3-4 units higher than the pKa of its corresponding acid. So Caro's acid (H2SO5) is a dibasic acid with protons attached to the oxygen of the hydroxyl group and perhydroxyl group. Its anion (HSO5') is a considerably weaker acid than the anion of sulfuric acid (HSO4"). Peroxymonosulfuric acid is a crystalline solid (melting point 45°C) and it is stable in air, if pure. It is soluble in alcohol, ether, and acetic acid, but the solutions are liable to explode. It explodes at once if mixed with aniline, benzene, or phenol. It behaves in aqueous solution as a monobasic acid. It is slowly hydrolyzed by water to H S 0 and H 0 (Caro, 1898). 2  4  2  2  The existence of peroxyacid was recognized nearly a century ago but it was not until 1898 that Caro demonstrated its existence (Caro, 1898). Caro knew that salts of peroxydisulfuric acid (H S 08) converted aniline to an insoluble dye (aniline black). He 2  2  used a solution of ammonium persulfate in concentrated sulfuric acid to treat aniline and obtained nitrosobenzene with no aniline black. Three years later, Baeyer and Villiger gave a conclusive evidence that Caro's acid was peroxymonosulfuric acid (Baeyer et al., 1901). However, the pure, anhydrous peroxymonosulfuric acid was not prepared until 1910 (d'Ans et al., 1910). On the basis of Raman and infrared spectra and of single-crystal xray diffraction, peroxymonosulfuric acid is given the configuration as shown in equation 3.  15  o II  HO—S-OOH II  (3) The oxygen atom, which is presumed, transferred in redox reactions and which is supposed to be released in many decomposition reactions is typed in bold face. Sulfuroxygen cleavage does occur in acid hydrolysis, then hydrogen peroxide is formed. The proton on the sulfonate group of H 2 S O 5 is similar in acidity to the first proton of sulfuric acid and completely ionizes in water solution. The peroxide proton of H 2 S O 5 has a pKa value near 9.4. The ultraviolet spectra of peroxides can give information on structure, ionization constants and useful wavelengths for quantification, but the spectra are not helpful in analysis of mixtures as the extinction coefficients are little dependent on the groups attached to the two peroxide oxygens. Extinction coefficients of aqueous solutions of peroxymonosulfate are presented in Table 4. As with almost all ROOR peroxides (HOOH,  S2O8  di-t-butyl peroxide, etc), and alkyl peroxides, the absorption starts near  300 nm with a low extinction coefficient and increases in intensity without maximum at least as far as 230 nm. Peracetic  acid  and  peroxymonosulfates,  together  with  the  parent  peroxymonosulfuric acid (also called Caro's acid) are powerful oxidizing agents. Table 5 compares the reduction potential of Caro's acid and peracetic acid with other oxidizing agents. Since the removal of lignin in chemical bleaching involves oxidation of lignin to form alkaline soluble fragments, peroxyacids are of interest as potential non-chlorine oxidizing agents. This combination of properties also enables peroxyacids to be useful in brightening oxidation reactions under more moderate conditions than those required with the use of hydrogen peroxide brightening.  16  Table 4: Ultraviolet spectra of PMS anions (Marsh et al., 1989) e^'cm"  1  X, nm  HS0 " (a)  HS0 " (b)  S0 - (a)  230 240 250 260 270 280 290 300  55.5 31.8 19.0 10.2 5.5 3.3 1.7 1.1  50 27 18 8.3 5.7 3.3 2.3 1.6  286 204 158 112 77 49 30 18  5  5  2  5  (a) Battaglia, C. J., Ph. D. thesis at Brown University (1962) (b) Rieche, A. "Alkyl Peroxide and Ozonide", Dresden and Leipaig (1931)  The use.of peroxyacids for the bleaching of pulps was described in the 1940s (Anderson, 1997). In order to develop environmentally friendly TCF bleaching in the 1990s, peroxyacids were reconsidered as bleaching agents (Kronis, 1995). Peroxyacids have been studied as a delignification agent, as an activation agent, and as a bleaching agent in late oxidizing stages (Liebergott, 1992). Table 6 shows where peroxyacids may be used in different stages of the pulping and multi-stage bleaching process for chemical pulps. Liebergott et al., (1992) has reviewed their applications and comparative performance. Peroxymonosulfate together with its parent peroxymonosulfuric acid (H2SO5) has been investigated as both a delignifying agent (to separate lignin components from the cellulose and carbohydrate materials) and a bleaching agent (to purify and whiten pulp). As a bleaching agent, it brightens without adversely affecting pulp properties. Peroxymonosulfate has been reported as a more effective delignifying and bleaching agent than hydrogen peroxide for kraft pulps (Cael, 1983; Anderson, 1997).  17  Peroxymonosullfuric acid was used in the bleaching sequence PxQP (where Px=Peroxymonosulfuric acid, Q=chelation, P=hydrogen peroxide) to delignify a softwood kraft pulp prior to bleaching with hydrogen peroxide (Basta et al., 1994). It was slightly less effective than peracetic acid in that the brightness of the bleached pulp was lower, the lignin content higher, and the viscosity of the pulp lower. On a mill scale, peroxymonosulfuric acid was used to improve the final brightness from an  OQ(EOP)(EOP)P  sequence by replacing the second  (EOP)  stage with a  peroxymonosulfuric acid stage This resulted in increase of brightness from 79% to 86% ISO (Seccombe et al., 1994). Peroxymonosulfuric acid has been evaluated as the replacement for Monox-L in the production of an MEopD sequence to produce a pulp of equivalent brightness and strength. The bleaching cost increased by approximately 50% (Anderson, 1997). A mixture of peracetic acid and peroxymonosulfuric acid (PXA) has been evaluated in a TCF sequence on an oxygen-delignified hardwood kraft pulp. A high dose of peroxide and a long retention time was required to achieve 88+% ISO brightness in the traditional oxygen/peroxide TCF sequence. The inclusion of PXA resulted in a significant decrease in the required peroxide charge and improved the final viscosity (Anderson, 1997).  18  Table 5: Reduction Potentials of Oxidants (Anderson, 1997)  Oxidation Reaction  Reduction Potential, V.SHE  0 + 2H + 2e" -> 0 + H 0  2.07  H 0 + 2H + 2e" -» 2H 0  1.78  +  3  2  2  +  2  HCIO2  2  2  + 3H + 4e" -> CI" + 2H 0  1.56  +  2  HOCl + H + 2e" -> C r + H 0  1.49  +  2  HSO5 + 2H  HSO4 +  + 2e ->  +  1.44  H 0 2  C l + 2e~^ 2C1"  1.36  2  C10 + H + e•-> HC10  1.15  +  2  C H 3 C O 3 H + 2H  CH3CO2H +  + 2e ->  +  2  H 0 2  1.06  CIO" + H 0 + 2 e ~ C 1 + 2 0 H "  0.90  HOO + H 0 + 2e -> 30H"  0.87  C10 " + 2H 0 + 4e" -> CI" + 40FT  0.78  0 + 2H 0 + 4e" -> 40H"  0.40  2  -  _  2  2  2  2  2  19  Table 6: Peroxyacid Use in the Bleaching Sequence (Anderson et al., 1995) 1. Delignification Agent: • •  Wood (pulping process such as Milox) Pulp (replacement or reinforcement for an initial chlorination stage, reinforcement for an Eop stage, or an ozone stage)  2. Activation Agent: • To activate lignin before or between an 0 stage(s) • To activate a peroxide stage together with oxygen under acidic or mildly basic conditions • To activate a subsequent peroxide brightening stage 2  3. Brightening Agent in later oxidizing stages: • ECF bleaching as a replacement for a chlorine dioxide or hypochlorite stage • ECF bleaching as a reinforcement of a chlorine dioxide stage • TCF bleaching as a replacement for a peroxide stage  An oxygen delignification stage was enhanced by an oxidative treatment of. the pulp with PMS (Springer et al, 1993). A prior treatment with a chelating agent or a mineral acid to remove transition metals was essential. Another important commercial use of peroxymonosulfuric acid is to activate lignin before or between oxygen stages. Oxygen delignification is increasing worldwide as an intermediate step between kraft pulping and final bleaching to remove up to 50% of the residual lignin in kraft pulp because of reducing the requirement for chlorine-based bleaching chemical. Extending conventional oxygen treatments to remove selectively more than 50% of the residual pulp lignin would decrease bleach chemical demands further and increase environmental benefits. Allison et. al (1995, 1997) have shown that pulp treatment with small amounts of peroxymonosulfate (P ) between two-oxygen delignification stages (OP O) improves x  x  20  overall selectivity. The OPxO sequence can selectively achieve greater than 70% delignification. They have shown that the inclusion of an interstage treatment with 2% H S 0 " reduced final OPxO pulp kappa number to less than 8 from 12.8 and pulp viscosity 5  increased from 20 to 25 mPa.s. Handsheet strength potential also improved by 5-10% over corresponding pulp without the interstage P treatment. Oxygen delignification is x  initiated through reactions of oxygen with ionized free phenolic groups in residual lignin. Thus pretreatments which increase the formation of free phenolic groups during oxygen delignification should improve overall selectivity. Pretreatments with molecular chlorine achieve this through either demethylation or p-ether cleavage reactions. Hydroxylation of lignin ring structures with acidic peroxide treatment has also been shown to improve oxygen delignification. But, the formation of free phenolic groups in lignin by pretreatments may not be the only mechanism to improve oxygen delignification selectivity. Chlorine dioxide has been shown to improve the selectivity although it reacts with and consumes free phenolic groups and ultimately reduces the overall content of free phenols in lignin. OPxO delignification is an extension of the O X O (X= chlorine) concept developed by Lachenal et al., 1990. Peroxymonosulfuric acid, like chlorine, degrades lignin and creates new sites for chemical attack, such as free phenolic groups. Thus PMS treatment has the potential to improve the second stage oxygen delignification in the same manner as chlorine treatments during O X O bleaching. More importantly, a sulfur-based chemical would be compatible with the kraft recovery system. At the same time, the O P O sequence does suffer from the high cost of peroxymonosulfuric acid and associated x  problems with excess sulfur inputs into .the kraft chemical recovery cycle. These drawbacks can be minimized if PMS can be generated cheaply and at high yield. Peroxymonosulfate applied in the acid (Peroxymonosulfuric acid) or salt (PMS) form also shows promise as a color-stripping agent (Kapadia et al., 1993). Because of the chemical constitution of the common dyes, oxidative agents alone usually cannot adequately remove color from dyed recycled papers. Therefore, color stripping is generally a separate operation as part of recycled paper bleaching. 21  Effective disposal of mill broke is an important decision facing a paper mill. The mills employ a variety of solutions such as selling it to a broker, land filling, or using it in a different grade. However, the preferred and perhaps the most economical option is to repulp the broke and incorporate it back in the furnish. Often, due to regulatory, business, or environmental reasons, this is the only solution possible for the mill. Similarly, for those mills that use secondary fibers in their furnish, repulping is a necessary processing step. Various additives are incorporated into the paper stock to impart specific properties to the final paper product. These include fillers, sizing agents, retention and drainage aides, dry and wet strength resins, and dyes. When the paper is recycled either in the broke system or in secondary fibers furnish, the additives now become contaminants that must be removed or rendered ineffective. White fillers, sizing agents, retention and drainage aids, and dry strength resins generally do not interfere with the de-fibering of paper and do not impart any undesirable color to the fibers. Using good agitation can repulp paper containing these contaminants. However, wet strength resins bond the fibers together through covalent bonds and dyes impart color to the paper that is not easily washed out. It is, therefore, necessary  to use special chemical treatments to repulp paper containing these  contaminants. The most commonly used repulping chemical for both wet strength resins and dyes is sodium hypochlorite. But the environmental concerns regarding the use of chlorine compounds in the paper mills have directed to find alternative repulping aides. Peroxymonosulfuric acid salts are reported such non-chlorine repulping aides for paper containing dyes and wet strength resins (Kapadia et al., 1993). For ionic oxidation reactions, potassium peroxymonosulfate (PPMS) works effectively in the alkaline region. Higher pH, temperature, and PPMS concentration favor repulping reactions. However, the largest commercial use of peroxyacids is for epoxidation of olefins and for oxidation of cyclohexane. Peroxyacids are also in commercial use in laundry bleaching. One specifically promising area of application is the substitution of peroxymonosulfuric acid for chlorine to decompose cyanide residues from electroplating  22  operations. Other oxidative applications include various bleaching processes, aluminum treatment, and wool shrinkproofing (Chiang, 1977). The drawback for the use of peroxyacids, particularly in the pulp and paper industry, is the relatively low current manufactured volume and high cost of manufacture. Peroxyacids are generally not sufficiently stable to be shipped safely or stored for long periods of time. On-site generation of peroxyacids is a necessary consideration for its use in bulk. These obstacles can be overcome if a safe and economical generation process is found. 1.3.1  Generation of Acidic Peroxymonosulfate (PMS)  Despite their recognized efficiency as oxidizing agents, peroxymonosulfates are not generally available on the chemical market since as a class they tend to be very unstable except for the triple salt of potassium peroxymonosulfate, a crystalline complex salt of the formula 2KHSO5.KHSO4.K2SO4 which is sold commercially. However, only two out of four molecules in the triple salt of potassium peroxymonosulfate are active, the remaining two are inert. Further, the triple salt is stable only when it is completely dry. In commercial practice, this salt is kept in the presence of a desiccant material in particulate form to protect against the presence of free water due to moisture picked up from the atmosphere. This desiccant material further diminishes the activity of this product on a weight basis. Most references indicate the use of 70% hydrogen peroxide in order to achieve satisfactory conversion to PMS acid via equation (1). In practice, there is an equilibrium mixture produced, containing PMS acid and the equilibrium is highly dependent on water content. Thus high conversion of peroxide to peracid is achieved by minimizing the amount of water entering the system and by applying sulfuric acid in excess. Consequently a reasonable yield of PMS is not achieved unless 70%  H2O2  is used as a  feedstock. The equilibrium conditions are such that the reaction is incomplete and large excess of reagent must be used. For instance, in order to prepare a 20% solution of  23  H2SO5,  one must use a five fold amount of sulfuric acid and a 12% excess of hydrogen  peroxide over the stoichiometric requirements (Chiang, 1977). The hazards for storage and handling of 70% hydrogen peroxide are significantly greater than for 50% hydrogen peroxide and the use of 70% hydrogen peroxide may be more highly regulated in certain jurisdictions (Anderson, 1997). The high cost of using 70%  H2O2  and limited (at most 70%) conversion from H 0 to 2  2  H2SO5  due to the  chemical equilibrium limits its application. The reaction evolves so much heat that it poses large-scale operation difficulties in keeping the temperature within the desired limits. The field effectiveness for yield of peroxymonosulfuric acid will be determined by the capability for removing heat and for minimizing the effect of process water. An on-site generator must have adequate heat exchange capability. Other common methods to prepare peroxymonosulfates are given by Chiang (1977) and Darbee et al., (1960): 1. From chlorosulfonic acid and hydrogen peroxide. This is not suitable for large-scale preparation because of heat evolution during the reaction (Heat of reaction not reported, however). HSO3CI  + H 0 -> 2  2  H2SO5  + HCI  (4)  2. From the hydrolysis of peroxydisulfates: H S 0 + H 0 -> H S 0 + H S 0 2  2  8  2  2  5  2  4  (5)  3. From the reaction of H2O2 with H2S2O8 H S 0 + H 0 -> 2H S0 2  4.  2  8  2  2  2  (6)  5  The procedure of electrolyzing a sulfuric acid solution that yields PMS acid in the  anolyte by hydrolysis of peroxydisulfuric acid is not commercially useful because the conditions for efficient electrolysis and proper hydrolysis are different. Attempting to do both reactions simultaneously results in low current efficiency and poor yields.  24  1.3.2 Generation of Basic Peroxymonosulfate (PMS) Most previous research was directed towards generation of acidic PMS. With the increase of oxygen delignification throughout the world, a cheaper method to generate alkaline PMS is needed. The demand for alkaline PMS is growing due to the fact that PMS improves overall selectivity when applied as an interstage treatment between two oxygen delignification stages (McGrouther et al., 1994). For example, by using 1.1% HS05" in the interstage treatment and 2% sodium hydroxide in the final oxygen stage, the Kappa number of kraft-oxygen pulp was reduced from 17 to 7 while pulp viscosity was only reduced from 24 to 21 mPa.s. On its own, oxygen delignification could only lower the Kappa number to 10 for the same loss in pulp viscosity. The inclusion of an alkaline PMS stage between two oxygen stages is economically important. An acidic PMS stage, if placed between two alkaline oxygen treatments, would require that alkali be added to the pulp in order to reach the pH 10 in the second oxygen stage. The additional cost of the added alkali would significantly increase the total cost associated with an acidic PMS stage application. Allison et al. (1995) also revealed that the efficiency of active oxygen transfer from PMS to lignin chromophores increased at high pH. Therefore, a greater chromophore removal would be expected at pH 10 compared with pH 4. It is evident that the commercialization of alkaline PMS is not likely to come about given the present state of the art. What is needed is a simple method of generating peroxymonosulfates at the point of application with minimal active oxygen losses from side reactions. A cheaper alternative route for peroxymonosulfate generation would be highly desirable to improve economics for both DMD and peroxymonosulfate bleaching processes. An alternative process would be to directly oxidize sulfite with air or oxygen to peroxymonosulfate, which was previously studied but was considered to be inefficient for bleaching application. On the other hand, this process seems very attractive in context of mill system closure. Sodium sulfite could be generated from green liquor at kraft mills using a carbonation process leading to H 2 S and then followed by incineration of H 2 S to 25  SO2 (Lee et al., 1994). Generated PMS solution can be potentially used in an HOP stage replacing the use of H2O2 and/or in pre- and post-bleaching treatment of chemical pulps to reduce demand for other bleaching chemicals (Chen et al., 1996). The alkaline PMS produced in such a way, if in high enough concentration, can be directly used in pulp bleaching, while spent sodium sulfate is recycled into kraft recovery. Scientists have been engaged in finding methods for the generation of PMS due to its demand for producing other organic chemicals, treating industrial wastes, bleaching textiles and pulps, and as an ingredient in cleaning powders and cleansing solutions. However, all of these processes suffered from drawbacks such as low efficiency or high complexity. The constant interest of researchers resulted in findings that adding calcium or sodium sulfite to vigorously aerated water could produce peroxymonosulfate. Devuyst et al. (1979) reported that the air oxidation of sodium sulfite under strong alkaline conditions (pH = 12.5) could generate sodium peroxymonosulfate according to the equation 7. Na S0 + 0 2  3  -> Na S0  2  2  (7)  5  The concentration of N a 2 S O s produced by this method was only about 0.35 g/L. However, such work was of interest because it was a landmark to generate alkaline PMS and was desirable to cheaply produce a non-chlorine bleaching agent. Chlorine and chlorine-containing oxidizing agents, although much used then in bleaching, lead to difficulties to dispose bleach spent liquor because the resultant chlorinated compounds were mutagenic and darkly colored. The discovery that the catalytic amount of cupric ions could improve the yield and concentration of peroxymonosulfate in situ was useful. Springer et al. (1988) discovered a method for producing peroxymonosulfate in which copper ion was used as a catalyst in aeration of sodium sulfite according to equation 8. Cu Na S0 2  3 4  + 0  +  +  **"  2  DH = 8-14  Na S0 2  1  b  5  (8)  Although the pH of the solution was varied from 8 to about 14 by sodium hydroxide addition, the best results were obtained when the pH was maintained between 26  12.0 and 12.9. The temperature of the reaction mixture ranged from about 0°C to 80°C. The higher temperatures provided lower conversions due to the fact that it allowed less oxygen dissolution, but was convenient in applications which involved simultaneous peroxymonosulfate generation and pulp bleaching. The cupric ion was supplied using cupric sulfate and its concentration was varied from 0.01 to 100 parts per million, with concentrations from 0.04 to 0.8 ppm, providing good results. The results of seven experiments are given in Table 7. As shown in Table 7, the concentration described in the Springer's patent was only slightly higher than that of Devuyst, around 0.65 g/L Na S0 . 2  5  Springer et al. also used such in-situ generated Na S05 from the catalytic sulfite oxidation 2  to study pulp bleaching. The equipment and procedure involved were essentially the same as that for PMS generation alone. Pulp was added to the alkaline solution immediately before sulfite addition was started. Using cupric sulfate as the catalyst, the efficiency of producing peroxymonosulfate by direct air oxidation of sulfite was not improved enough to meet the requirement for the bleaching process. Chen et al. (1996) also made an attempt to explore the catalytic sulfite air oxidation process to increase final Na S05 concentration and yield. They were successful 2  to markedly improve the oxidation efficiency under high oxygen pressure (100-120 psig) and intense mixing. Within a very short reaction time (1-2 minutes), peroxymonosulfate was produced at a concentration of 3.8 g/L Na SOs with a yield of about 20%. Although 2  this was a significant improvement, it was not sufficient for pulp bleaching. As with other chlorine-free systems, if the effluent were recycled to the recovery cycle, the extra sodium and sulfur would build up in the system and would eventually reach unacceptable levels. Therefore, both effective generation of PMS and a good mill chemical balance is required. In order to recover the sodium and sulfur, a process scheme has been proposed which recovers these chemicals from the green liquor (Lee et al., 1994). Chen et al. (1996) proposed that a yield of at least 50% and 10.0 g/L Na S0 from Na S0 is 2  5  2  3  desirable to reduce inorganic deadload of Na S04. In an economic assessment of PMS 2  generation via catalytic sulfite oxidation, they proposed $6.0 million (Can.) as the total capital cost that included S 0 generation (2.0 million), PMS generation from Na S0 at 2  2  27  3  20% yield and 3.8 g/L (3.0 million) and press for thickening (1.0 million). If the concentration of PMS could be increased, the money spent from the press would be saved. It was also reported that the inclusion of 2% PMS between two oxygen delignification stages achieved greater than 70% delignification (Allison et al., 1995). Further increase in PMS yield and concentration will also help to minimize the amount of dilution to 0 stage. 2  Table 7: Generation of Peroxymonosulfate Using Sodium Sulfite (Springer et al. 1988)  1  2  3  Run Number 4 5  6  7  Temperature (°C) PH  2  2  2  2  2  55  55  12.0  12.0  12.0  12.5  12.9  12.5  12.5  CuSC>4 (ppm)  0.1  0.8  2.0  0.4  0.4  0.4  0.4  Na S0 (M)  1.19  1.19  1.19  1.19  1.19  0.397  0.794  0.16  0.14  0.17  0.15  0.137  0.175  0.150  3  3  3  3  3  3  3  105  120  130  120  180  90  120  19.4  20.3  26.6  21.4  29.3  6.25  14.3  1.60  2.10  2.00  3.50  4.60  0.36  1.00  8.2  10.3  7.5  16.4  15.7  5.8  7.0  2  3  Na S0 Addition Rate (ml/min) Air Rate (1/min) 2  3  Total Time (min) Total Na S0 Added (mmoles) Total PMS Generated (mmoles) 2  Percent Conversion (%)  3  Chen et al. (1996) found that the efficiency of generating Na S0 from 2  5  oxidation was extremely low in the absence of copper ion. In the presence of copper i 28  at ppm levels, the production of  Na S05 2  was dramatically increased. The amount of  copper ion from about 0.6 ppm to 10 ppm did not seem to affect the sulfite oxidation significantly. An implication of this is that only a very small amount of copper ion (~ 0.8 ppm) is required and the catalyst concentration might be low enough not to pose any significant effect on bleaching performance (Chen et al., 1996). Such low concentration of the catalyst can eliminate the threat to the environment by allowing good metal management of bleaching performance. They also did not find any major difference between runs using air and pure oxygen. If this is true, the oxygen solubility would not be a limiting factor to improve the PMS yield and concentration. Chen et al. (1996) also reported an inverse relationship between the concentration and the yield of Na2S0 . They explained this phenomenon by the following three major 5  competing reactions involved in the catalytic sulfite oxidation. Na S0 + 0 2  3  -» Na S0  2  2  (9)  5  Na S0 +Na S0 -> 2 Na S0 2  3  2  5  2  Na S0 -» 2 N a S 0 + 1/2 0 2  5  2  4  (10)  4  (11)  2  As explained, the first reaction is dominant at the beginning of the reaction leading to a low concentration of Na S05 but a higher yield. The produced Na SO;, builds 2  2  up in concentration with more Na S0 addition as the reaction continues. Since Na SOs 2  3  2  is a strong oxidizing agent and Na S0 a reducing agent, the second reaction then 2  3  competes with the first one and consumes the already generated Na SOs. The self2  decomposition of Na S05 is also another possibility. The net result is the decrease in 2  Na S0 yield. 2  5  An important finding by Chen and co-workers was the assessment of the catalytic effect of other metal ions on the oxidation of sulfite to generate peroxymonosulfate. In order to do this, a number of transition metal ions including Fe , Fe , N i , Co , Mn , 2+  and V° were used under the same conditions. +  29  3+  2+  2+  2+  Table 8: Peroxymonosulfate Generation from Sulfite Air Oxidation Catalyzed by Various Transition Metal Ions (Chen et al., 1996) Metal ion  a  Concentration in solution (ppm)  Na S0 produced  0 0.32 0.64 +4.0 0.32 0.64 2.0 0.32 0.64 2.0 0.32 0.64 2.0 0.64 1.3  0.08 0.04 0.04 0.13  2.1 1.2 1.2 4.3  0.04 0.06 0.02 0.08 0.04 0.04  2.2 4.1 0.6 4.3 2.4 2.9  0.04 0.03 0.04 0.04 0.03  2.7 2.2 2.5 2.8 1.4  0.32 0.64 1.3  0.01 0.08 0.04  No catalyst Ni  2 +  Co  2+  Fe  2+  Fe  3+  Mn  2+  2  5  (R/L)  b  Yield of Na S0 (%) 2  5  0.4 2.5 1.1 Conditions: The reaction temperature was 1-2°C. 0.5M Na S0 solution was added at a rate of 0.5-0.6 mL/min and air was sparged at 2 L/min. All the metal ions were applied in the form of sulfate except for V as V 0 . A small amount of H 0 was also produced as determined by eerie sulfate titrations except for Ni as a catalyst. V  from V 0 5 +  2  5  2  3  a  5 +  b  2  2  5  2  As shown in Table 8, none of them resulted in any significant concentration of Na S0 (< 0.1 g/L) or yield (< 5%) compared to that of 1.0 g/L and 25% by C u ion. No 2+  2  5  specific reason was given for the success of the cupric ion over others. However, Ermakov et al. (1997) also found no progress of the sulfite oxidation reaction in the absence of catalysts or sources of free radicals. They also studied the mechanism of  30  transition metal ions (Fe, Mn, Cu, etc.) intensively and their effect in the generation mechanism remained unclear to them. Chen et al. (1996) also studied the decomposition of S0 " in the absence and 2  5  presence of copper ion at pH 12.5 and room temperature by measuring PMS concentrations at various time intervals. As reported, the concentration decreased gradually. After 90 minutes, S 0  2_ 5  was decomposed by about 10% for the 4.8 g/L, K S 0 2  5  solution and 15% for the 8.7 g/L K S 0 solution. They also found that the decomposition 2  5  was not accelerated by the presence of copper ion. Table 9: Effect of Water Source on Na SO Generation (Chen et al., 1996) 2  s  Source of water  Resistivity (M-ohms-cm)  Cu added (ppm)  Na S0 added (mole)  Na S0 produced (g/L)  Na S0 yield (%)  DI  0.883  0.64  0.126  1.05  5.8  DI  0.191  0.64  0.126  2.40  13.3  Tap  0.050  0.64  0.126  3.85  21.3  Tap  0.050  1.28  0.126  3.88  21.4  0.061  0.64  0.126  3.65  20.2  Tap  j  2+  2  3  2  5  2  5  Conditions: Starting temperature was at 5-10°C. 84 mL of 1.5M Na S0 solution was added into 1.0 L of water at pH 12.5. Oxygen pressure was 120 psig and mixing time was 2 minutes. DI = Deionized 2  3  I Chen et al. (1996) also found that the source of water had a significant impact on Na S05 generation. Table 9 represents a comparison of the results reported by them from 2  runs using deionized and laboratory tap water. It was clearly shown that using tap water led to a higher concentration and yield of Na S0 . They also conducted a series of 2  31  5  experiments by varying oxygen pressure from 0 to 120 psig at a constant mixing speed of 1300 revolutions per minute (rpm) in a Quantum® Mark IV mixer. As opposed to the PMS generation under the ambient pressure, it not only resulted in much higher Na2S05 concentration and yield but also significantly reduced the reaction time to less than 2 minutes from 60-120 minutes. When the mixing intensity was increased from 430 to 3,900 rpm at constant oxygen pressure (lOOpsig) in the Quantum mixer, PMS yield increased from 6% at 1.2 g/L to 15% at 3.0 g/L. This indicates that the reaction may be mixing sensitive. Chemical analyses by Chen et al. (1996) were performed to identify and quantify other possible products besides the generation of Na2SOs in the copper catalyzed sulfite oxidation. As shown in Table 10, Na SOs was the predominant oxidation product, 2  although a small amount of peroxydisulfate (Na2S 0s) and a negligible amount of 2  hydrogen peroxide (H2O2) were produced. 2+  Table 10: Chemical Analysis of Cu -Catalyzed Sulfite Oxidation Liquor (Chen et al., 1996) Na S0 Added (mole)  Cu added (ppm)  Na S0 Produced (g/L)  Na S 0 Produced (g/L)  produced (g/L)  0.126  0.64  3.60  0.19  0.002  2  3  2+  2  5  2  2  8  H2O2  Conditions: Starting temperature was 5°C. 84 mL of 1.5M Na S0 solution was added into 1.0 L of water at pH 12.5. Oxygen pressure was 120 psig and the mixing time 2 minutes. 2  3  All of the above informations established by different researchers were helpful to understand and design reactors to further improve yield and concentration of Na S0 . 2  32  5  1.4 Chemistry and Kinetics of Sulfite Oxidation The review of "Sulfite Oxidation" by Ermakov et al. (1997) is important in order to understand chemistry and kinetics of air oxidation of sodium sulfite. The kinetics of this reaction has been studied extensively and a large number of papers (>100) devoted to studying this process have been published. Backstrom (1934) described this reaction to the chain processes on the basis of the strong inhibiting effect of small amounts of readily oxidizable organic substances on its rate. New experimental techniques have also confirmed the conclusion about the free-radical chain mechanism of the reaction. To explain the kinetics, including catalysis by copper and other metals of variable valence, Backstrom (1934) proposed a free-radical chain mechanism with the following steps: Initiation: S0 " + Cu  — C u  ++  3  + *S0 "  +  (12)  3  Propagation: *S0 " + 0 — * S 0 " 3  2  'S0 ' + S0 " 5  (13)  5  3  k}  > S 0 " + *S0 ' 5  (14)  3  Termination: S 0 " + S0 "" 5  3  2S0 "  (15)  4  According to Backstrom's assumption, a likely route of chain initiation is the reaction between SO3 " and trace contaminants of copper ions. Assuming that initiation is 2  due to trace contaminants of copper ions, Backstrom did not take into account their hydrolysis and precipitation as hydroxides with increasing pH (Ermakov et al.,1997). Ermakov and co-workers (1997) reported the rate constant of equation 13 (k = 1.5 x 10  9  2  L mol" s" at 293 K). Deister et al. (1990) proposed that the reaction between "SO5" and 1  1  SO3 " proceeds by two pathways. 2  •S0 " + S 0 " " - ^ - > S0 "" + *S0 "  (16)  'S0 "+ S0 ""  (17)  5  5  3  5  3  S0 "" + *S0 "  3  4  33  4  Deister et al. (1990) also found that the ratio of k^/ks = 3 for alkaline media in which the peroxymonosulfate radicals are present in the dissociated form (*SC)") only. 5  Poskrebyshev (1995) only reported the k value [(3.6 ± 2) x 10 L mof s" at 293 K] in 5  1  1  5  acidic aqueous solutions. Oxidation of Na S03 by peroxymonosulfate was studied in most detail by the 2  stopped flow method with the spectrophotometric monitoring of [Na S03] in the pH 2  region from 1 to 11 and at T=278-31 IK (McElroy et al. 1990; Betterton et al. 1988). For alkaline media, the mechanism of Na S0 oxidation by peroxymonosulfate involving the 2  3  formation of the (S 0g ) adduct with its subsequent decomposition under the action of 4  2  OH" was proposed (McElroy et al. 1990; Betterton et al. 1988): HS0 " + S0 " 2  5  - >S 0 " + H  k  4  3  2  S0 " + S 0 '  k  S 0 " + OH"  kprod  2  2  5  3  4  2  " >S 0 " k  8  (19)  4  2  8  > 2S0 ~ + OH"  (20)  2  4  Where k = 6x 10 , k \ = 120 + 30, k  = 1.0 ± 0.5 L mol" s" at 293 K. The formation  3  ac  (18)  +  8  a k  1  pwd  1  of S 08 " appears to accompany the reaction of the lone electron pair of the sulfur atom 4  2  with the peroxy group of SO5 ~ (HSO5"). The lower value of the rate constant of the reaction of SO3 ' with SO5 " compared to that of the reaction with HSO5" was explained by the effect of the electrostatic factor (Ermakov et al.,1997). For Na S0 oxidation by 2  3  peroxymonosulfate, the authors (McElroy et al. 1990; Betterton et al. 1988) were also successful in detecting the intermediate (S 0s ") absorbing light at X = 205 nm (e = 320 L 4  :  2  mof cm" ). 1  1  For alkaline media that includes the formation of the (S 08 ") adduct (McElroy et 4  2  al. 1990; Betterton et al. 1988), the mechanism put forward by Backstrom (1934) can be modified: SO3" + Cu •SO3-  C u + -S0 -  ++  +  3  + o —-so " 2  5  34  (21) (22)  *S0 " + S 0 " " — S 0 " " + ' S 0 "  (23)  S0 " + S0 "  (24)  5  3  2  5  2  4  3  °* > S 0 "  k  3  S 0 " + OH" 2  5  4  2  kprod  8  8  > 2S0 " + OH"  (25)  2  4  Ermakov et al. (1997) established a kinetic scheme of free-radical oxidation of Na2SC>3 in acidic aqueous solutions in which rate constants of possible reactions by different authors were given. But the values could not be used due to lack of information regarding temperature and pH. Thus in summary, in acidic media, the formation of sulfate follows directly from equation (15) while in alkaline media, it goes through an intermediate (S2O8 ") as shown 4  in Equation (19) and (20).  1.5 Mixing Sensitive Chemical Reactions The method of bringing together reactants that are to undergo reaction can have an influence on the course of the reaction for certain systems. If the reaction can result in only one product, the method can only influence the reaction rate. If more than one product is possible, contacting can influence the product distribution as well. These considerations apply to both homogeneous and heterogeneous reaction systems. When multiple reactions occur in a wide variety of combinations, each of them must be examined individually to determine whether mixing can affect selectivity. Two major types are generally recognized as the basis of most combinations: Equation (26) and (27) represent competitive parallel reactions: A + B  k  '  > R  (26)  S  (27)  ko  A + C  For this reaction system, where A is added to B in the presence of a substance C, the ratio of R to S should be a direct function of the ratio of the reaction rates.  35  Competitive consecutive reactions, with B added to A are of the type shown in equations (28) and (29). A + B  1  R + B  2  > R  (28)  > S  (29)  Here R, the intermediate product, will be considered the desired product and S, the undesired product. A and B react to form R by the first reaction. B can further react with the R formed to produce S. When a substantial amount of R has been formed, both A and R compete for the B present in the system to form either more R or S respectively. The outcome of this competition determines the product distribution. Mixing will not affect selectivity if both reactions are first order since the concentration of only one component is involved. If both reactions are second order, mixing can affect selectivity because of the local excess B concentration, which can result in local overreaction of R to S. Chemical reaction of such type is referred to as second order, competitive-consecutive since it is competing with respect to B and consecutive with respect to A and R. This type of mixing sensitive chemical reaction can be used to study mixing. One such reaction is the azo-coupling between 1-naphthol (A) and diazotized sulfanilic acid (B) forming the mono (R) and bis (S) azo dyestuffs (Bourne et al., 1985). The product distribution (Xs), a measure of selectivity used here, is the ratio of the amount of B that reacts to form byproduct S to the total amount of B reacted (Bourne et al., 1985). For the reaction scheme in equations (28) and (29) is given by the equation: 2C X s =  5c7c:  (30)  Where Cs and C are the volumetric mean concentrations of R and S, respectively. The R  product distribution is measured when reaction is complete. When the reaction is much faster than the mixing rate (mixing controlled regime), the primary product of competitive-consecutive reactions does not survive and only A and S are found in the final mixture i.e. yield of R = 0 and Xs = 1. On the other hand, when the mixing rate is 36  much faster than the reaction (chemical controlled regime) negligible amount of S are formed (X tends to zero). Mixing rate in the intermediate regime therefore influences the s  product distribution. When the reactant B is added at low feed rate, the product distribution is independent of it's addition time. However, at high feed rate, mesomixing effects, i.e mixing at scales larger than microscale but smaller than macroscale may also influence the product distribution (Bourne and Thoma, 1991). The product distribution at long feed times is controlled by micromixing, which depends upon a) molecular diffusion, b) fluid deformation, and c) mutual engulfment of A-rich and B-rich regions (Baldyga and Bourne, 1989). Under conditions when (i) Sc « 4000, where the Schmidt number is v/D (v =kinematic viscosity, D = diffusion coefficient) and (ii) CBO » CAO, where CBO is the concentration of B (limiting reagent) and C o is the initial concentration of A, only the engulfment step (step c) is important. A  The engulfment rate coefficient, E, is related to the kinematic viscosity (v) and the rate of dissipation of the kinetic energy (s) by the following equation (Baldyga and Bourne, 1984). E = 0.058 (s/v)  (31)  05  In order to quantify the relationship between the product distribution and mixing intensity, a well-established micromixing model can be used. Bourne's engulfment model is one such model that relates the rate of energy dissipation and the product distribution in turbulent mixing and is solved by the following equation: ^  dt  = E(<Ci >-Ci) +  r i  (32)  Where, E = engulfment rate coefficient, <Q> = average concentration of substance i, Q = local concentration of substance i, rj = rate of formation of i by the reaction. When the engulfment model is expressed in dimensionless form, the product distribution, Xs depends upon  37  x =f s  ^N  k V rf,D >reactor type a  Km N  2  (33)  B  When the experimental conditions are fixed, Xs depends only on the Damkoler number, a mixing modulus that is the ratio of mixing time to reaction time. The Damkoler number, D is related to the engulfment rate coefficient, E by the following equation. a  D  ~ = % " E(l ^ ) J  i  2  (34)  +  As the engulfment rate coefficient (E) is related to the turbulent energy dissipation rate (s) by equation (31) that can be estimated from power input to the reactor. Thus from solution of equation (32), corresponding product distribution can be obtained at a given energy dissipation rate.  38  2  OBJECTIVES From the expected benefits of PMS as a bleaching agent, it is necessary to find a  suitable economic method for its generation. Consideration must be given to its limited stability if to be shipped safely or stored for long periods of time. On-site generation is thus important for its use in bulk. The catalytic oxidation of sodium sulfite with oxygen or air is one possible method for the on-site generation of PMS. A number of attempts have been made to improve both the yield and concentration of PMS by this method. But to date its generation in the laboratory is limited to 20% yield and 3.8 g/L Na SOs. Its 2  yield and concentration needs to be further improved for bleaching application. Allison and co-workers used I - 2 % PMS application in between two oxygen bleaching stages. The effect of increased yield and concentration of PMS on final pulp consistency are shown in Figure 2.1. For example, for medium consistency (12%») 0  2  stage with  interstage 2% PMS application at lg/L, the consistency would be lowered to 3.5%, clearly too low for medium consistency operation. But with 1 Og/L application, the consistency would drop to only 9.7%, whence the need for higher PMS concentration. Of course the higher the yield, the less sulfite is required in the production of PMS. When the bleaching chemicals are recycled to recovery, the higher the yield the less sulfur added to the recovery system and the better chances for achieving successful sulfur balance. From the reaction mechanism (equations 12 - 15) it is clear that the main competitive products in the catalytic oxidation of sulfite under alkaline conditions are PMS (desired) and sulfate salts (undesired). Conditions for effective mixing in PMS generation must therefore reduce the formation of undesired product. Thus the main objective of this work is to explore the catalytic sulfite air oxidation to increase both the yield and concentration of PMS. From preliminary indications, the reactions could be mixing-sensitive, it was therefore desired to verify and establish conditions for maximizing yield or concentration of PMS.  39  40  3  EXPERIMENTAL In order to achieve the highest yield and concentration of PMS, experiments were  performed in a number of different reactors due to the attributes of each reactor.  3.1 MIXERS USED The different equipment used in PMS generation are introduced in Table 11. Most experiments were performed in the medium-intensity mixer, with the key parameters that control the yield of PMS (feed-time, energy dissipation rate, sodium sulfite concentration, oxygenated water and sodium sulfite solution volume ratio, temperature and oxygen pressure) being varied. As it was found that the yield was increased with increasing energy dissipation, a high-shear mixer and a rotor-stator mixer were also used to further investigate the effect of this parameter. A PHT mixer was mainly used to investigate the effect of reactor pressure on product yield. Table 11: Different Mixers Used Mode  Liquid Volume (ml)  Impeller Speed (rpm)  Mean energy dissipation (W/kg)  SB  250-2000  N/A  0.01  0 - 2400  0-230  870  0 - 2400  0-950  Equipment Beaker Medium-Intensity Mixer A  SB  B  3000  PHT Reactor  SB  3000  1000-4000  240 - 850  High-Shear Mixer  SB  3400  1000-4000  263 - 5300  Rotor-Stator Mixer  SB  100-400  C  150  5000  4000-5000  SB= semi-batch; C=continuous  41  3.1.1  Experiments in a beaker  The apparatus used to carry out the oxidation reactions is shown in Figure 3.1. Reactions were carried out in a beaker, where a two-vaned laboratory propeller was used to achieve good mixing. In all runs, a desired amount of pre-chilled tap water was placed in a beaker. The pH of the solution was adjusted to 12.5 with 1 N NaOH. A proportional amount of  CUSO4  (0.8 ppm) as a catalyst was then added. Oxygen was sparged  continuously into the reaction mixture. After 5 minutes (to ensure the availability of dissolved oxygen into the reaction mixture), 0.5 M Na2S03 was added continuously (0.5 ml/min) to the reaction mixture . Na2S03 was injected at the top of the liquid surface as shown in Fig. 3.1. The reaction temperature was controlled to 0-4°C by using a cooling ice-water bath. The peroxymonosulfate generated was quantified using iodometric analysis as described in section 3.3.  42  Burette containing sodium sulfite Thermometer To oxygen cylinder  pH probe  Beaker Cooling system Sparger  Figure 3.1 Experimental Setup for Beaker, showing use of stirrers.  43  3.1.2  Medium-Intensity Mixer  A medium-intensity mixer was constructed from a cylindrical stainless steel tank that was enclosed in a cylindrical cooling bath filled with cooling water to maintain desired temperature (<5°C). Four baffles spaced at 90 degrees on the side of the vessel were provided to eliminate swirling effects in the vessel. The rotor consisted of fourblades mounted on a cylindrical hub extending the full length of the vessel, giving a power number of 5.4. Both ends were supported by thrust bearing to ensure minimum frictional losses. A 0.2kW motor provided the driving power and the power input to the mixer was measured using a power meter (model # 982521; Mitutoyo Mfg. Co., Ltd. of Japan). The impeller rotational speed measured using a digital tachometer (model # 08212; Cole Parmer, Chicago, IL). The experimental setup is illustrated in Fig. 3.2. A recirculation loop was provided to return the reactor liquid that was carried out of the vessel with sparged gas. In order to achieve maximum mixing efficiency, a feed pipe (1 mm ID) was located at a distance of 5mm from the rotor tip and 55mm from the top of the vessel. To perform an experiment, a desired amount of ice-cold water was first poured in the mixer. Enough sodium hydroxide was added to obtain the required pH of 12.5, followed by the catalytic copper sulfate (0.8ppm as Cu ). Oxygen from oxygen cylinder 2+  (20 psi, 20°C) was continuously sparged into the solution throughout the reaction progress. Sodium sulfite from a graduated cylinder was fed through selected feedpoint into the reaction mixture using a constant flow micro-pump (model 7617-70, Ismatec Instruments, Switzerland).  44  Figure 3.2: Photograph of Medium-Intensity Mixer Showing Operational Setup [(A) Mixing chamber, (B) pH Probe, (C) Pump (D) Oxygen Cylinder]  45  120mm 50mm -*  Figure 3.3: Schematic of Medium-Intensity Mixer showing Feed Point Location 46  Oxygen was bubbled continuously throughout the reaction progress. At the end of the reaction, a 10-ml sample was withdrawn for iodometric titration to determine the yield and concentration of peroxymonosulfate. Two medium-intensity mixer configurations with dimensions given in Table 12 were used to perform different experiments.  Table 12: Dimensions of Medium-Intensity Mixers Mixer configuration  3.1.3  A  Vessel Volume (ml) 3000  Vessel Height (mm) 170  Vessel Diameter (mm) 160  Rotor Diameter (mm) 90  B  870  170  115  90  Pur Reactor The PUT reactor is a versatile high-shear mixer designed for mixing pulp at  medium consistency (10-12%) under high-shear conditions. This was chosen for automatic control and to verify the effect of pressure on the P M S yield and concentration. The reactor consists of a zirconium bowl and impeller with provisions for heating and addition of chemicals. The installed bowl has a volume of 3.0L and has a two-vaned rotor that provides a high-shear mixing zone. The impeller is belt-driven by a 10 H . P . motor and is capable of attaining an operational speed of 2400 rpm in 0.5 sec. An electronic brake permits rapid deceleration of the motor and impeller.  47  48  The impeller shaft passes through a water lubricated mechanical seal at the bottom of the bowl. One of the two provided injection ports with "spark plug" tips (d=4mm) is used for the injection of chemical (sodium sulfite). The cover of the reactor contains two ports, one for venting the bowl pressure at the end of the reaction, and the other for pressurizing the bowl through the addition of a gas (oxygen). The bowl is equipped with an electronic pressure sensor to provide pressure readings to the computer. In addition a pressure gauge is also connected to the reactor to allow visual observation when the operator is not at the computer screen. The pressure sensor provides information to the computer for addition of gas for reactions carried out under pressure. The reactor bowl sits on a thick steel table and is enclosed in a Plexiglas cabinet with the motor drive mounted under the table in a separate insulated cabinet. The hydraulically controlled reactor cover is set to provide the reactor bowl with a maximum pressure capacity of 150 psi. However, the recommended maximum pressure should not exceed 135 psi. A safety interlock prevents the reactor cover from being opened manually until internal bowl pressure has been vented to less than 2 psi to prevent spraying of the bowl contents. The control panels for the motor and reactor are mounted on the right side of the machine and are interfaced with a personal computer running a control software operating on a Pentium 200 platform in a Windows 95 environment. Using the software program "Mixer", the entire reaction is controlled automatically after entering the desired reaction parameters. Two cold-water bands encircling the walls of the bowl are used for cooling the system but was not sufficient to achieve desired temperature (4°C). The chemicals were added to ice-cold water in the same sequence as described for medium-intensity mixer. Oxygen was sparged continuously into the reaction mixture using a specially designed ring. Required volume of sodium sulfite from a graduated cylinder was then added to the solution of dissolved oxygen with the aid of constant-flow micro-pump (model 7617-70, Ismatec Instruments, Switzerland). At the end of the reaction, a 10-ml sample was withdrawn for iodometric titration to determine the yield and concentration of peroxymonosulfate.  49  3.1.4  High-Shear Mixer  A 3.4L high-shear mixer was also chosen to increase the mixing intensity in the reaction.  A 22.5kW variable speed driven concentric-cylinder device designed by  Bennington (1988) was used for this purpose. Rotational speeds to be determined using remote optical sensor over the range 50 - 5000+1 rpm and torque measurement through built-in strain gauges over the range 0 - 43±0.3 N.m. A transparent plate enclosing the mixing chamber allows visual examination of the contents. A feed tube (d=lmm) was located at a point closest to the rotor (h=50mm, r=55mm) as shown in Figure 3.6. The method of performing experiments and iodometric titration to determine PMS yield and concentration were exactly the same as described for the medium-intensity mixer.  50  Figure 3.5: Photograph of High-Shear Mixer Showing Rotor-Stator Configuration Used.  51  Figure 3.6: Schematic of feed point location and dimensions of the High-Intensity Mixer  52  3.1.5 Dynamic Rotor-Stator Mixer A dynamic rotor-stator mixer (Polytron, Kinematica Gmbh, Switzerland) was used to impart high-energy dissipation rates over a very small shear gap. This is achieved in a closely-fitted assembly of rotor and a slotted stator that allows to draw fluid through the rotors and push it out of the stator, thus creating a continuous stream of fluid in and out of the high-shear zone. A feed tube of 1mm diameter was used for both cases and the feed point was located 5mm directly below the mixing head entrance. The mixer could be operated in two modes, namely semi-batch and continuous. In semi-batch mode, the mixer head is placed in a beaker containing water, sodium hydroxide, copper sulfate and dissolved oxygen. Oxygen was sparged continuously in the beaker before and during the reaction to ensure that it was available throughout the reaction progress. For continuous operation, the mixer head was enclosed in a cylindrical vessel with in- and outlet ports that enables the reacted mixture to be drawn from mixing chamber after passage through the slotted stator while allowing fresh solution to be drawn into the mixing head continuously. In the case of continuous operation, oxygen was sparged into the reaction mixture before passage through the mixing chamber.  53  GENERATED PMS Q. 5 => OL  in  ft  0 ZUI uj  o  O  t/>  Q_  s ZJ Q_  o  I  I O C0 J N O Z O X + + +  Sodium sulfite feed line  Rotor  View A - A  Figure 3.8: Schematic of feed point location of Rotor-Stator mixer  55  3.2 Chemicals Used Chemicals used in this study are shown in Table 13 below. All chemicals were purchased from Fisher Scientific (Fairlawn, New Jersey). They were used directly with the exception of sodium sulfite, cupric sulfate and sulfuric acid where appropriate solutions were made prior to use. Oxygen cylinders were purchased from PRAXAIR. Table 13: Chemicals used Chemical  Formula  Assay  F.W.  Batch No.  Sodium sulfite  Na S0  anhydrous, 98.1%  126.04  S430-3  Cupric sulfate  CuS0 .5H 0  100%  249.68  C489-500  Sodium hydroxide  NaOH  1 .ON solution  40  SS266B-4  Sulfuric acid  H S0  95.0-98.0%  98.075  A300-4  Potassium iodide  KI  Granular  166.01  P412B  Sodium thiosulfate  Na S 0  0.1N solution  158.06  SS368B-1  Starch indicator solution Oxygen  (CgHioOs),!  162n  SS408-1  32  -  2  3  4  2  2  2  o  2  4  3  2  Gas  3.3 Quantification Method When the reactions complete, a 10-ml sample from the reaction mixture was withdrawn and the yield and concentration of generated Na SOs was determined by iodometric 2  titration. The procedure of iodometric titration for peroxymonosulfate is given in Appendix A. Appendix A similarly shows calculation for both the yield and concentration of peroxymonosulfate that includes the stoichiometry of the reaction. Error percentage in  56  any experiment due to measurement criteria was determined by the method shown in Appendix D . The maximum relative error in yield due to measurements was found to be ±5%.  3.4 Mass balance of S0 ', SO/ lodometric titration 2  5  and S0 ' ions using FTIR and 2  3  In order to confirm the number o f moles o f SO5 " evaluated by titration method, 2  F T I R analysis o f the final product generated was undertaken. From stoichimetry, the final 2  2  2  product should contain SO5 ", SO4 " and unreacted SO3 ". Appendix F - l shows the FTIR spectra for P M S , sodium sulfite and oxone taken by Chen et al. (1996). It appears that S 0 " , S 0 " and S 0 " have peaks at 1230, 1100 and 940cm" respectively. The spectrum 2  2  5  2  4  1  3  ^  2  obtained from this study (Appendix F-2) also shows peaks for S0 ~~ and SO5 ". N o peak 4  2  2  was observed at 940 as expected, since SO3 " cannot co-exist with SO5 ". According to the sulfite oxidation chemistry, the total moles o f SO5 " and SO4 " produced must match the 2  SO3 " moles in the mass balance. 2  number o f  SO3 " 2  2  From Table E - X X X I I I it can be seen that the total  moles added matches with total number of SOs " (calculated by titration 2  method) and SO4 " moles (calculated from FTIR spectrum). 2  57  4 RESULTS AND  DISCUSSIONS  4.1 Test Reproducibility A series of tests were conducted under identical experimental conditions to determine the reproducibility with which PMS yield and concentration could be determined. The first set of tests were conducted in medium-intensity mixer A (f=4°C; volume of water=2.0L; NaOH=50mL; CuS0 (0.8ppm)=40mL; 0.5M Na S0 =45mL; 4  2  3  pH=12.5) at feed times between 55s and 60s. The results of five determinations gave yield of 53.3% ±2.66 (s.d.) and concentration at 0.888 g/L ± 0.04 (s.d.), given in Table EXXXVI. A second set of five tests were made in medium-intensity mixer B (t=4°C; volume of water=0.45L; NaOH=9.8mL; CuS0 (0.8ppm)=12mL; 2.0M Na SO =50mL; 4  2  3  pH=l2.5) at feed time between 22s and 36s. Yield of 31.5% ± 1.19 (s.d.) and concentration at 9.53 g/L ± 0.36 (s.d.) were obtained as reported in Table E-XXXVII.  4.2 Confirmation of Previous Tests in Beaker The sulfite oxidation reactions under ambient pressure were first carried out in a beaker with laboratory propeller to verify if the results from previous laboratories were reproducible. Although the results were found to be reproducible, the yields we achieved in the beaker were limited to 9.6 % and 0.5 g/L PMS yield and concentration, respectively at 2-4°C. Running reactions in a graduated cylinder, Chen and co-workers obtained 14 to 17% PMS yield at 1.4 to 1.3 g/L respectively at 1-2°C (Chen et al., 1996). The difference can be explained by the fact that when reactions were run in the graduated cylinder a better oxygen dissolution could be attained. When the reaction was run without using any catalyst, the generation of Na SOs was either extremely low (4%> at 0.2 g/L) and 2  sometimes no PMS could be generated at all. But using copper ion at ppm levels (0.8 ppm), the PMS yield was increased significantly. Since our target was to maximize the PMS yield or concentration, we then used medium-intensity reactors for this purpose.  58  4.3 Tests in Medium-Intensity Mixer As discussed earlier, the sodium sulfite oxidation reactions follow a radical mechanism (Ermakov et al., 1997) and a competitive-consecutive scheme (eqns 21-25). The mechanism is complex but can be simplified to 0  + Na S0  2  2  Na S0 2  5  k l 3  > Na S0 2  + Na S0 2  3  ^  2  (35)  5  > 2Na S0 2  (36)  4  This simplification was made from the following information: 1. In all cases, the rate of  *S0 " generation by reaction between 3  S0 " and 2  3  trace amounts of copper ion (eqn. 21) is not clear. However, it is assumed to be very rapid and hence not rate limiting (Poskrebyshev, 1998). 2. Reaction (22) is very much faster than the other reactions in the chain (k « 1.5xl0 L/mol.s) 9  3. Chain propagation (eqn. 22, 23) and chain termination (eqn. 24, 24) were proposed as the rate-limiting reactions in the oxidation (Ermakov et al., 1997) 4. Reaction 25 is not important (Poskrebyshev, 1998) It has been reported that the formation of sodium peroxymonosulfate (eqn 35) is faster than that of sodium sulfate (eqn 36) in the acidic medium (Ermakov et al., 1997). As discussed previously, for 2 order competitive-consecutive chemical reactions nd  where one reaction is faster than the second one, mixing can play a crucial role to increase the relative yield. The selectivity of such chemical reactions is determined not only by their kinetics, but also by the way in which the reagents are mixed. If the reaction is not controlled by oxygen transfer, the relative amount of peroxymonosulfate or sulfate formed from the competition will depend on the mixing intensity.  59  The product distribution o f sodium sulfate can be defined as the concentration ratio o f sodium sulfate to the sum of sodium sulfate and sodium peroxymonosulfate. Since two moles o f sodium sulfite generates two moles o f sodium sulfate as opposed to Bourne's reaction scheme (equations 28 and 29), we can express them as follows:  X  Na,S0  n ^Na,S0  4  +  4  (37)  C ^Nn,S0  +  5  The product distribution can also be expressed in terms of moles in the similar fashion.  C  Na,S0  N ! N  X  Na,S0  4  = C V ^Na,S0 4  C Na,S0  (39)  v  N  Na,SO  (J Na,S0 N  4  (38)  y  4  4  V  j(  +  ^Na,S0  N 5  Na SO, + Na S0 1 2  N  2  5  V  (40)  TT: The yield and product distribution are interchangeable with one another in the following manner: Yield, Y = R / B  =>  B = R/Y  Product distribution, X s = S / (S + R) S +R=B X  s  or S = B - R  = (B - R) / (B - R + R) = ((R/Y) - R) / (R/Y) = (R/Y)(1 - Y ) / ( R / Y ) = 1- Y  Product distribution = 1 - Yield Yield = 1 - Product distribution  60  (41)  Consideration of the reaction scheme (Eqns 35 and 36) showed that the generation of P M S should be mixing sensitive. Accordingly we designed our experiments to investigate those parameters important in mixing controlled reactions. These variables include the feed time, energy dissipation, volume ratio, initial reactant concentrations, and temperature.  4.3.1  Effect of Feed Time The effect o f feed time (addition rate o f sodium sulfite) was studied in P M S  generation  by keeping other  variables  constant. A reaction  was  carried out in  configuration A at 4°C using average energy o f 0.33'W/kg (N=l 66rpm) by changing only the addition time o f 1.5M sodium sulfite. The mean energy dissipation was calculated from measured power and total liquid volume in the reactor. The addition rate of sodium sulfite was found to affect both the P M S yield and concentration. For very short feed times, there is a point o f where the yield o f P M S is a maximum. This phenomenon can be explained in terms of coupling o f mesomixing effects, which decreases the yield and the energy input from jet stream of feed solution that increases the yield at the local mixing zone. For longer feed times, the energy o f the feed stream is negligible, and mesomixing effects are overcome. Once the feed time effect on P M S yield was found, similar experiments were then performed at higher mixing intensity. The results are plotted in Figures 4-1 and 4-2 as P M S yields versus feed time and P M S yields and concentration versus energy dissipation rates respectively. The figures clearly indicate that P M S yields increase with the increase of mixing intensity. Higher mixing intensity also allows achieving higher yield at considerably reduced feed time.  61  40 V = 1L; V A  30  C  H  B O  B  = 45mL  =1.5M;T=4°C  -o- E=33.4 W/kg (N==2323rpm) e=2.67 W/kg (N==560rpm) —n—e=0.33 W/kg (N==166rpm) -v1000  1500  Feed time, t , [s] f  Figure 4.1: Effect of feed time on PMS yield at different mixing intensities under atmospheric pressure  62  2000  4 0  3 5  ~i  1  1  1  1  '  1 0  r  A 8  4  TJ  >-  46 3 0  CO  o  4  O CO  _ 0_ 2 5  4  H  1 0  ,  1 5  ,  ,  ,  1 0  1 1 5  ,  ,  ,  2 0  , 2 5  ,  1  ,  !  3 0  1  U  3 5  Energy dissipation, W/kg  Figure 4.2: The Effect of Energy Dissipation on PMS Yield and Concentration (1.5M Na S0 =45ml, VT=1.09L, T=4°C) 2  3  The effect of feed time in the medium-intensity mixer in configuration B was verified by using 0.5 M Na S03 at average energy of 900W/kg (N=2323rpm) and the 2  results are reported in Table E-IV. A similar trend as using 1.5M sodium sulfite was found. These experiments indicate that for each set of reaction conditions and optimum feed time must be found to achieve maximum PMS yield. The use of configuration B that is of smaller volume allowed us to achieve higher energy dissipation rates for the same power input. 4.3.2  Effect of Energy Dissipation  The effect of energy dissipation rate was observed for different molar (0.5, 1.5, 2.0) sodium sulfite solution. A number of experiments using 0.5 M Na S03 (45 ml), 1.0 L 2  water at pH 12.6 and at 4°C were performed at different mixing intensities in reactor A. First, a reaction was performed at a given intensity to find its best feed time to achieve the  63  highest yield and then the actual reactions were run at different mixing intensities with corresponding best feed time. The results are summarized in Table E-V.  <-u  1  1  1  0  .  ,  10  .  20  ,  —,—  -  30  Energy Dissipation, W / k g  Figure 4.3: Effect of Energy Dissipation on PMS Yield (0.5M Na S0 =45ml, V =1.09L, T=4°C) 2  3  T  Figure 4.3 clearly indicates that the PMS yields are substantially increased with increase in energy dissipation. However what would be the effect of energy dissipation rate if the feed time were also kept constant? In order to verify that, another set of experiments were performed with a feed time of 12.3 seconds. As shown in Figure 4.4. (Table E-VI), the PMS yields were still increased with increasing energy dissipation. When 1.5 M Na2SC>3 was used and the feed time was changed to 11.25 s, a similar trend was observed as tabulated in Table E-VII.  64  0  200  400  600  800  1000  Energy Dissipation, W/kg  Figure 4.4: Effect of Energy Dissipation on PMS Yield (T=4C, pH=12.5, V =0.542L, 0.5M Na SO =20 ml, tf=12.3 seconds) T  2  3  It is clearly revealed from Tables E - V to E-VII that an increase of the energy dissipation resulted in increase of the PMS yield. It is also important to note that Na2S03 concentration used in reaction played an important role to increase both the yield and concentration of PMS. Once a dependency of PMS yield on the energy dissipation rate was established, the engulfment model could be used to predict the PMS yield at higher energy dissipation. Unfortunately, the reaction rates of sulfite oxidation reactions for alkaline media were not reported (Ermakov et al., 1997). However, the reported reaction kinetics for acidic media may be used in the place of alkaline rate constants because they are expected to be similar (Poskrebyshev, 1998). The ratio of the reaction rates in acidic media («40-48) were useful and suggested that a ki/k ratio of 44 could be applied in the 2  model. With ki/k ratio set at 44, k was varied in the E-model until an agreement was 2  2  65  found with experimentally determined yield values at one set of conditions (CA 0.5M , =:  T=4°C, V /V =44.5, C o=2.19 mol/m ). 3  A  B  B  As shown in Figure 4.5, the experimental PMS yields were close to what the engulfment model predicted with k2 = 4.5 m .mol" .s" and ki/k2 = 44. With these values, 3  1  1  the E-model therefore can be used to predict the PMS yield at higher energy dissipation rates. Table E-XXVII shows model predictions found using these constants at highenergy dissipation rates. The model predicts that we could achieve 64% yield PMS at 10 g/L. From personal communications, Poskrebyshev (1998) expressed his opinion that for any transition metals the rate for generation of SO3" radicals is not clear. The rate and products of sodium sulfite oxidation depend upon many conditions (rate of generation and source of radicals, pH, temperature etc.). He also informed that the mechanism of sodium sulfite oxidation does not have strong difference for alkaline and acidic medium. In that case the literature values in alkaline media for ki = (3.6±2)xl0 m .molds' and k2 2  3  3  3  1  1 1  = (120±30)xl0~ m .mof .s" respectively are reported (Ermakov et al., 1997). For such reaction rates, the E-model predicts a PMS yield over 90% at 2.6 g/L. However, we were only able to generate maximum 55% yield at 1.6 g/L of PMS.  66  i  i 111in  1  i i 11 i n  |  I  I " I I I 11 U j  I I I I 111|  I I I 111111  I  rrj  I I I 11 l l |  i—i i  I I II  100'  111  >: 1 0 rate constants Expt.  o •  literature  fit  Na S0 2  - --  -  3  -  0.5M  -  1.5M  -  -  H  "I  10"  1 I • 1  1111  10"  2  1—I I I I 1  1111  10°  1—I I I I III]  10  1  1  1 I I I T  10  111  1—1  2  ^  10'  I I IMill  1  10*  I 1 1 11 III  10  3  10°  e, W/kg  Figure 4.5: Comparison of E-model prediction and experimental values for the effect of energy dissipation on the yield of PMS (medium-intensity mixer config. B; V /V =44.5, T=4°C, H=12.5; V=545mL; tf=41-60s; Cu =0.8ppm; Fit: k,/k =44; k2=4.5m /(mol.s); Literature: ki=7.2m /(mol.s), k2=0.15 m /(mol.s) 2+  A  B  P  3  2  3  67  3  4.3.3  Effect of Sodium Sulfite Concentration  In order to remove any ambiguity about the effect of Na S03 concentration, a few 2  experiments were performed in reactor B by varying the Na S03 concentrations from 2  0.5M to 2.0M when other variables were kept constant (tf d 12 seconds, =  ee  V ier 500 =  wa  £ e 980 =  a v  W/kg,  ml). It was not possible to prepare sulfite solution of higher than 2M, because  of the solubility limitation of sodium sulfite which is 2.48M at 298K (Merck index). Thus sodium sulfite solubility will limit PMS generation as PMS yields depend on volume ratios of oxygenated water to sodium sulfite solution. Table E-VIII represents a comparison of the yield and concentration of PMS generated by using different (0.5, 1.5, 2.0) molar sodium sulfite solution. When the volume of water in reactor B was changed to 400 ml, the average energy was changed to 610 W/kg. Using 11.8 seconds as sodium sulfite addition time, the effect of Na SC>3 concentration on the PMS yield was recorded in Table E-IX. 2  These experiments clearly indicated that a compromise between the yield and concentration of PMS must be found. A higher yield always resulted in lower concentration of PMS and vice versa as shown in Fig. 4.6. This is of course important when the priority of yield or concentration is known, although a higher value for both is desirable. Since our target was to maximize the PMS yield or concentration, a 2.0M Na S0 2  3  was used in order to increase the PMS concentration and the results are given in Table EX. The concentration was improved at the expense of yield. When the reactions were performed with 0.5M Na SC>3 using same amounts of chemicals as for the previous 2  experiment (Table E-X), a considerable increase in yield (57%) was achieved but the concentration decreased to 0.9 g/L as reported in Table E-XI.  68  J U  -1  '  1  0.0  1  1  0.5  •  1.0  1  •  1  1.5  '  1  2.0  2.5  N a S 0 Concentration, M 2  3  Figure 4.6: Effect of Sodium Sulfite Concentration on PMS Yield (Medium-intensity mixer B, V =438mL, t cd=11.8 sec, T=4°C, s =610 W/kg) T  T  J U  1  -i  avc  fc  1  1  10  20  •  1  1  1  1  1  30  r  1 40  1  1  •  50  1  1  60  F e e d time, s  Figure 4.7: Effect of feed time on PMS Yield (Medium-intensity mixer B, 0.5M Na SO =20mL, V =542mL, pH=12.55, T=4°C, e =905 W/kg) 2  3  T  avc  69  Even by changing sodium sulfite concentration, the effect of feed time was still found to hold the same trend. Figure 4.7 represents experimental results showing the effect of feed time on PMS generation using 0.5 M Na2S03. 4.3.4  Effect of Volume Ratio of Oxygenated water to Sodium Sulfite  In order to find the effect of volume ratio of oxygenated water to sodium sulfite, the concentration and volume of sodium sulfite were changed by maintaining the number of moles constant. The results are reported in Table E-XIII. The lower volume ratio was found to give higher PMS yield. The effect of volume ratio was also studied by changing volume of oxygenated water (500,400,300,200 ml) for 20 ml sodium sulfite solutions of different concentrations (2.0, 1.5, 0.5M). The results are given in Tables E-XIV to E-XVI. From these tables, it can be observed that the yield of PMS decreases with decreasing amount of oxygenated water. In other words, by decreasing the volume ratio of oxygenated water to sodium sulfite solution, the yield of PMS decreases. Although a lower volume ratio should give better yield, an opposite trend was observed. The change in the volume ratio affects energy dissipation rate as well as gas hold up which are important for this reaction. From the Tables E-XIV to E-XVI, it can be seen that the changes in the energy dissipation rate are more dramatic by changing the amount of oxygenated water. As shown in Figure 4.8, using 0.5M sodium sulfite, it was possible to achieve 54.4% yield of PMS at 1.58 g/L concentration (pH=12.5, T=4°C, 20mL Na S0 , V =545 2  3  T  mL, N=2300rpm). However when 2.0M sodium sulfite was used at the same conditions, a 40% yield of PMS at 4.66 g/L was obtained.  70  5 4 ^  0.5M  1.5M  2.0M  Na S0 2  3  Figure 4.8: Effect of sodium sulfite concentration on PMS yield (pH=12.5,T=4°C, 20mL Na S0 , V =545 mL, N=2300 rpm) 2  3  T  From the previous experiments it was found that a higher volume ratio and a higher concentrated sodium sulfite could be useful to increase the PMS concentration. With this in mind, feed times were varied to run experiments using 2.0M Na S0 with two 2  3  volume ratios (10.5 and 9.4) by keeping average energy dissipation rate at around 770 W/kg (Volume of water = 450 ml IN NaOH = 9.8 ml; CuS0 (0.1 g/L) = 12 ml; 2.0 M 4  Na S0 = 45 or 50 ml; Total volume, V = 0.5168 or 0.5218 L; pH = 12.50; Reaction 2  3  T  temperature = 4°C; N=2323rpm). Table E-XVII represents these results. A volume ratio of 9.4 using 2.0M Na S0 could easily increase the PMS concentration up to 9.8 g/L, 2  3  although a yield of only 32.5% was achieved. 4.3.5  Effect of Temperature  PMS yields was found to depend on temperature. A comparison was made at different temperatures (4°C and 22-23 °C), using 500 ml cold tap-water and 20 ml sodium sulfite at different concentrations in reactor B as are observed from Tables E-XIV to  71  E-XVI.  Figure 4.9 summarizes these results (0.5M or 1.5M or 2.0M N a S 0 2  3  = 20 ml, p H  = 12.5, N = 2300 rpm). P M S Yield is higher (54.4%) at lower temperature ( 4 ° C ) .  60-1  1  1  1  -i  1.5M  0.5M  r  2.0M  Na S0 Cone. 2  3  Figure 4.9: Effect of temperature on PMS generation under atm. Pressure (NaSO=20 mL, pH = 12.5, N = 2300 rpm) 2  Table  E-XVIII  3  was generated from different experiments to understand more  clearly the effect of temperature on P M S generation. A s it indicates the P M S yields are decreased by about 20% when experiments are performed at room temperature.  4.3.6  Other Tests The P M S could not be generated when the solvent was changed to acetonitrile to  increase oxygen solubility. W e also investigated the ionic and buffer effects on this reaction. For the ionic effect 2 M sodium chloride was added to sodium hydroxide whereas sodium hydroxide was replaced by I M  sodium carbonate for the buffer effect. Both  attempts failed to generate a detectable amount of P M S .  72  From experiments, we have observed that both the formation and yield of PMS could be increased substantially as the energy dissipation rate is increased. Chen et al. (1996) also reported an increase in both PMS yield and concentration at constant mixing speed (1300 rpm) by varying oxygen pressure from 0 to 120 psi as summarized in Table E-XX. The medium intensity mixer could only achieve 54.4% yield at 1.58 g/L or 32.5% yield at 9.8 g/L, with a maximum energy dissipation of 950 W/kg. Experiments for conditions of high-energy dissipation and high pressure required were carried out in the PHT reactor in order to see if we could increase the yield further.  4.4 P Reactor HT  In the PHT mixer, the effects of energy dissipation rate, feed time (addition rate of sodium sulfite) and pressure were studied. In order to vary energy dissipation rate in the P r mixer, reactions were carried at H  2000 and 3000 rpms by keeping other variables constant. From the Table E-XXIII, it can be seen that PMS yield is practically constant at both mixing intensity. This result may be due to the absence of baffles in the PHT mixer. An increase in rotational speed may not increase the local turbulence level around reaction zone and hence there was no effect of rotational speed on the PMS yield. In order to find the effect of feed time, a reaction was run at 20 psi at temperatures 10-14°C at 240 W/kg (N=2000rpm). From the Table E-XXVIII, it is observed that the effect of feed time is similar to what was found in medium-intensity mixers. The pressure effects were verified using 2.0M and 0.5M sodium sulfites. As shown in Tables E-XXIX to E-XXX, a pressure effect was found for 2.0M sodium sulfite at 80 and 60 psi respectively. However no pressure effect was observed for 0.5M sodium sulfite as shown in Table E-XXXI. The availability of dissolved oxygen in the reactions can explain different behavior of using 2.0M and 0.5M sodium sulfite respectively. An  73  oxygen meter was used in these reactions before and after the reactions to determine the amount of dissolved oxygen. The results are given in Table E-XXXII. It is shown that enough dissolved oxygen are available when reactions take place. This is true for both 2.0M and 0.5M sodium sulfite. However oxygen depletion can be a limiting factor with 2.0M sodium sulfite because PMS yields were always increased at higher pressures. Although the oxygen meter found no deficiency in dissolved oxygen, it measures the amount of dissolved oxygen in the reactor, not in the reaction zone.  By using mass  transfer calculations (Appendix C) it is shown mathematically that when 0.5M Na S03 is 2  used, there is sufficient oxygen in the reaction zone. However, for 2 M Na S03 there is 2  insufficient oxygen in the reaction zone. The PHT mixer was good to verify the effect of pressure on PMS yield. However, PMS yield obtained by this mixer was not higher than that obtained in medium-intensity mixer. One reason for not achieving better yield in the PHT mixer is possibly due to the size of the feed nozzle (d=4mm). Another limitation of the P r mixer is the absence of H  temperature control. Attempts to use external cooling could only achieve 10±0.5°C. It should be possible to get further yield improvements if the P r mixer is designed with H  proper feeding port, cooling system and baffles to effectively utilize the power.  4.5 High-Shear Mixer Without achieving significant improvement in yield or concentration of PMS in the PHT mixer, we turned to a high-intensity mixer to take advantage of high energy dissipation. Here, 50 ml 2.0M Na S03 were added to mixer containing 2.5L water, 65mL 2  NaOH and 80mL C U S O 4 . Gas was sparged for 5 minutes before reaction was started, and also continuously during the reaction. The reaction was carried at 23°C. Table E-XXIV shows the results for varying feed times at a N=2000rpm (263 W/kg). When the same reaction was run at N=4000rpm (2080 W/kg) using feed time for 22 sec, the yield was practically constant as was observed in the PHT mixer as shown in Table E - X X V . However, the effect of feed time was observed as shown in Table E - X X V I . Running reactions only at two feed times is not conclusive. The maximum yield obtained in the  74  high-shear mixer was only 32% at 1.9 g/L PMS. The main draw back of the high shear mixer is the absence of effective cooling system to remove heat due to high-energy dissipation.  4.6 Rotor-Stator Mixer Failure to obtain further improvements in PMS yield by using different mixers that progressively increased energy dissipation led us to use the dynamic mixer. Diazo coupling reactions were carried out in the dynamic mixer in order to estimate the local energy dissipation rates (Bennington and Mmbaga, 1996). This mixer was good enough to achieve high values of local energy dissipation rates (5000W/kg). Fig. 3.8 Shows experimental set-up for rotor-stator mixer operation in batch mode. Experiments were performed in beakers to change total volume of the reaction mixture. The feed nozzle was located directly below the mixing head such that the stream of fluid from feed nozzle will be drawn through the mixer. Yields and concentrations obtained from a few experiments at room temperatures are reported in Table E-XXXIV and no improvement is observed. However an expected feed time effect was found as shown in Table E-XXXV. The maximum PMS that could be generated in the dynamic mixer was 33.7% at 2.84 g/L. The preliminary experiments using continuous operation generate low yield (810%) of PMS. This may be due to the higher ratio of flow rate of sodium sulfite to the oxygenated water entering into the reaction zone. The excess sodium sulfite concentration depletes the dissolved oxygen concentration in reaction zone. Further experimentation is necessary to establish the best condition under which this reactor to be used.  75  5 SUMMARY AND  CONCLUSIONS  The catalytic oxidation reaction of sodium sulfite by oxygen has been found to be sensitive to the mixing conditions available in mixer. Different mixing devices and mixing conditions have shown that substantial improvement in the yield and concentration of PMS (Na2SOs) is possible. The yield of PMS has been found to increase with increasing energy dissipation in medium-intensity mixers. In the PHT mixer, it was observed that oxygen pressure has no effect when 0.5M sodium sulfite solution was used. However significant improvement with increasing oxygen pressure was observed for 2.0M sodium sulfite solution. An inverse relationship between PMS yield and concentration was observed with sodium sulfite concentration. With 0.5M sodium sulfite solution, a higher yield of PMS with low concentration was obtained. Using 2M sodium sulfite, the concentration of PMS was increased, but the yield was decreased. For example, with 0.5M sodium sulfite, the yield of PMS in medium intensity mixer was 54% at 1.6 g/L. When 2.0M sodium sulfite was used, the concentration of PMS was increased to 9.8 g/L, but the yield dropped to 32.5%. Mathematical modeling shows that by adjusting mixing intensity, both the yield and concentration can be increased. PMS generation limits are expected to about 90% yield at 2.6 g/L PMS or 64% yield at 10.0 g/L or 45% yield at 18.0 g/L PMS. We have used different reactors but each of them had limitations. Under present condition, 32.5% PMS yield at 9.8 g/L concentration was achieved. The target yield of PMS (50% and 10.0 g/L) should be achievable with a proper construction of reactor by taking account of all the mixing variables.  76  6 FUTURE •  WORK  A combination of the medium-intensity mixer and the PHT reactor that will take into account of important variables for mixing sensitive reaction may achieve the target yield and concentration of PMS. The medium-intensity mixer could not be pressurized due to its construction from Plexiglas. The P T reactor contained no H  proper injection point and it failed to achieve good mixing due to lack of significant baffles in it. Moreover an effective cooling-system needs to be introduced in this reactor. The high shear mixer suffers from the lack of cooling system and more extensive study is needed for this reactor. A proper design of a reactor is essential to observe the combined effect of all mixing variables to achieve the maximum PMS yield and concentration in future. Preliminary mass transfer calculations have shown that oxygen depletion may occur in the reaction zone for a 2 M sodium sulfite application. Since we can achieve target concentration with 2 M sodium sulfite, what remains is to improve the yield of PMS. Other means of improving the oxygen availability in the reaction zone must be looked into. More studies in the dynamic mixer. Although it was not possible to achieve as high yield as desired, it is still felt that more investigations may reveal otherwise.. Determination of appropriate rate constants for alkaline media is required. A l l reported rate constants are for acidic medium. This will facilitate more accurate modeling results. In order to facilitate online analysis of PMS by FTIR methods, calibration data for PMS quantification based on oxone spectra should be obtained.  77  7 A B Cj D D E k\, N Nj N P V{ R 5" V V Xs a  p  A  B  NOMENCLATURE oxygen sodium sulfite concentration of component i, mol/m rotor diameter, m Damkohler number engulfment rate coefficient, s" reaction rate constants, m /mol.s impeller rotation speed, rev/s number of moles of component i, mol power number power, W rate of formation of substance i by reaction sodium peroxymonosulfate sodium sulfate combined volume of water, sodium hydroxide and copper sulfate, volume of sodium sulfite, m product distribution 3  1  3  3  Greek letters  p s Save  v  density, kg/m energy dissipation rate, W/kg average energy dissipation rate, W/kg kinematic viscosity, m /s 3  Abbreviations  AOX DMD DSL ECF FAO ISO PMS TCF  adsorbable organic halogen dimethyl dioxirane domestic substances list elemental chlorine-free food and agricultural organization International Organization for Standardization peroxymonosulfate totally chlorine-free  Bleaching Sequence Nomenclature  A  dimethyldioxirane 78  c  D E  H 0 P PA  Px PXA  Q X  z  DC E  0  EOP  E  P  PO  chlorine chlorine dioxide alkaline extraction hypochlorite oxygen hydrogen peroxide peroxyacetic acid peroxymonosulfuric acid mixture of peracetic and peroxymonosulfuric acid chelation enzymes ozone a combination of chlorine and chlorine dioxide alkaline extraction under mild oxygen pressure alkaline extraction under both mild oxygen pressure and hydrogen peroxide alkaline extraction under hydrogen peroxide hydrogen peroxide under oxygen pressure  79  8  REFERENCES  Allison, R. and K. McGrouther, "Improved Oxygen Delignification with Interstage Peroxymonosulphuric Acid Treatment", Tappi J., 78(10), 134-142 (1995). Allison, R., K. McGrouther and M . Ellis, "Optimizing the Effects of Interstage Peroxymonosulphate Treatment on Two-Stage Oxygen Delignification", Journal of Pulp and Paper Science, 23(9), J433-438 (1997). Anderson, J. R., "Peracids in Pulp Bleaching", Tech 97: Pulp Bleaching Course (Nanaimo, BC), September 14-19, CPPA Technical Courses, CPPA, Montreal, QC, II110 (1997). Anderson, J. R., W. Wilkinson and B. Amini, "On-Site Generation and Use of Peroxyacids in Chemical Pulp Bleaching", Proceedings of Technical Section, 81 CPPA Annual Meeting, Montreal, Canada, B59-B67 (1995). st  Argyropoulos, D. S., Y. Sun., R. M . Berry and J. Bouchard, "Reactions of Dimethyldioxirane with Lignin Model Compounds", Intern. Pulp Bleaching Conference Proceedings (Vancouver, B. C ) , Tech. Sect. CPPA, Montreal, Vol. 2, 83-91 (1994). Backstrom, H. L. J., "The Mechanism of the Autoxidation of Sodium Sulfite Solutions", J. Phys. Chem., B25, 122-138 (1934). Baeyer, A. and V. Villiger, "Ueber die Sulfomonopersaure (Caro'sche Saure)", Ber. Dtsch. Chem. Ges., 34, 853-862 (1901). Baldyga, J. and J. R. Bourne, " A Fluid Mechanical Approach to Turbulent Mixing and Chemical Reaction", Chem. Eng. Commun, 28, 231-259 (1984). Baldyga, J., J.R. Bourne, "Simplification of Micromixing Calculations. I. Derivation and Application of New Model", Chem Eng. J. 42, 83 - 92 (1989). Basta, J., L. Holtinger, A. -S. Hallstrom, and P. Lundgren, "Peroxo Compounds in TCF Bleaching", Proc. Tappi Pulping Conf., Tappi Press, Atlanta, GA, 953-956 (1994). Beal, M., "Clearing up the Bleach Market", Pulp Paper Int., 39(10), 37-39 (1997). Bennington, C.P.J, "Mixing in Pulp Suspensions" PhD Thesis, University of British Columbia, Vancouver, B.C. (1988).  80  Bennington, C.P.J, and J.P. Mmbaga, "The use of Mixing-Sensitive Chemical Reactions for The Study of Mixing in Pulp Fibre Suspensions", in N.P. Cheremisnoff (ed.), Advances in Engineering Fluid Mechanics, Gulf Publishing Co., Houston Chapter 17. (1996). Bennington, C.P.J., G. Owusu and D.W. Francis, "Gas-Liquid Mass Transfer in Pulp Suspension Mixing Operations", Can. J. Chem. Eng. 75, 56-61 (1997) Betterton, E. A. and M. R. Hoffman, "Oxidation of Aqueous S O 2 by Peroxymonosulfate", J. Phys. Chem., 92(21), 5962-5965 (1988). Bourne, J. R., C. Hilber and G. Tovstiga, "Kinetics of the Azo Coupling Reactions Between 1-Naphthol and Diazotized Sulfanilic Acid", Chem. Eng. Commun., 87, 293314(1985). Bourne, J. R. and S. Thoma, "Some Factors Determining the Critical Feed Time of a Semi-Batch Reactor", Chem Eng. Res. Des., 69, 321 - 326 (1991). Cael, J. J., "Bleaching of Lignocellulosic Materials with Monopersulphuric Acid or Its Salts", US Patent 4,404,061, September 13 (1983). Caro, H. L., "Zur Kenntniss der Oxydation Aromatischer Amine", Angew. Chem., 845850 (1898). Chen, J., "Enhanced Alkaline Peroxide Bleaching of Softwood Kraft Pulps Using a New Activator", CPPA Spring Conf. Jasper, Alberta, May 13-16 (1998). Chen, J., R. G. Robertson, J. T. Wearing, and R. Hogikyan, "Improved Generation of Peroxymonosulphate by Catalytic Oxidation of Sodium Sulphite", Paprican Report, PPR (1996), J. Indust. Eng. In Chem. Res. (to be submitted). Chiang, J. S., "Process of Preparing Peroxymonosulfate", US Patent 4,049,786, Sept. 20, (1977). Chirat, C. and D. Lachenal, "Pulp Producers Race to Keep up with Technology", Pulp Paper Int., 39(10), 29-32 (1997). Cockram, R., "Demand from Papermakers will Grow", Pulp Paper Int., 33(6), 49-56 (1991). Croon, I., "Kraft pulping: The next leap", Paper Maker, 79(8), 41-44 (1997). d'Ans, J. and W. Friedrich, "Oxygen and Oxy-Radicals in Chemistry" Ber, 43, 1880-1884 (1910). 81  Darbee, L. and J. Kolczynski, "Method of Preparing Monopersulfate", 2,955,020 (1960).  US Patent  Deister, U. and P. Warneck, "Photooxidation of SO3 " in Aqueous Solution", J. Phys. Chem., 94, 2191-2198 (1990). 2  Devuyst, E. A. P., V. A. Ettel and M . A. Mosoiu,"An Oxidant of Unexpected Utility" Chemtech, 9(7), 426-427 (1979). Ermakov, A. N., G. A. Poskrebyshev and A. P. Purmal, "Sulfite Oxidation: The State-ofthe-Art of the Problem", Kinetics and Catalysis, 38(3), 295-308 (1997). Fossum, G. K. and S. L. Haggstrom, US Patent 4,222,819, Sept. 16, (1980). Gratzl, J. S., "General Principles of Delignification and Bleaching with Oxygen, Hydrogen Peroxide and Ozone", Proc, 1991 Workshop on Emerging Pulping and Chlorine-Free Bleaching Technologies, Sponsored by North Carolina State University, Raleigh, NC. March 18-20 (1991). Hammer, B., H. Michaud, and S. Weiss, "Process for Bleaching and Delignifying Cellulose-Containing Products", US Patent 5,034,096, July 23, (1991). Kapadia, P. C , H. G. Tessier, and S. Langlois, "A Non-Chlorine Repulping Aide for Wet Strength Resin Destruction", CPPA Tech. Sect. Annual Mtg., Montreal, QC, A62-A66 (1993). Kronis, J. D., "The Continued Evolution of the Bleaching Sequence via ECF to TCF Using the DegOXTM Process", TAPPI/NCSU Emerging Pulp and Bleaching Technologies Workshop Notes, TAPPI PRESS, Atlanta, GA. (1995). Lachenal, D., L. Bourson, M . Muguet and A. Chauvet, "Lignin Activation Improves Oxygen and Peroxide Delignification", Cell. Chem. Tech., 24(5), 593-601 (1990). Lachenal, D. and N . B. Nguyen-Thi, "Rationalization of Chlorine-Free Bleaching" Proc. Seventh Int'l Symp. On Wood and Pulping Chem., Beijing, P. R. China, May 25-28, 1, 166-169(1993). Lee, C-L., R. W. Murray and K. Hunt, "Activated Oxygen. A Selective Bleaching Agent for Chemical Pulps. Part I: Laboratory Bleaching with Isolated and in-situ Generated Activated Oxygen", Pulp Paper Sci., 20(5), J125 (1994). Lee, C-L., R. W. Murray and K. Hunt, "Dimethyldioxirane as a Non-Chlorine Agent for Chemical Pulp Bleaching", Tappi J., 76(11), 137-140 (1993). 82  Lee, C-L., R. Hogikyan, A. Skothos, G. Sacciadis, J. T. Wearing, C. W. Oloman, B. Amini, W. W. Teetzel, D. L. Carmichael, P. Fetissoff, L. Tench and S. Harper, "Activated Oxygen - A Selective Bleaching Agent for Chemical Pulps. Part IV. Totally Chlorine-Free Bleaching: Pilot Trial and Process Designs", Proceedings of 1994 Intern. Pulp Bleaching Conference, Vancouver, B. C , Canada, June 13-16, 273-279 (1994). Lee, C-L., R. W. Murray, K. Hunt, J. T. Wearing, R. Hogikyan, C. W. Oloman. and J. Chen, "Bleaching of Lignocellulosic Material with in-situ Generated Dioxirane", US Patent 5,366,593, November 22, (1994). Liebergott, N., B. van Lierop, B. C. Garner and G. J. Kubes, "Bleaching a Softwood Kraft Pulp without Chlorine Compounds", Tappi J., 76(8), 76-86 (1984). Liebergott, N . , B. van Lierop and A. Skothos, "A Survey of the Use of Ozone in Bleaching Pulps, Part 1", Tappi J., 75(1), 145-163 (1992). Linek, V. and V. Vacek, "Chemical Engineering Use of Catalyzed Sulfite Oxidation Kinetics for the Determination of Mass Transfer Characteristics of Gas-Liquid Contactors", 36(11), 1747-1768 (1981). Marsh, C. and J. O. Edwards, "The Free-radical Decomposition of Peroxymonosulfate", Prog. Reaction Kinetics, 15, 35-75 (1989) McDonough, T. J., "Oxygen Delignification" in "Pulp Bleaching, Principles and Practice", (C. W. Dence and D. W. Reeve, Eds.), TAPPI PRESS, Atlanta, GA, 215-237 (1996). McDonough, T. J., A. Marquis and A. J. Ragauskas, "Totally Chlorine-free Bleaching with Dimethyldioxirane", Intern. Pulp Bleaching Conference Proceedings (Vancouver, B. C ) , CPPA Tech. Sect., Montreal, QC, 47-52 (1994). McElory, W. J. and U. Deister, "Kinetics of the Oxidation of Sulfur (IV) by Peroxymonosulfate", National Power Research Report No. ESTB/L/0173/R90 (1990). McGrouther, K. G. and R. W. Allison, "Improved Oxygen Delignification with Peroxyacid Treatment", Appita J., 47(3), 238-242(1994). Munro, F. C , "The Impact of Oxygen Delignification in Espanola's Softwood Bleaching Sequence", Pulp Paper Can., 88(7), 34-39 (1987). Nelson, P. J., "Elemental Chlorine Free (ECF) and Totally Chlorine Free (TCF) Bleaching of Pulps" in "Environmentally Friendly Technologies for the Pulp and Paper  83  Industry", (R. A. Young and M . Akhtar, Eds.), John Wiley and Sons, Inc., 215-251 (1998). Paul, E. L. and R. E. Treybal, "Mixing and Product Distribution for a Liquid-Phase, Second-Order, Competitive-Consecutive Reaction", AIChE J., 17(3), 718-724 (1971). Poskrebyshev, G. A., Private communications (1998) Ragauskas, A. J., "Investigation of Dimethyldioxirane as a Bleaching Reagent for Kraft Pulp", Tappi J., 76(7), 87-90 (1993). Roy, B. P., B. van Lierop, A. Audet and R. Berry, "High-Temperature Alkaline Peroxide Bleaching of Kraft Pulps", Proceedings of 1995 Tappi Pulping Conference, Chicago, IL, October 1-5, Book 2, p. 771-778 (1995). Seccombe, R. C , H. Martens, and A. Haakana, "Practical Experience with the Use of Caro's Acid to Produce High Brightness TCF Pulps", Proc. 48 Appita Annual General Conference, Melbourne, Australia, 139-146 (1994). th  Springer, E. L. and J. D. McSweeny, "Method for Producing Salts of Monoperoxysulfuric Acid and Simultaneously Bleaching Pulp", US Patent No. 4,756,800, (1988). Springer, E. L. and J. D. McSweeny,"Treatment of Softwood Kraft Pulps with Peroxymonosulfate before Oxygen Delignification", Tappi J. 76(8), 194-199 (1993). Troughton, N . A., F. Desprez and J. Devenyns, " Peracids: The Pathway to High Brightness Kraft Pulps" * Proceedings of 1994 International Non-Chlorine Bleaching Conference, Session 10-1, Amelia Island, Florida, USA, March 6-10, (1994).  84  9  Appendices  85  A. Procedures of iodometric titration for peroxymonosulfate analysis •  Place 50 ml deionized water, 15 ml 20% H S 0 and 5 ml I N KI in 2  4  a 250 ml erlenmeyer flask. •  Pipet 10 ml of peroxymonosulphate sample solution into the erlenmeyer flask.  •  Titrate the above solution with 0.1N Na S 0 solution to the end 2  2  3  point using starch solution as the indicator.  Calculation of both the PMS yield and concentration of PMS Suppose 1.0L water, 20 ml CuS04, 84 ml 1.5 M Na S0 were used to 2  3  generate peroxymonosulfate. Actual number of moles of sodium sulfite used: = molarity * volume of Na S0 used 2  3  = 1.5 mol/L* 0.084L = 0.126 moles. Suppose it required 4.91 ml of 0.1 N Na S 0 solution (as an example) to 2  2  3  titrate 10 ml sample from the reaction mixture, then we can use the formula below: M = Mi W V 2  2  Where, M = molarity of the reaction mixture, M i = molarity of sodium 2  thiosulfate solution = 0.1 moles/L, Vj = volume of sodium thiosulfate used = 4.91 ml, V = volume of reaction mixture used = 10 ml 2  86  It is evident from the reactions below that one mole of iodine reacts with two moles of sodium thiosulfate.  S 0 - + 2KI+H S0  •  2S0 - + l2 + K2S04  •  2Nal + Na S 0  2  5  2  + 2Na S 0 2  2  4  3  S 0 - + 2KI + H S0 + 2Na S 0  2  4  2  2  4  2  2  6  • 2S0 " + + K S 0 + N a S 0  2  5  4  2  3  4  2  4  2  4  6  Hence from the stoichiometry of this reaction, we have: M = M i Vi/2V 2  2  M = (0.1 mole/L * 4.91 ml)/(2 * 10 ml) = 0.02455 mol/L 2  Concentration of Na S05: 2  = 158 g/mol * 0.02455 mol/L = 3.88 g/L Yield of Na SO : 2  s  = S05 "moles produced / S03 "moles reacted 2  2  = 0.02455 mol/L* (1.0 L water + 0.02 L CuS0 soln. 4  +0.084 L Na S0 soln.) * 100 / 0.126 mol 2  = 21.51%  87  3  B. Calculation of 0.8 ppm C u  Concentration of C11SO4 = 0.8 ppm = 0.8 mg Cu/L Molecular weight of CuS0 .5H 0 = 249.68 g 2  4  and for Cu = 63.54 g It means 0.8 mg Cu is present in (249.68 g * 0.8 mg / 63.54 g) = 3.14 mg CuS0 .5H 0. So 3.14 mg CuS0 .5H 0 in 1.0 L water is required to get 0.8 2  4  2  4  ppm Cu concentration.  C . Preliminary estimation of the ratio of moles of 0 Na S0 2  3  to the moles of  2  available  Experimental  conditions  for  sample  calculation  (0.5M  NagSOj,  54.4%PMS yield at 1.58 g/L)  0.5MNa SO = 20 ml 2  3  Feed time = 11.8 sec Reactor volume of medium-intensity mixer B = 870 ml Total volume of liquid = V 2 o + V o H + V so4 + V 2 s o H  Na  Cu  Na  3  = (500+ 12+ 13 +20) ml = 545 ml Gas hold-up = (870 - 545) ml = 325 ml = 0.325 L Reaction temperature = 4°C Pressure = 1 atmosphere Average energy dissipation rate = 900 W/kg Actual power = (900 W/kg) X 0.545 kg = 490.5 W Power per unit reactor volume = (490.5 W / 0.87 L) X (1 L / 10" m ) 3  88  3  = 563.8 X 10 W/m 3  3  Gas-Liquid volumetric mass transfer coefficient, ki,a is given by the following equation (Bennington et al., 1997): k a = 0.0006 s ^ X g  1 5 6  L  In this case, Power per unit reactor volume, s = 5 63.8 X 10 W7 m 3  3  Gas void fraction in the mixer, X = 0.325 L g  k a = 0.0006 (563.8 X 10 )°- (0.325) 3  72  156  L  = 1.439 s"  1  The rate of oxygen mass transfer can be measured using either the change in sulfite or sulfate concentration with time and is given by Bennigton et al. (1997): dC, dt—  =  k  L  a  \  c  -C  b  n  as bulk oxygen concentration is equal to 0, oxygen solubility at 4°C is determined by the following correlation (Linek and Vacek, 1981):  a = 5.909 x 10-6 exp{1602.1/T - 0.9407C 2so4 / (1+0.1933  C  Na  Na2  so4}  M/atm where T = 273 + °C Using above correlation, oxygen solubility at reaction condition = 0.0014 mol/L ( d C / dt) = 1.439 s" x 0.0006 mol.L" x 0.545 L = 1.09 x 10" mol.s" 1  1  3  1  02  (dCS03 " / dt) = (20 ml / 12 s) = 1.667 ml.s" = 8.335 x 10" mol.s" 2  1  4  1  Ratio of oxygen to sulfite = (1.09 x 10" / 8.335 x 10" ) = 1.31 3  89  4  Experimental conditions for sample calculation (2.0M Na^SOa, 32.5% PMS yield at 9.8 g/L) 2.0M Na S0 = 50 ml 2  3  Feed time = 23 sec Reactor volume of medium-intensity mixer B = 870 ml Total volume of liquid = V o + V o H + Vcuso4 + H 2  Na  V  N a 2  so3  = (450 + 9.8 + 12+ 50) ml = 522 ml Gas hold-up = (870 - 522) ml = 348 ml = 0.348 L Reaction temperature = 4°C Pressure = 1 atmosphere Average energy dissipation rate = 770 W/kg Actual power = (770 W/kg) X 0.522 kg = 401.94 W Power per unit reactor volume = (401.94 W / 0.87 L) X (1 L / 10" m ) 3  = 462X 10 W/m 3  k a = 0.0006 e°- X 72  L  3  3  L56 g  In this case, Power per unit reactor volume, s = 462 X 10 W/ m 3  3  Gas void fraction in the mixer, X = 0.348 L g  k a = 0.0006 (462 X 10 ) (0.348) ' 3  a72  1 56  L  = 1.387 s"  1  (dC / dt) = 1.387 s" x 0.0006 mol.L" x 0.522 L = 1.01 x 10" mol.s" 1  1  3  1  02  (dCS03 - / dt) = (50 ml / 23 s) = 2.174 ml.s" = 4.348 x 10" mol.s" 2  1  3  1  Ratio of oxygen to sulfite = (1.052 x 10" / 4.348 x 10" ) = 0.23 3  90  3  D . Error analysis  Quantities measured: V i = Tank volume (stock solution) V2 = Sodium sulfite solution V 3 = Sample aliquot V4 = Sample titration Errors in measured quantities: Stock solution, AVi = 1000 ± 10 ml (large measuring cylinder) Sodium sulfite solution, AV2 = 45 ± 2 ml (pump metering) Sample aliquot, AV = 10 + 0.1ml (pipette) 3  Sample titration, AV4 = 3 ± 0.05 ml, just an example (burette) .-. AVi = 10, AV = 2, AV = 0.1, A V = 0.05 2  % Error =  3  + (^) + (^) . + (*f f 2  2  91  4  = 0.0495 = «5%  E. List of Tables Table E-I: Effect of feed time in the medium intensity mixer A (N=166 rpm; 8 e 0-33 W/kg; Reactor temperature = 4°C; Volume of water = 1.0 L; Volume of NaOH = 25 ml; Volume of CuS0 = 20 ml; 1.5 M Na S0 = 4.5 ml; Total volume = 1.09 L) =  aV  4  2  Feed Time (sec)  Titration (ml)  Yield (%)  63 320  0 1.1  0 8.88  Cone, of Na S0 (g/L) 0 0.869  620 960 1720  2.95 2.5 2.3  23.8 20.2 18.6  2.33 1.97 1.82  2  3  5  Table E-II: Effect of feed time in the medium intensity mixer (N=560 rpm (Save = 2.67 W/kg); Reactor temperature = 4°C; Volume of water = 1.0 L; Volume of NaOH = 25 ml; Volume of CuS0 = 20 ml; 1.5 M ; Na S0 = 45 ml; Total volume = 1.09 L 4  2  Feed time (sec)  Titration (ml)  Yield (%)  6  0  0  Na S0 (g/L) 0  60  1.8  14.5  1.42  170 295 455 855 1660  3.6 3.4 3.2 2.8 2.4  29.1 27.4 25.8 22.6 19.4  2.84 2.69 2.53 2.21 1.90  92  Cone, of 2  5  3  Table E-III: Effect of feed time in the medium intensity mixer A (N=2323 rpm (e e = 33.4 W/kg); Reactor temperature = 4°C; Volume of water = 1.0 L; Volume of NaOH - 25 ml; Volume of CuS0 = 20 ml; 1.5 M Na S0 = 45 ml; Total volume = 1.09 L aV  4  2  Feed time (sec)  Titration (ml)  Yield (%)  0  0  0  Cone, of Na S0 (g/L) 0  48.3 53.94  4.0 4.3  32.30 34.72  3.16 3.397  61.3 116.94 127.88 137.53 950 1900  4.4 3.9 3.7 3.6 2.9 2.7  35.53 31.49 29.87 29.07 23.41  3.476 3.089 2.923 2.844 2.175  21.8  2.017  2  3  5  Table E-IV: Effect of feed time in the medium intensity mixer B (s —720 W/kg; N-23 lOrpm; Reactor temperature = 4°C; Volume of water = 0.5 L; Volume of NaOH = 12 ml; Volume of CuS0 = 10 ml; 0.5 M Na S0 = 20 ml; Total volume = 0.542 L ave  4  2  Feed time (sec)  Titration (ml)  Yield (%)  Cone, of Na S0 (g/L)  9.05  1.5  40.6  1.18  10.1  1.6  43.4  1.26  11.8  1.9  51.5  1.50  12.0 13.5 15.0  2.0  54.2 51.5  1.58 1.50  48.8  1.42  46.1 40.6 40.6  1.34 1.18 1.18  15.8 24.0 58.0  1.9 1.8 1.7 1.5 1.5  93  2  5  3  Table E-V: Effect of energy dissipation in the medium intensity mixer A (Reactor temperature = 4°C; Volume of water = 1.0 L; Volume of NaOH = 25 ml; Volume of CuS0 = 20 ml; 0.5 M Na S0 = 45 ml; Total volume = 1.09 L 4  N  Save  (rpm)  (W/kg) 1  2  3  Feed time  Titration  Yield  Cone, of  (sec)  (ml)  (%)  Na SO (g/L) 2 5  166  0.33  630  1.1  26.6  0.869  560  2.67  182  1.4  33.9  1.11  1360  16.7  153  1.6  38.8  1.26  2323  33.4  62  1.8  43.6  1.42  Table E-VI: Effect of energy dissipation in the medium intensity mixer B (Reactor temperature = 4°C; Volume of water = 0.5 L; Volume of NaOH = 12 ml; Volume of CuS0 = 10 ml; 0.5 M Na S0 = 20 ml; Total volume, V = 0.542 L; Feed time, t d = 12.3 s 4  2  3  T  f e e  N (rpm)  Save  (W/kg)  Titration  PMS  Cone, of  (ml)  Yield  Na SO (g/L) 2 5  (%) 536  11  1.3  35.2  1.03  1160  130  15  40.6  1.18  1620  348  1.65  44.7  1.30  1900  742  1.9  51.5  1.50  2100  852  2.0  54.2  1.58  94  Table E-VII: Effect of energy dissipation in the medium intensity mixer B (Reactor temperature = 4°C; Volume of water = 0.5 L; Volume of NaOH = 12 ml; Volume of CuS0 = 10 ml; 1.5 M Na S0 = 20 ml; Total volume, V = 0.542 L; Feed time, t d = 11.25 s 4  2  3  T  fee  N  Save  Titration  Yield  Cone, of  (rpm)  (W/kg)  (ml)  (%)  535  11  0  0  Na SO (g/L) 2 5 0  1138  145  2.6  23.5  2.05  1610  328  3.9  35.2  3.08  2100  864  4.2  37.9  3.32  2250  890  4.4  39.7  3.47  Table E-VIII: Effect of Na S0 cone, in the medium intensity mixer B (s = 900 W/g, N=2300rpm; Reactor temperature = 4°C; Volume of water = 0.5 L; Volume of NaOH = 12 ml; Volume of CuS0 = 13 ml; (0.5,1.5, 2.0 M) Na S0 = 20 ml; Total volume, V = 0.545 L; Feed time, t d = 12 s 2  3  ave  4  2  3  T  Na S0 Molarity (M) 0.5 1.5 2.0 2  3  fee  Titration (ml)  PMS Yield (%)  PMS Cone. (g/L)  2.0 4.9 5.9  54.5 44.5 40.2  1.58 3.87 4.66  Table E-LX: Effect of Na S0 cone, in the medium intensity mixer B (s e = 610 W/kg, N=2300rpm; Reactor temperature = 4°C; Volume of water = 0.4 L; Volume of NaOH = 10 ml; Volume of CuS0 = 8 ml; (0.5,1.5, 2.0 M) Na S0 = 20 ml; Total volume, V = 0.438 L; Feed time, t d = 11.8 s 2  aV  3  4  T  Na S0 Molarity (M) 0.5 1.5 2.0 2  3  2  fee  Titration (ml)  PMS Yield (%)  PMS Cone. (g/L)  2.1 5.5 6.3  46.0 40.1 34.5  1.66 4.35 4.98  95  3  Table E-X: The effect of energy dissipation (medium-intensity mixer A) on PMS yield (Reactor temperature = 4°C; Volume of water = 2.0 L; IN NaOH = 50 ml; CuS0 (0.1 g/1) = 40 ml; 2 M Na S0 = 45 ml; pH = 12.6; Total volume, V = 2.135 L) 4  2  3  T  N (rpm)  (W/Kg)  2040 2145  197 230  Save  Feed time (sec)  Titration (ml)  PMS Yield (%)  PMS Cone.  3.6 3.7  42.7 43.9  2.84 2.92  60.2 61.9  Table E-XI: The effect of energy dissipation (medium-intensity mixer A) on the yield of PMS (Reactor temperature = 4°C; Volume of water = 2.0 L; I N NaOH = 50ml; CuS0 (0.1 g/L) = 40 ml; 0.5 M Na S0 = 45 ml; pH = 12.65; Total volume, V = 2.13 5 L) 4  2  3  T  N (rpm)  (W/kg)  2085 2030  227 211  time  (sec)  Titration (ml)  PMS Yield (%)  PMS Cone. (g/L)  59.6 58.03  1.2 1.1  56.9 52.2  0.948 0.869  Feed  Save  Table E-XII: The effect of feed time (medium-intensity mixer B) on PMS generation (Volume of water = 0.5 L; IN NaOH = 12 ml; CuS0 (0.1 g/L) = 10 ml; 0.5 M Na S0 = 20 ml; Total volume, V = 0.542 L; pH = 12.55; Reaction temperature = 4°C; s e = 905 W/kg, N=2400rpm) 4  2  3  T  aV  Feed time (sec) 58 20 15.8 15 13.5 12.3 11.8 9.0  Titration (ml)  PMS Yield (%) 40.6 40.6 46.1 48.8 51.5 54.2 54.2 40.6  1.5 1.5 1:7 1.8 1.9 2.0 2.0 1.5  96  PMS Cone. (g/L) 1.19 1.19 1.30 1.42 1.50 1.58 1.58 1.19  Table E-XIII: The effect of volume ratio (medium-intensity mixer B) on PMS yield (Volume of water = 0.5 L; IN NaOH = 12 ml; CuS0 (0.1 g/L) = 13 ml; pH= 12.53; N=2400rpm) 4  Molarity (M) 2.0 1.0 0.5 0.2  Na S0 added (ml) 5 10 20 50 2  3  Volume ratio (VA/VB)  100 50 25 10  Save  (W/kg) 845 845 850 875  Feed time (sec) 6 10.8 19.3 45  Titra. (ml)  PMS Yield %  0.8 1.6 1.7 1.95  21.2 42.8 46.3 56.1  PMS Cone. (g/L) 0.632 1.26 1.34 1.54  Table E-XIV: The effects of initial volume (medium-intensity mixer B) on PMS yield (20mL 2.0 M Na S0 , pH=12.5, N=2300rpm) 2  3  Feed Time (sec) 11.7  Save  Titra.  4  Volume NaOH (ml) 12  (W/kg) 915  400  4  9.5  12.2  300 500  4 22  7.2 13.5  400  22  12  Volume Water ml 500  Temp. °C  (ml) 5.9  PMS Yield (%) 40.1  PMS Cone. (g/L) 4.66  522  6.3  34.8  4.98  11.56 11.7  194 910  6.2 4.5  26.3 30.7  4.90 3.59  12  662  4.3  23.9  3.40  Table E-XV: The effects of initial volume (medium-intensity mixer B) on PMS yield (20mL 1.5 M Na S0 , pH=12.5, N=2300rpm) 2  3  (ml)  PMS Yield (%)  PMS Cone. (g/L)  905  4.9  44.4  3.87  610 476 587  5.5 5.8 4.25  40.5 32.8 31.4  4.34 4.58 3.36  Volume Water (ml)  T °C  Volume NaOH (ml)  Feed Time (sec)  Save  Titra.  (W/kg)  500  4  12  11.6  400 300 400  4 4 22  9.5 7.2 12  11.6 11.7 11.6  97  Table E-XVI: The effects of initial volume (medium-intensity mixer B) on PMS yield (20mL 0.5 M N a S 0 , pH=12.5, N=2300rpm) 2  3  Feed Time (sec) 12.0  Save  Titra.  4  Volume NaOH (ml) 12  (W/kg) 900  400  4  9.5  11.8  300  4  7.2  500  22.5  400  22.5  Volume Water ml 500  T (°C)  (ml) 2.0  PMS Yield (%) 54.4  PMS Cone. (g/L) 1.58  700  2.1  46.4  1.66  11.7  410  2.5  42.4  1.97  14.5  11.2  905  1.5  41.0  1.19  12  11.7  782  1.7  37.7  1.34  Table E-XVII: The effect of changing volume and feed time in mediumintensity mixer B (Volume of water = 0.45 L; IN NaOH = 9.8 ml; CuS0 (0.1 g/L) = 12 ml; 2.0 M Na S0 = 45 or 50 ml; Total volume, V = 0.5168 or 0.5218 L; pH = 12.50; Reaction temperature = 4°C; s e = 770 W/kg; N=2323rpm) 4  2  3  T  aV  Volume Na S0 (ml) 45 45 45 45 50 50 50 50 50 2  3  Feed Time (sec) 24 34 46.3 60 36.5 30.6 28 25.8 22.6  Titration (ml) 9.0 10.9 10.4 10.0 11.3 12 12.2 12.4 12.4  98  PMS Yield (%) 25.8 31.3 29.8 28.7 29.5 31.3 31.8 32.5 32.5  PMS Cone. (g/L) 7.11 8.61 8.22 7.90 8.93 9.48 9.64 9.80 9.80  Table E-XVTII: The effect of temperature on the yield of PMS in mediumintensity mixer B Molarity (M) 2.0  0.5  Temperature 4°C  Temperature 22-23 °C  40.1% yield  30.7% yield  4.66 g/L concentration  3.59 g/L concentration  54.4% yield  41.0% yield  1.58 g/L concentration  1.19 g/L concentration  Table E-XIX: Oxygen Solubility in Water at 23°C calculated from "Handbook for Engineers" Pressure (atm) 1 5.26 9.21 10.7  Oxygen Solubility (mol/m ) 1.31 6.25 10.5 12.0  Oxygen Solubility (g/L) 0.042 0.199 0.336 0.383  3  Table E-XX: Comparison of PMS yields  Chen et al.  CURRENT  Pressure (psi) 50  PMS Yield (%) 18.0  PMS Cone. (g/L) 3.20  120  22.0  3.88  54.4  1.60  32.5  9.80  atmospheric  99  Table E-XXI: PHT Mixer (Reaction temperature = 23 °C; Volume of water = 1.5 L; IN NaOH = 36 ml; CuS0 (0.1 g/L) = 40 ml; 1.5 M Na S0 = 60 ml; Total volume, VT - 1.636 L; Feed time =11.5 sec; s = 240 W/kg; N=2000rpm) 4  2  3  ave  Pressure (psi)  Oxygen Solubility mg/L 15.6 22.5 39.3  20 40 100  Titration (ml)  PMS Yield (%)  PMS Cone.  3.8 4.1 3.8  34.5 37.3 34.5  3.0 3.24 3.0  (g/L)  Table E-XXII: The effect of feed time in PHT Mixer (Reaction temperature = 23°C; Volume of water = 1.5 L; IN NaOH = 36 ml; CuS0 (0.1 g/L) = 40 ml; 1.5 M Na S0 = 60 ml; Total volume, VT = 1.636 L; Pressure =100 psi; e = 240 W/kg, N=2000rpm) 4  2  3  ave  Feed time (sec)  Titration (ml)  PMS Yield (%)  PMS Cone. (g/L)  11.5 15.2 30.0 60.0  3.8 .4.0 3.5 3.4  34.5 36.4 31.8 30.9  3.01 3.16 2.77 2.69  Table E-XXIII: PHT Mixer (Reaction temperature = 23 °C; Volume of water = 1.5 L; IN NaOH = 36 ml; CuSO (0.1 g/L) = 40 ml; 1.5 M N a S 0 = 60 ml; Total volume, VT = 1.636 L; Pressure = 20 psi; Feed time = 20 sec) 4  N (rpm) . 2000 3000  2  3  (W/kg)  Titration (ml)  PMS Yield (%)  PMS Cone. (g/L)  240 813  4.4 4.5  40.0 40.9  3.48 3.56  Save  100  Table E-XXIV: The effect of feed time in high-intensity mixer (Reaction temperature = 23°C; Volume of water = 2.5 L; IN NaOH = 65 ml; CuS0 [0.1 g/L] = 80 ml; 2.0 M Na S0 = 50 ml; Total volume, V = 2.695 L; e 263 W/kg, N=2000rpm) 2  3  T  4  a1  Feed time (sec)  Titration (ml)  PMS Yield (%)  PMS Cone. (g/L)  22 99 105 115  2.1 1.5 1.4 1.4  28.3 20.2 18.9 18.9  1.66 1.19 1.11 1.11  Table E-XXV: Effect of energy dissipation in high-intensity mixer on PMS yields (Reaction temperature = 23°C; Volume of water = 2.5 L; IN NaOH = 65 ml; CuS0 [0.1 g/L] = 80 ml; 2.0 M Na S0 = 50 ml; Total volume, V t = 2.695 L; s = 263 W/kg, N=2000rpm, tf=22s) 4  2  3  ave  N (rpm)  (W/kg)  Titration (ml)  PMS Yield (%)  PMS Cone. (g/L)  2000 4000  263 2080  2.1 2.2  28.3 29.6  1.66 1.74  £ ve a  Table E-XXVI: The effect of feed time in High intensity mixer (Reaction temperature = 23°C; Volume of water = 2.5 L; IN NaOH = 65 ml; CuS0 (0.1 g/L) = 80 ml; 2.0 M Na S0 = 50 ml; Total volume, V = 2.695 L; e 2080 W/kg, N=4000rpm) 2  3  T  Feed time (sec)  Titration (ml)  PMS Yield (%)  PMS Cone. (g/L)  22 18  2.2 2.4  29.7 32.3  1.74 1.90  ,  101  4 n  Table E-XXVTI: Model predictions at high-energy dissipation rate (V /V =44.5; T=4°C; pH=12.5; V=545mL;ki/k =44; C =2.0M) A  B  2  B0  (W/kg)  Pressure (atm)  Yield (%)  Cone. (g/L)  111 111 5000 5000 5000  1 10.72 1 10.72 10.72  23.9 24.1 39.5 45.4 64.0  9.20 9.48 15.5 17.8 10.0  Save  Table E-XXVIII: Effect of feed time in PHT reactor (Volume of water = 1.8 L; IN NaOH = 50 ml; CuS0 (0.1 g/L) = 50 ml; 2.0 M Na S0 = 60 ml; Total volume, V = 1.96 L; e = 240 W/kg, N=2000rpm) 4  T  Feed time (sec) 12 17 20 27.2 34.5 46 78  2  3  ave  Temp. (°C) 12.8-15.0 10.0-13.3 10.5-14.0 10.6-13.7 10.3-13.3 10.9-15.2 10.9-14.6  Titration (ml) 0.5 2.9 3,8 3.95 4.2. 4.0 3.6  PMS yield (%) 4.10 23.7 31.0 32.3 34.3 32.7 29.4  PMS cone. (g/L) 0.401 2.29 3.01 3.12 3.32 3.16 2.84  Table E-XXIX: Effect of pressure in PHT reactor on PMS yield (Volume of water = 1.8 L; IN NaOH = 50 ml; CuS0 (0.1 g/L) = 50 ml; 2.0M Na S0 = 60 ml; Total volume, V = 1.96 L; e = 240 W/kg; N=2000rpm; V / V = 31.67) 4  T  Pressure (psi) 20 40 80 100  Feed time (sec) 13.0 13.8 16.3 18.3  2  ave  Temp. (°C) 10.4-14.0 9.2-11.7 9.0-12.3 8.5-12.0  A  Titration (ml) 2.5 4.0 5.3 4.7  102  Yield (%) 20.4 32.7 43.3 38.4  3  B  Cone. 1.98 3.16 4.19 3.71  Table E-XXX: Effect of pressure in P T reactor on P M S yield (Volume of water = 1.8 L; IN NaOH = 50 ml; CuS0 (0.1 g/L) - 50 ml; 2.0 M Na S0 = 60 ml; Total volume, V = 1.96 L; e = 240 W/kg; N=2000rpm; V / V = 31.67) H  4  T  Pressure (psi) 20 25 60 80 100  Feed time (sec) 9.87 12.28 11.62 12.58 13.4  2  ave  A  Titration (ml) 2.25 2.80 4.50 4.40 4.40  Temp. (°C)  10.6-14.5 8.7-12.3 9.6-14.1 9.9-13.0 8.5-12.1  Yield (%) 18.4 22.9 36.8 35.9 35.9  3  B  Cone. (g/L) 1.78 2.21 3.56 3.48 3.48  Table E-XXXI: Effect of pressure in P T reactor on P M S yield (Volume of water = 1.8 L; IN NaOH = 50 ml; CuS0 (0.1 g/L) = 50 ml; 0.5 M Na S0 = 60 ml; Total volume, V = 1.96 L; s e = 240 W/kg; N=2000rpm; V / V = 31.67) H  4  T  Pressure (psi) 20 40 60 80 80  Feed time (sec) 15.40 20.20 25.12 25.05 15.97  2  A  aV  Titration (ml) 1.30 1.25 1.25 1.30 1.40  Temp. (°C) 13.5-16.0 10.8-13.9 10.3-12.6 9.9-11.3 9.8-13.1  Yield (%) 42.5 40.8 40.8 42.5 45.7  3  B  Cone. (g/L) 1.03 0.99 0.99 1.03 1.11  Table E-XXXII: Oxygen solubility measured by oxygen meter in P T Reactor H  (psi)  Initial solubility (mg/L)  Initial reaction temperature (°C)  Solubility after reaction (mg/L)  Final reaction temperature (°C)  20 60 80  17.2 30.9 38.0  10.2 10.6 10.4  15.6 29.5 34.0  14.5 14.0 14.1  Pressure  103  Table E-XXXIII: Mass balance of S 0 \ S0 ", and S0 " ions using FTIR and Iodometric Titration (P T mixer, N=2000rpm, pH = 12.5, T = 15°C) 2  2  3  2  5  4  H  V (ml)  Save  T  (W/kg)  2045 1995 1955 2020 1955  Sulfite added (ml)  S0 " moles added  150 100 60 125 60  0.147 0.100 0.061 0.124 0.061  195.6 200.5 204.6 198.0 204.6  2  3  Generated SO5 " moles by titration method 0.034 0.029 0.021 0.032 0.018 2  7  SO4 " moles by FTIR spectrum 0.119 0.075 0.043 0.089 0.040  Total &  SO5 "  S O 4 " moles 2  in product 0.152 0.104 0.064 0.121 0.058  Table E-XXXIV: Dynamic rotor-stator mixer (Reaction temperature = 23°C; pH =12.5; Mixer setting # 7) V (L) T  0.3542 2.3 1.17 1.405  0.5M Na S0 (ml) 37 200 125 150 2  (W/kg)  t (sec)  Titra (ml)  3058 5465 4133 4472  44 80 87 56  3.4 2.2 3.2 3.6  Save 3  f  PMS Yield (%) 32.6 25.3 30.0 33.7  PMS Cone. (g/L) 2.69 1.74 2.53 2.84  Table E-XXXV: Effect of feed time in dynamic mixer (Reaction temperature = 23°C; Volume of water = 0.5 L; IN NaOH = 12 ml; CuS0 (0.1 g/L) = 14 ml; 2.0 M Na S0 = 20 ml; Total volume, V = 0.546 L; e = 3380 W/kg; Mixer setting #7) 4  2  Feed time (sec) 7 11 15 25 40  3  T  Titration (ml) 0.5 3.7 4.7 1.5 1.0  PMS Yield (%) 3.4 25.2 32.1 11.2 6.8  104  2  ave  PMS Cone. (g/L) 0.4 2.8 3.6 1.2 0.79  Table E-XXXVI: Reproducibility of yield and concentration of PMS (Medium-Intensity reactor config. A; Conditions: Reactor temperature = 4°C; Volume of water = 2.0 L; IN NaOH = 50ml; CuS0 (0.1 g/L) = 40 ml; 0.5 M Na S0 = 45 ml; pH = 12.65; Total volume, V = 2.135 L; tf=554  2  3  T  60s;Save=219W/kg; N=2050rpm) Run no.  Titration (ml)  1 2 3 4 5  1.2 1.1 1.15 1.12 1.05  Mean Standard Deviation  PMS Cone. (g/L) 0.948 0.869 0.909 . 0.885 0.829 0.909 0.04  PMS Yield (%) 56.9 52.2 54.6 53.1 49.8 53.3 2.66  Table E-XXXVTI: Reproducibility of yield and concentration of PMS (Medium-Intensity reactor config. B; Conditions: Reactor temperature = 4°C; Volume of water = 0.45 L; IN NaOH = 9.8ml; CuS0 (0.1 g/L) = 12 ml; 2.0 M Na S0 = 50 ml; pH = 12.50; Total volume, V = 0.5218L; tf=2236s;s =219W/kg; N=2323rpm) 4  2  3  T  :  ave  Run no.  Titration (ml)  1 2 3 4 5  11.3 12.0 12.2 12.4 12.4  Mean Standard Deviation  PMS Yield (%) 29.5 31.3 31.8 32.5 32.5 31.5 1.19  105  PMS Cone. (g/L) 8.93 9.48 9.64 9.80 9.80 9.53 0.36  F. FTIR spectra of PMS F.l: FTER Spectra of Sodium Sulfite, Oxone and Generated PMS  i  i  1200  1100  1  jj  1000  900  wavenumber (cm- ) 1  F.2: FTIR Spectra of PMS Sample  i  \  i  ,  .  1400  1300  1200  1100  1000  wavenumber (cm- ) 1  106  1  900  

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.831.1-0058948/manifest

Comment

Related Items