@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix dc: . @prefix skos: . vivo:departmentOrSchool "Forestry, Faculty of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "An, Yuxian"@en ; dcterms:issued "2009-07-06T17:11:09Z"@en, "2000"@en ; vivo:relatedDegree "Master of Science - MSc"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """The first objective of the study was to determine whether drying could improve the treatability of sawnwood with respect to CCA and borate preservatives. Heartwood boards of lodgepole pine (Pinus contorta Dougl.), white spruce (Picea glauca [Moench.] Voss.), amabilis fir (Abies amabilis [Dougl] Forbes), western hemlock (Tsuga heterophylla Raf. Sarg), and Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco) were used in this study. The treatability of each board was measured and the boards ranked for permeability. The seven drying processes were: air drying (used as reference), dehumidification drying, conventional drying, presteaming plus conventional drying, high temperature drying, radio frequency/vacuum drying, and superheated steam/vacuum drying. After drying, the boards were pressure treated with either 2.5% CCA or 4.2% disodium octaborate tetrahydrate, and analysed for preservative penetration and retention. It was concluded that: • No single drying regime consistently improved the treatability o f all wood species, for either chemical; • in most cases, CCA and borate preservative penetrations were similar: • for Douglas-fir and western hemlock RF/V drying gave the best penetration for both CCA and borate; The second objective was to examine the microdistribution of the CCA components in different cells of the various softwoods using light microscopy, and scanning electron microscopy (SEM) coupled with energy dispersive X-ray spectrometer (EDX). From the visible and SEM - EDX microscopic examination it was found that: the Cu:As:Cr ratio differed markedly for specific analysis locations in treated wood, confirming the different fixation reactions in specific cell types; ray cells, resin canals (if present) and pits played key roles in enhancing CCA penetration in sawnwood; copper and chromium were located in the highly lignified cell corners and compound middle lamellae; arsenic was precipitated in the secondary wall; enhanced copper content was found in the resin canals, suggesting a preferential reaction with extractives; Copper and arsenic concentrations were enhanced in the pit areas; a high concentration o f chromium was found at the crassulae, which is highly lignified; deposits at the tracheid lumens of amabilis fir were shown to contain mainly arsenic and copper."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/10189?expand=metadata"@en ; dcterms:extent "7081436 bytes"@en ; dc:format "application/pdf"@en ; skos:note "INFLUENCE OF DRYING PROCESSES ON THE TREATABILITY AND CCA DISTRIBUTION IN THE HEARTWOOD OF FIVE CANADIAN SOFTWOODS by YUXIAN A N M . S c , Northeast Forestry University, China, 1986 B . S c , Northeast Forestry University, China, 1983 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E OF M A S T E R O F S C I E N C E in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Department o f Wood Science) We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A March, 2000 © Yuxian A n , 2000 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract The first objective o f the study was to determine whether drying could improve the treatability o f sawnwood with respect to C C A and borate preservatives. Heartwood boards o f lodgepole pine (Pinus contorta Dougl.), white spruce (Picea glauca [Moench.] Voss.), amabilis fir (Abies amabilis [Dougl] Forbes), western hemlock (Tsuga heterophylla Raf. Sarg), and Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco) were used in this study. The treatability o f each board was measured and the boards ranked for permeability. The seven drying processes were: air drying (used as reference), dehumidification drying, conventional drying, presteaming plus conventional drying, high temperature drying, radio frequency/vacuum drying, and superheated steam/vacuum drying. After drying, the boards were pressure treated with either 2.5% C C A or 4.2% disodium octaborate tetrahydrate, and analysed for preservative penetration and retention. It was concluded that: • N o single drying regime consistently improved the treatability o f all wood species, for either chemical; • in most cases, C C A and borate preservative penetrations were similar: • for Douglas-fir and western hemlock R F / V drying gave the best penetration for both C C A and borate; The second objective was to examine the microdistribution of the C C A components in different cells of the various softwoods using light microscopy, and scanning electron microscopy ( S E M ) coupled with energy dispersive X-ray spectrometer ( E D X ) . From the visible and S E M - E D X microscopic examination it was found that: ii the Cu:As:Cr ratio differed markedly for specific analysis locations in treated wood, confirming the different fixation reactions in specific cell types; ray cells, resin canals (if present) and pits played key roles in enhancing C C A penetration in sawnwood; copper and chromium were located in the highly lignified cell corners and compound middle lamellae; arsenic was precipitated in the secondary wall; enhanced copper content was found in the resin canals, suggesting a preferential reaction with extractives; Copper and arsenic concentrations were enhanced in the pit areas; a high concentration o f chromium was found at the crassulae, which is highly lignified; deposits at the tracheid lumens o f amabilis fir were shown to contain mainly arsenic and copper. iii Table of Contents Abstract i i Table o f Contents iv List o f Tables ix List o f Figures xi i Acknowledgments xx Glossary xxi Chapter 1. Background 1 1.1. Objectives 5 Chapter 2. Literature Review 6 2.1. Improving wood permeability 6 2.2. Drying processes and wood permeability 7 2.3. Species influence on wood permeability to C C A and borates 8 2.4. Factors influencing the C C A distribution in softwoods 11 2.4.1. C C A microdistribution 11 2.4.2. C C A distribution in cell walls 13 2.4.3. C C A distribution in the pits 16 2.4.4. C C A distribution in the resin canals 19 Chapter3. Influence of Drying on C C A and Borate Penetration and Retention in Five Softwood Species 21 3.1. Introduction 21 3.2. Methodology 23 iv 3.2.1. Overview 23 3.2.2. Sample preparation 23 3.2.3. Sample selection-evaluation o f the initial treatability 24 3.2.4. Drying procedures 25 3.2.4.1. Drying processes 26 3.2.4.2. Measurement of Initial moisture content 28 3.2.5. Preservative treatment 29 3.2.6. Determination of preservative uptake 29 3.2.7. Measurement o f preservative penetration 30 3.2.7.1. Measurement o f C C A penetration 30 3.2.7.2. Measurement o f borate penetration 31 3.2.8. Measurement o f preservative retention 31 3.2.8.1. Determination o f C C A retention by X-ray analysis 31 3.2.8.2. Determination o f borate retention 32 3.2.9. Data analysis 33 3.3. Results 34 3.3.1. Penetration and retention data 34 3.3.1.1. Lodgepole pine 34 3.3.1.2. White spruce 36 3.3.1.3. Western hemlock 36 3.3.1.4. Ainabilis fir 39 3.3.1.5. Douglas-fir 39 3.4. Discussion 43 v 3.4.1. Comparison of two preservative treatment 43 3.4.2. The treatability of five species 47 3.4.3. Penetration pathways and preservative distribution 53 Chapter 4. The Distribution of C C A in Five Softwood Based on Light and Scanning Electron Microscopy 60 4.1. Introduction 60 4.1.1. Light microscopic examination o f treated wood 61 4.1.2. Analysis o f the microdistribution o f preservatives in wood by S E M - E D X . . . 62 4.2. Methodology 64 4.2.1. Sample preparation 64 4.2.2. Light microscopy 65 4.2.2.1. Thin section preparation 65 4.2.2.2. Staining reagent selection 66 4.2.2.3. Thin section staining 66 Copper (Cu 2 + ) 66 Chromium (Cr 6 + ) 66 Chromium (Cr 3 + ) 67 Arsenic (As 5 + ) 67 4.2.2.4. Dehydrating process 68 4.2.2.5. Microscopic observation and photography 68 4.2.3. S E M - E D X 69 4.2.3.1. Sample preparation 69 4.2.3.2. S E M - E D X examination and analysis location selection 69 vi 4.2.3.3. E D X data analysis 72 4.2.3.4. Reference sample preparation 73 4.3. Results 74 4.3.1. The pathways o f C C A solution penetration as determined by light microscopy 74 4.3.1.1. Copper distribution 74 4.3.1.2. Chromium distribution 78 4.3.1.3. Arsenic distribution 85 4.3.2. The C C A component distribution as determined by S E M - E D X 89 4.3.2.1. The C C A component distribution in tracheid walls 89 4.3.2.2. The C C A component distribution in pit region 93 4.3.2.3. The C C A component distribution in ray cells 96 4.3.2.4. C C A component distribution in resin canals 96 4.4. Discussion 102 4.4.1. The pathways o f C C A solution 102 4.4.2. C C A distribution 102 4.4.2.1. The C C A component distribution in tracheids 105 4.4.2.2. The C C A component distribution in pit regions 109 4.4.2.3. The C C A component distribution in ray cells 112 4.4.2.4. The C C A component distribution in resin canals 113 4.4.2.5. C C A distribution in the longitudinal parenchyma 119 Conclusions 120 Literature 122 vii Appendix 131 Appendix 1. The treatment analysis data 131 Appendix 2. The treatment statistical analysis results 138 Appendix 3. The treatment statistical analysis results 144 viii List of Tables Table 4-1. The summary o f C C A component distribution in elements o f treated wood 104 Table A l - 1 . Process details o f six o f the drying regimes for lodgepole pine and white spruce - Phase 1 132 Table A l - 2 . Process details o f six o f the drying regimes for Douglas-fir - Phase II 133 Table A l - 3 . Process details o f six o f the drying regimes for western hemlock -Phase II 134 Table A l - 4 . Process details of six o f the drying regimes for amabilis fir - Phase II 135 Table A l - 5 . Process details radio frequency/vacuum drying o f lodgepole pine and white spruce - Phase 1 136 Table A l - 6 . Process details radio frequency/vacuum drying o f western hemlock, amabilis fir and Douglas-fir - Phase II 137 Table A 2 - 1 . Chemical uptake, penetration and retention by analysis in lodgepole p i n e - C C A 139 Table A2-2 . Chemical uptake, penetration and retention by analysis in lodgepole pine - borate 139 Table A2-3 . Chemical uptake, penetration and retention by analysis in white spruce - C C A 140 Table A2-4 . Chemical uptake, penetration and retention by analysis in white spruce - borate 140 ix Table A2-5 . Chemical uptake, penetration and retention by analysis in western hemlock - C C A 141 Table A2-6. Chemical uptake, penetration and retention by analysis in western hemlock - borate 141 Table A 2-7. Chemical uptake, penetration and retention by analysis in amabilis fir - C C A 142 Table A2-8 . Chemical uptake, penetration and retention by analysis in amabilis fir - borate 142 Table A2-9 . Chemical uptake, penetration and retention by analysis in Douglas-fir - C C A 143 Table A2-10. Chemical uptake, penetration and retention by analysis in Douglas-fir - borate 143 Table A3-1 . C C A penetration - lodgepole pine 145 Table A3-2 . C C A retention - lodgepole pine 145 Table A3-3 . Borate penetration - lodgepole pine 146 Table A3-4 . Borate retention - lodgepole pine 146 Table A3-5 . C C A penetration - white spruce 147 Table A3-6. C C A retention - white spruce 147 Table A3-7 . Borate penetration - white spruce 148 Table A3-8 . Borate retention - white spruce 148 Table A3-9 . C C A penetration - western hemlock 149 Table A3-10. C C A retention - western hemlock 149 Table A3-11. Borate penetration - western hemlock 150 Table A3-12. Borate retention - western hemlock 150 Table A3-13. C C A penetration - amabilis fir 151 Table A3-14. C C A retention - amabilis fir 151 Table A3-15. Borate penetration - amabilis fir 152 Table A3-16. Borate retention - amabilis fir 152 Table A3-17. C C A penetration - Douglas-fir 153 Table A3-18. C C A retention - Douglas-fir 153 Table A3-19. Borate penetration - Douglas-fir 154 Table A3-20. Borate retention - Douglas-fir. 154 xi List of Figures Figure 3-1. The processing pattern for each board (the size was measured at the green moisture content) 23 Figure 3-2. C C A uptake, penetration and retention by analysis in lodgepole pine 35 Figure 3-3. Borate uptake, penetration and retention by analysis in lodgepole pine 35 Figure 3-4. C C A uptake, penetration and retention by analysis in white spruce 37 Figure 3-5. Borate uptake, penetration and retention by analysis in white spruce 37 Figure 3-6. C C A uptake, penetration and retention by analysis in western hemlock 38 Figure 3-7. Borate uptake, penetration and retention by analysis in western hemlock 38 Figure 3-8. C C A uptake, penetration and retention by analysis in amabilis fir 40 Figure 3-9. Borate uptake, penetration and retention by analysis in amabilis fir 40 Figure 3-10. C C A uptake, penetration and retention by analysis in Douglas-fir 41 Figure 3-11. Borate uptake, penetration and retention by analysis in Douglas-fir 41 Figure 3-12. C C A and borate penetration in white spruce 44 Figure 3-13. C C A and borate penetration in western hemlock 44 Figure 3-14. C C A and borate penetration in amabilis fir 45 Figure 3-15. C C A and borate penetration in Douglas-fir 45 Figure 3-16. C C A and borate penetration in lodgepole pine 46 Figure 3-17. C C A penetration in five wood species after drying with the seven drying processes ; 48 Figure 3-18. C C A retention in five wood species after drying with the seven drying processes 48 xii Figure 3-19. Borate penetration in five wood species after drying with the seven drying processes 49 Figure 3-20. Borate retention in five wood species after drying with the seven drying processes 49 Figure 3-21. High temperature dried lodgepole pine lumber treated with C C A 51 Figure 3-22. High temperature dried lodgepole pine lumber treated with borate 51 Figure 3-23. Presteaming plus conventional dried Douglas-fir lumber treated with C C A 52 Figure 3-24. Presteaming plus conventional dried Douglas-fir lumber treated with borate 52 Figure 3-25. The type o f the cross-field pit pairs occurring in wood. (a), pinoid pit pairs (b). piceoid pitpairs (C). Taxodioid pit pairs (d). cupressoid pit pairs 55 Figure 3-26. Superheated steam/vacuum dried western hemlock treated with C C A 57 Figure 3-27. Superheated steam/vacuum dried western hemlock treated with borate 57 Figure 3-28. High temperature dried amabilis fir treated with C C A 58 Figure 3-29. High temperature dried amabilis fir treated with borate 58 Figure 4-1. The E D X spectrum of treated wood 71 Figure 4-2. Abies amabilis. A micrograph o f copper distribution (dark green color) at the limit o f penetration. Magnification 200x. Dc: copper deposit. M : middle lamella. Pb: bordered pit pair. R: ray 75 xiii Figure 4-3. Abies amabilis. A micrograph of copper distribution (dark green color) at the limit o f penetration. Magnification 200x. Dc: copper deposit. M : middle lamella. Pb: bordered pit pair. Ps: semibordered R: ray 75 Figure 4-4. Picea glauca. A micrograph o f copper distribution (dark green color) at the wood surface. Magnification 50x. M : middle lamella. Ps: semibordered R: ray. Rc: resin canal 76 Figure 4-5. Pseudotsuga menziesii. A micrograph o f copper distribution (dark green color) at the limit o f penetration. Magnification 200x. P b l : bordered pit on radial section. Pb2: bordered pit on tangential section. R: ray. S: spiral thickening 76 Figure 4-6. Pinus contorta. A micrograph o f copper distribution (dark green color) at the limit o f penetration. Magnification lOOx. R: ray. Rc: radial resin canal... 77 Figure 4-7. Pinus contorta. A micrograph o f copper distribution (dark green color) at the limit o f penetration. Magnification 50x. Rc: resin canal 77 Figure 4-8. Tsuga heterophylla. A micrograph o f copper distribution (dark green color) at the limit o f penetration. Magnification 200x. P: longitudinal perenchyma 78 Figure 4-9. Tsuga heterophylla. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. Magnification 200x. M : middle lamella. Pb: bordered pit pair. R: ray 79 Figure 4-10. Picea glauca. A micrograph o f chromium distribution (orange-red color) at the wood surface. Magnification lOOx. M : middle lamella. R: ray. Rc : resin canal 80 xiv Figure 4-11. Picea glauca. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. Magnification 50x. M : middle lamella. R: ray. Rc : resin canal 80 Figure 4-12. Picea glauca. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. Magnification 200x. M r : radial middle lamella. M t : tangential middle lamella 81 Figure 4-13. Tsuga heterophylla. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. Magnification 50x. Ps: semi-bordered pit pair. R: ray 81 Figure 4-14. Tsuga heterophylla. A micrograph o f chromium distribution (orange-red color) at the limit of penetration. Magnification lOOx. Pb: bordered pit pair 82 Figure 4-15. Pinus contorta. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. Magnification 200x. C: crassulea. Pb: bordered pit pair 82 Figure 4-16. Picea glauca. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. Magnification lOOx. Rc : resin canal 83 Figure 4-17. Pseudotsuga menziesii. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. Magnification lOOx. Rc: resin canal.... 83 Figure 4-18. Tsuga heterophylla. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. Magnification lOOx. P: longitudinal parenchyma 84 X V Figure 4-19. Tsuga heterophylla. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. Magnification lOOx. P: longitudinal parenchyma. Pb: bordered pit pair 84 Figure 4-20. Tsuga heterophylla. A micrograph o f arsenic distribution (brown color) at the limit of penetration. Magnification 200x. Da: arsenic deposit. P: longitudinal parenchyma. Pb: bordered pit pair. Ps: semibordered pit pair. R: ray cell 86 Figure 4-21. Picea glauca. A micrograph o f arsenic distribution (brown color) at the limit o f penetration. Magnification 200x. M : middle lamella. Pb: bordered pit pair. Rc : resin canal. Ws: secondary wall 86 Figure 4-22. Pinus contorta. A micrograph o f arsenic distribution (brown color) at the limit o f penetration. Magnification 200x. M : middle lamella. Ws: secondary wall 87 Figure 4-23. Picea glauca. A micrograph o f arsenic distribution (brown color) at the limit o f penetration. Magnification 200x. Pb: bordered pit pair 87 Figure 4-24. Abies amabilis. A micrograph o f arsenic distribution (brown color) at the limit o f penetration. Magnification 200x. Pb: bordered pit pair 88 Figure 4-25. Pinus contorta. A micrograph o f arsenic distribution (brown color) at the limit o f penetration. Magnification 200x. R: ray. Rc: resin canal 88 Figure 4-26. The arsenic, copper and chromium peak areas on the tracheid and ray cell walls o f amabilis fir 89 xvi Figure 4-27. A scanning electron microscope photograph o f Douglas-fir. The location o f a resin canal, ray cells and tracheids which were analysis by E D X are marked 90 Figure 4-28. The arsenic, copper and chromium peak areas on the tracheid o f Douglas-fir 91 Figure 4-29. The scanning electron microscopic photos o f the deposits at the lumen surface o f tracheids o f amabilis fir. A : earlywood. B : latewood 91 Figure 4-30. The arsenic, copper and chromium peak areas at the deposits at the lumen surface o f the tracheids o f amabilis fir 92 Figure 4-31. A scanning electron microscope photograph of bordered pits on the tracheid wall o f amabilis fir 93 Figure 4-32. The arsenic, copper and chromium peak areas on the bordered pits, the tracheid walls and lumen o f amabilis fir 94 Figure 4-33. A scanning electron microscope photograph o f bordered pits on the tracheid wall o f Douglas-fir 95 Figure 4-34. The arsenic, copper and chromium peak areas on the bordered pits on the tracheid walls o f Douglas-fir 95 Figure 4-35. The scanning electron microscope photograph o f dehumidification dried lodgepole pine 97 Figure 4-36. The arsenic, copper and chromium peak areas in the double resin canal and tracheids o f dehumidification dried lodgepole pine 97 Figure 4-37. A scanning electron microscope photograph o f a transverse resin canal in radial section o f conventionally dried lodgepole pine 98 xvii Figure 4-38. The arsenic, copper and chromium peak areas in the transverse resin canal, ray parenchyma wall, ray tracheid wall and longitudinal tracheid walls o f conventionally dried lodgepole pine 99 Figure 4-39. The scanning electron microgroph o f superheated steam/vacuum dried white spruce 100 Figure 4-40. The arsenate, copper and chromium peak areas on ray cell wall, resin canal and tracheid walls of superheated steam/vacuum dried white spruce. 100 Figure 4-41. The arsenic, copper and chromium peak areas on tracheid wall and resin canal cell wall o f Douglas-fir 101 Figure 4-42. The ratios o f copper and chromium to arsenic peak area on the tracheid walls and ray cell walls o f amabilis fir 106 Figure 4-43. The ratio o f chromium to copper peak area for the tracheid and ray cell walls o f amabilis fir 107 Figure 4-44. The ratios o f copper and chromium peak area to arsenic peak area on the bordered pits, the tracheid walls and lumens o f amabilis fir 109 Figure 4-45. The ratios o f chromium peak area to copper peak area on the bordered pits, the tracheid walls and lumens o f amabilis fir 110 Figure 4-46. The ratios o f copper and chromium peak area to arsenic peak area in resin canals and tracheids o f dehumidification dried lodgepole pine 112 Figure 4-47. The ratios o f copper and chromium peak area to arsenic peak area in transverse resin canals, ray parenchyma wall, ray tracheid wall and longitudinal tracheid walls o f conventionally dried lodgepole pine 113 xviii Figure 4-48. The ratios o f copper and chromium peak area to arsenic peak area on ray cell wall, resin canal and tracheid walls o f superheated steam/ vacuum dried white spruce 114 Figure 4-49. The ratios o f copper and chromium peak area to arsenic peak area on tracheid walls, ray cell wall and resin canal cell wall o f Douglas-fir 115 xix Acknowledgments I would like to express my sincere appreciation for all the support received during my studies at the University o f British Columbia. I am pleased to acknowledge the financial support provided by the Forest Renewal o f British Columbia which funded the major part o f my research program. M y research supervisor, Prof. J. N . R. Ruddick, deserves special thanks, not only for his knowledge, encouragement and dedicated guidance, but also for his understanding and additional financial support. I am also grateful to my co-supervisor, Prof. S. Avramidis for his valuable advice. I would like to express my thanks to Dr. P. Morris for his suggestions and support in organizing the research conducted at Forintek Canada Corp. The technical assistance provided by M r . S. McFarling and Dr . I. Hartley are gratefully acknowledged, as they were responsible for the generation o f the drying data (except R F / V drying) at Forintek Canada Corp. I owe a lot o f thanks to my friends who have studied in Prof. Ruddick's Wood Preservative group, for their assistance during my experimental work and their helpful discussions. Last, I wish to express my special thanks to my dear husband, Yingzhong L i n . Without his support and understanding, this thesis would not have been completed. X X Glossary Air : air drying D H : dehumidification drying Conv.: conventional drying Pre-steaming: pre-steaming plus conventional drying High T: high temperature drying R F / V : radio frequency/vacuum drying SS /V: superheated steam/vacuum drying S E M : scanning electron microscope E D X : energy dispersive x-ray spectrometer R: ray Ps: semi-bordered pit pair P: longitudinal parenchyma Pb: bordered pit pair M : middle lamella Rc: resin canal Dc : copper deposit Da: arsenic deposit R W : ray cell wall T W : tracheid wall Ws: secondary wall T W C L : cell corner o f latewood T W C E : cell corner o f earlywood xxi T W T E : tracheid tangential wall o f earlywood T W T L : tracheid tangential wall o f latewood T W R E : tracheid radial wall o f earlywood T W R L : tracheid radial wall o f latewood T L : tracheid lumen T D : deposit at the tracheid lumen surface T R W : ray tracheid wall S: spiral thickening TP-a: the border edge o f a bordered pit close to pit opening TP-b: the border o f a bordered pit TP-m: the membrane o f bordered pit C : crassulea R L D : deposit at the ray cell lumen surface resin: resin canal ray T W : ray tracheid wall xxii Chapter 1. Background Wood is an ideal building material in many respects - it is easily workable, has an excellent strength to weight ratio, is a renewable resource and can be used on a sustainable basis. Ideally, the service life o f wood products should match the rotation period o f the forest. However, only the heartwood o f a few Canadian wood species, such as western red cedar (Thuja plicata Donn), yellow cedar (Chamaecyaris nootkatensis [D. Don] Spach) and white cedar (Thuja occidentalis L. ) are considered to be durable, with most wood species being either moderately durable, for example, Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco), or slightly-durable, examples o f which are lodgepole pine (Pinus contora Dougl.), white spruce (Picea glauca [Moench.] Voss) and amabilis fir (Abies amabilis [Dougl] Forbes). The pathways and distribution o f preservatives in wood treated with oilborne preservatives are different from those impregnated with waterborne formulations. Oilborne and organic-solvent preservatives can penetrate lumber via the open capillaries and the pit systems connecting them (Walters and Cote, 1960), whereas aqueous solutions can penetrate and swell the cell wall (Liese and Cote, 1960). When the waterborne preservative, C C A , is forced under pressure into wood, a series o f complex chemical reactions take place, during which the active ingredients are strongly fixed in the wood (Anderson, 1990). These reactions can influence the preservative penetration and cause uneven distribution o f components in treated wood. The precipitation o f the chemicals in the cell-wall pit membrane may further impede penetration (Nicholas, 1972). In order to achieve adequate improvement to wood properties, the wood protecting chemicals must also penetrate to a significant depth into the wood. Unfortunately, the heartwood o f most Canadian wood species can not be treated easily due to its poor permeability (Ruddick, 1991). Many factors impact on wood treatability. These include the treatment process, the choice o f chemical and the wood permeability, with the last being the key factor. Considerable research has focused on approaches to improving wood permeability, o f which the most common technique is incising. This technique, while very beneficial, can damage the wood surface, reducing wood strength. In western Canada, o f the commercial species, spruce, pine, hemlock, true firs and Douglas-fir, only western hemlock and amabilis fir are moderately permeable. The research done by the author is based on wood samples produced during a major study into the effect o f various drying regimes on the treatability o f heartwood o f Canadian species undertaken by Forintek Canada Corp. (Forintek) in collaboration with U B C Department o f Wood Science. Forintek staff coordinated and carried out the drying and treatment aspects o f the study, including sampling and chemical analyses o f the test material with some technical assistance provided by the author. Professor Stavros Avramidis directed the radio frequency vacuum drying with technical assistance provided by the author. To determine the impact i f any o f the drying processes on the preservative treatment, the author undertook a statistical analysis o f preservative penetration and retention data developed by Forintek. This provided a basis for subsequent sample selection for the main research focus, which was a study o f preservative penetration pathways through wood supported by the microscopic examination o f preservative 2 distribution at the tissue level. This aspect o f the research was done under the supervision of Professor JohnN.R. Ruddick. Accordingly the research described in this thesis is compiled in two main parts. In the first, the drying and treatment aspects undertaken by Forintek are briefly described, together with the sample recovery from the treated lumber. The discussion focuses on the statistical analysis made by the author o f the influence o f the drying strategies on the chromated copper arsenate ( C C A ) and borate penetration and retention data provided by Forintek. The interpretation o f the C C A treatment is based solely on the penetration o f copper using standard assessment strategies. Since easily treated samples were removed from the study, the data can not provide an indication of the relative treatability o f commercial lumber o f these wood species. Also as part o f the discussion, a visual examination o f cross sections o f treated lumber, stained with reagents for copper and boron, enabled the avenues o f preservative penetration from the wood surface to be recorded. The results are interpreted in terms of the macro-distribution of, and the pathways taken by, the C C A and borate to penetrate the softwood lumber. In the second part, the microdistributions o f the chromium, copper and arsenic are investigated between different cell types, using light microscopy and stained thin sections. The findings were discussed with respect to the penetration pathways and C C A component distribution in treated wood. A semi-quantitative analysis o f the distribution o f the C C A components in different cell walls is presented based upon scanning electron microscopy coupled with x-ray analysis. While the literature contains many studies o f the C C A distribution in sapwood, this is the first reported examination o f C C A distribution in the heartwood of softwoods. In 3 addition, since the microscopic examination was made at the limit o f penetration o f the C C A in the softwood lumber, it provides a unique insight into the pathways taken by the preservative. It also highlights interactions that take place, between the individual components, as well as between the chromium, copper and arsenic and the wood cell components. The research on penetration pathways and distribution was entirely the work o f the author. 4 1.1. Objectives The research had three main objectives. These were: 1. To evaluate the effect o f different drying regimes on the treatability o f five Canadian softwood species, with respect to water-based treatments. 2. To identify the principal pathways by which chromated copper arsenate ( C C A ) preservative penetrates kiln dried wood. 3. To determine the distribution o f chromium, copper and arsenic in different cell types o f treated heartwood at the limit o f preservative penetration, using light microscopy, and semi-quantitatively within cell walls using scanning electron microscopy coupled with energy dispersive spectrometer (microdistribution). 5 Chapter 2. Literature Review 2.1. Improving wood permeability The heartwood o f several western softwood species, most notably Douglas-fir, white spruce, and lodgepole pine, is very difficult to treat with preservatives, because it has low permeability (Cooper, 1973; Graham, 1956; Ruddick, 1980). The major characteristics o f wood, which affect its penetrability by preservatives, are anatomical structure, pore size, degree o f pit aspiration and moisture content (Jewell et ah, 1990). Stone and Green (1959) described solution penetration as the movement o f liquid into the wood capillaries under the influence of a pressure gradient. A number o f pretreatment procedures are currently applied to lumber to ensure an effective treatment. These include incising to increase preservative retention and depth o f penetration, and drying to enhance preservative uptake (Morris, 1991; Ruddick, 1986 and 1989). Incising involves making incisions or small slits in the wood surface (Morris et al., 1994). Compression, achieved by passing boards through rollers to reduce their thickness, has been reported to improve both seasoning and treatability (Cooper, 1973). Removing extractives from the heartwood of some species by solvent extraction increased permeability, presumably by removing encrustation from pit membranes (Krahmer and Cote, 1963). Steaming can improve wood treatability, but was found to cause changes in the physical and mechanical properties o f wood, as well as in its structure and chemical composition (MacLean, 1953; El lwood and Erickson, 1962; Hildebrand, 1970; Kubinsky and Ifju, 1973; Chen and Workman, 1980; Voulgaridis and Tsoumis, 1982). O f the physical methods for improving permeability, incising, precompression and steaming all 6 reduce wood strength (Cooper, 1973; Eaton and Hale, 1993; Nunomura and Saito, 1983; Perrin, 1978), while removal o f extractives is not practiced (Gunzerodt et al., 1988). It is also known that bacterial and fungal colonisation o f wood can improve sapwood permeability, by destroying the pit structure (Dunleavy, et al., 1973; Rosner et al., 1998). 2.2. Drying processes and wood permeability Drying is necessary to provide void space for the injection o f preservative (Cooper, 1973). Some researchers (Booker, 1990; Booker and Evans, 1994; Moldrup, 1995) have suggested that wood permeability may be enhanced through specific drying methods. Booker and Evans (1994) studied the effect o f air drying, kiln drying, and high temperature drying on radiata pine (Pinus radiata). They reported that severe drying schedules increased the radial permeability by 100% in sapwood and 280% in heartwood. It was suggested that in heartwood the resistance to liquid flow from resin canals into tracheids, must be much larger than in the sapwood. This may explain why air dried sapwood wil l take up more preservative solution than high temperature dried heartwood. The increase in permeability was attributed to movement and modification o f resin in the resin canals which were responsible for transporting fluid in radiata pine (Booker and Evans, 1994; Cobham and Vinden, 1994). Vinden (1985) and Booker (1990) also observed that the ease of heartwood impregnation was related to the drying schedule of the lumber. High temperature dried lumber showed better heartwood penetration than air-dried lumber. Cobham and Vinden (1993) also found that a combination o f high temperature drying and pressure impregnation o f radiata pine was necessary to guarantee greater than 80% treatment o f 7 heartwood. The best preservative uptake was achieved following steaming or high temperature drying. Similar claims are made for the effect o f steam/vacuum drying on permeability (Moldrup, 1995). Presteaming prior to pressure treatment has been shown to improve the treatability o f some species grown in B C , such as Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco) and western hemlock (Tsuga heterophylla Raf. Sarg.) (Lebow and Morrell , 1993; Morris et ah, 1997), while kiln drying to moisture content below 20% can reduce the treatability o f spruce (Picea sp.) and lodgepole pine (Pinus contorta Dougl.) (Morris, 1991). However, no comprehensive study is available that clearly compares the impact o f the different drying methods on the treatability o f the main commercial wood species in British Columbia. 2.3. Species influence on wood permeability to C C A and borates Since it was developed over 60 years ago, C C A has gained widespread acceptance as a reliable, safe wood preservative, although problems related to the treatability o f some wood species such as Douglas-fir have been encountered. The moisture content o f C C A treated wood in service fluctuates in the same way as untreated wood, creating checks that may penetrate through the treatment. This may create difficulties in species such as spruce, i f the treatment is confined to the relatively permeable sapwood, since the heartwood is not durable. The rapid reaction o f the C C A with heartwood extractives during treatment, may also contribute to limiting the zone of protection (Cooper et al, 1997) by preventing the chemical from moving deep into the timber. 8 Different species have different permeabilities. Kumar and Morrel l (1989) found the following order o f decreasing heartwood treatability: Pacific silver fir, white fir, grand fir, noble fir, western hemlock, and Douglas-fir. The differences in the permeability and fluid flow through earlywood and latewood have also been a topic o f discussion. In one study the lateral movement o f preservative in white spruce was recorded to be the same in earlywood and latewoood (Keith and Chauret, 1988), while other authors cited more random variability, with earlywood being both more and less permeable than the latewood (Baines and Saur, 1985; Baines, 1986). The length o f latewood ray tracheids is about half that o f the earlywood cells, and may explain why preservative penetration in the earlywood o f spruce is often better than in the latewood (Baines and Saur, 1985). The use o f preservative treated western hemlock has increased considerably in the past decade. Previously, the presence o f wet-wood in the heartwood zone has caused problems. These wet pockets have a slower drying rate compared to normal heartwood, as well as a higher concentration of extractives (Schroeder and Kozl ik , 1972). A major portion o f the extractives in western hemlock heartwood is lignans with the dominant ones being a-conidendrin, hydroxymatairesinol, and matairesinol. Although floccosoids in western hemlock contain large amounts o f a-conidendrin (Barton, 1963), Krahmer et al. (1970) showed that considerable concentrations o f extractives, assumed to be lignan, lined most tracheid walls throughout the heartwood and often incrusted the bordered pits o f dried wood. Upon drying, these deposits could be observed on the end surface o f the small wood samples. Western hemlock has good axial permeability compared to most Canadian species and generally the species receives a high retention o f preservative on treatment. However, the depth o f preservative penetration in sawn material depends considerably on 9 slope o f grain, resulting in generally inconsistent treatments (Schroeder and Kozl ik , 1972). Pizzi 's study (1983) has suggested that tannins and flavonoids present in some wood species precipitate copper before it can react with lignin so that studies o f the distribution o f C C A in heartwood could be expected to be different from that found in sapwood. Although the heartwood o f coastal Douglas-fir can be treated reasonably consistently by pressure processes with oil-borne preservatives, treatment with C C A is much less successful (Schroeder and Kozl ik , 1972). One study o f Douglas fir permeability found irreversible effects o f liquid movement in wood on the flow rates (Bailey, 1965). The general reason cited for this change in flow rates includes blockage o f the pathways by particles or air bubbles and pit aspiration. Boron-based wood preservatives have been used world-wide for a number o f decades (Schoeman et al., 1998). One o f the major advantages o f boron is its ability to move with moisture throughout wood. This mobility o f boron, however, has posed a substantial challenge to those attempting to develop reliable estimates o f the toxic threshold to fungi, since most tests require the wood to be in contact with a wet substrate (Morrell et al., 1998). Borates are capable o f migrating long distances through wet wood and can diffuse through the heartwood o f many wood species, such as Douglas-fir, pine and western hemlock, some o f which are generally considered impermeable to conventional wood preservatives (Lebow and Morrel l , 1989; Roff, 1974). Despite the current basic interest in using boron as wood preservative component, no studies o f the penetration pathways taken by boron during treatment have been reported. Some o f the reasons for this could be its high mobility (making sample 10 preparation very difficulty), and the inability to measure boron using conventional S E M -E D X , since boron is a light element the first row o f the Periodic Table. 2.4. Factors influencing the C C A distribution in softwoods 2.4.1. C C A microdistribution Greaves and Nilson (1982) studied preservative treatment o f hardwoods. They noted a disproportion o f C C A treatments based on electron probe microanalysis o f bulk specimens. It was suggested that radial penetration primarily occurred along the rays with subsequent movement into tracheids (Greaves, 1974; Levy and Greaves, 1978). The copper.xhromium.arsenic ratios differed from that in the treating solution. It was noted that while comparisons o f x-ray data can establish the main features o f microdistribution between fibers, rays and vessels (Greaves and Nilsson, 1982), care must be taken when interpreting data. Drysdale et al. (1980) carried out a spot analysis o f the lumen/S3 layer. They found that the actual composition, or ratio o f elements, varied from one location to another. This is not unusual, as there is often a relatively large variability in X-ray micro-analytical data from biological materials. This can be due to chemical reaction with wood components as well as variation in preservative content in different cells and tissues, i.e. real variability. In addition, electron excitation parameters as well as specimen-detector geometry can also produce large variations in count rates, i.e. artificial variability (Greaves and Nilsson, 1982). Hulme et al. (1976) examined C C A treated Eucalyptus maculata, Betula alba, Populus robusta and P. radiata by S E M - E D X , and found that the peak to background ratio o f elements differed according to the wood species. Levy and Greaves (1978) also 11 confirmed that each timber has its own characteristic effects, not only upon the penetration o f preservative, but also on the disproport ionate o f the C C A within the wood structure. Analytical data were expressed as peak height to background (P/B) ratios and as elemental ratios. Since the relationship o f P /B ratio to element concentration is essentially linear in thin sections (Russ, 1984), a comparison of P /B ratio suffices to compare results from one preparation procedure to another. In a study by Lee et al. (1992), all three elements, arsenic, copper and chromium, located within the middle lamella and the cell corners were found in greater concentrations than within the secondary walls. Rudman (1966a) also observed that chromium, copper, and arsenic were all located at the analysis locations following passage o f treating liquid across the cell wall, with an indication that the maximum concentration o f metal occurred in the middle lamella region. However, the location o f the treated wood sampled is important. I f the sample is removed from the limit o f C C A penetration, not all o f the components may be present. The application o f chemical reagent to treated wood to measure preservative penetration has shown different results for copper and arsenic reaction. Measured depth o f copper penetration in most species, was less than that o f arsenic, based on observations in treated Douglas-fir, pacific silver fir, noble fir and grand fir. Higher chromium content in C C A - A may give more rapid fixation, resulting in more copper and arsenic deposited near to the surface (Rudman, 1966a). According to Cooper et al. (1997), chromium (Cr 6 + ) is the last component o f C C A to be immobilized during the fixation process. Thus by maximising chromium fixation, the immobilization o f the copper and arsenic is increased. But Jansen et al. (1985) concluded 12 that chromium penetrates and fixes faster than either arsenate or copper. This is in agreement with the results obtained by chemical kinetics (Pizzi, 1981 and 1982). 2.4.2. C C A distribution in cell walls Flynn (1995) has suggested that the deposition o f high C C A preservative concentrations in pressure-treated material is evidence o f the main path o f preservative flow. Increased gross C C A retention did not influence distribution within cell walls, but did increase the amount o f preservative deposited in discrete locations. In a visual examination o f C C A treated wood sections using transmission electron microscopy, Drysdale et al. (1980) observed a dark deposit lining the cell lumen o f each tissue type. This indicated that the preservative solution reached the lumen o f all cells during treatment. Bailey (1965) proposed that \"the transient cell wall capillary network contributes significantly to flow from cell to cell\", while Fleischer (1950) noted that \"the treating chemicals existed not as deposits within the cell cavity, but were instead relatively uniformly distributed throughout the cell wall\". Preservative has been located on the surface o f the cell wall o f some coniferous species (Fleischer, 1950; Hosl i , 1986; Kumar and Jain, 1978), especially at high retentions (Hosli, 1986). Belford (1959) used polarized light microscopy, X-ray contact micro-radiography and electron microscopy to show that in C C A treated wood, the copper is located within the cell wall and not simply deposited on the lumen surface. Yata et al. (1982) measured the metal distribution in wood using S E M - E D X and based on a point analysis to determine the concentration o f metals in the various morphological regions o f wood. They clearly demonstrated that the lateral movement o f 13 ions such as C^O? 2 \" , C u 2 + and Z n 2 + , from the initially penetrated cells to adjacent cells, occurred mainly by ionic diffusion through the cell wall, rather than by the liquid flow through the pit pairs. In both hardwoods and softwoods pressure treated with C C A , the region o f the middle lamella-primary wall between cells tends to be better treated than the adjacent S2 layers (Drysdale et al, 1980; Greaves, 1974). Yata et al. (1979) investigated the permeability o f the various layers within the cell wall (compound middle lamella, S i , S2 and S3 layer) and the copper distribution across the cell wall when equilibrium was reached. They found that in a cell wall the compound middle lamella layer and the boundary between Si and S2 were the most permeable layers. The copper concentration in the compound middle lamella layer was much larger than that in the S2 layer. This result was also reported by DeGroot and Kuster (1984). Based on electron microscope examination o f thin sections, and studies by electron diffraction, it has been proposed that copper may be complexed with hydroxyl groups on the surface o f the cellulose microfibrils (Belford et al, 1959). Examination o f copper levels within individual latewood tracheid walls o f softwoods suggested that the copper content varied widely, even for the same location within the S2 layer chosen for analysis (Hulme and Butcher, 1977). When Belford et al. (1959) examined C C A treated Douglas-fir timber by optical microscopy, they were unable to detect any o f the blue copper-dithio-oxamide complexes in the cell lumen. The middle lamella was stained very deeply, and the color was noticeable throughout the whole o f the cell wall thickness. It was, however, patchy over any section, suggesting that the copper was not distributed uniformly. Using a similar approach, Greaves (1972 and 1974) also 14 found that the copper content was greatest in the middle lamellae and the outer region o f the secondary wall. In contrast to this, stained sections cut from blocks showed little coloration o f the cell wall except adjacent to the lumen, the ray cells, and in the region o f the pits. In wood treated with either potassium chromate or potassium dichromate solution, DeGroot and Kuster (1984) found that the chromium content o f the compound middle lamella was higher than that o f the secondary wall. Klein and Bauch (1979) concluded that the Cr 2 07 2 \" ion becomes almost completely adsorbed and reduced to a lower oxidation state (Cr 3 + ) within the wood cell walls. Higher amounts o f chromium-ions were adsorbed on the cell wall compared to copper. There are very few reports on the distribution o f arsenic in C C A treated wood. Drysdale et al., 1980 reported that the occurrence o f arsenic was poorly distributed in both cell wall and middle lamella (Fagus. sylvatica). However, most researchers concluded that arsenic is located the same regions as copper and chromium (Rudman, 1966a; Chou et al; 1973; Lee et al, 1992). Jansen et al (1985) observed that the arsenic penetration profile follows closely that o f chromium, indicating that the majority o f chromium and arsenic are present as chrome arsenate, equally distributed between lignin and cellulose. He also noted the arsenic penetration profile was always lower than that o f chromium, suggesting that the latter starts to fix first. Chou et al (1973) also showed that in each o f the analyzed areas, copper was invariably associated with either chromium or arsenic. The marked change in the distribution pattern o f metal preservative components from the S2 layer to the middle lamella, parallels a marked difference in the ultrastructure o f the two regions 15 A s the chemical constituents are not uniformly distributed throughout the cell wall, insoluble C C A products may not be uniformly formed. When the concentration o f the C C A components becomes low at the limit o f penetration, the fixation reaction slows, so unfixed chemical deposited deep within the wood may continue to migrate and extend the penetrated zone (Yata et al., 1978). 2.4.3. C C A distribution in the pits The three primary causes o f the difference o f liquid flow in heartwood and sapwood are pit aspiration, and the amount, and character o f the extractives (Erickson, 1970). In heartwood, the solution penetrates from tracheid to tracheid via bordered pits that are often aspirated and heavily encrusted with extractives which tend to reduce the permeability (Erickson, 1970; Krahmer and Cote, 1963). Latewood pits in conifers have less tendency to aspirate than earlywood pits (Erickson and Crawford, 1959). Pit aspiration, during the initial drying phase prior to treatment, would greatly reduce permeability (Comstock, 1967). The majority (99%) o f the pits in spruce specimens examined by Flynn and Goodell (1996) were aspirated. This aspiration in spruce, takes place during heartwood formation (Richter, 1990). Wood also can be dried by \"low surface tension\" methods, e.g., by freeze-drying or solvent-drying, which result in much less pit aspiration. However these methods are not commercially feasible for large wooden members (Vinden, 1985; Booker, 1990). Diffusion o f solution from the lumen into the cell wall surrounding the bordered pit was found to be poorer than that in other regions, confirming that the cell wall adjacent to a simple pit pair also was impermeable to diffusion. The chemicals diffusing in the cell wall 16 make a detour around this portion o f the wall. Since these parenchyma cells do not have an amorphous layer on the inner surface o f the cell wall, this impermeable layer is considered to be made o f a cellulosic substance (Yata et ah, 1981). A considerable amount o f research has been done on the pathway o f liquid flow through the cell lumen and the pit membrane pores of wood (Yata et al., 1978, 1979, 1981a and b, and 1982; Rudman 1966a and b; Booker, 1990). However, there is very little good information on the pathway taken by liquids into the cell wall from the lumen. The distribution o f C C A is affected by wood density, wood ultrastructure and solution concentration. L o w density wood species with thin cell walls, should allow preservative solution to flow easily into wood through the pits and cell lumens, resulting in good distribution o f preservative. Brennan et al. (1995) found that in softwoods, the open pit structure in sapwood produced a more uniform distribution o f preservative, compared to heartwood where the pits are aspirated. It has also been observed that the C C A distribution in latewood with thicker cell walls containing smaller pits, is better than that in earlywood which has a thinner cell wall, with more and larger pits. This is because the pits in earlywood are easily aspirated, preventing liquid from passing through them. The thicker pit margo in the latewood is too rigid to allow easy aspiration. Apart from aspiration, the diameter o f the pit opening, the number o f pits, as well as the size o f the pores in the pit membranes, also significantly influences softwood permeability. Cross-field pits are very important in lateral penetration because they connect ray cells to the longitudinal tracheids. The type, number and size o f cross-field pits directly influence the lateral penetration o f preservatives. Cross-field pit membranes receive the heaviest treatment although tracheids bordered pits were also well treated 17 (DeGroot and Kuster, 1986). B y studying the path taken by liquid gallium in wood using scanning electron microscopy, Trenard and Gueneau (1984) found that the window-like pit pairs o f Scots pine (Pinus sylvestris) were easily ruptured under pressure. The same authors concluded that the smaller pits found in alpine fir (Abies sp.) should limit liquid flow. The same may be said for the pits on the ray tracheid. However, it has been reported that one o f the more treatable species in western Canada is alpine fir (Morris, 1991 and 1995). Rudman (1966a) proposed that a polar liquid may flow from cell to cell in two ways, namely via the pit or the cell wall capillary system. For bulk liquid movement, the pits (despite the fact that a number appear to be blocked) are the more important o f the two. Hence, flow under pressure should favor movement through the larger pores in the pit membranes, rather than through the finer cell wall capillaries. However, such a conclusion does not rule out diffusion or flow through cell wall capillaries. In some instances, it was noted that the wall on one side o f a cell had been penetrated, but that the opposite wall had not been penetrated (Rudman, 1966b). Rudman (1966a) has shown that the middle lamella o f eucalypts does not prevent flow from a cell wall to the adjacent cell wall, since unblocked pits were few in number and not much flow o f liquid from lumen to lumen could occur. In hardwoods, Greaves (1972) noted that the penetration pathway is usually longitudinal via vessels and then through either cell walls or pits into adjoining fibers. This more tortuous route may partly account for the uneven distribution o f preservative in hardwoods (Greaves, 1972 and 1974; Dickinson, 1974). 18 The distribution o f chemicals reacting with lignin was also examined by Wardrop and Davies (1961). It was established that in P. radiata the penetration o f preservative proceeds from tracheid to tracheid via the bordered pits and spreads laterally through the rays. Penetration in easily treated softwoods is essentially radial, and since every tracheid has at least one pit connection with a ray, all wood tissues should be rapidly and uniformly treated (Greaves, 1972; Chou et al. 1973). In softwoods pressure-treated with C C A , the preservative content o f ray cells was much greater than that o f tracheids (Greaves, 1974). When copper was detected by staining with dithio-oxamide, Belford et al. (1959) found that the cell wall adjacent to the region o f the pits was stained more heavily than regions o f the cell wall well removed from pits. However, Yata et al. (1981b) found that chromium (Cr 6 + ) diffusion from the lumen into the cell wall around the bordered pit was poorer than that in other regions. The cell wall neighboring the pit cavities o f a simple pit-pair also was impermeable to diffusion. This impermeable layer is considered to be made o f a cellulosic substance. 2.4.4. C C A distribution in the resin canals Booker (1990) developed on impregnation modal for radiata pine sapwood. He injected dye into radiata pine and made photomicrographs o f the pathways. These photomicrographs clearly showed that the resin canals play an important part in fluid movement in wood. During preservative impregnation, the main flow into radiata pine sapwood occurred along the radial resin canals into adjoining axial resin canals, while practically no flow occurred along rays that do not contain a resin canal. I f few intercellular spaces are created in the heartwood by cell collapse, preservative solution can 19 only flow from the resin canals into the tracheids by first passing through the resin canal wall into the ray parenchyma cells and then penetrating the undamaged ray-to-tracheid pits (Krahmer and Cote, 1963). The resin canals also facilitated longitudinal penetration. The path o f penetration was similar in both sapwood and heartwood (Wardrop and Davies, 1961). 20 Chapter 3. Influence of Drying on CCA and Borate Penetration and Retention in Five Softwood Species 3.1 Introduction In many applications, an acceptable service life can only be achieved through the use o f a supplementary chemical treatment designed to enhance durability, increase hardness or improve dimensional stability. One o f most effective preservatives used in Canada is chromated copper arsenate ( C C A ) , a waterborne preservative, which fixes to the wood. This fixation, while reducing the environmental impact o f the treated wood, may contribute to limiting the penetration o f the preservative, i f fixation begins before the potential penetration has been achieved. Recently, there has been increased interest in the use o f borate preservatives, such as disodium octaborate tetrahydrate (DOT). This preservative does not fix to the wood and is only suitable for protection against insects, fungi and termites in a non leaching environment. These two chemicals represent two extremes in terms o f their movement with the solution, as it penetrates into wood. They are, therefore, useful as model systems for evaluating the effect o f pretreatment processes, such as drying, on the permeability o f wood to chemical treatment. Opinions differ about the effects of drying wood on its subsequent treatability. Some researchers consider that the drying process may improve wood treatability, while others do not. In 1997 the Forest Renewal o f British Columbia sponsored a project at Forintek Canada Corp. that focused on the role o f drying in enhancing wood treatability. In the first stage o f the current research program, the penetration and retention data 21 collected by Forintek were statistically analysed, in order to determine whether specific drying regimes enhanced the preservative treatment. 22 3.2. Methodology 3.2.1. Overview The drying and treatment aspects o f the project were primarily carried out by the technical staff o f Forintek with some technical assistance by the author. Preparation o f the samples for retention determination and data analysis was done by the author. 3.2.2. Sample preparation In phase I, pre-sorted to #2 grade or better, lodgepole pine (Pinus contorta Dougl.) (750 pieces) and white spruce (Picea glauca [Moench.] Voss.) (600 pieces) rough green boards (48 mm x 100 mm x 3.7 m) were obtained from Rustad Brothers. The boards were sorted and those with major checks and red heart were discarded. The remaining boards were individually numbered. Two pieces, approximately 1.2 m and about 48 mm 50mm 100 mm Measure penetration and retention MC control sample Z 2.45 m 3.70 m Evaluate initial permeability 1.20 m Figure 3-1. The processing pattern for each board (the size was measured at the green moisture content). 23 50 mm (moisture content sample) long were cut from each board (as shown in Figure 3-1). The 1.2 m long sections were retained for testing the initial treatability o f the samples before drying, following the procedure described in Section 3.3.2. The moisture content ( M C ) sections were oven dried, as described in Section 3.2.4.2. The remaining pieces from each board (approximately 2.45 m long) were covered with a tarpaulin, and stored outside prior to sorting and drying. After drying and treatment, a 50 mm section was cut from each board for measuring the chemical penetration and retention, as described in Sections 2.3.7 and 2.3.8. For phase II, the lumber was sourced from Interfor in the Fraser Valley. Coastal amabilis fir (Abies amabilis [Dougl.] Forbes), western hemlock (Tsuga heterophylla Raf. Sarg,) and Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco) green logs were obtained. The logs were shipped to a portable sawmill in Maple Ridge for processing into lumber. The rough, green lumber (55 mm x 104 mm x 3.7 m), was shipped to Forintek. The boards were individually numbered, covered with a tarpaulin and left outside until sawn into test samples. They were processed as described for Phase 1. 3.2.3. Sample selection - evaluation of the initial treatability In order to allow for the inherent variability in wood permeability, it was necessary to group the boards within each wood species by their treatability, before drying. In this way, boards representing the different permeability groups could be included in each drying regime. The 1.2 m sections were pressure treated with 2.5% C C A using a 30 minute vacuum (-75 kPa), 2 hours at 1000 kPa and a 15 minute final vacuum (-75 kPa). Following treatment, boards with high permeability (fully treated) were identified and 24 eliminated from the experiment. The rerraining test pieces for each species were sorted into four groups based on heartwood permeability. The test pieces in each o f the four groups were evenly divided among seven subgroups, with 60 pieces in each subgroup. Each subgroup was dried with one drying regime to a target moisture content o f 22-25%. 3.2.4. Drying Procedures For the study, seven commercial, or semi-commercial, drying processes were selected by Forintek. They were air drying, dehumidification drying, conventional drying, Presteaming plus conventional drying, high temperature drying, superheated steam/vacuum (SS/V) drying and radio frequency/vacuum (RF/V) drying. The air-dried samples served as a reference treatment, with conventional drying acting as a secondary reference process. Commercial drying schedules were applied, where available. For each wood species, seven sets o f 2.45 m long boards were prepared. Each set comprising 60 replicates, was dried with one of the seven drying regimes. Since the drying schedules o f lodgepole pine and white spruce are similar, they were dried together, except for the R F / V drying, where the unit capacity prevented this. The remaining species were dried in species sets. Details o f the drying regimes are given in Appendix 1 (Tables A l - 1 to A l - 6 ) . 25 3.2.4.1. Drying Processes Air drying o f lumber involved exposing stickered piles o f lumber under covered storage to outside air. The air drying schedules are shown in Appendix 1 (Table A l - 1 for white spruce and lodgepole pine, Table A l - 2 for Douglas-fir, Table A l - 3 for western hemlock and Table A1-4 for amabilis fir). A dehumidification dry kiln lowers the vapor pressure of water in the air surrounding the wood, to establish a moisture gradient that causes water to leave the wood. Dehumidification drying schedules are shown in Appendix 1 (Table A l - 1 for white spruce and lodgepole pine, Table A l - 2 for Douglas-fir, Table A l - 3 for western hemlock and Table A1-4 for amabilis fir). In conventional kiln drying, the vapor pressure o f the water in the wood is increased by applying heat. The difference in the vapor pressure o f the air in the wood and that surrounding the wood, drives the moisture out o f the wood. The conventional kiln drying schedules used in the project are shown in Appendix 1 (Table A l - 1 for white spruce and lodgepole pine, Table A l - 2 for Douglas-fir, Table A l - 3 for western hemlock and Table A1-4 for amabilis fir). A small pilot plant was developed at U B C for research on the R F / V kiln process. It consisted o f a 2.75 m long cylinder, with doors at both ends. Two platens (2.24 m) running the length of the kiln, applied the radio frequency field to wood placed between them. The maximum dimension o f wood to be dried was 250 x 250 mm. The upper platen was adjustable to minimize the air gap between the platens and the lumber, thereby improving heating efficiency. The chamber could evacuate to accelerate the rate o f drying. During drying (or heating), the temperature at various locations inside the timber was 26 monitored using fiber optic temperature probes. A l l process data was collected through a data acquisition system and stored by a computer. R F / V drying takes advantage of the strong dielectric properties of water, allowing selective heating o f the water in hygroscopic materials. Because the average dielectric constant for water is about 20 times greater than that o f a dry wood cell wall, exposure to the same frequency range (1 to 10 M H z ) wi l l cause the water to heat at a much more rapid rate than the wood (Avramidis and Zwick, 1992.). The rapid temperature rise, i f uncontrolled can lead to localized overheating o f the wood, with possible charring or surface discoloration. The drying schedules for the R F / V drying processes are shown in Appendix 1 (Table A l - 5 for white spruce and lodgepole pine and Table A l - 6 for Douglas-fir, western hemlock and amabilis fir). During S S / V drying, lumber is dried at temperatures above the boiling point o f the water. In this project the heating o f wood was carried out under a vacuum o f approximately 20kPa. Throughout the S S / V process, the boiling point o f water is usually below 100°C (between 50°C and 70°C), so that the lumber can be dried at relatively low temperatures. The S S / V drying schedules are shown in Appendix 1 (Table A l - 1 for white spruce and lodgepole pine, Table A l - 2 for Douglas-fir, Table A l - 3 for western hemlock and Table A1-4 for amabilis fir). In this project, the high temperature drying was similar to the conventional drying, except that a much higher temperature was used. The high temperature drying schedules are shown in Appendix 1 (Table A l - 1 for white spruce and lodgepole pine, Table A l - 2 for Douglas-fir, Table A l - 3 for western hemlock and Table A1-4 for amabilis fir). 27 The final drying process involved presteaming coupled with a conventional drying process. It was similar to the conventional drying, except that the wood was steamed at 79°C for 4 hours before drying. Drying schedules are shown in Appendix 1 (Table A l - 1 for white spruce and lodgepole pine, Table A l - 2 for Douglas-fir, Table A l - 3 for western hemlock and table A1-4 for amabilis fir). 3.2.4.2. Measurement of Initial Moisture Content The initial moisture content o f each sample board was determined before drying. The moisture content sample cut from each board (Figure 3-1) was weighed (Wi) , oven dried for 24 hours at 105°C, and reweighed (W 2 ) . The moisture content was computed from W - W MC(%) = V2 2 X l Q° where MC is the moisture content (%), Wl is initial weight o f the test block (g) and W2 is weight o f the oven dried test block (g). 28 3.2.5. Preservative treatment After drying, all samples were planed at Forintek to cross section dimensions o f 38 x 89 mm. Each group o f 60 lumber samples was then divided into two subgroups o f 30 pieces for subsequent treatment with either 2.5% C C A or 4.2% borate. The treating cylinder containing the lumber to be treated with C C A solution was first evacuated to -75 kPa and the vacuum held for 30 minutes. The solution was introduced into the retort, and the system pressured to 1035 kPa for 2 hours. A 15 minute final vacuum (-75 kPa ) was applied at the end o f the treatment cycle to remove excess solution. After removing the C C A treated samples from the treating cylinder, they were immediately placed for two weeks in a covered storage area at 20°C, to allow fixation to take place. The pilot plant was cleaned to allow the borate treatment to be done. The 4.2% disodium octaborate tetrahydrate solution (5% boric acid equivalent - B A E ) was applied using the same treating schedule as C C A . However, no storage time was allowed before sampling the borate treated wood. 3.2.6. Determination of preservative uptake Each board was weighed before (W 3 ) and immediately after (W 4 ) treatment to determine the solution uptake. The uptake in terms o f weight o f chemical per unit volume was computed from where UTl is the uptake o f chemical (kg/m 3), W3 is weight o f the wood sample before treatment (kg), 29 WA is weight o f the wood sample after treatment (kg), V is volume (m 3) o f the wood sample after drying, but before treatment and C is preservative solution concentration (%). 3.2.7. Measurement of preservative penetration After measuring chemical uptake by weight, a 50 mm section was cut at the center o f each treated board for measuring the chemical penetration and retention (Figure 3-1). 3.2.7.1. Measurement of CCA penetration Traditionally, when assessing the C C A penetration in wood, the penetration o f the copper component is measured, because it is the principal fungicidal component, and is readily determined. To measure the copper penetration, a 10 mm slice was cut from each 50 mm section (the remaining 40 mm sections were kept for measuring retention and observing chemical distribution) and oven dried. The indicator was prepared by dissolving 0.5 g chrome azurol S and 5.0 g sodium acetate in 80 ml o f water, and diluting the solution to 500 ml with distilled water. When the chrome azurol S solution was sprayed onto the C C A treated wood, a deep blue color revealed the presence o f copper ( A W P A , 1996a). The penetration was measured as the extent o f the blue colored wood in the defined assay zone at the center of heartwood face o f each section, to a maximum penetration o f 18 mm. 30 3.2.7.2. Measurement of borate penetration For the borate treated lumber, a 50 mm thin slice was cut from the center o f each board immediately after the treatment and oven dried at 50°C to constant weight. A 10 mm slice was then cut from each o f these 50 mm sections to assess the boron penetration. A two-part curcumin indicator was sprayed sequentially onto the freshly cut face. Part A was prepared by dissolving 0.28 g curcumin in 100 ml o f ethanol. Part B was a saturated solution o f salicylic acid in 90% ethanol/10% concentrated hydrochloric acid. The cut face was first sprayed with part A and, after two minutes, sprayed with part B . A red or pink color indicated the presence o f boron. After waiting 30 minutes for the color to develop fully, the boundary o f the treated zone was marked with a permanent felt-pen. The penetration o f borate was measured at the center o f the heartwood face, from the block surface to the edge o f the line marking the limit o f borate penetration. This rather elaborate sampling procedure was used to avoid the potential for redistribution o f borate associated with boring or sawing o f wet treated wood (Morris et al., 1996). 3.2.8. Measurement of preservative retention 3.2.8.1. Determination of CCA retention by X-ray analysis A small block (20 mm (L) x 20 mm ( T) x 16 mm (R))was cut from the center o f the heartwood face o f each penetration and retention sample and ground to 40 mesh sawdust. A 0.4 g sample o f the sawdust was thoroughly mixed with 0.1 g o f cellulose powder. The mixture was placed in a die and compressed for three minutes at 200 M P a to produce a pellet 19 mm in diameter. The pellet was then analyzed using a Tracor X-ray energy 31 dispersive X-ray spectrometer which had been calibrated for chromium, copper and arsenic ( A W P A , 1996b). The reference density provided by the American Wood Preserver's Association standard ( A W P A , 1996c) for lodgepole pine, western hemlock, amabilis fir, Douglas-fir, and white spruce were used to convert results from a weight per weight basis, to the weight per volume unit (kg/m 3) used to express preservative retention. Wt Wt Wt RTrr = — x p RTr„ = — x p RT. = x p Cr Wt Wt A s Wt , r 1 wood \" 1 wood \" 1 wood where: RTCr, RTCu, and RTAs are the retentions (kg/m 3) o f chromium, copper and arsenic, WtCr, WtCu, WtM are the weights (g) o f chromium, copper and arsenic measured by the x-ray spectrometer, p is the reference density o f the wood sample (kg/m 3) and Wtwood is the weight o f wood sample (g). 3.2.8.2. Determination of borate retention For each sample a 16 mm analysis sample was cut from the center o f the heartwood face and ground into 40 mesh sawdust. The borate was extracted from the sawdust, using hot water at 95°C for four hours. The leachate was titrated using the mannitol titration method ( A W P A , 1996d), and the retention expressed as the boric acid equivalent ( B A E ) . The B A E was converted into boric oxide (kg/m 3) by the following equation, RTbo = BAE x p, where RTBO is retention o f boric oxide (kg/m 3), p is the reference density o f wood (kg/m 3), and BAE is the retention as boric acid equivalent (%). 32 3.2.9. Data analysis Following the measurement o f preservative penetration and retention using the standard protocols, the data was compiled on a computer for analysis. The effect o f the various drying regimes on the penetration o f each wood preservative was evaluated using A N O V A . The multiple significant (p = 0.05) effects of the drying regimes on C C A or borate treatment were noted. The retention and penetration data for the air dried specimens, treated with either C C A or borate were set as the primary reference. The data for conventional drying served as a secondary reference. The impact o f the various drying regimes for each wood species and preservative combination was determined with respect to the preservative penetration and retention. In this way, the effect o f each drying regime on heartwood permeability for the five softwood species was determined. Where the effects o f drying regimes on the retention and penetration were examined for significance using the A N O V A method, a Duncan Multiple Range Test was also used to further analyze the difference among the effects o f all the tested drying regimes. The statistical analysis identified which drying methods had an effect either on preservative retention or penetration into wood. To analyze whether the two tested preservatives ( C C A and borate) had different effects on penetration and retention under each drying regime, a t-test was performed on the experimental data for each wood species tested. 33 3.3. Results 3.3.1. Penetration and retention data The mean C C A and borate retentions for the five wood species, dried with the seven drying processes, are presented as histograms in Figures 3-2 to 3-11 and as tables in Appendix 2 (A2-1 to A2-10). In the figures, the seven drying regimes are presented from left to right in order of increasing severity with respect to anticipated temperature and pressure gradients. Air-drying was used as the reference, since it is the most common commercial method used to condition wood prior to treatment, and it does not normally have a detrimental effect on treatability. Conventional drying served as the secondary reference, because it is the most common artificial drying method in North America. Treatment results showed marked differences in both the retention and penetration, among the five wood species (Appendix 3). However, no artificial drying method produced uniform effects on treatability in all species. Indeed for most wood species, the best results were found for air drying, with artificial drying often causing reductions in chemical penetration or retention. 3.3.1.1. Lodgepole pine Compared to air drying, artificial drying did not significantly affect C C A penetration and retention, while R F / V drying and dehumidification drying produced a significant reduction in borate penetration. Dehumidification drying, presteaming plus conventional drying, high temperature drying, conventional drying, and R F / V drying all significantly reduced borate retention. 34 • weight uptake (kg/mA3) • retention (kg/mA3) • penetration (mm) Air DH Conv. Pre- High T steaming RF/V SS/V Figure 3-2. C C A uptake, penetration and retention by analysis in lodgepole pine. • weight uptake (kg/m*3) • retention (kg/m^) O penetration (mm) as Air DH Conv. Pre-stearning HighT RF/V SS/V Figure 3-3. Borate uptake, penetration and retention by analysis in lodgepole pine. 35 Compared to conventional drying, the different drying regimes did not significantly affect C C A penetration or C C A retention or borate retention. R F / V drying and dehumidification drying caused a significant reduction in borate penetration. 3.3.1.2. White spruce Comparing air drying to the other drying regimes, no significant difference was shown in C C A penetration and borate penetration in white spruce. A significant reduction in C C A retention was found for high temperature drying and S S / V drying. A i r drying also showed a significantly higher borate retention compared to prestearning plus conventional drying. Using conventional drying as a reference process, there was no significant difference in C C A penetration or retention, and borate penetration or retention. 3.3.1.3. Western hemlock Compared to air drying, radio frequency vacuum drying and S S / V drying produced a significant increase for both C C A penetration and retention in western hemlock. R F / V drying and prestearning plus conventional drying gave significant reduction in borate penetration. The borate retention data showed air drying to be significantly higher than dehumidification drying, conventional drying and high temperature drying. 36 Conv. Pre- HighT RF/V steaming Figure 3-4. C C A uptake, penetration and retention by analysis in white spruce. 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 cmJj • weight uptake (kg/m^) • retention (kg/m7 )^ • penetration (mm) Air DH Conv. Pre- HighT RF/V steaming SS/V Figure 3-5. Borate uptake, penetration and retention by analysis in white spruce. 37 • weight uptake (kg/mA3) • retention (kg/nY )^ • penetration (mm) Air DH Conv. • Pre- HighT RF/V SS/V steaming Figure 3-6. C C A uptake, penetration and retention by analysis in western hemlock. • Weight uptake (kg/trr^ ) • Retention (kg/rrr^ ) u reneirauon \\p J d 1—!—B«H - i Air DH Conv. Pre- HighT RFA/ SSA/ steaming Figure 3-7. Borate uptake, penetration and retention by analysis in western hemlock. 38 When the drying processes were referenced to conventional drying, R F / V drying and S S / V drying showed significant increases in C C A penetration and retention in western hemlock. Presteaming plus conventional drying, high temperature, R F / V drying and S S / V drying reduced borate penetration, while R F / V drying, and S S / V drying increased borate retention. 3.3.1.4. Amabilis fir In amabilis fir, when all o f the drying processes were compared with either air drying or conventional drying, none showed any effect on either the penetration or retention o f both C C A and borate. 3.3.1.5. Douglas-fir In Douglas-fir, when using air drying as the reference process, R F / V drying significantly increased the C C A penetration, but there was no difference in the C C A retention for all o f the drying regimes. Conventional drying, R F / V drying and S S / V drying increased borate penetration compared to air drying, but only conventional drying and R F / V drying increased borate retention. 39 14 12 10 • w eight uptake (kg/nf^) • retention (kg/m*3) • penetration (mm) Air DH Conv. Pre- HighT RF/V SS/V steaming Figure 3-8. C C A uptake, penetration and retention by analysis in amabilis fir. 14 12 10 • weight uptake (kg/mA3) • retention (kg/mA3) • penetration (mm) __ Jm IB JB_ , IB. H i Air DH Conv. Pre- HighT RF/V SS/V steaming Figure 3-9. Borate uptake, penetration and retention by analysis in amabilis fir. 40 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 • weight uptake (kg/mA3) • retention (kg/nV )^ •penetration (mm) Air DH Conv. Pre- HighT RF/V SS/V steaming Figure 3-10. C C A uptake, penetration and retention by analysis in Douglas-fir. 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 • weight uptake (kg/mA3) • retention (kg/rn )^ • penetration (mm) Z L Air DH Conv. Pre- HighT RF/V SS/V steaming Figure 3-11. Borate uptake, penetration and retention by analysis in Douglas-fir. 41 Compared to conventional drying, none o f the different drying regimes showed any influence on the C C A penetration or retention in Douglas-fir. For the borate treatment, high temperature drying, prestearning plus conventional drying, air drying and dehumidification drying all reduced the penetration, compared to conventional drying. Dehumidification drying, S S / V drying, prestearning plus conventional drying and air drying reduced the borate retention. 42 3.4. Discussion 3.4.1 Comparison of two preservative treatments In general, there was no consistent pattern for the C C A penetration and retention, or the borate penetration and retention data. N o drying regime improved both the penetration and retention for a preservative in all species. Borates do not fix to wood during treatment. Therefore, the borate penetration pattern should be similar to that o f water and act as a reference. On the other hand, C C A fixes to wood, and this fixation may influence the C C A penetration. It is useful to compare the C C A penetration pattern with that o f borate. It would be expected that the mean C C A penetration should be significantly lower than that o f borate. This was indeed observed for some wood species and drying combinations, white spruce with dehumidification drying, high temperature drying, R F / V drying and S S / V drying (Figure 3-12); western hemlock with air drying, dehumidification drying and conventional drying (Figure 3-13); amabilis fir with conventional drying (Figure 3-14); and Douglas-fir with conventional drying, R F / V drying and S S / V drying (Figure 3-15). However, in most cases, there was no difference between the C C A penetration and borate penetration. Indeed, in some species, the C C A penetration was actually higher than the borate penetration. Examples o f this are western hemlock (Figure 3-13) dried by R F / V drying and S S / V drying, and Douglas-fir (Figure 3-15) dried by prestearning plus conventional drying and lodgepole pine (Figure 3-16) dried by R F / V drying. This observation may be caused by over-drying the timber. The target moisture content had been set at 22-25%. But for some drying processes, the moisture content o f the lumber may have been lower. 43 • CCA Air DH Conv. Pre- HighT RF/V SS/V steaming Figure 3-12. C C A and borate penetration in white spruce. Where: Y* means that C C A penetration is significantly different from borate penetration (at 95% level o f significance). i 6 1 4 Figure 3-13. C C A and borate penetration in western hemlock. 44 Air DH Conv. Pre- HighT RF/V SSA/ steaming Figure 3-14. C C A and borate penetration in amabilis fir. Air DH Conv. Pre- HighT RFA/ SSA/ steaming Figure 3-15. C C A and borate penetration in Douglas-fir. 45 • CCA • borate Air DH Conv. Pre- HighT RF/V SS/V steaming Figure 3-16. C C A and borate penetration in lodgepole pine. For example, lodgepole pine was dried by R F / V drying to 19%, and western hemlock was dried by R F / V drying and S S / V drying to 17.7% and 19.3% respectively. A t these lower moisture contents, the borate wil l not diffuse well. Despite the presorting for treatability and for moisture content, there was still a very high degree o f variation among sub-groups. Some o f the final moisture contents, after planing and before treating, were higher than the target moisture content o f 22%, e.g. western hemlock, amabilis fir and Douglas-fir (Appendix 1, Tables A l - 3 , A l - 4 , and A l - 6 ) . This was mainly due to uneven drying o f the rough, green lumber. Planing the rough lumber (38 mm x 89 mm) would remove the drier wood located at the surface. The freshly exposed surface may then have a higher moisture content. For the white spruce and lodgepole pine, the smaller dimension green lumber required less planing, so it would have less o f an effect on the final moisture content. 46 Although target moisture contents were not met (after planing) in all cases, the average moisture contents were still within the range normally considered ideal for pressure treatment (20-30%). N o relationship could be identified between the C C A penetration or retention and the average moisture content, but it would be expected that moisture content wi l l influence borate diffusion after preserve treatment. When the moisture content is lower than about 20%, the borate diffusion wil l be reduced (Schoeman, 1998). The degree o f diffusion may then also depend upon the time between the treatment and the removal and oven-drying o f the slice recovered for the chemical analysis. The C C A penetration data was recorded after allowing two weeks for fixation. During this time, some additional movement o f copper might occur. However, for the borate, the penetration data was recorded immediately after treatment, so the borate had little chance to diffuse. 3.4.2. The treatability of five species The C C A and borate treatability o f the five species decreased in the order (Figures 3-17 to 3-20): amabilis fir> western hemlock> lodgepole pine> white spruce = Douglas-fir. Amabilis fir has been recognized as one o f Canada's most treatable heartwood species in a previous study (Morris, 1995). It is, therefore, very difficult to achieve further improvement in penetration and retention. White spruce and Douglas-fir are so refractory (difficult to treat) that the anticipated degree o f improvement may be o f little practical benefit. White spruce required very high density incising (Morris, 1991; Ruddick, 47 Figure 3-17. C C A penetration in five wood species after drying with the seven drying processes. • amabilis fir • hemlock • lodgepole pine • I • Douglas-fir l l l y i white spruce HI SSA/ Figure 3-18. C C A retention in five wood species after drying with the seven drying processes. 48 • amabilis fir Air DH Conv. Pre-steaming HighT RF/V SSA/ Figure 3-19. Borate penetration in five wood species after drying with the seven drying processes. • amabilis fir steaming Figure 3-20. Borate retention in five wood species after drying with the seven drying processes. 49 1986 and 1989), and long treating schedules (Morris, 1991) to treat successfully to North American standards. Any comparison o f average penetration and retention data for each wood species must be made with caution, when the penetration is not uniform. Such errors can be reduced by sampling each specimen in several locations, and by examining a large number o f specimens. It is very interesting that the C C A penetration in lodgepole pine was higher than that in Douglas-fir (Figure 3-17), but the retentions for these two wood species were similar (Figure 3-18). The borate penetration and retention in lodgepole pine were higher than those in Douglas-fir (Figures 3-19 and 3-20). Thus, for lodgepole pine, the overall C C A retention must be higher, and the C C A treatment more erratic. In figure 3-21, ray cells in C C A treated lodgepole pine were penetrated, but some tracheids were not, particularly at the limit o f penetration. This was different from boards treated with borate (Figure 3-22). For the Douglas-fir treated with C C A , all the tracheids and ray cells were well penetrated in the treated region, but sometimes latewood was better penetrated than earlywood for both chemical treatments. In this study, not all o f the boards had a tangential grain orientation in which the annual ring is about 45° angle to the board edges (Figure 3-23 and 3-24). Some had flat grain and others vertical grain. This variation in grain direction may cause lower penetration in some boards. It might be anticipated that species like western hemlock, which are considered to be moderately treatable, could benefit more from processes, such as drying, than refractory species, such as white spruce. However, none o f the drying regimes gave consistently higher penetrations in western hemlock with both C C A and borate solutions. 50 Figure 3-21. High temperature dried lodgepole pine lumber treated with C C A . Figure 3-22. High temperature dried lodgepole pine lumber treated with borate. 51 Figure 3-23. Prestearning plus conventional dried Douglas-fir lumber treated with C C A . Figure 3-24. Prestearning plus conventional dried Douglas-fir lumber treated with borate. 52 It should be noted that the most treatable material was culled from the mix during the initial assessment o f permeability. In addition, none o f the lumber was incised, so that the values recorded were much lower than that usually achieved in a commercial treatment plant. 3.4.3. Penetration pathways and preservative distribution In this trial, boards being pressure treated were more than two meters long. The samples for examination were recorded from the center o f each board. The penetration recorded on the cross sections was, therefore, lateral penetration, which on the heartwood face, would be predominantly radial penetration. Thus the differences in the penetration recorded for the five wood species may reflect mainly differences in radial permeability o f the heartwood o f the five wood species. The penetration o f solution into wood can be considered as taking place in two stages - the first when the solution enters the cell lumen and the second, the diffusion o f the solution into the cell wall. Considering the first stage - the movement o f the solution into the wood. I f this penetration pathway is mainly in the radial direction, one may identify three pathways o f penetration. The first is that some solution enters into ray cell lumens at the wood surface and then travels along the ray cells or directly into ray cells at the wood surface. The solution can then pass into tracheids lumens which are adjacent to ray cells, by way o f the cross-field pit pairs. A second pathway is that some solution moves into the tracheid lumens at the wood surface and then passes into interval tracheid lumens via bordered pit pairs connecting the tracheids. A third pathway may occur where there are resin canals in the wood species. The solution may penetrate into wood along 53 resin canals. During the second stage, the solution may diffuse into the cell wall from the cell lumen, and even move through the primary wall and middle lamella into walls o f adjacent cells in which no solution had penetrated into the lumen. Resin canals, bordered pits and cross-field pits (especially for radial penetration) are the principal anatomical features, which influence preservative penetration in wood. It would appear that the main penetration pathways are through the surface tracheids, and from there via the pits to adjacent tracheids, or alternatively through radial resin canals to longitudinal resin canals and from these, by way o f the pits to adjacent tracheids. The rays provide a path for lateral penetration, while the pits provide channels between adjacent cells (Wardrop and Davies, 1961). The solution moves into wood along the ray cells much faster than across the tracheid walls or along the middle lamella. In order to penetrate from the wood surface, the solution from the ray cell passes into adjacent tracheids via cross-field pit pairs. The type and number o f cross-field pit pairs in softwood can influence the radial penetration. In lodgepole pine, the cross-field pits are pinoid pit pairs [Figure 3-25(a)], while both white spruce and Douglas-fir have piceoid pit pairs [Figure 3-25(b)]. In amabilis fir, taxodioid pit pairs are found [Figure 3-25(c)], while in western hemlock, they are piceoid or cupressoid pit pairs [Figure 3-25(b) and (d)]. From Figure 3-25, it is clear that pinoid and taxodioid pit pairs have larger openings than the others, while the piceoid pit pairs have the smallest openings in cross-field pit pairs. One can then conclude, that based upon the size o f pit openings, amabilis fir or lodgepole pine should be more treatable than hemlock, with white spruce and Douglas-fir having the poorest penetration. 54 (c) (d) Figure 3-25. The type o f the cross-field pit pairs occurring in wood. (a), pinoid pit pairs (b). piceoid pit pairs (c). taxodioid pit pairs (d). cupressoid pit pairs O f the five wood species, only three, lodgepole pine, white spruce and Douglas fir contain resin canals. The resin canals should enhance treatability, since the longitudinal resin canals and radial resin canals are connected to form a network that can allow preservative solution to penetrate through the resin canals into adjacent tracheids. In lodgepole pine, there are many large diameter resin canals (80-90 um in diameter), which are surrounded by the thin walled epithelium. One may, therefore, expect resin canals to be an important pathway in lodgepole pine. However, In white spruce and Douglas-fir, there are only a few, small diameter resin canals (60-90 um in diameter), which are 55 surrounded by a thick walled epithelium. The contribution o f resin canals wil l , therefore, be reduced in white spruce and Douglas-fir. Despite large cross-field pits and more and large resin canals, lodgepole pine did not receive uniform penetration for C C A preservatives. From Figure 3-21, we can see that ray cells were penetrated far ahead o f tracheids. Booker (1990) has suggested that the reason for the high radial permeability in radiata pine is the resin canals, which provide a pathway for the solution. The solution penetrates wood from the radial surface via the large number o f radial resin canals, after which the flow is redistributed axially to other radial resin canals (Booker, 1990). This would result in good penetration in this species. White spruce was not treated very well by either preservative, after applying the seven drying regimes. Western hemlock received a heavy, uneven C C A treatment. Generally, C C A penetration o f latewood was better than earlywood (Figure 3-26), while borate treatment was quite uniform (Figure 3-27). Amabilis fir was the most permeable o f the five species. However, both C C A and borate treatments were quite erratic, with the preservative generally following the earlywood in some annual rings, but failing to penetrate adjacent rings to any degree. Within each annual ring, earlywood was better treated than latewood (Figure 3-28 and 3-29). 56 Figure 3-26 Superheated steam/vacuum dried western hemlock treated with C C A . Q ^ ^ B l - 111. 1 Figure 3-27. Superheated steam/vacuum dried western hemlock treated with borate. 57 Preservative penetration is the critical factor in determining the ability o f a wood treated to meet wood preservation standards, because retention is quite easily enhanced by increasing the solution strength. It is penetration that is ultimately most seriously affected by permeability. Where there were positive effects on penetration compared to air drying, these were found with S S / V and/or R F / V drying. It therefore seems likely that the incorporation o f vacuum in these processes may be opening up, or preventing the closure, o f pathways for preservative penetration. R F / V drying was particularly effective on Douglas-fir. 59 Chapter 4. The Distribution of C C A in Five Softwoods Based on Light and Scanning Electron Microscopy 4.1 Introduction When C C A solution penetrates wood under pressure, two main processes occur. One is the bulk flow through the interconnected voids found in wood, such as cell lumens and pits. The second is diffusion, which consists of two types: intergas diffusion, which includes the transfer o f water vapor through the air in the lumens o f cells, and bound-water diffusion, which takes place within the cell walls o f wood (Saul, 1984). The intergas diffusion does not help C C A solution diffuse into cell wall because the C C A components do not move with water vapor. This kind o f diffusion can only deposit or concentrate the C C A components at the cell lumen surface. The bound-water diffusion is more important and determines where the C C A components wi l l fix in the cell wall. Several methods have been developed which can be used to determine the C C A component distribution in wood. These include light microscopy (Belford et al., 1959), X -ray microanalysis ( E D X ) (Petty and Preston, 1968), transmission electron microscopy ( T E M ) (Dickinson, 1974) and scanning electron microscopy (SEM) equipped with x-ray microanalysis (Greaves, 1974), and laser microprobe mass analysis ( L A M M A ) (Klein and Bauch, 1979). The most commonly used methods are light microscopy and S E M - E D X . This is because microscopy can provide direct observation o f the chemical location in thin sections after staining, while S E M - E D X is able to study the distribution o f several chemical elements at a time in the wood cell wall. 60 4.1.1. Light microscopic examination of treated wood Although CCA-treated wood generally darkens on treatment, this change in color is too weak to allow the distribution o f the preservative to be observed in thin sections. Also the color can not be related to the distribution o f the individual components o f C C A . Therefore, the fixation and staining o f the target chemicals are essential prior to attempting to determine the distribution o f the C C A within the wood (Yata et al, 1978). Specific staining reactions must, however, be used with caution as the C C A components can interfere with the reactions o f each other (Belford et al, 1959). For instance, copper, manganese, nickel, and cobalt salts wi l l interfere with the chromium (Cr 6 + ) reaction with diphenyl carbazide (Vogel, 1979) causing erroneous assessment of the chromium. Belford et al. (1959) reported the optical microscopic examination o f stained sections to identify the distribution o f copper, using dithio-oxamide to stain C C A treated wood. Another approach that has been adopted is to use diphenyl carbazide as a staining reagent for chromium (Cr 6 + ) and measure the absorbency at the specified wavelength in a light spectrometer (Wardrop and Davies, 1961). Yata et al. (1979, 1981a, b, and 1982) developed a series o f papers describing morphological studies on the movement o f substances into the wood cell walls. Thin sections o f wood were treated with copper sulfate or copper formate solution, and then stained with dithio-oxamide to detect copper (Yata et al, 1979). In another investigation, thin sections o f wood treated with potassium chromate or potassium dichromate solution were stained with an aqueous solution o f silver nitrate to detect hexavalent chromium (Cr 6 + ) (Yata et al, 1981b and 1982). However, there have been no reports on studies o f the distribution o f all three C C A components using light microscopy. A better understanding o f the mechanism o f 61 C C A penetration and reaction with wood may be gained by investigating the distribution o f all three components in wood. 4.1.2. Analysis of the microdistribution of preservatives in wood by S E M - E D X Analysis o f micro-distribution o f the preservative in relation to wood structure is carried out using an S E M coupled with an E D X . It has the advantage o f being able to study several chemical elements at a time, providing elemental profiles o f samples. There are, however, many uncertainties associated with electron probe X-ray analysis so that it usually provides only semi-quantitative measurements (Hall and Hohling, 1968). The question o f incomplete fixation is not always adequately addressed. Different wood species are used. It is not surprising then to find many apparent differences between published results. Even the X-ray analysis technique is subject to errors. For example, loss or movement o f preservative elements may occur during specimen preparation. Therefore, when the microdistribution o f preservatives is studied, minimal sample preparation is preferred, in order to limit any chemical redistribution which may occur (Ryan, 1986). There is little danger o f element redistribution with S E M samples when they are air dried and carbon coated before examination (Russ, 1984 and Ryan, 1986). Problems can arise where the sections are wetted prior to microtoming, as this can remove mobile preservative components. Other difficulties can be experienced when S E M - E D X is used to analyze preservative distribution. Yata et al. (1983) estimated that, with an incident beam o f 20 keV, the diameter o f the surface analyzed is approximately 3.6-4.5 um, with a corresponding volume o f up to 50 um 3 . X-rays may then be generated in adjacent regions, 62 some o f which may contain chemical deposits lining the lumen wall (Hulme and Butcher, 1977). In some cases, such as ray cells, this lining may be so dense as to completely fill the cell lumen. Thus S E M - E D X bulk analysis may not reflect the amount o f chemical at a specific location due to inclusion o f adjacent deposits, which may be found on the cell wall. Consequently, the S E M - E D X data may show some variabilities, during which individual observation may not be consistent with overall trends recorded for the C C A microdistribution in different cell types. 63 4.2. Methodology 4.2.1. Sample selection The cross sections for the analysis o f the microdistribution were removed from the midpoint o f the lumber, so that in the absence o f incising to aid penetration, the movement of the C C A solution was generally restricted to the lateral direction, i.e. either radial or tangential movement from each sawn surface. This study wil l tend to focus on radial penetration, since sections were prepared from the heartwood face. Penetration in the radial direction is generally considered to be better than that in the tangential direction, due to the higher permeability of the ray cells and resin canals. O f the five species, lodgepole pine, white spruce and Douglas-fir had resin canals, while western hemlock and amabilis fir did not. A section (ca. 20 mm thick) was cut from the midpoint of representative pieces of lumber (see Figure 3-1). For the lodgepole pine and spruce, samples for microscopy were prepared from representative sections for all drying regimes examined. For the Douglas-fir, western hemlock and amabilis fir, sections were chosen to show the full range o f penetrations observed for the species, irrespective o f the drying regime. In this way the microdistribution o f C C A both in completely treated wood as well as at the limit o f penetration were recorded. However, the most useful information on the pathway taken by the preservative was found from the samples at the limit o f the C C A penetration. The sections were oven-dried to 10% moisture content at 50°C after which radial and cross sections were prepared from the heartwood face for subsequent microtoming into thin sections. Observations o f the distributions o f C C A components in different cell types were made using stained sections and a light microscope. The examination of 64 preservative distribution within the wood cell wall was done using an S E M coupled to an E D X . This can be readily accomplished for C C A since all three elements provide a strong response to X-ray excitation with a relatively high binding energy so that detection o f the relevant X-rays is efficient. 4.2.2. Light microscopy 4.2.2.1 Thin section preparation In preparing samples for light microscopy, all cutting and microtoming was carried out on dry samples, in order to rninimize preservative movement and loss o f C C A components. A small cube (10 mm(R) x 10 1 1 1 % ) x 20 mni(L)) was trimmed with a disposable razor to produce a smooth radial or cross sectional surface. The cube was mounted onto the sample holder o f a Spencer sliding microtome, with the smooth surface facing up. The blade o f the microtome was adjusted to produce 18 um thick sections. A few passes were made with a disposable blade to provide a uniform and flat surface, after which the blade was replaced with a new one. The sample surface was covered with a thin layer o f distilled water using a small brush and then thin sections were microtomed. The dimensions o f the transverse sections were 5(T) mm x 10(R) mm x 18(d urn, while the radial longitudinal sections were 8 (L) mm x 10(R) mm x 18 (T) um. The thin sections were removed immediately and placed into one o f the prepared staining solutions. The remaining portion o f each block was retained for the S E M - E D X examination. 65 4.2.2.2. Staining reagent selection Several reagents have been used in classical qualitative tests for chromium, copper and arsenic (Vogel, 1979). Following initial testing o f several reagents for each component, using thin sections o f treated wood, the most effective o f these were identified and used in the subsequent research. The most effective reagents were: dithio-oxamide for copper (Cu 2 + ) ; alizarin yellow R for chromium (Cr 3 + ) ; and silver nitrate for arsenic (As 5 + ) . Diphenyl carbazide reacts with chromium (Cr 6 + ) to give a pink or violet color. Since the chromium (Cr 6 + ) interferes with the arsenic reagent - silver nitrate, the samples were first tested with diphenyl carbazide to corifirm the completion o f fixation and the absence o f chromium (Cr 6 + ) . 4.2.2.3. Thin section staining Copper (Cu 2 +) The thin sections were soaked in 0.2% dithio-oxamide ethanol solution overnight, after which they removed and washed with ethanol until the wash-ethanol was free of color. They were kept in 1:1 ethanol and glycerin solution for 10 minutes, after which they were immersed for 10 minutes in pure glycerin. Each section was mounted between a microscopic glass slide and glass cover. The cells containing copper (Cu 2 + ) were stained dark green, while untreated cells retained their original color. Chromium (Cr 6 +) The thin sections were soaked in sulphuric acid (0.5 % H2SO4) for two to three seconds. This served to destroy the interference produced by copper, after which the thin 66 sections were washed with distilled water to remove the acid (until p H test paper did not change its color in the wash water). The thin sections were dipped momentarily in 1.5% diphenyl carbazide acetone/water (2:1) solution. I f the treated cells were colored pink, the C C A had not completely fixed. In this study, none o f the treated cells became pink in color, confirming that C C A had completely fixed in all cases. Chromium (Cr 3 +) The thin sections were stained overnight with a saturated aqueous alizarin yellow R, after which they were dipped momentarily in 0.5% H 2 S 0 4 . This removed the interference produced by other metal elements (e.g. copper). The thin sections were washed with distilled water to remove the acid (until p H test paper did not change its color in the wash water) and then placed in 2% ammonium hydroxide solution for 5 minutes. I f chromium (Cr 3 + ) from the C C A treatment was present, the wood became stained an orange-red color. I f chromium (Cr 3 + ) was absent, the dyestuff turned to orange and dissolved in the N H 4 O H solution. The stained sections were washed with distilled water until the washing water was free o f color. Arsenic (As s +) After making sure that the C C A in representative treated thin sections had completely fixed (i.e. they did not contain any chromium (Cr 6 + ) , by testing with diphenyl carbazide), matching thin sections were stained with 10 % aqueous silver nitrate solution for 5 hours. The arsenate created a brown precipitate. The stained sections were washed five times by immersing each section for two or three minutes in distilled water. 67 4.2.2.4. Dehydrating process Following staining, the sections were dehydrated by immersing sequentially for 10 minutes in 30 %, 50 %, 75 % and 100 % ethanol solution, after which they were placed in a 50 % glycerin/ethanol solution for a further 10 minutes. Finally, they were placed in pure glycerin for 10 minutes, before being mounted with glycerin between a microscopic glass cover and slide for observation under the light microscope. 4.2.2.5. Microscopic observation and photography In order to check the C C A component distribution, the stained thin sections were examined using a Jenamed 2 (Carl Zeiss) light microscope. Samples o f characteristic observations were recorded with a camera (Pentacon) attached to the microscope, using 135 Fuji 100 film. For the calculation o f the final image magnification, a correction factor must be established for each photograph based on the enlargement o f the negative. A clear characteristic feature on each film negative was measured to provide the actual size o f the feature. The same feature was then measured on the final print made from the negative. The enlargement due to the printing process was determined by dividing the size on the picture by that on the negative film. The total magnification o f a feature on the picture, was calculated by multiplying the magnification provided by the microscope with that for the printing o f the image. 68 4.2.3. S E M - E D X 4.2.3.1. Sample preparation Sample preparation for S E M - E D X analysis is very important, since it directly affects the result o f the chemical distribution measured. The sample surface should be as clean and smooth as possible. The sample moisture content was maintained below 10% to minimize chemical redistribution. Additional cross sections (5 mm x 5 mm by 2 mm thick) and radial sections (5 mm x 5 mm by 2 mm thick) were cut from the smooth surface o f each block, following removal o f the thin sections for the visible microscopic study. In this way the S E M - E D X observations provided relative information that can be compared to the visible microscopic observations. Transverse sections and radial sections, with the smooth surface upper-most, were fixed on aluminum stubs using double-sided tape and coated with carbon film (thickness o f 300-400 A ) in a vacuum thermal evaporator. •4.2.3.2. SEM-EDX examination and analysis location selection A n initial visual examination o f sections was made by using a Hitachi S-500 scanning electron microscope (SEM) . This preliminary assessment o f the sections was performed in order to identify similar locations for a more detailed analysis o f different cell types located both at the surface and at the limit o f preservative penetration. The electron microscopic analysis utilized an Hitachi S E M coupled with a Kevex 700 energy dispersive x-ray spectrometer ( E D X ) . The operating conditions, i.e. the working distance (35mm) of the X-ray detector from the samples, the accelerating voltage (20 K e V ) , illuminating current (200 uA) and tilting angle o f specimen (0°), were held constant throughout the entire study. The E D X analysis windows for chromium, copper 69 and arsenic and their corresponding background regions were created (Figure 4-1). Using computer software, each X-ray analysis was recorded for 200 seconds. The X-ray beam can penetrate about 7 um into the sample, over an area o f 5 um wide in wood samples. This resulted in data representing the average from the total volume penetrated by the X -ray beam. Selection o f the analysis locations is very important, because it directly affects the analysis results. In order to reduce the data variation and to increase comparability, the analysis locations were chosen to be as similar as possible, for the same characteristic feature in the different samples. For example tracheid walls and ray cell walls at specific locations from the treated wood surface were selected, so that the trends in chemical microdistribution could be compared in different species. When assessing variation in C C A from the surface to the limit o f penetration, the interval (radial) between two sequential measurements was different for each wood species, because o f their different treatabilities. For a more refractory wood species, the number o f tracheids in each interval was small; for example, the interval between two analysis levels for white spruce and Douglas-fir was five tracheids, while that for lodgepole pine, western hemlock and amabilis fir was ten tracheids. The interval o f tangential measurement was two tracheids for all five species. Measurements were also made on the tracheids, ray cells, resin canals, pits and lumen surfaces at the limit o f C C A penetration since this provided the best evidence o f the pathways taken by the preservative when penetrating the heartwood. 70 O © S i (XI 0 : 01 : U> : CD : tS) • 0J : (VI CSD I D m i ~o 0> • H Oj I* II Q CO f -.—# - .—t r~ O rrJ OJ 4-> c ZJ u o o o r^-i n CO CJ O \"O *0 O ii +> 3 c a H i 4> % Pt4 71 4.2.3.3. EDX data analysis To establish the relative concentrations o f each preservative component (copper, chromium and arsenic) in C C A treated wood, cross and radial sections were examined by a scanning electron microscope coupled with an energy dispersive spectrometer ( S E M -E D X ) . The corrected peak to background ratios o f arsenic, copper and chromium were calculated to represent the relative concentration o f the C C A components. X-ray counts o f arsenic, copper and chromium were converted into corrected peak to background ratios (peak area) using the following equation: Pb B Where: Rpb: Corrected peak to background ratio (peak area), P: X-ray peak count and B: X-ray background count. The absolute magnitude o f the chromium, copper and arsenic peak to background ratio is influenced by many factors, such as wood density at the analytical location, and matrix effects due to the variable amounts o f the C C A components present. It is likely therefore to vary from location to location. This difficulty can be overcome by comparing the ratios o f the corrected peak to background ratios for any two elements. In this way, trends in changes in element concentration can be observed. To estimate the relative amount o f each C C A component in the treated wood, the copper and chromium peak areas were compared with the arsenic peak area (referred to as Cu:As and Cr :As ratios), while the chromium peak area was also compared with the 72 copper peak area (referred to as Cu:Cr ratio). A l l data were displayed as histograms. While variation in the magnitude o f individual results for the different locations was to be expected for lodgepole pine and spruce, the relationship between arsenic, copper and chromium peak areas and the trend in the Cu:As, Cr :As and Cr :Cu ratios were found to be similar for all seven drying regimes. On the other hand, the influence o f the wood species on the C C A distribution was very significant. Consequently, it was more important to examine the microdistribution in each wood species, rather than that in every drying regime. Therefore, only two drying regimes were used to illustrate the relationship o f the arsenic, copper and chromium peak areas and the trend o f the Cu:As, Cr :As and Cr :Cu ratios in lodgepole pine and white spruce, and only one drying regimes was used to illustrate the characteristics o f the other species. 4.2.3.4. Reference sample preparation A few drops o f C C A solution were placed on a small glass slide, and allowed to evaporate to dryness. The glass slide was fixed, with the C C A precipitate upper-most, on aluminum stubs using double-sided tape. It was coated with carbon film (thickness about 900 A ) in a vacuum thermal evaporator. The C C A solid was analyzed by E D X . The corrected peak to background ratio o f the C C A solid served as a reference during examination o f the treated wood sections. 73 4.3. Results 4.3.1. The pathways of C C A solution penetration as determined by light microscopy 4.3.1.1. Copper distribution Figures 4-2 to 4-8 show the reaction o f copper with dithio-oxamide. For all five wood species, copper deposits were observed in some of the tracheid lumen (Dc in Figures 4-2 and 4-3) that were near to the ray cells. Copper was unevenly distributed within the tracheid wall. The compound middle lamellae and cell corners in particular, were stained very deeply ( M in Figures 4-2 to 4-4) compared to secondary walls. Examination o f copper deposited in the bordered pits showed that it varied markedly from one pit to another, at the limit o f penetration. Some pits were strongly colored while others were not (Pb in Figures 4-2, 4-3 and 4-5). In Douglas-fir, the spiral tWckenings on the lumen surface o f tracheids were clearly visible and were often strongly colored (S in Figure 4-5). The ray cells (R in Figures 4-2 to 4-6) were heavily stained a dark green color, identifying their importance in transporting the C C A solution. The semi-bordered pit pairs were deeply stained (Ps in Figures 4-3 and 4-4) for all o f the species examined. For the lodgepole pine, white spruce and Douglas-fir, the resin canals were stained more intensely than the surrounding tracheids (Rc in Figures 4-4, 4-6 and 4-7). In the western hemlock, longitudinal parenchyma (P in Figure 4-8) were observed at the boundary o f earlywood and latewood. The copper was present in much higher concentration in the longitudinal parenchyma, than the surrounding tracheid cell walls. 74 Figure 4-2. Abies amabilis. A micrograph o f copper distribution (dark green color) at the limit o f penetration. Magnification 200x. Dc: copper deposit. M : middle lamella. Pb: bordered pit pair. R: ray. Figure 4-3. Abies amabilis. A micrograph o f copper distribution (dark green color) at the limit o f penetration. Magnification 200x. Dc: copper deposit. M : middle lamella. Pb: bordered pit pair. Ps: semibordered pit pair. R: ray. 75 Figure 4-4. Picea glauca. A micrograph o f copper distribution (dark green color) in cells at the wood surface. Magnification 50x . M : middle lamella. Ps: semibordered pit pair. R: ray. Rc: resin canal. Figure 4-5. Pseudotsuga menziesii. A micrograph o f copper distribution (dark green color) at limit o f penetration. Magnification 200x. P b l : bordered pit on radial section. Pb2: bordered pit on tangential section. R: ray. S: spiral thickening. 76 Figure 4-6. Pinus contorta. A micrograph o f copper distribution (dark green color) at limit of penetration. Magnification lOOx. R: ray. Rc : radial resin canal. r r ~ o 3 c - D O O O U ( IS nob q . o; 6 o. o o o 013 U U Figure 4-7. P/wms contorta. A micrograph o f copper distribution (dark green color) at the limit o f penetration. Magnification 50x. Rc : resin canal. 77 Figure 4-8. Tsuga heterophylla. A micrograph o f copper distribution (dark green color) at the limit o f penetration. Magnification 200x. P: longitudinal parenchyma. 4.3.1.2. Chromium distribution Figures 4-9 to 4-19 show the distribution o f chromium (III). Like the copper distribution in the tracheid wall, the compound middle lamellae and cell corners were deeply colored orange-red, compared to the secondary walls ( M in Figures 4-9 to 4-11). They were stained unevenly, both in latewood and earlywood and were colored more intensely in the radial direction ( M r in Figure 4-12) than in the tangential direction (Mt in Figure 4-12). Tracheids associated with ray cells (at the point o f the section taken) were more heavily stained than those further from the ray cells (Figure 4-13). This pattern was present in all five wood species. Some bordered pit pairs were stained heavily, while others were not (Pb in Figures 4-9 and 4-14). The crassulea (C in Figure 4-15) present at the tracheid lumen surface o f lodgepole pine, white spruce and amabilis fir were stained deeply compared to the pit areas (Pb in Figure 15). 78 The ray cells were heavily stained an orange-red color (R in Figures 4-9 to 4-11 and 4-13), which indicates the increase to chromium (III) at this location. The semi-bordered pit pairs were also stained deeply (Ps in Figure 4-13) For lodgepole pine and white spruce, the resin canals were stained more intensely than the surrounding tracheids (Rc in Figures 4-10, 4-11 and 4-16). However, in Douglas-fir (Figure 4-17) the resin canal was lightly stained compared to the ray cells and surrounding tracheids. In western hemlock, chromium was present in much higher concentration in the longitudinal parenchyma (P in Figures 4-18 and 4-19) than the surrounding tracheids. Figure 4-9. Tsuga heterophylla. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. Magnification 200x. M : middle lamella. Pb: bordered pit pair. R: ray. 79 Wood surface Figure 4-10. Picea glauca. A micrograph o f chromium distribution (orange-red color) in cells at the wood surface. Magnification lOOx. M : middle lamella. R: ray. Rc: resin canal. Figure 4-11. Picea glauca. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. Magnification 50x. M : middle lamella. R: ray. Rc: resin canal. 80 I Figure 4-12. Picea glauca. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. X200. M r : radial middle lamella. M t : tangential middle lamella. 81 Figure 4-14. Tsuga heterophylla. A micrograph of chromium distribution (orange-red color). Magnification lOOx. Pb: bordered pit pair. Figure 4-15. Pinus contorta. A micrograph o f chromium distribution (orange-red color) at the limit penetration. Magnification 200x. C: crassulae. Pb: bordered pit pair. 82 Figure 4-16. Picea glauca. A micrograph o f chromium distribution (orange-red color) at the limit penetration. Magnification lOOx. Rc : resin canal. Figure 4-17. Pseudotsuga menziesii. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. Magnification lOOx. Rc : resin canal. 83 Figure 4-18. Tsuga heterophylla. A micrograph o f cliromium distribution (orange-red color) at the limit of penetration. Magnification lOOx. P: longitudinal parenchyma. Figure 4-19. Tsuga heterophylla. A micrograph o f chromium distribution (orange-red color) at the limit o f penetration. Magnification lOOx. P: longitudinal parenchyma. Pb: bordered pit pair. 84 4.3.1.3. Arsenic distribution Figures 4-20 to 4-25 show the reaction o f arsenic with silver nitrate. For all five wood species, arsenic deposits were observed in some o f the tracheid lumen (Da in Figure 4-20) adjacent to, or near to the ray cells. Arsenic was unevenly distributed in the tracheid cell wall. The compound middle lamellae and cell corners in particular were stained very slightly ( M in Figures 4-21 and 4-22) compared to the secondary walls (Ws in Figures 4-21 and 4-22). Some o f the bordered pits were strongly colored while others were not (Pb in Figures 4-20, 4-21, 4-23 and 4-24). High concentrations o f arsenic were present at the torus of some bordered pits (Figure 4-23). The ray cell (R in Figures 4-22, 4-24 and 4-25) was heavily stained a brown color. The semi-bordered pit pairs were also deeply stained (Ps in Figure 4-20) for all five species. For the lodgepole pine, white spruce and Douglas-fir, the resin canals were stained more intensely than the surrounding tracheids (Rc in Figures 4-21 and 4-25). In western hemlock, arsenic was present in much higher concentration in the longitudinal parenchyma (P in Figure 4-20) than in the surrounding tracheids. 85 Figure 4-20. Tsuga heterophylla. A micrograph o f arsenic distribution (brown color) at the limit o f penetration. Magnification 200x. Da: arsenic deposit. P: longitudinal parenchyma. Pb: bordered pit pair. Ps: semibordered pit pair. R: ray cell. Wood surface Figure 4-21. Picea glauca. A micrograph o f arsenic distribution (brown color) in cells at the wood surface. Magnification 200x. M : middle lamella. Pb: bordered pit pair. Rc : resin canal. Ws: secondary wall. 86 Figure 4-22. Pinus contorta. A micrograph o f arsenic distribution (brown color) at the limit o f penetration. Magnification lOOx. M : middle lamella. Ws: secondary wall. Figure 4-23. Abies amabilis. A micrograph o f arsenic distribution (brown color) at the limit o f penetration. Magnification 200x. Pb: bordered pit pair. 87 Figure 4-25. Pinus contorta. A micrograph o f arsenic distribution (brown color) at the limit o f penetration. Magnification lOOx. R: ray. Rc: resin canal. 88 4.3.2. The C C A component distribution as determined by S E M - E D X 4.3.2.1. The CCA component distribution in tracheid walls The chromium, copper and arsenic peak areas for the tracheid walls o f all five wood species showed a higher concentration o f C C A components at the surface, than at the limit o f penetration (Figure 4-26). The relationship between the C C A component peak areas changed as the sampling zone moved into the wood. For example, the copper peak areas were lower than arsenic at the first two analysis levels and higher at levels three to five (Figure 4-26). The chromium peak areas measured on the tracheid walls decreased with increasing distance from the surface. As with chromium, arsenic and copper peak areas decreased gradually over the first four levels, but increased slightly at level five, confirming that the C C A distribution was uneven in this species. The peak areas o f arsenic, copper and chromium for solid C C A were 5.51, 6.74 and 12.44, respectively. A s expected they were greater than the treated wood. o 1 \"D C 3 O s_ O) o 1 o 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 • As[(P-B)/B] »Cu[(P-B)/B] ®Cr[(P-B)/B] • i i i L i 1 I 1 1 i , i i i 1 p l 1 i l 1 1 1 1 i n m i i 1 1 1 1 1 1 1 1 1 1 I 1 1 1 1 P I 1 P r i i • • • • • rrirp 1 r n 1 1 1 1 1 • • • I • • • r r r n 1 1 1 1 i i i i 1 1 1 1 f 1 1 1 1 T - T - CM CO CM T - CM £ r» r? 2 * f? r? CO CO 1 - CM CO CM CM > CO CO CO £ £ i 5 § § S ^ t - C M C O l O T - C M C O 1 a. f. •* 5 Figure 4-26. The arsenic, copper and chromium peak areas on the tracheid and ray cell walls o f amabilis fir. 89 A detailed analysis o f the C C A distribution was made o f the tracheid radial wall (TWRE) , tracheid tangential wall (TWTE) and cell corner ( T W C E ) o f earlywood, as well as the tracheid radial wall ( T W R L ) , tracheid tangential wall ( T W T L ) and cell corner ( T W C L ) o f Douglas-fir latewood (Figure 4-27). The latewood that was analyzed was closer to the surface so that the C C A component peak areas were higher than those in the earlywood (Figure 4-28). Some deposits were observed on the lumen surface o f earlywood (Figure 4-29-A) and latewood tracheids (Figure 4-29-B) o f amabilis fir. The deposits appeared to be similar. Such deposits have not been reported on the tracheid lumen surface o f untreated wood. They were therefore analyzed at the locations shown in Figure 4-29. Figure 4-27. A scanning electron microscope photograph o f Douglas-fir. The locations o f a resin canal, ray cells and tracheids which were analysed by E D X are marked. 90 • As [(P-B)/B] TWCL TWRL TWTL TWCE TWRE TWTE Figure 4-28. The arsenic, copper and chromium on peak areas on the tracheid walls o f Douglas-fir. Figure 4-29. The scanning electron microscopic photos o f the deposits on the lumen surface o f tracheids o f amabilis fir. A : earlywood. B : latewood. 91 The relationship of the C C A component peak areas for the deposit on the earlywood lumen surface (TD1) was similar to that on the lumen surface o f latewood (TD2). The copper and arsenic peak areas were very high for these two points (Figure 4-30). The chromium peak areas at these two points were similar (slightly enhanced) compared to the other sampling locations. The C C A component peak areas on the tracheid walls (TW1, TW3 and TW4) , and the lumen surface (TL1 , T L 2 and TL3) were very similar. The C C A content of the tracheid wall associated with a ray cell (TRW) contained slightly higher copper and arsenic peak areas. The closer the sampling locations were to the cell with the deposits, the higher were the copper and arsenic peak areas. Figure 4-30. The arsenic, copper and chromium peak areas at the deposits at the lumen surface o f the tracheids o f amabilis fir. 92 4.3.2.2. The CCA component distribution in the pit region Figure 4-31 shows the E D X analysis locations on a tracheid lumen surface and the associated bordered pits in amabilis fir. TP-a was located on the border edge o f a bordered pit close to the pit opening. In the corresponding E D X analysis data shown in Figure 4-32, the first three points were taken from one earlywood tracheid on level three, the second three points were from an earlywood tracheid on level four and the last three from one earlywood tracheid on level five (See section 3.3.2 o f the methodology). The three tracheids came from different samples. The corresponding E D X analysis o f the C C A components present in the cell walls were presented in Figure 4-26. The relative Figure 4-31. A scanning electron microscope photograph o f bordered pits on the tracheid wall o f amabilis fir. 93 o 3.5 4.0 • As [(P-B)/B] • Cu[(P-B)/B] Cr [(P-B)/B] TP-a1 TW1 TP- TL2 a2 TW2 TP-a3 TL3 TW3 Figure 4-32. The arsenic, copper and chromium peak areas on the bordered pits, the tracheid walls and lumens o f amabilis fir. proportion o f the C C A components for the three locations analyzed in each tracheid were the same, although there were differences between the component peak areas in the different tracheids. S E M - E D X analyses were also performed on bordered pits on the tracheid walls o f western hemlock, and Douglas-fir (Figures 4-34 and 4-35). The results showed that the copper and arsenic peak areas were relatively high on the pit area, compared to those on the tracheid walls. 94 Figure 4-33. A scanning electron microscopic photo o f the bordered pits on the tracheid wall o f Douglas-fir. • As ratio of P/B TL1 TP-m TP-b TP-a TL2 Figure 4-34. The arsenic, copper and chromium peak areas on bordered pits on the tracheid walls o f Douglas-fir. 95 4.3.2.3. The CCA component distribution in ray cells The relationship between C C A components on the ray cell walls, like that on the tracheid walls changed from the cells located at the surface to those at the limit o f penetration. The chromium peak areas o f the ray cell walls were slightly higher than those in the tracheid walls at the same analysis level. They slowly decreased with increasing distance from the surface (Figure 4-26). The arsenic peak areas also gradually decreased, while the copper peak areas remained fairly constant from R W 2 to R W 5 . Both arsenic and copper peak areas were very low at the surface o f treated wood (RW1 in Figure 4-26). 4.3.2.4. CCA component distribution on resin canals The component distribution arising from interactions between C C A solution and the resin canals o f treated lodgepole pine, was detennined from S E M - E D X measurements. These were made at different locations in the cell walls o f the resin canals (identified as double resin 1, double resin 2 and double 3) and the tracheids (TW1, T W 2 , T W 3 and TW4) surrounding the resin canals (Figure 4-35). The relationship between arsenic, copper and chromium in the resin canals was clearly quite different from that in the associated tracheids. The copper concentrations, relative to that o f arsenic and chromium, were extremely high in the resin canals. The tracheids showed a similar composition with respect to copper, chromium and arsenic in which the chromium was much lower than that was found in C C A solid. While copper peak areas were similar to that in solid C C A . 96 Figure 4-35. The scanning electron micrograph o f dehumidification dried lodgepole pine. double double double TW1 TW2 TW3 TW4 resinl resin2 resin3 Figure 4-36. The arsenic, copper and chromium peak areas in the double resin canal and tracheids o f dehumidification dried lodgepole pine. 97 The effect o f the transverse resin canal on the C C A distribution in the associated cells was determined from an E D X analysis o f selected cell areas. These were made on the transverse resin canal (resin), the ray parenchyma cell walls (RW) associated with the resin canal, the transverse tracheid wall (Ray T W ) , and the tracheid wall (TW1 and TW2) on each side o f the resin canal, as shown in Figure 4-37. The copper uptake was also greatly enhanced in the transverse resin canal. The ray cells associated with the resin canal also showed a very high copper retention, and the relative concentrations of chromium and Figure 4-37. A scanning electron microscope photograph o f a transverse resin canal in a radial section o f conventionally dried lodgepole pine. 98 arsenic were similar in both. The peak areas o f copper, chromium and arsenic measured for the longitudinal (TW) and transverse tracheids (Ray TW) were similar to each other (Figure 4-38). • As[(P-B)/B] rayTW resin RW TWl TW2 Figure 4-38. The arsenic, copper and chromium peak areas in the transverse resin canal, ray parenchyma wall, ray tracheid wall and longitudinal tracheid wall o f conventionally dried lodgepole pine. Figure 4-39 shows a resin canal associated with ray cells and tracheids in white spruce. Two analysis points were taken from the resin canal cell walls (resin ball and resin), while the other locations ( T W l and TW2) were taken from the tracheid walls and R W from the ray cell adjacent to the resin canal. The C C A component peak areas for each location are presented in Figure 4-39. 99 Figure 4-39. The scanning electron microgroph o f superheated steam/vacuum dried white spruce. 1.6 • As[(P-B)/B] • As[(P-B)/B] »Cr[(P-B)/B] resin resin ball RW TW1 TW2 Figure 4-40. The arsenate, copper and chromium peak areas on ray cell wall, resin canal and tracheid walls o f superheated steam/vacuum dried white spruce. 100 The arsenic, copper and chromium peak areas for white spruce are shown in Figure 4-40. The relationship between the peak areas o f C C A components in resin canal was different from that for the resin canals of lodgepole pine (Figure 4-38). Figure 4-27 shows a single resin canal and associated ray cells and tracheids in Douglas-fir. Various analyzed regions are marked, including the ray cell wall (RW) and TW3 and TW4 on the tracheid walls, close to the resin canal. The E D X analysis revealed high copper retention in the resin canal and the associated ray cell wall (Figure 4-41). This is similar to observations in lodgepole pine. However, the same data suggest that the chromium and arsenic content both in the resin canal, and ray cell wall, as well as in the walls o f the tracheids bordering the resin canal (TW3 and TW4) is very low compared to that on the trachieds more distant from resin canals. • As ratio of P/B Figure 4-41. The arsenic, copper and chromium peak areas on tracheid walls, ray cell wall and resin canal cell wall o f Douglas-fir. 101 4.4. Discussion 4.4.1. The pathways of C C A solution During pressure treatment o f timber, there are three main features that contribute to the bulk flow of solution from the wood surface into the wood. The first are the ray cells, which allow solution to move quickly into the wood. They also allow C C A solution quickly to penetrate into the tracheids adjacent to the ray parenchyma, through semi-bordered pit pairs that connect the two (Ps in Figures 4-2, 4-13 and 4-20). The solution can then penetrate into the next tracheids through the bordered pit pairs connecting the two tracheids (Pb in Figures 4-2, 4-13 and 4-20). So the pits are another important feature. The resin canals are the third influential feature, in those species where they are abundant, e.g. lodgepole pine. Together the interlinked radial resin canals (Figure 4-6) and longitudinal resin canals (Figure 4-7) form a network, through which the solution can penetrate into wood. The importance of the frequency o f the resin canals as an aid to penetration, is comparable to the influence o f the ray cells in those species where they are present (Figures 4-6 and 4-7). 4.4.2. C C A distribution When wood is treated with C C A under pressure, the resulting pressure gradient plays a more important role at the wood surface, than at the limit o f penetration. The major difference between the surface and limit o f penetration is the fact that at the surface, the C C A concentration is high and all components are uniformly present in all o f various cell walls. A t the limit o f penetration, the C C A concentration wil l be low so that the rate 102 of fixation slows and the solution is out o f balance. Therefore, the C C A component distribution in the cell walls at the limit o f penetration is uneven. A t the limit of penetration, a high concentration o f the three C C A components is present in the ray cells compared to the surrounding tracheids. Copper and chromium dominated in the compound middle lamella and cell corners, while arsenic, on the other hand, was most commonly found in the secondary walls. The resin canals absorbed more copper than the surrounding tracheids and ray cells. A l l these findings are summarized in Table 4-1, based on the light microscopic observations and the S E M - E D X examination. The copper:chromium:arsenic ratio in the treated wood differed from that in the treating solution. It varied from one location to another because o f chemical reaction with wood components (including extractives) as well as variation in preservative content in different cells. 103 Table 4-1. Summary o f C C A component distribution in elements o f treated wood. Species pine spruce hemlock Douglas--fir amabilis fir C C A component C u Cr A s C u Cr A s C u Cr A s C u Cr A s C u Cr A s Cel l corner + + - + + - + + - + + - + + -Middle lamella + + - + + - + + - + + - + + -Secondary wall o 0 + o 0 + 0 0 + o 0 + 0 0 + Bordered pit + 0 + + o + + 0 + + 0 + + o + Crassulea - + - - + - N N N N N N - + -Spiral thickening N N N N N N N N N + - - N N N Ray cell + + + + + + + + + + + + + + + Resin canal + o 0 + 0 o N N N + - 0 N N N Longitudinal N N N N N N + + + N N N N N N parenchyma Where: + : relatively strong concentration, o : medium concentration. - : relatively low concentration. N : no element or feature in wood species. 104 4.4.2.1. The CCA component distribution in tracheids The microscopic appearance o f impregnated wood sections under light microscopy after staining with dithio-oxamide to detect the copper, staining with alizarin yellow R to detect the chromium and staining with silver nitrate to detect arsenic, was most revealing. At the surface o f the treated wood, all the tracheids were uniformly treated (Figures 4-4, 4-10 and 4-21), but at the limit o f penetration, the tracheids associated with ray cells were more heavily stained than those further removed from the ray cells (Figure 4-2, 4-9 and 4-27). In addition, at both the surface and limit o f penetration, the stain was not uniform across the whole cell wall. The compound middle lamella and cell corner in particular were stained very deeply compared with the secondary wall ( M in Figure 4-3 and 4-9 to 4-11, Table 4-1) by dithio-oxamide or alizarin yellow R. However, the arsenic was associated with the secondary walls o f tracheids, with only a small amount o f arsenic being detected in the middle lamella and cell corners (Figures 4-21 and 4-22, Table 4-1) o f the same tracheid. This observation could be confirmed by the S E M - E D X data (Figures 4-27 and 4-28). Lower arsenic peak areas were present at the cell corner ( T W C L and T W C E in Figure 4-28) compared with that in the same tracheid wall ( T W R L and T W T L in Figure 4-28) in latewood or in earlywood. The high concentration o f chromium and copper in the compound middle lamella and cell corner may be explained by the high lignin content. It has been suggested that chromium reacts with lignin to form chromium esters (Ostmeyer et al., 1989). Copper also reacts with phenolic components containing adjacent hydroxyl groups, to produce a chelated structure. According to the microscopic observations, copper and chromium were not distributed uniformly throughout the cell wall. Secondary walls were not stained uniformly 105 in either latewood or earlywood. The dark green color (copper) and orange-red color (chromium) were more intense in the radial direction than in the tangential direction (Figures 4-2, 4-3, 4-8, 4-9 and 4-12). This was more noticeable (Figure 4-8), with increasing distance from the sample surface, but did not show very clearly in the arsenic distribution. According to the results o f the S E M - E D X analysis o f the latewood or earlywood, the arsenic, copper and chromium peak areas in the radial tracheid wall ( T W R L or T W R E in Figure 4-28) were higher than those in the tangential tracheid wall ( T W T L or T W T E in Figure 4-28). This pattern was present in all five wood species and arises because the radial walls produce a continuous diffusion pathway, whereas the tangential walls are separated by tracheid lumens, and time is required for the slow diffusion o f the solution throughout the cell wall. The arsenic, copper and chromium concentrations in tracheid walls o f treated wood differed from that in the C C A solid according to the peak area data observed in Figure 4-26 and those in the C C A solid (peak area is 5.51, copper 6.74 and chromium 12.44). The relationship between the C C A component peak areas changed as the sampling zone moved into the wood. The arsenic peak areas were higher at the surface analysis levels and lower further from the surface. The chromium peak areas remained at all analysis locations. Comparing the peak areas o f copper and chromium to that o f arsenic (Cu:As ratio or Cr :As ratio) in Figure 4-42, it is clear that the Cu:As ratio is fairly constant (and lower than that in the C C A solid) near to the surface, but increases slightly further from the surface. A similar pattern was observed for the Cr :As ratio, but the Cr :As ratio decreased at level five. This implies either that both chromium and copper concentration increased relative to that o f arsenic or that the arsenic concentration relative 106 to copper and chromium decreased. It should be noted that the absolute concentration o f all elements decreased with increasing distance from the wood surface and arsenic peak areas decreased faster than copper peak areas, so it is certain that the arsenic concentration decreased relative to copper and chromium. 4.0 3.5 -3.0 --2.5 --2.0 1.5 1.0 0.5 | 0.0 --III I ratio of Cu As i ratio of Cr:As •As ratio = 1 i I | . mrrrrmiiiiiiiiiiii 3 § o £ T - CM CO CM T — | 1 | I I CM CO Tf CO N CM CM g n j s i - m n M M lO i - CM CO i l l I tn z z S. Figure 4-42. The ratios o f copper and chromium to arsenic peak area on the tracheid walls and ray cell walls o f amabilis fir. Examination o f the Cr :Cu ratio (Figure 4-43) is helpful in confirming the changes in the composition in the various analytical regions. The C r : C u ratios increased slightly from level one to level four and were slightly higher than that in the C C A solid, implying a relative decrease in the copper content. Finally, the C r : C u and Cr :As ratios showed a marked reduction in level five, corresponding to a lower relative chromium content. 107 35 -i m n ratio of CrCu Cu ratio = 1 3.0 -3 2.5-o (5 2.0-•s o 1.5 -s S 1.0 -0.5 -0 0 • CCA RW1 TW11 TW12 TW13 RW2 TW21 TW22 TW23 TW24 RW3 TW31 TW32 TW33 TW34 RW4 TW41 TW42 TW43 RW5 T - CM CO 2 11 Figure 4-43. The ratio o f chromium to copper peak area for the tracheid and ray cell walls o f amabilis fir. Published electron microscope microanalytical research has suggested that copper is invariably found associated with either chromium or arsenic (Chou et al. 1973). A l l three elements were present in each o f the submicroscopic volumes o f wood examined. The significant change in the distribution o f preservative components deposited, from the secondary wall to the middle lamella region reported here, parallels a marked difference in the ultrastructure o f these two regions. In 1992, Lee et al. noted that in C C A treated wood, all three elements were located within the middle lamella and the cell corners in greater concentrations than within the secondary wall. Rudman (1966a), using the electron probe x-ray microanalyser on Eucalyptas maculata sapwood, concluded that the three elements chromium, copper, and arsenic were all observed at any point in the cell wall, suggesting that there was no screening o f any component during the passage o f treating liquid across the cell wall. He also noted that the maximum concentration o f the three C C A components occurred in the middle lamella region. However, in the current studies, 108 of the softwoods, copper and chromium appear at different locations from the arsenic (Figures 4-3, 4-8, 4-9, 4-12 and 4-22). Deposits were observed on the lumen surface o f earlywood (Figure 4-29-A) and latewood tracheids (Figure 4-29-B) o f amabilis fir. The deposits which appeared to be similar, have not been reported on the tracheid lumen surface o f untreated wood. The C C A component peak areas for the deposit on the earlywood lumen surface (TD1) were similar to that on the lumen surface o f latewood (TD2) with the copper and arsenic peak areas being very high (Figure 4-30). The chromium peak areas for the deposit were higher than those for all the other analysis points. This suggests that the deposits were comprised o f all C C A components. During the light microscopic study o f Douglas-fir, the spiral thickenings on the lumen surface o f tracheids were clearly visible after staining with the reagent for copper. Since the spiral thickening wi l l cause the lumen surface to be rough, producing greater resistance to liquid flow, this may be one o f the reasons that Douglas-fir is not so easily treated. 4.4.2.2. The CCA component distribution in pit regions The light microscopic examination o f the preservative deposited in the bordered pits varied markedly from one pit to another. In addition, since the location o f the area sampled was at the limit o f C C A penetration, it was not surprising that some o f pits were strongly colored while others were not (Figures 4-14 and 4-23). This can also be seen in the cross section (Figures 4-3, 4-9 and 4-24). Comparing Figure 4-23 with Figure 4-24, 109 the arsenic was located at the torus, whereas the chromium was located at the margo. Copper was found relatively uniformly throughout the pit region. Comparing the peak areas o f copper and chromium to arsenic (from Figure 4-32), the Cu:As ratios at all the locations were similar to that in C C A solid (Figure 4-44). The Cr: A s ratios at TP-a were lower than those at the T W and T L for three cells. This msm ratio of Cu:As mmm ratio of Cr:As As ratio = 1 Figure 4-44. The ratios o f copper and chromium peak area to arsenic peak area on the bordered pits, the tracheid walls and lumens o f amabilis fir. suggested that either the chromium content was lower at the TP-a or that both copper and arsenic were higher at that location. The Cr :Cu ratios for the three locations on each tracheid (Figure 4-45) showed a consistent trend. The Cr :Cu ratio was lowest at the bordered pit area (TP-a) and highest at the cell wall (TW) for each tracheid. This suggested that there was a higher copper concentration at TP-a. Figure 4-45. The ratios o f chromium peak area to copper peak area on the bordered pits, the tracheid walls and lumens o f amabilis fir. S E M - E D X analyses were performed on bordered pits on the tracheid walls o f western hemlock, Douglas-fir and amabilis fir. The results showed that the Cu:As ratios were relatively constant and Cr :As ratios were low in the pit area (Figures 4-44 and 4-45), compared to those on the tracheid walls. This means that more copper and arsenic relative to chromium was distributed in the bordered pit area. This is consistent with the visible microscopic observation in which arsenic was observed on the bordered pit. This is expected since the pits are primary pathways for C C A solution penetration between cells. In addition, there is a high proportion o f pectin in the bordered pit areas which can react with copper. This result was similar to that found on the deposit observed on the lumen surface. Crassulea were associated with a higher concentration o f intercellular substances (Table 4-1). From the stained sections (Figure 4-15), it appeared that chromium(VI) can react with the intercellular substances at the crassulea forming insoluble chromium(III) i l l products. There was no evidence that copper and arsenic were concentrated in the crassulea. Analysis o f stained sections showed few pits contained chromium, but many pits were penetrated by arsenic (Figure 4-23 and Table 4-1). 4.4.2.3. The CCA component distribution in ray cells A t the wood surface, all the C C A components were located in the ray cell walls and lumen (Figures 4-6 and 4-11, Table 4-1). However, at the limit o f penetration, copper and chromium were mainly observed in the ray cell walls (Figures 4-2 and 4-13), while arsenic appeared at both ray cell walls and lumen (Figure 4-20). Therefore, the ray cells generally showed a positive reaction to all C C A components (Table 4-1), while the adjacent tracheid cells often showed a reaction only in those cells adjacent to the ray cells. This confirms the importance o f the ray cells in facilitating C C A penetration. These observations mirrored those at the interface o f the treated and untreated wood in the treated samples. The relationship between C C A components, like that on the tracheid walls changed from the surface to the limit o f penetration. But the arsenic and copper peak areas at the RW1 (Figure 4-26) were extremely low compared to those on the tracheids in the same level. This could be caused by copper and arsenic being fixed much less rapidly at the surface, and insolublized further from the surface o f the sample. Excess arsenic and copper are mobile, and are either removed from the ray cells at the surface during the final evacuation or they migrate further in the wood during treatment. 112 4.4.2.4. The CCA component distribution in resin canals Since the resin canals are one o f the pathways by which solution penetrates pressure treated wood, it is not surprising that they were treated more intensely than the surrounding tracheids (Figures 4-6, 4-7, 4-11, 4-16 and 4-25). According to the E D X -S E M analysis, the copper concentration relative to those o f arsenic and chromium was extremely high in the resin canals (Figure 4-38). The arsenic and chromium peak areas at double resin2 were higher than those at double resinl and 3. While the copper peak area at double resin3 was higher than those at double resinl and 2. This shows that the C C A distribution varies even in the same analysis feature in wood. Compared to surrounding tracheids, the copper peak areas at the resin canals were extremely high, while the chromium and arsenic peak areas were both higher and lower. This means that the C C A distribution in the resin canal differed totally from that in the trachieds. Analysis o f the ratios o f chromium and copper peak areas to that o f arsenic (Figure 4-22) showed that the Cu:As ratios were very high in the resin canals compared to those in C C A solid and in the surrounding tracheids, while the Cr :As ratios were lower than that in the C C A solid but slightly higher than that in the tracheids (Figure 4-46). The enrichment o f the copper in the resin canals is likely caused by selective reaction o f the copper in the C C A solution with the extractives (resin acid) in the resin canals. It is known that pine contains polyphenolic extractives, such as pinobanksin and pinosylvin, that can effectively chelate copper forming very stable complexes (Ruddick and Xie , 1994). 113 4.5 4.0 + wmm ratio of Cu:As msm ratio of Cr.As As ratio = 1 0.0 CCA double resinl double resin2 double resin3 TW1 TW2 TW4 Figure 4-46. The ratios o f copper and chromium peak area to arsenic peak area in resin canals and tracheids o f dehumidification dried lodgepole pine. The same distribution pattern with enrichment o f the copper appeared in both the longitudinal and transverse resin canals o f lodgepole pine (Figures 4-36 and 4-38). This suggests that similar reactions with extractive occurs in these two kinds o f resin canals in lodgepole pine. The higher copper concentration is also detected in the ray cell wall (RW) adjacent to the transverse resin canal, and the proportions o f the three C C A components at R W is similar to those at the cell wall o f the resin canal (Figure 4-47) and at double resinl (Figure 4-46). This could be caused by the extractives (resin acid) in the resin canal being redistributed into the adjacent ray cell during the drying process. It is noticed that a similar effect was not observed in the ray tracheid wall. The relative chromium content in the tracheid cell walls was lower than that present in the original C C A solution. 114 CCA rayTW resin RW TW1 TW2 Figure 4-47. The ratio o f copper and chromium peak area to arsenic peak area in transverse resin canal, ray parenchyma wall, ray tracheid wall and longitudinal tracheid walls o f conventionally dried lodgepole pine. The C C A distribution pattern in the resin canals o f white spruce (Figure 4-40) was quite different from that in the lodgepole pine (Figures 4-36 and 4-38). The concentration o f the copper in the resin canal o f white spruce was very low compared to that in lodgepole pine. Consequently, the Cu:As ratio in the resin canal was only slightly greater than that in the C C A solid (Figure 4-48), which was similar to that in the ray cell and tracheids. This observation may be explained by the lack o f polyphenolic extractives in white spruce. A slight enrichment o f the chromium in the resin canal and the associated ray cell may also be present, based on the slightly higher Cr :As ratio in the resin canal. However, all o f the Cr: A s ratios are lower than that found in the C C A treating solution. 115 2.5 » 2.0 + 5 1.5 + 2 1.0 § 2 0.5 + 0.0 I ratio of Cu:As ! ratio of Cr:As •As ratio = 1 CCA resin resin ball RW TW1 TW2 Figure 4-48. The ratios o f copper and chromium to arsenic on ray cell wall, resin canal and tracheid walls o f superheated steam/vacuum dried white spruce. The S E M - E D X analysis o f CCA-treated Douglas-fir sections revealed extremely high copper retentions in both the resin canal and the associated ray cell wall (Figure 4-41). However, the chromium content in the resin canal, ray cell wall, as well as in the walls of the tracheids adjacent to the resin canal (TW3 and TW4) was quite low. This is consistent with the visible light microscopy o f stained thin sections, which also showed low chromium distribution on the resin canal o f Douglas-fir (Figure 4-17). The C u : A s ratios on the resin canal cell walls and the ray cell walls were extremely high compared to that in C C A solid (Figure 4-49). In the associated tracheids, all the locations were slightly higher than that for C C A . The Cr :As ratios in the resin canals and ray cells was similar to that in C C A . The Cr :As ratios for the tracheids far from the resin canal were similar to that in C C A , while those on the tracheids (TW3 and TW4) close to 116 the resin canal were very low. This was caused by the low concentration o f arsenic and chromium in resin canals and the surrounding tracheids. The magnitude o f the copper peak areas was even greater than that observed in lodgepole pine, suggesting a more effective reaction between the copper and the polyhydroxphenohc component in the Douglas-fir. Figure 4-49. The ratio o f copper and chromium peak area to arsenic peak area on tracheid walls, ray cell wall and resin canal cell wall o f Douglas-fir. Reviewing the reaction o f C C A with resin canal and associated ray cells and tracheids, the enhanced uptake o f copper is primarily associated with the ray cells and occasionally the adjoining tracheids. This suggests that during fixation, copper reacts with extractives in the resin canal where it is fixed without reacting with either chromium or arsenic. The permanence o f such copper-extractive compounds is not known, but they 117 could play a vital role in protecting wood from decay. This may also explain why when some wood species are treated with copper sulphate solution, a proportion o f the copper can not be leached out, suggesting that it has reacted strongly with the wood (Jiang and Ruddick, 1997). Localization o f the C C A elements in the resin canals can be seen clearly for lodgepole pine (Figures 4-6 and 4-7), white spruce (Figures 4-4, 4-10, 4-11 and 4-16) and Douglas-fir (Figure 4-17). In general, the light microscopy o f stained thin sections o f lodgepole pine and white spruce showed a strong reaction between the copper and the epithelial cells o f the resin canals (Ruddick and Xie , 1994) (Table 4-1). This suggests that copper readily reacts with the resin acids present in the epithelial cells o f the resin canals, to form stable complexes. The resin acids are dominant constituents in pine wood. Douglas-fir contains taxifolin, a polyhydroxyphenolic compound, which reacts very effectively with copper (Table 4-1). This is consistent with the relative ionic strengths o f the phenol proton compared to a carboxylic acid proton. The intensity o f the copper stain in the epithelial cell walls o f white spruce was lower than the lodgepole pine and Douglas-fir. The resin canals affect the C C A distribution on the associated tracheids. In this study, it was noted that these tracheids absorb more copper than the tracheids further from the resin canals (Figures 4-38, 4-54 and 4-55). Based on the E D X data, the composition o f the fixed C C A in the resin canals, was different from that o f the tracheid walls. 118 4.4.2.5. CCA distribution in the longitudm^parenchyma In western hemlock, longitudinal parenchyma were treated more intensely than surrounding tracheids (Table 4-1), which in turn were better treated than tracheids which were more distant from the parenchyma (Figures 4-8, 4-14 and 4-18 to 4-20). This may lead to uneven C C A distribution in hemlock, due to penetration o f C C A along the longitudinal parenchyma. 119 Conclusions From the examination o f the influence o f seven drying processes on the treatability o f sawnwood with C C A and borate, it was concluded that: • N o single drying regime improved the treatability o f all wood species, for either chemical solution; • In most cases, C C A and borate penetration were similar; • For white spruce, lodgepole pine and amabilis fir, air drying resulted in the best permeability; • For Douglas fir, R F / V drying gave the best penetration for both chemical solutions; • For western hemlock, S S / V and R F / V drying improved the treatability; • Amabilis fir has already been recognized as one o f Canada's most treatable heartwood species. It was not possible to further improve preservative penetration by artificial drying; • The C C A and borate treatability of the five species decreased in the order: amabilis fir > western hemlock > lodgepole pine > white spruce = Douglas-fir; • Despite having large cross field pits and many large resin canals, lodgepole pine did not treat well with C C A . c From the examination o f the microdistribution o f C C A in dried and pressure treated heartwood lumber, it was concluded that: • The Cu:As:Cr ratio diffused markedly for different analysis locations in treated wood. This suggested that the fixation reactions were not the same in different cell types; 120 ray cells and resin canals (if present) played key roles in enhancing C C A penetration in sawnwood; pits aided C C A penetration from the ray cells into the tracheids, and between tracheids; the longitudinal parenchyma in western hemlock were well treated, indicating that they contribute to increasing C C A penetration; copper and chromium were located in the highly lignified cell corners and compound middle lamellae; in lodgepole pine, white spruce and amabilis fir, a high concentration chromium was found at the crassulae where is high lignified; arsenic was precipitated in the secondary wall; enhanced copper content in the resin canals, suggested a preferential reaction with extractives; copper and arsenic concentration were higher in the pit areas, due to either precipitation or reaction with chemicals in the pit components; deposits at the tracheid lumens of amabilis fir were shown to contain mainly arsenic and copper. 121 Literature A W P A , 1996a. American Wood Preservers' Association A3-96. Standard methods for analysis o f treated and treating solution by X-ray spectroscopy. A W P A Woodstock M D . A W P A , 1996b. American Wood Preservers' Association A9-96. Standard methods for determining penetration o f preservatives and fire retardants. A W P A Woodstock M D . A W P A , 1996c. American Wood Preservers' Association A12-96. Wood densities for preservative retention calculations by standards A 2 , A 9 and A l 1. A W P A Woodstock M D . A W P A , 1996d. American Wood Preservers' Association A2-96. Determination o f boron trioxide in sodium borate wood preservatives and the treating solutions made thereof. A W P A Woodstock M D . Anderson, D . G . 1990. The accelerated fixation o f chromated copper preservative treated wood. Proc. Amer. Wood Preserv. Assoc. 86:129-151. A n i , S. and S. Salamah. 1995. Preliminary study on relation o f wood structure to copper-chrome-arsenic ( C C A ) distribution in Kempas (Koompassia malaccensis). Internat. Res. Group Wood Preserv. Document N o . IRG/WP/95-40054. Avramidis S. and R . L . Zwick. 1992. Exploratory radio-ffequency/vacuum drying o f three B . C . coastal softwood. For. Prod. J. 42 (7/8): 17-24. Bailey, P.J. 1965. Some studies on the permeability of wood in relation to timber preservation. Rec. Ann. Conv. Brit . Wood Preserv. Assoc. 2:31-66. Baines, E .F . 1986. The current status o f commercial waterborne preservative treatment o f spruce. Rec. Ann. Conv. Brit. Wood Preserv. Assoc. 80:14-26. Baines, E .F . and J. Saur. 1985. Preservative treatment o f spruce and other refractory wood. Proc. Amer. Wood Preserv. Assoc. 81:136-147. Barton, G . M . 1963. Conidendrin in floccosoids o f western hemlock. For. Prod. J. 13(2):30-44. Belford, D.S . , R . D . Preston, C D . Cook and E . H . Nevard. 1959. The impregnation o f timber by waterborne preservatives I. General survey. J. Appl . Chem. 9(3):192-200. Bodner, J. and G . Pekny. 1991. Distribution o f wood preservatives in the cell wall. Holzforschung und Holzverwertung 2(l):46-48. Booker, R . E . 1990. Changes in transverse wood permeability during the drying o f Dacrydiym cupressinum and Pinus radiata. N . Z . J. For. Sci. 20(2):231-244. 122 Booker, R . E . and J . M . Evans. 1994. The Effect o f Drying Schedule on the Radial Permeability o f Pinus radiata D . Don. Holz als Roh-und Werkstoff. 52:150-156. Brennan, G . K . , G.R. Siemon, J. A . Pitcher and K . J . White. 1995. Effects o f end treatment on C C A preservative distribution in regrowth karri transmission poles. CALMScience . l(3):245-249. Chen, P . Y . S . and E . C . Workman. 1980. Effect o f steaming on some physical and chemical properties o f black walnut heartwood. Wood Fiber Sci. 11(4):218-227. Chou, C . K . , J. A . Chandler and R . D . Preston. 1973. Microdistribution o f metal elements in wood impregnated with a copper-chrome-arsenic preservative as determined by analytical electron microscopy. Wood Sci. and Technol. 7(1): 151-160. Cobham, P. and P. Vinden. 1994. The treatment o f radiata pine heartwood: A problem revisited. Proceedings, International Treated Pine Conference. N e w Zealand Timber Industry Federation. Fiji. Comstock, G . L . 1967. Longitudinal permeability o f wood to gases and nonswelling liquids. For. Prod. J. 17(10):41-46. Comstock, G . L . and W . A . Cote. 1968. Factors affecting permeability and pit aspiration in coniferous sapwood. Wood Sci. and Technol. 2(4):279-291. Cooper, P . A . 1973. Effect o f species, precompression and seasoning on heartwood preservative treatability o f six western conifers. For. Prod. J. 23(7):51-59. Cooper, P .A . , Y . T . Ung and G . Zarjani. 1994. Comparison o f methods for monitoring C C A fixation. Internat. Res. Group Wood Preserv. Document N o . IRG/WP/94-40023. Cooper, P . A . , Y . T . Ung and D.P . Kamden. 1997. Fixation and leaching o f red maple (Acer rubrum L . ) treated with C C A . For. Prod. J. 47(2):70-74. DeGroot, R . C . and T . A . Kuster. 1986. S E M X-ray microanalysis o f tracheid cell walls in southern yellow pine sapwood treated with water dispersable pentachlorophenol. Wood Fiber Sci. 18(l):58-67. Dickinson, D .J . 1974. The microdistribution o f copper-chrome-arsenate in Acer pseudoplatlnus and Eucalyptus maculata. Material und Organismen. 9(l):21-33. Dickinson, D.J . , N . A . A . H . Sorkoh, and J.F. Levy. 1976. The effect o f the microdistribution o f wood preservatives on the performance o f treated wood. Rec. Ann. Conv. Br . Wood Preserv. Assoc. 72:25-40. 123 Drysdale, J .A. , D.J . Dickinson and J.F. Levy. 1980. Microdistribution o f a C C A preservative in five timbers o f varying susceptibility to soft rot. Material und Organismen. 15(4): 287-303. Dunleavy, J .A. , J.P. Moroney and J .E. Russell. 1973. The association o f bacteria with increased permeability o f water stored spruce wood. Rec. Brit. Wood. Pres. Assoc. Ann. Conv: 127-148. Eaton, R . A . and M . D . C . Hale. 1993. Wood decay, pests and protection. London, Glasgow. 88-109. Ellwood, E . L . and R . W . Erickson. 1962. Effect o f presteaming on seasoning stain and drying rate o f redwood. For. Prod. J. 12(7):328-332. Erickson, H . D . 1970. Permeability o f southern pine wood-A review. Wood Sci. 2(3): 149-158. Etickson, H . D . and R J . Crawford. 1959. The effects o f several seasoning methods on the permeability o f wood to liquid. Proc. Amer. Wood Preserv. Assoc. 55:210-220. Fernandez-Golfin, J.I. and H . A . Noves. 1994. Some limiting factors affecting the quality in superheated vapour drying. Forest Research Centre (CIFOR-INIA) , Forest Ind. Area, P .O. 8.111 28080 Madrid, Spain. Fleischer, H . O . 1950. A n anatomical comparison o f refractory and easily treated Douglas-fir heartwood. Proc. Amer. Wood Preserv. Assoc. 46:152-157. Flynn, K . A . 1995. A review o f the permeability, fluid flow, and anatomy o f spruce (Picea spp.). Wood Fiber Sci. 26(4):278-284. Flynn, K . A . and B . C . Goodell. 1996. Physical effects o f the pulsation preservative treatment process on northeastern red spruce (Picea rubens Sarg.). For. Prod. J. 46(1):56-62. Graham, R . D . 1956. The preservative treatment o f eight Oregon conifers by pressure processes. Proc. Amer. Wood. Preserv. Assoc. 52: 118-139. Greaves, H . 1972. Structural distribution o f chemical components in preservative treated wood by energy dispersion X-ray analysis. Material und Organismen.7(4):277-286. Greaves, H . 1974. The microdistribution o f copper-chrome-arsenic in preservative-treated sapwood using X-ray micro-analysis in scanning electron microscopy. Holzforschung 28(6): 193-200. 124 Greaves, H . and J.F. Levy. 1978. Penetration and distribution o f copper-chrome-arsenic preservative in selected wood species. Holzforschung 32(6):200-208. Greaves, H . and T. Nilsson. 1982. Soft rot and the microdistribution o f water-borne preservatives in three species o f hardwood following the field test exposures. Holzforschung 36(4):207-213. Gunzerodt, H . , J. Walder and K . Whybrew. 1988. Compression rolling o f Sitka spruce and Douglas-fir. For. Prod. J. 38(2):16-18. Hall , T . A . and H.J . Hohling. 1968. Fifth Internat. Cong, in X-ray optics and microanalysis. Tubingen. Ed . Mollenstendt and Caukler: Springer-Verlag. Hildebrand, R. 1970. K i l n drying of sawn timber. Hildebrand Maschinenbau GmbH. Hosli , J.P. 1986. Fundamentals o f testing wood permeability. Internat. Res. Group Wood Preserv. Document N o . IRG/WP/86-4002248. Hulme, M . A . and J .A. Butcher. 1977. Soft-rot control in hardwoods treated with copper-chrome-arsenic preservatives. I. treatment problems. Material und Organismen 12(2):81-93. Hulme, M . A . , J .A. Butcher and J. Drysdale. 1976. Reasons for poor soft rot control in hardwoods treated with chromated copper arsenate preservatives. N . Z . For. Serv. Res. Inst., Wood Pres. Rep. N o . 13. Jansen, A . , A . Pizzi and W . E . Conradie. 1985. The penetration characteristics o f C C A preservatives in wood-radial/tangential, processes and species effects. Ho lz als Roh-und Werkstoff.43(5):181-186. Jewell, R . A . , P . H . Mitchell and N . P . Kutscha. 1990. Species effects on the penetration o f C C A in southern pine lumber. Proc. Amer. Wood Preserv. Assoc. 86:22-30. Jiang, X . and J.N.R. Ruddick. 1997. A comparison o f the leaching resistance o f diammine-copper complexes and copper carbonate precipitated in wood. Internat. Res. Group Wood Preserv. Document N o . IRG/WP/86-300158. Keith, C.T. and G . Chauret. 1988. Anatomical studies of C C A penetration associated with conventional (tooth) and with micro (needle) incising. Wood Fiber Sci. 20(2): 197-208. Klein, P. and J. Bauch. 1979. On the Localization o f Ions in the cell walls layers of treated wood baced on a laser-microprobe-mass-analyser ( L A M M A ) . Holzforschung. 33(1): 1-6. Knuffel, W . E . 1985. The effect o f C C A preservative treatment on the compression strength o f South African pine structural lumber. Holzforsch. Holzverwert. 37(5):96-99. 125 Kollmann, F.P. and W . A . Cote. 1968. Principles o f Wood Science and Technology. Springer-verlag. N e w York. 460-463. Krahmer, R. L . and W. A . Cote. 1963. Structural Factors Affecting the Permeability o f Wood. J. Poly. Sci. part C. 2:231-242. Krahmer, R . L . , R . W . Hemingway, and W . E . Hillis. 1970. The cellular distribution o f lignans in Tsuga heterophylla wood. Wood Sci. Technol. 4(1):122-139. Kubinsky, E . and G . Ifju. 1973. Influence o f steaming on the properties o f red oak. Part I. Structural and chemical changes. Wood Sci. 6(1):87-94. Kumar, S. and V . K . Jain. 1978. Effects o f wood preservatives on physical properties o f wood II. Effects o f different salt loadings o f copper-chromate-arsenic composition. Wood Fiber Sci. 9(4):262-269. Kumar, S. and J.J. Morrell . 1989. Penetration and absorption o f different C C A compositions in six western conifers. For. Prod. J. 39(10):19-24. Lebow, S.T. and J.J. Morrell . 1989. Penetration o f boron in Douglas-fir and western hemlock lumber. For. Prod. J. 39(l):67-70. Lebow, S.T. and J.J. Morrell . 1993. Effect o f steaming on treatability o f Douglas-fir heartwood with sodium octaborate tetrahydrate. For. Prod. J. 43(4):35-38. Lee, J.S., I. Furukawa and T. Sakuno. 1992. Microdistribution o f elements in wood after pretreatment with chitosan and impregnation with chrome copper arsenic preservative. Mokuzai Gakkaishi (J. Jap. Wood Res. Soc.) 38(2): 186-192. Levy, J.F. and H . Greaves. 1978. Penetration and distribution o f copper chrome arsenic preservative in selected wood species. 2. Detailed microanalysis o f vessels, rays and fibers. Holzforschung 32(6):209-213. Liese, W. and W . A . Cote. 1960. Electron microscopy o f wood: results o f the first ten years o f research. Proc. 5 t h World Forestry conference 2:1288-1298. Mackay, J .F .G. 1973. Surface checking and drying behavior o f Pinus radiata sapwood boards treated with C C A preservatives. For. Prod. J. 32(9):92-94. Maclean, J.D. 1953. Effect o f stearning on the strength o f wood. Proc. Amer. Wood Preserv. Assoc. 49:88-112. Maturbongs, L . and M . H . Schneider. 1996. Treatability and C C A preservative distribution within ten Indonesian hardwoods. Wood Fiber Sci. 28(2):259-267. 126 Moldrup, B . 1995. Advances in drying and preservative treating o f wood. Proc. Can. Wood Preserv. Assoc. 15:107-118. Morrell , J. J., C . Freitag and S. Unger. 1998. Development o f threshold values for boron compounds in above ground exposure preliminary trials. Internat. Res. Group Wood Preserv. IRG/WP/98-3019. Morris, P.I. 1991. Improved preservative treatment of spruce-pine-fir at higher moisture contents. For. Prod. J. 41(11/12):29-32. Morris, P.I. 1995. Pacific silver fir is the more-treatable component o f hem-fir from coastal British Columbia. For. Prod. J. 45(9):37-40. Morris, P.I., A . Byrne and D.R. Minchin. 1996. Achieving shell or complete penetration o f western hemlock and Pacific silver fir with borates by pressure/diffusion treatement. For. Prod. J. 46(3):51-55. Morris, P.I., A . Byrne, J. Mackay and S. McFarling. 1997. The effect o f steaming prior to pressure treatment on the penetration o f borate into western hemlock. For. Prod. J. 47(3):62-65. Morris, P.I., J.J. Morrel l and J.N.R. Ruddick. 1994. A review o f incising as a means o f improving treatment o f sawnwood. Internat. Res. Group Wood Preserv. Document N o : IRG/WP/94-40019. Nicholas, D . D . 1972. Characteristics o f preservative solutions which influence their penetration into wood. For. Prod. J. 22(5):31-36. Nunomura, A and M.Saito. 1983. The effects o f incising on the strength and preservative penetration o f the lumber o f Hokkaido softwood (II). J. Hok. For Prod. Res. Inst. 378: 12-16. Ostmeyer, J .G. , T.J. Elder and J.E. winandy. 1989. Spectroscopic analysis o f southern pine treated with chromated copper arsenate. II. Diffuse reflectance fourier transform infrared spectroscopy. J. Wood Chem. and Tech. 9(1):105-122. Perrin, P .W. 1987. Review of incising and its effects on strength and preservative treatment o f wood. For. Prod. J. 28(9): 27-33. Petty, J .A. and R . D . Preston. 1968. Electron probe microanalysis o f metals in cell walls o f conifer wood treated with preservative. Holzforschung. 22(6):174-177. Pizzi , A . 1981. The chemistry and kinetic behavior o f Cu-Cr -As /B wood preservatives. Part 1: Fixation o f copper in wood. J. Polym. Sci. Chem. E d . 19:3093-3121. 127 Pizzi , A . 1982. The chemistry and kinetic behavior o f Cu-Cr -As /B wood preservatives. Part 4: Fixation o f copper in wood. J. Polym. Sci. Chem. Ed . 20:739-764. Pizzi , A . 1983. Practical consequences of the clarification o f the chemical mechanism o f C C A fixation to wood. Internat. J. Wood Preserv. 3(2):89-94. Roff, J .W. 1974. Boron diffusion treatment o f packaged utility grade lumber arrests decay in storage. Canad. Forest. Serv. Western Forest Prod. Lab. Report N o . V P - X - 1 2 5 . Vancouver, 14pp. Rosner, B . , K . Messner, E . Tucker and A . Bruce. 1998. Improved prservative penetration o f spruce after pre-treatment with select fungi. I. Fungal pre-treatment o f pole sections. Internat. Res. Group Wood Preserv. Document N o . IRG/WP/98-400117. Ruddick, J .N.R. 1980. Treatability o f lodgepole pine lumber with A C A and C C A . For. Prod. J. 30 (2): 28-32. Ruddick, J .N.R. 1986. A comparison o f needle and Nor th American incising technique for improving treatment o f spruce and pine lumber. Holz als Roh-und Werkstoff. 44:109-113. Ruddick, J .N.R. 1987. A field evaluation o f four wood pole groundline preservative formulations using western Canadian wood species. Report for the Canadian doctrinal association research and development. C E A N o . 068 D129-phase 2. Ruddick, J .N.R. 1989. Multihead incisor for lumber, timber and the like. U . S . Patent N o . 4,836,254. Ruddick, J .N.R. 1991. The influence o f C C A penetration in sawnwood on its performance against decay fungi. For. Prod. J. 41(11/12):19-24. Ruddick, J .N.R. and C. X ie 1994. Why does Douglas-fir heartwood turn black when treated with ammoniacal copper preservatives? For. Prod. J. 44(1):57-61. Rudman, P. 1966a. Studies in wood preservation. Pt. II. Movement o f aqueous solutions through the pits and cell walls o f Eucalypt sapwoods. Holzforschung. 20(2):57-61. Rudman, P. 1966b. Studies in wood preservation. Pt. III. The penetration o f the fine structure o f wood by inorganic solutions, including wood preservatives. Holzforschung. 20(2):62-67. Russ, J .V. 1984. X-ray microanalysis in the biological sciences. J. Submicr. Cytal. 6:55-79. 128 Ryan, K . G . 1986. Preparation techniques for X-ray analysis in the transmission electron microscope o f wood treated with copper-chrome-arsenate. Material und Organismen. 21(3):223-230. Saul, J.F. 1984. Transport Processes in Wood. Springer-verlag. N e w York . 76-89. Schoeman, M . W . J .D. Lloyd and M . J . Manning. 1998. Movement o f borates in a range of timber species at various moisture contents. Internat. Res. Group Wood Preserv. Document N o . IRG/WP/98-30181. Schroeder, H . , and C.J . Kozl ik . 1972. The chracterization o f wetwood in western hemlock. Wood Sci. and Technol. 2(6):85-94. Smith, D . N . and A . L . Williams. 1969. Wood preservation by the boron diffusion process-the effect o f moisture content on diffusion time. J. Inst. Wood Sc. 4(4):3-10. Stone, J .E. and H . V . Green. 1959. Penetration and diffusion into hardwood. Tappi. 42(8):700-709. Trenard, Y . and P. Gueneau, 1984. Penetration pathways of liquid gallium in wood seen by scanning electron microscopy. Wood Fiber Sci. 16(3):403-410. Vinden, P. 1985. Optimisation of light organic solvent preservative (LOSP) treatment o f radiata pine. Proc. 26 t h N e w Zealand Wood Preserv. Assoc. Ann. Conf. 87-104. Vogel . A . I . 1979. Textbook o f macro and semimicro qualitative inorganic analysis. Fifth edition. Longman. Harlow, Essex, England. 223-227. Voulgaridis, E . and G . Tsoumis. 1982. Effect o f steaming on the properties o f beech wood. Science Annals. School For. & Nat. Env., V o l . K E , 5:165-181. Walters, C.S . and W . A . Cote. 1960. Distribution o f pentachlorophenol in the microstructure o f basswood. Holzforschung 14(4): 183-189. Wardrop, A . B . and G .W. Davies. 1961. Morphological factors relating to the penetration o f liquids into wood. Holzforschung. 15(5):129-141. Williams, D . H . and J .A. Kininmonth 1984. High-temperature kiln drying o f radiata pine sawn timber. For. Res. Bulletin N° 73. N e w Zealand Forest Service pp:16. Yata, S., J. Mukudai and H . Kajita. 1978. Morphological studies on the movement of substances into the cell wall o f wood. I. Methods for microscopical detection o f diffusing substances in the cell wall. J. Jap. Wood Res. Soc. 24(9):591-597. 129 Yata, S., J. Mukudai and H . Kajita. 1979. Morphological studies on the movement of substances into the cell wall o f wood. II. Diffusion o f copper compounds into the cell wall. J. Jap. Wood Res. Soc. 25(3):171-176. Yata, S., J. Mukudai and H . Kajita. 1981a. Morphological studies on the movement o f substances into the cell wall o f wood. III. Diffusion o f zinc compounds into the cell wall. J. Jap. Wood Res. Soc. 27(ll):761-766. Yata, S., J. Mukudai and H . Kajita. 1981b. Morphological studies on the movement o f substances into the cell wall o f wood. I V . Diffusion o f hexavalent chromium into the cell wall. J. Jap. Wood Res. Soc. 27(12):821-827. Yata, S., J. Mukudai and H . Kajita. 1982. Morphological studies on the movement o f substances into the cell wall o f wood. V . Distribution o f chromium in the cell wall. J. Jap. Wood Res. Soc. 28(1): 10-16. Yata, S. and K . Nishimoto. 1983. Distribution o f metal elements in wood impregnated with aqueous solutions o f metal salts as determined by S E M - E D X A . Wood research, Kyoto university, No.69:80-88. 130 A p p e n d i x 1 The drying schedules 131 CM or 1 1 •I O <=4 .a (4-1 O T3 CM U • H U M < o.> CO B I J J .S c j as B O 8 i I I 00 4> , 2PI 2 o CO U & 00 a o e oo .3 M 6 a »/-> i n 1 j i f £ 8 00 E 1 EC 3 B .2 Ha B O U o IT) © fN B 00 00 S « .s B ^ 3 fN O 00 o B 'S .2 § o B 00 00 si 4> u 5 & o> (« ^ 1 132 8>g $ .5 •a si 00 o o o O N S C/3 CN *§ o O < k. V u 00 CN oo CN O CN 1.9 J © CO 00 CN 00 ro » 60 <5 .£ ^ r ) I * ( J to —^/ CN O N CN O N CN O N 5 O N C O i n oo C O c o 8 2 O U a CN 5 i n CN .3 00 e 3 3 09 a 1 i \"E« •S g i f £ = o« 8 g i l | g §5 133 J * o CO 4> 60 W H O 3 ^ Q u 60 -•-» 21 •o o kH u u CO B « 60 IS* c o o o i I 3* CN CN < oo CN vq T i -ro ON CN C O s »/-> CN 93 a o O U CN < 00 CN C O IT) CN co 8 i n CN 2?« CN 00 CN 00 00 CN 00 CN i n O N vo M • 1-4 6 m CN 4> 9 a ON VO 00 ON CN m CN CN O o VO 00 vS 00 CN 00 CN 00 CN C O on •a o U < o CN C O o 00 j3 3 C O O N CN VO M 00 00 S 1 § is 111 o . a o 00 oo .S CO CO C I . | | HE- 1 1-1 co ML o CO CO « CO CO n P» O v© S in CN .9 O N OO o O N vol o o NO 00 s 00 CN OO n &0 a to j _ . *-> 03 135 So1 .ts XJ CM -cr o o £ 1 3 i I §1 13 CO c/5 U i n Q 5P/-V Q ' 00 g , Q 2 ! H U D • 5 U c« < D , > C O C — i N ? a o o o oo\"2 as a CN ro CN 00 C O o o V O o i o vo o i o o I CO V O vo m r - | m OS O N 3 .3 Q CN o C O C O a. C O 136 g 3 U N/M N/M N/M N/M N/M N/M N/M N/M N/M N/M N/M N/M N/M N/M N/M 45-55 50-60 50-60 40-50 40-50 50-90 60-70 50-60 50-60 50-60 55-65 40-50 60-70 60-70 60-70 • — Phase II Voltage (kV) 0.3-1.0 0.5-0.8 0.6-1.0 0.7-1.1 0.7-0.8 0.6-1.5 0.8-1.5 0.9-1.2 0.9-1.1 0.9-1.3 0.6-0.8 0.6-0.7 0.6-0.7 0.5-0.7 0.5-0.7 Douglas-fii Drying Pressure (kPa) 80-90 60-70 60-70 60-70 i 60-70 50-90 40-50 40-50 40-50 40-50 60-70 60-70 60-70 60-70 60-70 ock, amabilis fir and Drying Frequency (MHz) 13.56 13.56 13.56 13.56 13.56 13.56 13.56 13.56 13.56 13.56 VO i n CO 13.56 13.56 13.56 13.56 ock, amabilis fir and Air Speed m/sec o o o o o o o o o o o o o o o ern hem' Time (hr) 23.17 10.50 7.67 7.33 4.50 27.50 20.20 8.00 17.33 22.50 6.50 5.00 4.84 3.16 3.84 Table Al-6: Process detail for radio frequency/vacuum drying for, west after planing M.C. (%) 27.7 27.7 27.7 27.7 27.7 22.5 22.5 22.5 22.5 22.5 21.2 21.2 21.2 21.2 21.2 Table Al-6: Process detail for radio frequency/vacuum drying for, west Final M.C. (%) 20.4 ON NO co © 00 VO VO » n VO r» vd co ON co ON 00 00 r-Table Al-6: Process detail for radio frequency/vacuum drying for, west Initial M.C. (%) 129.5 84.2 62.1 48.4 37.8 112.4 71.4 54.8 36.4 45.0 58.6 41.2 38.4 36.2 32.9 Table Al-6: Process detail for radio frequency/vacuum drying for, west Location RF/V Kiln c 2 e 1 c 2 & c 2 c 2 & c 2 c 2 c 2 & c 2 1 c 2 & c 2 I 2 -S 2 & -S 2 Table Al-6: Process detail for radio frequency/vacuum drying for, west Drying Method (RF/V) Westem-hemlock 1 Westem-hemlock 2 Westem-hemlock 3 Westem-hemlock 4 Westem-hemlock 5 Amabilis fir 1 Amabilis fir 2 Amabilis fir 3 Amabilis fir 4 Amabilis fir 5 Douglas-fir 1 Douglas-fir 2 Douglas-fir 3 • CO « c Q Douglas-fir 5 137 Appendix 2 The treatment data 138 Table A2-1 : Chemical uptake, penetration and retention by analysis in lodgepole pine - C C A . Drying method Weight Uptake (kg/m 3) Heartwood Penetration (mm) C C A Retention (kg/m 3) Mean SD Mean SD Mean SD A i r 0.75 0.38 5.70 6.30 1.71 1.24 Dehumidification 0.89 0.38 3.56 3.92 1.99 1.15 Conventional 0.82 0.45 5.13 5.79 2.52 2.00 Presteaming plus conventional 0.87 0.49 4.44 3.65 1.95 1.17 High temperature 0.90 0.44 4.84 4.11 1.65 0.85 R F / V 0.61 0.34 5.38 5.70 2.63 2.56 S S / V 0.84 0.44 6.13 6.15 2.54 1.94 Table A2-2 . Chemical uptake, penetration and retention by analysis in lodgepole pine - borate Drying method Weight Uptake (kg/m 3) Heartwood Penetration (mm) Boric Oxide Retention (kg/m 3) Mean S D Mean SD Mean S D Ai r 0.74 0.54 7.66 6.79 2.49 2.28 Dehumidification 0.62 0.35 1.93 4.23 0.90 1.00 Conventional 0.73 0.48 5.87 5.83 1.34 0.92 Presteaming plus conventional 0.73 0.48 3.97 4.62 0.84 0.53 High temperature 0.79 0.51 4.50 5.47 1.14 1.42 R F / V 0.75 0.33 1.83 3.04 1.06 0.64 S S / V 0.77 0.51 6.84 7.59 2.52 2.89 139 Table A2-3 : Chemical uptake, penetration and retention by analysis in white spruce - C C A Drying method Weight Uptake (kg/m 3) Penetration Heartwood (mm) C C A Retention (kg/m 3) Mean SD Mean S D Mean S D Ai r 0.69 0.47 2.09 4.27 1.41 1.80 Dehumidification 0.51 0.32 0.70 0.84 1.05 0.85 Conventional 0.54 0.26 1.27 2.85 1.17 1.25 Presteaming plus conventional 0.51 0.25 2.27 3.73 1.31 1.01 High temperature 0.35 0.16 0.84 0.90 0.66 0.30 R F / V 0.50 0.18 1.22 2.43 1.33 1.33 S S / V 0.46 0.26 0.91 1.03 0.72 0.35 Table A2-4. Chemical uptake, penetration and retention by analysis in white spruce - borate Weight Uptake Heartwood Boric Oxide Drying method (kg/m 3) Penetration (mm) Retention (kg/m 3) Mean S D Mean S D Mean SD Air 0.53 0.27 3.14 4.42 1.40 0.97 Dehumidification 0.44 0.19 2.04 2.22 0.72 0.50 Conventional 0.41 0.25 2.07 1.98 0.69 0.61 Presteaming plus conventional 0.32 0.13 1.68 1.62 0.47 0.22 High temperature 0.55 0.31 3.20 3.75 0.69 0.43 R F / V 0.43 0.31 3.17 5.46 0.69 0.61 S S / V 0.43 0.27 2.27 3.08 1.09 1.34 140 Table A2-5 . Chemical uptake, penetration and retention by analysis in western hemlock - C C A Drying method Weight Uptake (kg/m 3) Heartwood Penetration (mm) C C A Retention (kg/m 3) Mean S D Mean SD Mean SD Ai r 2.72 1.20 4.73 4.55 4.33 3.43 Dehumidification 2.09 0.91 4.87 3.56 4.38 3.39 Conventional 2.11 1.16 5.50 5.01 3.96 3.67 Presteaming plus conventional 3.11 0.85 6.53 5.07 4.91 3.30 High temperature 2.16 1.01 4.90 5.45 3.70 3.87 R F / V 2.28 0.84 8.17 5.80 6.73 3.67 S S / V 3.54 1.22 8.07 5.76 7.03 3.90 Table A2-6. Chemical uptake, penetration and retention by analysis in western hemlock - borate Drying method Weight Uptake (kg/m 3) Heartwood Penetration (mm) Boric Oxide Retention (kg/m 3) Mean SD Mean SD Mean SD Ai r 2.05 1.06 7.67 5.96 3.03 2.37 Dehumidification 1.00 0.41 9.83 5.85 1.89 1.24 Conventional 0.92 0.39 9.03 5.16 1.09 0.71 Presteaming plus conventional 1.76 0.72 5.07 4.48 2.14 2.26 High temperature 1.17 0.49 5.50 4.57 1.65 1.90 R F / V 1.50 0.62 4.87 4.54 2.49 1.42 S S / V 1.72 0.93 5.90 4.47 2.36 2.16 141 Table A2-7 . Chemical uptake, penetration and retention by analysis in amabilis fir - C C A Drying method Weight Uptake (kg/m 3) Heartwood Penetration (mm) C C A Retention (kg/m 3) Mean SD Mean SD Mean SD A i r 3.87 0.75 11.43 6.41 8.39 4.05 Dehumidification 2.48 0.70 10.40 6.28 5.90 3.06 Conventional 3.90 0.92 9.40 6.31 6.24 3.34 Presteaming plus conventional 3.28 0.77 9.30 6.27 7.95 4.57 High temperature 2.90 0.82 10.33 6.47 6.34 3.84 R F / V 2.36 0.78 10.00 6.32 6.38 3.98 S S / V 3.01 0.56 8.20 5.91 7.59 4.30 Table A2-8 . Chemical uptake, penetration and retention by analysis in amabilis fir - borate Drying method Weight Uptake (kg/m 3) Heartwood Penetration (mm) Boric Oxide Retention (kg/m 3) Mean SD Mean SD Mean SD Ai r 3.04 0.80 13.50 6.28 4.18 2.44 Dehumidification 2.47 0.91 12.90 6.17 4.32 2.38 Conventional 2.57 0.81 13.03 6.31 3.91 2.29 Presteaming plus conventional 2.58 1.05 11.30 7.16 4.36 2.88 High temperature 2.54 1.07 9.40 8.25 4.09 2.37 R F / V 1.31 0.52 11.00 6.95 2.86 1.87 S S / V 2.29 0.95 10.03 7.45 3.77 2.76 142 Table A2-9 . Chemical uptake, penetration and retention by analysis in Douglas-fir - C C A Drying method Weight Uptake (kg/m 3) Heartwood Penetration (mm) C C A Retention (kg/m 3) Mean S D Mean SD Mean SD Ai r 1.37 0.65 1.53 1.53 2.25 1.50 Dehumidification 0.95 0.36 1.03 1.61 1.83 1.35 Conventional 1.63 0.38 2.17 2.42 2.70 2.22 Presteaming plus conventional 1.36 0.40 2.10 1.47 2.01 1.44 High temperature 1.07 0.43 1.37 1.43 1.89 1.26 R F / V 1.24 0.65 3.37 4.39 3.15 2.79 S S / V 1.02 0.37 1.73 1.55 2.63 2.27 Table A2-10. Chemical uptake, penetration and retention by analysis in Douglas-fir - borate Drying method Weight Uptake (kg/m 3) Heartwood Penetration (mm) Boric Oxide Retention (kg/m 3) Mean SD Mean SD Mean SD Ai r 0.51 0.18 0.83 1.18 0.55 0.30 Dehumidification 0.34 0.14 0.87 0.82 0.71 0.43 Conventional 0.63 0.24 4.60 3.22 1.25 0.91 Presteaming plus conventional 0.57 0.17 0.33 0.56 0.76 0.40 High temperature 0.41 0.17 1.57 1.77 0.93 0.48 R F / V 0.47 0.25 4.63 3.60 1.15 0.63 S S / V 0.45 0.17 3.07 1.76 0.76 0.41 143 Appendix 3 The treatment statistics analysis results 144 Table A3-1 . C C A penetration - lodgepole pine 1 2 3 4 5 6 7 DH Pre-steaming HighT Conv. RF/V Air SS/V 1 DH X N N N N N N 2 Pre-steaming N X N N N N N 3 High T N N X N N N N 4 Conv. N N N X N N N 5 RF/V N N N N X N N 6 Air N N N N N X N 7 SS/V N N N N N N X DH Pre-steaming HighT Conv. RFA/ Air SSA/ Table A3-2 . C C A retention - lodgepole pine 1 2 3 4 5 6 7 HighT Air Pre-steaming DH Conv. SSA/ RFA/ 1 High T X N N N N N N 2 Air N X N N N N N 3 Pre-steaming N N X N N N N 4 DH N N N X N N N 5 Conv. N N N N X N N 6 SSA/ N N N N N X N 7 RFA/ N N N N N N X HighT Air Pre-steaming DH Conv. SSA/ RFA/ 145 Table A3-3 . Borate penetration- lodgepole pine 1 RF/V 2 DH 3 Pre-steaming 4 HighT 5 Conv. 6 SS/V 7 Air 1 RF/Vac X N N N N Y Y 2 DH N X N N N Y Y 3 Pre-steaming N N X N N N N 4 HighT N N N X N N N 5 Conv. N N N N X N N 6 SS/V Y Y N N N X N 7 Air Y Y N N N N X RF/V DH Pre-steaming HighT Conv. SS/V Air Table A3-4 . Borate retention - lodgepole pine 1 2 3 4 5 6 7 Pre-steaming DH RF/V HighT Conv. SS/V Air 1 Pre-steaming X N N N N Y Y 2 DH N X N N N Y Y 3FR/V N N X N N Y Y 4 High T N N N X N N Y 5 Conv. N N N N X N Y 6 SS/V Y Y Y N N X N 7 Air Y Y Y Y Y N X Pre-steaming DH RF/V HighT Conv. SS/V Air 146 Table A3-5 . C C A penetration - white spruce 1 2 3 4 5 6 7 DH HighT SSA/ Conv. RFA/ Air Pre-steaming 1 DH X N N N N N N 2 High T N X N N N N N 3 SSA/ N N X N N N N 4 Conv. N N N X N N N 5 RFA/ N N N N X N N 6 Air N N N N N X N 7 Pre-steaming N N N N N N X DH HighT SSA/ Conv. RFA/ Air Pre-steaming Table A3-6. C C A retention - white spruce 1 2 3 4 5 6 7 HighT SSA/ DH Conv. Pre-steaming SSA/ Air 1 High T X N N N Y Y Y 2 SSA/ N X N N Y Y Y 3 DH N N X N N N N 4 Conv. N N N X N N N 5 Pre-steaming Y Y N N X N N 6 SSA/ Y Y N N N X N 7 Air Y Y N N N N X HighT SSA/ DH Conv. Pre-steaming SSA/ Air 147 Table A3-7 . Borate penetration - white spruce 1 Pre-steaming 2 DH 3 Conv. 4 SS/V 5 RF/V 6 HighT 7 Air 1 Pre-steaming X N N N N N N 2 DH N X N N N N N 3 Conv. N N X N N N N 4 SS/V N N N X N N N 5 RF/Vac N N N N X N N 6 HighT N N N N N X N 7 Air N N N N N N X Pre-steaming DH Conv. SS/V RF/V HighT Air Table A3-8 . Borate retention - white spruce 1 Pre-steaming 2 HighT 3 Conv. 4 DH 5 RFA/ 6 SSA/ 7 Air 1 Pre-steaming X N N N N N Y 2 High T N X N N N N N 3 Conv. N N X N N N N 4 DH N N N X N N N 5 RF/V N N N N X N N 6 SSA/ N N N N N X N 7 Air Y N N N N N X Pre-steaming HighT Conv. DH RFA/ SSA/ Air 148 Table A3-9 . C C A penetration - western hemlock 1 Air 2 DH 3 HighT 4 Conv. 5 Pre-steaming 6 SS/V 7 RF/V 1 Air X N N N N Y Y 2 DH N X N N N Y Y 3 HighT N N X N N Y Y 4 Conv. N N N X N Y Y 5 Pre-steaming N N N N X Y Y 6 SS/V Y Y Y Y Y X N 7 RF/V Y Y Y Y Y N X Air DH HighT Conv. Pre-steaming SS/V RF/V Table A3-10. C C A retention - western hemlock 1 HighT 2 Conv. 3 Air 4 DH 5 Pre-steaming 6 RF/V 7 SS/V 1 HighT X N N N N Y Y 2 Conv. N X N N N Y Y 3 Air N N X N N Y Y 4 DH N N N X N Y Y 5 Pre-steaming N N N N X Y Y 6 RF/Vac Y Y Y Y Y X N 7 SS/V Y Y Y Y Y N X HighT Conv. Air DH Pre-steaming RF/V SS/V Table A3-11. Borate penetration - western hemlock 149 1 2 3 4 5 6 7 RF/V Pre-steaming HighT SS/V Air Conv. DH 1 RF/V X N N N Y Y Y 2 Pre-steaming N X N N Y Y Y 3 HighT N N X N N Y Y 4 SS/V N N N X N Y Y 5 Air Y Y N N X N N 6 Conv. Y Y Y Y N X N 7 DH Y Y Y Y N N X RF/V Pre-steaming HighT SS/V Air Conv. DH Table A-12 . Borate retention - western hemlock 1 Conv. 2 HighT 3 DH 4 Pre-steaming 5 SS/V 6 RF/V 7 Air 1 Conv. X N N N Y Y Y 2 HighT N X N N N N Y 3 DH N N X N N N Y 4 Pre-steaming N N N X N N N 5 SS/V Y N N N X N N 6 RF/V Y N N N N X N 7 Air Y Y Y N N N X Conv. HighT DH Pre-steaming SS/V RF/V Air 150 Table A3-13. C C A penetration - amabilis fir 1 SS/V 2 Pre-steaming 3 Conv. 4 RF/V 5 HighT 6 DH 7 Air 1 SS/V X N N N N N N 2 Pre-steaming N X N N N N N 3 Conv. N N X N N N N 4 RF/V N N N X N N N 5 HighT N N N N X N N 6 DH N N N N N X N 7Air N N N N N N X SS/V Pre-steaming Conv. RF/V HighT DH Air Table A3-14. C C A retention - amabilis fir 1 DH 2 Conv. 3 RF/V 4 HighT 5 SS/V 6 Pre-steaming 7 Air 1 DH X N N N N N N 2 Conv. N X N N N N N 3 RF/V N N X N N N N 4 HighT N N N X N N N 5 SS/V N N N N X N N 6 Pre-steaming N N N N N X N 7 Air N N N N N N X DH Conv. RF/V HighT SS/V Pre-steaming Air 151 Table A3-15. Borate penetration - amabilis fir 1 HighT 2 SS/V 3 RF/V 4 Pre-steaming 5 DH 6 Conv. 7 Air 1 HighT X N N N N N N 2 SS/V N X N N N N N 3 RF/V N N X N N N N 4 Pre-steaming N N N X N N N 5 DH N N N N X N N 6 Conv. N N N N N X N 7 Air N N N N N N X HighT SS/V RF/V Pre-steaming DH Conv. Air Table A3-16. Borate retention - amabilis fir 1 Air 2 DH 3 SS/V 4 Pre-steaming 5 HighT 6 RF/V 7 Conv. 1 Air X N N N N N N 2 DH N X N N N N N 3 SS/V N N X N N N N 4 Pre-steaming N N N X N N N 5 HighT N N N N X N N 6 RF/V N N N N N X N 7 Conv. N N N N N N X Air DH SS/V Pre-steaming HighT RF/V Conv. 152 Table A3-17. C C A penetration - Douglas fir 1 DH 2 HighT 3 Air 4 SS/V 5 Pre-steaming 6 Conv. 7 RF/V 1 DH X N N N N N Y 2 HighT N X N N N N Y 3 Air N N X N N N Y 4 SS/V N N N X N N N 5 Pre-steaming N N N N X N N 6 Conv. N N N N N X N 7 RF/V Y Y Y N N N X DH HighT Air SS/V Pre-steaming Conv. RF/V Table A3-18. C C A retention - Douglas fir 1 DH 2 HighT 3 Pre-steaming 4 Air 5 Conv. 6 SS/V 7 RF/V 1 DH X N N N N N N 2 HighT N X N N N N , N 3 Pre-steaming N N X N N N N 4 Air N N N X N N N 5 Conv. N N N N X N N 6 SS/V N N N N N X N 7 RF/V N N N N N N X DH HighT Pre-steaming Air Conv. SS/V RF/V 153 Table A3-19. Borate penetration - Douglas-fir 1 2 3 4 5 6 7 Pre-steaming Air DH HighT SS/V Conv. RF/V 1 Pre-steaming X N N N Y Y Y 2 Air N X N N Y Y Y 3 DH N N X N Y Y Y 4 HighT N N N X N Y Y 5 SS/V Y Y Y N X N N 6 Conv. Y Y Y Y Y X N 7 RF/V Y Y Y Y N N X Pre-steaming Air DH HighT SS/V Conv. RF/Vac Table A3-20. Borate retention - Douglas-fir 1 Air 2 DH 3 SS/V 4 Pre-steaming 5 HighT 6 RF/V 7 Conv. 1 Air X N N N N Y Y 2 DH N X N N N Y Y 3 SS/V N N X N N N Y 4 Pre-steaming N N N X N N Y 5 HighT N N N N X N N 6 RF/V Y Y N N N X N 7 Conv. Y Y Y Y N N X Air DH SS/V Pre-steaming HighT RF/V Conv. 154 "@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "2000-05"@en ; edm:isShownAt "10.14288/1.0099424"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Forestry"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Influence of drying processes on the treatability and CCA distribution in the heartwood of five Canadian softwoods"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/10189"@en .