UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The gluability of certain hardwoods from Burma Chunsi, Khalid Salum 1973

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1973_A6 C48_5.pdf [ 6.06MB ]
Metadata
JSON: 831-1.0075322.json
JSON-LD: 831-1.0075322-ld.json
RDF/XML (Pretty): 831-1.0075322-rdf.xml
RDF/JSON: 831-1.0075322-rdf.json
Turtle: 831-1.0075322-turtle.txt
N-Triples: 831-1.0075322-rdf-ntriples.txt
Original Record: 831-1.0075322-source.json
Full Text
831-1.0075322-fulltext.txt
Citation
831-1.0075322.ris

Full Text

THE GLUABILITY OF CERTAIN HARDWOODS FROM BURMA by KHALID SALUM CHUNSI Sc. (Hons) Forestry, University of Wales 1969 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF FORESTRY in the Department of Forestry We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June, 1973 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 i t 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 representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Khalid Sal urn Chunsi Department of Forestry The University of British Columbia Vancouver 8, Canada Date June 20, 1973 ABSTRACT The gluing properties of pyinkado (Xylia dolabriformis Benth.), thitya (Shorea obtusa Wall.), ingyin (Pentacme siamensis Kurz.), padauk (Pterocarpus macrocarpus Kurz.), in (Dipterocarpus tuberculatus Roxb.), and kanyin (Dipterocarpus sp.) from Burma were investigated. Three room temperature curing glues, phenol-resorcinolQformaldehyde (exterior), urea-formaldehyde (interior), and casein were used. Specific gravity, shrinkage, pH and extractive content were determined for each wood species and their influence on the gluing properties discussed. Block shear specimens for phenol-resorcinol-formaldehyde were tested dry, cold soak, boil and vacuum-pressure treated. Percent delamination was measured. Wood glued with urea-formaldehyde was tested dry and after cold soaking, whereas that glued with casein was only tested dry. In and kanyin (specific gravity 0.67 and 0.59, respectively) showed wood failure values above 90 percent and good shear strength with all the three glues. Pyinkado (specific gravity 0.75) showed high shear strength with all the three glues. It developed 81 percent wood failure with phenol-resorcinol-formaldehyde, 86 percent with urea-for-maldehyde but only 22 percent wood failure with casein. Padauk (specific gravity 0.77) developed high shear strength with casein (2,709 psi) , but slightly lower shear strength values with phenol-resorcinol formalde-hyde (2,693 psi) and urea-formaldehyde (2,448 psi). It had 78 percent i i i i i wood failure with phenol-resorcinol-formaldehyde and 87 percent with urea-formaldehyde but only 67 percent wood failure with casein. Ingyin and thitya (specific gravity 0.81 and 0.80, respectively) showed low shear strength and low wood failure percent with all the three glues. In and kanyin showed low percentage of extractives soluble in ether (1.8 and 1.3 percent, respectively), ethanol (0.5 and 1.4 percent, respectively), hot water (0.5 and 0.4 percent, respectively), and acetone (2.5 and 2.2 percent, respectively). No direct relationship was observed between percentage extractives in the solvents used and shear strength or gluability. If the influence of specific gravity is excluded, a trend of increase in glue joint strength with increase in wood pH was observed. The pH of the wood was from 4.47 to 5.09. The vacuum-pressure treatment reduced the dry shear strength of all species more so for in (37 percent) and kanyin (34 percent). All species, except ingyin and thitya, s t i l l showed high wood failure. High delamination percent was observed for pyinkado (22 percent), thitya (68 percent), and ingyin (34 percent), and low delamination for in (8 percent), kanyin (8 percent), and padauk (11 percent) with phenol-resorcinol-formaldehyde. The boil test was observed to reduce glue joint strength most for all species. The reduction was 57 percent for ingyin, 53 percent for in , 50 percent for kanyin, 30 percent for pyinkado and thitya, and 19 percent for padauk. Wood failure percent was high for all species except thitya and ingyin. Padauk, in and kanyin meet the requirements for exterior structural lamination with phenol-resorcinol-formaldehyde. Ingyin, thitya and pyinkado i v failed to meet the requirements. Padauk, pyinkado, ingyin, in and kanyin meet the minimum requirements for interior structural lamination with urea-formaldehyde. Kanyin, in and padauk are considered suitable for interior structural lamination with casein. Further study is recommended for ingyin, thitya, and pyinkado for gluing with phenol-resorcinol-for-maldehyde for exterior structural lamination; pyinkado, ingyin and thitya for gluing with casein; and thitya for gluing with urea-formaldehyde for interior structural lamination. TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS . . . • • v LIST OF TABLES xi LIST OF FIGURES xi i i ACKNOWLEDGEMENT xv INTRODUCTION • . . • "1 1. Importance of Tropical Forests 1 2. Objective and Scope of the Study 3 LITERATURE REVIEW 5 1. Gluing, general 5 2. Specific Gravity . . . . 9 3. Effect of Extractives 12 4. Wettability 16 5. pH 19 6. Effect of Glue . . . . • 20 MATERIALS AND METHODS 23 1. Species . . . 23 2. Glues 24 2.1 Phenol-resorcinol-formaldehyde 24 2.2 Urea-formaldehyde 24 v vi Page 2.3 Casein 25 3. Experimental Procedure 25 3.1 Preliminary experiments . . 25 3.2 Main experiment gluing 26 3.2.1 Gluing with phenol-resorcinol-formaldehyde 27 3.2.2 Gluing with urea-formaldehyde 27 3.2.3 Gluing with casein 28 3.3 Specimen preparation 28 3.3.1 Block shear specimen . . 29 3.3.2 Delamination specimen 30 4. Testing Procedure . 30 4.1 Dry test , 30 4.2 Cold soak test 31 4.3 Boil test 31 4.4 Vacuum-pressure cyclic test 31 5. Determination of Specific Gravity and Shrinkage 33 5.1 Specific gravity 33 5.2 Volumetric shrinkage 33 5.3 Radial and tangential shrinkage 34 6. Extractive Content Determination 34 7. pH Determination 36 RESULTS 37 1. Basis of Judgement 37 2. Differences between Species within Glue 38 vii Page 2.1 Phenol-resorcinol-formaldehyde :,resin'adhesive 38 2.1.1 Dry test shear strength and wood failure 38 2.1.2 Cold soak test shear strength and percent wood failure . 39 2.1.3 Boil test shear strength and wood failure 40 2.1.4 Vacuum-pressure cyclic test shear strength, wood failure and percent delamination 40 2.2 Urea-formaldehyde resin adhesive 41 2.2.1 Dry test shear strength and wood failure 41 2.2.2 Cold soak test shear strength and wood failure 42 2.3 Casein glue 43 2.3.1 Dry test shear strength and wood failure 43 3. Variation within Species between Glues 44 3.1 Pyinkado 44 3.1.1 Dry test shear strength and wood failure 44 3.1.2 Cold soak test shear strength and wood failure 44 3.2 Thitya 45 3.2.1 Dry test shear strength and wood failure percent . . . . 45 3.2.2 Cold soak test shear strength and wood failure percent . 45 3.3 Ingyin 46 3.3.1 Dry test shear strength and wood failure percent . . . . 46 3.3.2 Cold soak test shear strength and wood failure percent . 46 3.4 Padauk 47 3.4.1 Dry test shear strength and wood failure percent . . . . 47 3.4.2 Cold soak test shear strength and wood failure percent . 47 3.5 In 47 vi i i Paoe 3.5.1 Dry test shear strength and wood failure percent . . . 47 3.5.2 Cold soak test shear strength and percent wood failure 48 3.6 Kanyin 48 3.6.1 Dry test shear strength and wood failure percent . . . 48 3.6.2 Cold soak test shear strength and wood failure percent 48 4. Effect of Treatment within Species 49 4.1 Phenol-resorcinol-formaldehyde adhesive 49 4.1.1 Pyinkado: Shear strength and wood failure percentage . 49 4.1.2 Thitya: Shear strength and wood ! failure percentage . 49 4.1.3 Ingyin: Shear strength and wood failure percentage . . 50 4.1.4 Padauk: Shear strength and wood failure percentage . . 50 4.1.5 In: Shear strength and wood failure percentage . . . . 50 4.1.6 Kanyin: Shear strength and wood failure percentage . . 51 4.2 Urea-formaldehyde adhesive 51 4.2.1 Pyinkado: Shear strength and wood failure percent . . 51 4.2.2 Thitya: Shear strength and wood failure percent . . . 51 4.2.3 Ingyin: Shear strength and wood failure percent . . . 52 4.2.4 Padauk: Shear strength and wood failure percent . . . 52 4.2.5 In: Shear strength and wood failure percent 52 4.2.6 Kanyin: Shear strength and wood failure percent . . . 52 5. Within Treatment between Species Variation 53 5.1 Phenol-resorcinol-formaldehyde: Shear strength and wood failure percent 53 5.2 Urea-formaldehyde adhesive: Shear strength and wood failure percent 54 ix Page 6. Specific Gravity 54 7. Radial, Tangential and Volumetric Shrinkage 55 8. pH 56 9. Extractives 56 9.1 Ether soluble extractives 56 9.2 Ethanol soluble extractives 57 9.3 Hot water soluble extractives 57 9.4 Acetone soluble extractives 57 DISCUSSION 58 1. Influence of Specific Gravity 58 1.1 Phenol-resorcinol-resin adhesive: Glue line dry shear strength and wood failure percent 58 1.2 Urea-formaldehyde resin adhesive: Glue line dry shear strength and wood failure percent 61 1.3 Casein: Glue line dry shear strength and wood failure percent 64 2. Influence of Extractives 67 2.1 Successive extraction . . . 71 2.1.1 Ether soluble extractives 71 2.1.2 Ethanol soluble extractives 72 2.1.3 Hot water soluble extractives 73 2.1.4 Total extract!ves= 73 2.2 Acetone soluble extractives 75 3. Influence of pH 77 4. Influence of Shrinkage on Shear Strength and Delamination 79 X Page 4.1 Shrinkage 79 4.2 Del ami nation percent 82 5. Comparison of Treatments 85 CONCLUSION 87 REFERENCES 91 APPENDIX 130 LIST OF TABLES Table Page 1 Moisture content change of wood during storage 96 2 Specific gravity, wood shear strength as tested and minimum acceptable glue line shear strength at 10 percent moisture content for the six Burmese woods . . . 97 3 Shear strength and wood failure of wood bonded with phenol-resorcinol-formaldehyde adhesive 98 4 Duncan's multiple range test for shear strength of laminated wood bonded with phenol-resorcinol-formaldehyde adhesive 100 5 Duncan's multiple range test for wood failure percent of wood bonded with phenol-resorcinol-formaldehyde " '\\ adhesive 101 6 Percent delamination for wood bonded with phenol-resorcinol-formaldehyde adhesive and statistical comparison . . . . 102 7 Shear strength and wood failure of wood bonded with urea-formaldehyde adhesive 103 8 Statistical ranking by Duncan's test for shear strength and wood failure percent for wood bonded with urea-formaldehyde adhesive . . 104 9 Shear strength and wood failure percent for wood bonded with casein glue 105 10 Statistical ranking by Duncan's test for dry shear strength and wood failure percent of wood bonded with casein glue 106 11 Comparison of shear strength and wood failure percent for four treatments and three adhesives (urea-formaldehyde, phenol-resorcinol-formaldehyde and casein) 107 12 Specific gravity, shrinkage percent and green moisture content, as determined in the study of the six Burmese woods 108 xi xii Table Page 13 Statistical ranking of specific gravity of the six Burmese woods . . . 109 14 Average pH value for cold water extractions of the six Burmese woods . 110 15 Statistical ranking by pH for cold water extractions of the six Burmese woods Ill 16 Percentage of extractives soluble in ether, ethanol, hot water and acetone for the six Burmese woods . . . . 112 17 Statistical ranking by percentage of extractives soluble in ether, ethanol, water and acetone 113 LIST OF FIGURES Figure Page 1 Form and dimension of block shear test specimen 114 2 Shearing tool 115 3 Dry test shear strength and percent wood failure for wood bonded with phenol-resorcinol-formaldehyde resin adhesive 116 4 Cold soak shear strength and percent wood failure for wood glue with phenol-resorcinol-formaldehyde resin adhesive . . . 117 5 Boil test shear strength and percent wood failure for wood bonded with phenol-resorcinol-formaldehyde resin adhesive 118 6 Vacuum-pressure cyclic test shear strength and percent wood failure for wood glued with phenol-resorcinol-formaldehyde resin adhesive 119 7 Dry test shear strength and wood failure percent for wood glued with urea-formaldehyde resin adhesive . . . 120 8 Cold soak shear strength and wood failure percent for wood glued with urea-formaldehyde resin adhesive . . . 121 9 Dry test shear strength and percent wood failure for wood bonded with casein glue 122 10 Phenol-resorcinol-formaldehyde, urea-formaldehyde and casein dry test shear strength 123 11 Relationship between specific gravity and glue line dry shear strength for wood glued with phenol-resorcinol-formaldehyde resin adhesive 124 12 Relationship between specific gravity and glue line dry shear strength for wood glued with urea-formaldehyde resin adhesive 125 13 Relationship between specific gravity and glue line dry shear strength for wood glued with casein glue . . . . 126 xi i i xiv Figure Page 14 Relationship between pH of wood and glue line dry shear strength for wood glued with phenol-resorcinol-formaldehyde resin adhesive . . . 127 15 Relationship between pH of wood and glue line dry shear strength for wood glued with urea-formaldehyde resin adhesive 128 16 Relationship between pH of wood and glue line dry shear strength for wood glued with casein glue 129 ACKNOWLEDGEMENT The author wishes to express his gratitude to Dr. S.-Z Chow of the Western Forest Products Laboratory, Canadian Forestry Service, Vancouver and to Dr. R. W. Wellwood of the Faculty of Forestry without whose constant supervision and advice the study would not have been possible. Appreciation is also due to Mr. Henry N. Mukai of the Western Forest Products Laboratory for his help in the experimental work. The author is also grateful for being allowed to use the Forest Products Laboratory faci l i t ies. The author is greatly indebted to Canadian International Devel ment Agency both for financing the author's study in Canada and for providing the Burmese hardwoods used in the study. The author is also appreciative of the financial assistance obtained from the Faculty of Forestry, University of British Columbia. Thanks are also due to the Director of Forestry, Tanzania, for making i t possible for the author to study in Canada. xv INTRODUCTION 1. Importance of Tropical Forests Tropical forests (defined by climate) extend over 2,100 million ha.whereas temperate zone forests cover about 1 ,800 million ha. The world average is 1.6 ha. of forest land per capita while the tropics average 2 ha. of forest land per capita (Wangaard, 1966). Total broad leaved wood removals for 1967 for the world was 3 3 1,091 million m of which 771 million m was fuelwood and 320 million o m was industrial wood (Pringle, 1969). For temperate countries the 3 3 total removal was 451 million m of which 226 million m was fuelwood and 225 million was industrial wood. Broad leaved wood removals for 3 industrial use in the tropical countries amounted to 95 million m , 3 3 fuelwood 545 million m , making a total removal of 640 million m . Some tropical hardwoods are already well established in markets of the industrialized countries. For example 60 percent of the hardwood plywood consumed in the United States is imported and at least 90 percent of this is manufactured from lauan (Shorea, Parashorea and Pentacme spp.) timber of Southeast Asia (Stadelman, 1969). In 1967, the Philippines furnished 64 percent of Japan's imports of hardwood logs, as well as 80 percent of the logs imported by Korea. The Philippines produced a total of 3.3 bill ion bd. ft. of lauan in 1967, over 2.8 bill ion bd. ft. of which 1 2 was exported to Japan, Korea and Taiwan all in the form of logs. The lauan plywood produced by Japan, Korea and Taiwan was in turn exported, chiefly to the United States. There is a tremendous diversity in growth which is regarded as a vast reserve for man's utilization. The level of utilization of tropical forests is very low because of a number of factors, including: a) the fact that a substantial part of the forest is isolated from centres of population and is physically or economically inaccessible at present; b) the fact that innmany densely"p'6pul:a'tedmparts" o'f^the"tropics the most accessible forests, have been rendered unproductive through over-cutting and abuse over long periods, in extreme cases to the point of conversion to barren land subject to erosion by wind and water; c) the fact that tropical forests are typically heterogenous and have a multiplicity of species, of which relatively few are currently marketable. This is a major limiting factor for their economic use. Under these circumstances, Wangaard (1966) reports that usable timber is actually scarce in many parts of the tropics; this, coupled with a low level of income, has resulted in a per capita consumption of processed or industrial wood that is generally far below the world average. Expressed per 1,000 population, the consumption of sawn wood for 1960, 3 3 for example, was approximately 10 m in Tropical Africa, 69 m in Latin 3 America and 12 m in the Tropical Asia Pacific region, as compared with 3 an average world wide consumption of 145 m . The usefulness of the tropical forest must therefore be increased to contribute more to the comfort and economic development of the peoples of these regions. 2. Objective and Scope of the Study 3 One of the big problems in the utilization of tropical species is their great variation in properties. This necessitates a proper understanding of properties of individual species before they can be introduced for commercial processing. The problem is further magnified by the fact that there is usually a profusion of species within a given area, only a few of which can be removed and used economically. Research in the properties of tropical species is therefore a slow and costly process which many of the tropical countries can barely afford. This study was undertaken in an effort towards adding to infor-mation on some properties of tropical hardwoods. It is proposed to investigate the gluing properties of six species from Burma (Southeast Asia). The selection of these species was made by the Burmese Timber Authority - based on their commercial and economic availability as well as their differences in physical properties. Some of these woods are already being used for construction activities and are therefore familiar to the local people. The species which will be found suitable for gluing may be used for manufacturing laminated beams in a factory to be built by Canadian International Development Agency as part of industrial aid to Burma. The following will be investigated: * a) The gluing characteristics of the six wood samples using phenol-resorcinol-formaldehyde, urea-formaldehyde and casein glue formulations. All the glues are room temperature setting. * Any reference to the term 'species' used in this study refers only to the wood samples received and not necessarily..representative of the species in Burma. 4 b) The influence of wood specific gravity on the gluability of the wood samples. c) The amount of extractives in the individual wood samples and their influence on the gluing characteristics. d) The pH of the wood and its influence on gluability. e) Shrinkage properties of the wood samples and their effect on durability of the glue bonds. f) Effect of four different treatments on the strength of the wood glue bond. LITERATURE REVIEW 1. Gluing, general The gluing of hardwoods has been undertaken in Europe and North America for some time, but.comprehensive studies on the gluing of tropical hardwoods are not many. Troop and Wangaard (1950) studied the gluing properties of certain tropical American woods. The results of their preliminary survey of gluing properties of selected tropical woods indi-cated that many of them could be adapted to a number of industrial or structural uses. The gluing properties of each wood were based on joint strength and percentage of wood failure as developed in the standard block shear test. It is emphasized in their report that the gluing of these woods is important not only from the standpoint of assembly gluing as employed conventionally in fabricated wood products and structures, but in addition, from the standpoint of the more complete utilization that may be made possible through laminated and edge glued construction. Twenty-nine wood species were investigated using two adhesives, resorcinol-formaldehyde and phenol-resorcinol-formaldehyde. The authors recognised the fact that other types of adhesives and other gluing tech-niques may yield results different from theirs. Narayanamurti (1957) studied the gluability of Diospyros melanoxy- lon Roxb. and the various problems involved in gluing this species noted for its high density and high extractive content. Extractive removal 5 6 improved gluability. Narayanamurti et a l . (1962) reported on the gluing of Tectona grandis L. and Acacia catechu Willd. and the effect of extrac-tives from these species on the properties of animal glue and of urea-formaldehyde. Extractives from the two species affected gelation time of the glues. The gluing characteristics of secondary hardwoods, including red alder (Alnus rubra Bong.), cottonwood (Populus trichocarpa Torr. and Gray) and aspen (Populus tremuloides Michx.) were investigated by Carstensen (1961). In the same study he investigated the gluing characteristics of softwood veneers. He concludes that within any one species, wide variations exist in density, grain configuration, moisture content and veneer surface characteristics and that gluability will vary accordingly. Accumulation of resinous material on the surface and high percentage of summer wood may add to the gluing problems of the woods noted. Dost and Maxey (1964) reported on the gluing charactistics of some California hardwoods including black oak (Quercus kelloggii Newb.), chinkapin (Quercus muehlenbergii Engl em.), madrone (Arbutus menziesii Pursh) and tanoak (Lithocarpus densiflora Rehd.). They concluded that all four species glue well with phenol-formaldehyde adhesive and with exterior polyvinyl-acetate resin emulsion. Also, joint strength by species and "stock treatment" appeared closely related to "stock density." In recent years many investigations on the gluing of tropical hardwoods have been carried out in Japan. The species investigated are from Southeast Asia. Among the notable contributions in this field are those discussed below. 7 The effect of specific gravity, wettability, pH and percentage of extractives on the glliability of eighteen tropical woods using urea-formaldehyde, phenol-formaldehyde adhesive and polyvinyl»acetate emulsion, was reported on by Goto et a l . (1967). The durability of the bond was tested by the drying-wetting cycle. They concluded that glue-joint strength increased with increase of specific gravity. There is a high degree of reverse correlation between wettability and specific gravity. There was no significant relationship between glue-joint strength and pH. They also reported that the relationship between glue-joint strength and the percentage of extract either by cold water or by hot water is not significant. Moriya et al . (1968) reported the gluing faculties of laminated wood made of fourteen species of Kalimantan woods with resorcinol resin adhesive, phenol-formaldehyde adhesive, urea-formaldehyde adhesive, polyvinyl-acetate resin emulsion adhesive, and casein adhesive. Glue bond durability was tested by del ami nation test and the dry strength by block shear test. The gluing properties of laminated wood made of keruing (Diptero- carpus spp.) grown in Malaya were investigated by Moriya et a l . (1969). The results showed that the gluing characteristics of keruing were un-satisfactory. Red lauan (Shorea negrosensis Foxworthy.) from the Philippines was used by Moriya et al . (1971) in their study on the gluing characteris-tics of laminated wood. The glues used included resorcinol-formaldehyde, phenol-formaldehyde, urea-formaldehyde adhesive, polyvinyl-acetate resin emulsion and casein. Block shear test and del ami nation test were used 8 to evaluate the strength and durability of the glue bond. One of the conclusions reached was that glue joint strength increases with increase in specific gravity of the wood. Sakuno and Goto (1970b) studied the gluability of thirty-six tropical woods using urea-formaldehyde resin adhesive and phenol-formaldehyde. They reported the coefficient of correlation between glue joint strength, wood failure and various properties including specific gravity, pH, ether soluble extractives and wettability. Their findings are elaborated below. Extractive removal from wood surface by solvent led to an increase in joint strength but the increase in strength varied with species irres-pective of glue. This was shown by Chen (1970) in his study on the effect of extractive removal on adhesion and wettability of eight tropical woods. The solvents used included a 10 percent solution of sodium hydroxide, acetone, and alcohol benzene. Urea-formaldehyde and resorcinol-formaldehyde resin adhesive were used to glue the wood. Extractive removal improved wettability and increased the pH of the wood surface. Some extractives of kapur (Dryobalanops spp.) were shown by Imamura et al . (1970) to inhibit the adhesion of veneer with phenol-formaldehyde adhesive and also the curing of paint film with unsaturated polyester resin varnish. Yamagishi and YoshvhirS (:]972)/oreport':6n; the use'''o'f two different species in bonding some tropical woods. They concluded that in using mixed species for plywood, careful consideration should be given to the specific gravity of the species. The difference in specific gravity between the species should be within +0.12. 9 2. Specific Gravity The effect of specific gravity on the gluing of wood has been studied by various workers. * Eickner (1942) investigated the gluing of fjfiteew species- using tr.e"a*.¥briDa:l'dehydeJ,adhesi've:can.d.:casei.n.--.He. concluded that there, was less difference between gluing characteristics of woods of different densities when glued with urea-formaldehyde adhesive than when glued with casein. Twenty-nine Tropical American woods were used by Troop and Wan-gaard (1950) in their study on gluing characteristics. Resorcinol and phenol-resorcincol-formaldehyde adhesives were used to glue the wood species. They observed that the general trend is that of increasing joint strength and decreasing wood failure with increase in specific gravity. They also observed some uneven character of trend of values for wood failure and shear strength in certain species. This was caused by interference with the adhesive. Such interference may have been introduced as a result of defective surfacing or i t may have been related to the character of chemical components, such as gums, resins, oils and waxes which occur in varying amounts as extractives. The effect of such interference, they contend, is partly obscured in shear strength data because of deviations of individual species from the strength values anticipated on the basis of specific gravity alone. Freeman (1959) studied the effect of specific gravity, pH and wettability on the strength of glue joints. It was found that specific gravity is of prime importance as far as failure in wood is concerned. High specific gravity increased bond strength. Original not seen. Cited from Troop and Wangaard (1950). 10 In his report on the gluing characteristics of softwood veneers and secondary hardwoods Carstensen (1961) concluded that within any one species, wide variations exist in density, grain configuration, moisture content and surface characteristics; gluability will therefore vary accordingly. Dost and Maxey (1964) report that joint strength by species and "stock treatment" appeared closely related to "stock density." They found this trend in their study of the gluing characteristics of Cali-fornia hardwoods. The effect of specific gravity, wettability, pH and percentage of extractives on the gluability of eighteen tropica:! woodstwas-studied by Goto et a l . (1967). They found a high degree of correlation between glue joint strength and specific gravity. The glue joint strength in-creased with the increase of specific gravity. Of the four factors investigated, specific gravity was shown to have the most important effect on glue-joint strength followed by wettability. Yagishita and iKarasawa (.1969), in their study of fourteen species of Kalimantan woods, reported that the relation between bond strength and the. apparent specific gravity was that the species with higher specific gravity showed higher values of strength with phenol-formaldehyde adhesive and melamine-formaldehyde. This trend was not observed with urea-formaldehyde. In the literature reviewed above no attempt has been made [except generally for Troop and Wangaard (195' )] to distinguish the point up to which the statement that glue joint strength increases with the increase in specific gravity of the wood. Sakuno and Goto (1970b) in their study 11 on wood gluing established this point. They studied thirty-six species using urea-formaldehyde and phenol-formaldehyde glues. They concluded that the correlation between specific gravity and glue line shear strength was significant at the 5 percent level for both urea-formaldehyde and phenol-formaldehyde glues for values of specific gravity of 0.8 and below. For values above 0.8, the correlation between specific gravity and glue-line shear strength was not significant. For wood failure, however, the opposite was found to be the case. There was no correlation between wood failure and specific gravity for values of specific gravity of 0.8 and below, but above 0.8 there was significant correlation at the 5 percent level. Moriya et al . (1971), in their study on the gluing faculties of laminated wood made of red lauan sawn boards from the Philippines, reported on the effect of specific gravity on shear strength. They used ten species and five adhesives including resorcincol-formaldehyde, phenol-formaldehyde, urea-formaldehyde, polyvinyl-acetate resin emulsion adhesive and casein adhesive. Their results follow essentially the same pattern as those of Sakuno and Goto (1970b). Species with apparent specific gravity of above 0.8 exhibit lower strength properties than expected by extrapolation in all the five glues. Species with specific gravity of 0.8 or below follow the general pattern that glue line strength increases with increase in specific gravity. 12 3. Effect of Extractives Extractives have been known to interfere with gluing properties of wood for a long time but until now very l i t t le was known about the nature of these extractives and their mode of action, especially in tropi-cal hardwoods. Different species have different types of extractives. Extractive quantities vary within species and within tree. Heartwood normally has more extractives than sapwood. * Rapp (1948) studied the gluing characteristics of lignum vitae (Guaiacum officinale L.), a species which is known to be among the most diff icult to glue. He investigated the possibility of various surface treatments to improve the gluing performance by removal of at least part of the resinous extractives contained in this wood. The solvents used included carbon tetrachloride, benzene, acetoneand ethyl alcohol. None of these solvents was as successful as an application of 10 percent caustic soda solution wiped on the surface, allowed to remain for 10 minutes, and removed by washing with water. The results show an increase in shear strength and wood failure values, especially in the combination of sanding and caustic soda treatment. In one study carried out by Gamble Brothers Inc. i t is stated that the washing of the surface of teak, which contains an oily extractive, with acetone improved joint shear strength and wood failure. Troop and Wangaard (1950), on the other hand, report that Burma teak does not it Original not seen. Cited from Troop and Wangaard (1950). Original not seen. Cited from Troop and Wangaard (1950). 13 necessarily require preliminary treatment when glued with resorcinol adhesive. Narayanamurti (1957) summarised the importance of extractives in wood. He points out that the distribution of extractives varies both vertically and horizontally in a tall tree. Extractives affect the hygroscopicity, swelling and shrinkage of wood. The effect of ex-tractives on the gluing of wood is of special importance. He noted that different researchers observed during testing of glue joints that sapwood, which has a lower extractive content, can be glued better under some conditions and with certain adhesives than heartwood. He found that the gluability of Diospyros melanoxylon is improved by extraction and that other species which were treated with extractions from Diospyros  melanoxylon lost considerably in glue joint strength. Narayanamurti et a l . (1962) discussed the influence of extractives on the setting of adhesives. Wood extractives affect the bonding of wood but their effect may vary from glue to glue. Extractives affect the viscosity and rigidity of glue. The effect of various extractives from Acacia catechu and teak on the gelation time and rigidity modulus of adhesive, "animal" glue and urea-formaldehyde was undertaken. They concluded that extractives affect gelation and rigidity of glues but the effect depends on the species being glued. Teak extractives were more effective than those of Acacia. Extractives from afrormosia' (Afrormosia elafa Harms.) i.nhibit the setting of most adhesives, particularly animal glue,acid set phenol-formaldehyde and polyvinyl acetate. This phenomenon was observed by 14 Chugg and Gray (1965). The same authors reportthat extractives lower the surface tension of the wood surface and reduce wettability, which is essential for a good glue bond. Most wood species contain traces of low molecular weight fatty acids or resins which could easily migrate to the surface during a drying or hot press operation forming an oriented surface layer. If the surface tension of the wood is sufficiently low the glue tends to display a definite receding angle of contact on some parts of the surface so that these remain completely free from the glue. This behaviour is more likely with glues possessing a high surface tension such as cold setting phenol-formaldehyde and urea-formaldehyde. Hancock (1964) reports that Douglas f i r veneer dried at high temperatures had fatty acids concentrated at the surface. The fatty acids were shown to reduce the wettability of veneer,and:affectt^ate"'and "depth of. penetration of glue. In a study on surface inactivation of wood at high temperatures, using wood microsections of white spruce [Picea glauca (Moench) Voss] that had had extractives removed to varying degrees, Chow (1971), con-cludes that extractives serve only as catalysts for oxidation and that when drying wood at temperatures over 180°C, in addition to oxidation, pyrolytic degradation occurred. Since no high temperatures were used either in drying the wood or in the hot press in this study, i t is not necessary to review in detail the effect of high temperature on wood surface inactivation. Goto et al . (1967), in their study of tropical woods, conclude that the relationship between the glue joint strength and percentage of extract either by cold water or by hot water was not significant. 15 But glue joint strength increased with the decrease of percentage of extract in the ether extraction when the effect of specific gravity is excluded. They further state that the value of pH and percentage extract have less important effects on the glue joint strength than wettability and specific gravity. Similar results were obtained by Sakuno and Goto (1970b) in their investigation of thirty-:s'tx species. --They report that tP$tifi£ngJue rjoyi,t strength ^she^ar strength jof figTue^oint/spe^'if-ic gravity) was significantly correlated with percent ether extract for urea-formaldehyde for woods of specific gravity 0.8 and below. There was no significant correlation at the 5 percent level of significance between glue joint strength and ether extract percent for wood species with specific gravity higher than 0.8. S ^ h J p - f r ^ :tne~-eJffecr of -•&U afMd?ar eM v&l;>9& iadhesri on, and wettabi iSiiywrT.Sojdi^ um: hydrox^dei S'ol ution (10 percent), acetone and alcohol benzene were the solvents for the extractives. The wood was glued with urea-formaldehyde and resorcinol-formaldehyde adhesives. Glue-joint strength was improved by all treatments in all but one species. In the case of the 10 percent solution of sodium hydroxide, some species were affected more by one treatment than the other, without regard for the adhesive used. Extractive removal improved wettability and increased pH of the wood in all species examined. A positive linear correlation existed between wettability and joint strength eojf.: blocks glued with urea-formaldehyde; however, no such correlation was observed for resorcinol-formaldehyde. Wood extractives of kapur wood (Dryobalanops spp.) inhibit the adhesion of veneer with phenol-formaldehyde adhesive. This was shown 16 by Imamura et al . (1970). They also found that the extractives inhibited the curing of paint film with unsaturated polyester resin varnish. It was shown in the same study that only certain portions of the extractives had inhibitory effect. Other extractives in the wood had no inhibition effect. Onishi and Goto (1971) report the effect of wood extractives on the gelation time of urea-formaldehyde resin adhesive. The influence of wood extractives with cold water, boiling water and alcohol benzene on the gelation time of urea-formaldehyde, and the compressive strength of setting material, was investigated. The results show that gelation time of urea-formaldehyde to which wood extractives were added increases with the increased amount of wood extractives with cold and boiling water. For cold water extractives i t was observed that the gelation time decreased with decrease in pH. Alcohol-benzene extracts also have effect on the gelation time of urea-formaldehyde though to a lesser extent than water extract. 4. Wettability An effective glue joint for structural purposes cannot be made unless a liquid adhesive spreads completely over the surface to be joined. Air bubbles from surface irregularities have to be displaced and chemical contact with the wood substrate made so that a permanent, high strength bond is developed between the glue and the wood. The physical quantities which determine effectiveness of spreading and adhesion are the surface 17 tension of the wood surface and the interfacial tension between the adhesive and the wood. Freeman (1959), Bodig (1962), Freeman and Wangaard (1960), Marian and Stumbo (1962) have demonstrated that glue bond strength can be cor-related with surface wettability. Patton (1970) reviews adhesion theory based on surface energetics. Gary (1962) has reviewed the concept of equilibrium contact angle and the methods of measuring i t on wood sur-faces. Most of the wettability measurements on wood have been made using homogeneous liquids such as water. Herczeg (1965), however, inves-tigated the wetting of wood by liquids having various surface tensions and pointed out the probability that bond strength is closely dependent upon wetting, spreading and surface tension of the adhesive. Bryant (1968) showed that bond quality is influenced by the wettability of wood by the resin and the wood surface. Freeman and Wangaard (1960) observed that during the period of closed assembly, prior to the application of gluing pressure, glue line solids content and viscosity increase more rapidly in the woods of high wettability than in woods of low wettability. Changes in wettability are due to contamination of the wood surface with active chemicals which reduce surface tension (Chugg and Gray, 1965). These chemicals may be traces of low molecular weight fatty acids or resins which migrate to the surface. They lower the surface tension of the wood surface and reduce wettability. Hse (1972) reports on the wettability of southern pine (species not mentioned) veneers by _3fhi^ty^si:x,'-exteteiaitegRtftfe: ph :erfofliEc'resiir adhesives having a wide range of physico-chemical properties. He also discusses 18 the effects of resin formulation on wettability as measured by contact angle, as well as the relationship between contact angle and bond quality. One of the important findings of the study is that contact angle was positively correlated with glue bond quality as tested by wet shear strength, percent of wood failure and percent del ami nation. Extractive removal improved wettability and increased pH of the wood in some tropical hardwoods studied by Chen (1970). He reports that a positive linear correlation existed between wettability and joint strength of blocks glued with urea-formaldehyde but no such correlation was observed for blocks glued with resorcinol-formaldehyde. A reverse correlation between wettability and specific gravity was observed by Goto et a l . (1967). Wettability increased with decrease in specific gravity. Glue joint strength increased with increase in wettability when the effects of specific gravity are excluded. Sakuno and Goto (1970a) studied the wettability of thirty,-six tropi woods and related wettability to glue joint strength and wood failure. They found that specific glue joint strength (shear strength/specific gravity) and wettability had significant correlation at 5 percent level for all species with specific gravity 0.8 and below, for both urea-formaldehyde and phenol-formaldehyde. No correlation was observed between specific glue joint strength and wettability at the 5 percent level for all species with specific gravity higher than 0.8. Significant corre-lation is reported at the 5 percent level between wood failure percent and wettability for phenol-formaldehyde for all species with specific gravity less than 0.8. There was no significant correlation at 5 percent 19 level between wood failure and wettability for species with specific gravity higher than 0.8. Since no actual wettability measurements were made in the present investigation on the gluing of the six Burmese hardwoods, no review of the methods of measuring wettability will be done. 5. pH Acidity or alkalinity of the wood have an effect on the setting of adhesives. The effect of pH on the strength of particleboard was studied by Kitahara and Mizumo (1961). They showed that the strength of particleboard was correlated with pH of the wood chips and that in some cases acid in the wood is sufficient to cure the urea resin adhesive. Chugg and Gray (1965) report that joints made with resorcinol adhesives display a marked change in strength with change of pH. Goto et a l . (1967) found that the relation between glue joint strength and pH is not significant, and that specific gravity and wett-ability are more important factors. Sakuno and Goto (1970a) report that pH affects the wettability of woods with specific gravity lower than 0.8. In a study of some tropical woods Chen (1970) reports that extrac-tive removal improved wettability and increased the pH of the wood in all species tested. The gelation time of urea-formaldehyde to which wood extractives were added increases with increased amount of extractives with cold or boiling water (Onishi and Goto, 1970). For cold water 20 extractives, a certain relationship was observed between the gelation time and the pH of wood, the gelation time decreasing with a decrease in pH of each wood. 6. Effect of Glue The chemical processes whereby most glues cure to give a final bond are influenced by a number of factors among which are the minor .. chemical components of the wood. Wood adhesives harden either by heating, cooling, by loss of water, by chemical action or by a combination of two or more of these. Premature solidification of an adhesive may in-terrupt the sequence in the glue line development at any point because each successive phase of this development requires mobility for optimum results. If solidification occurs before the transfer of the adhesive has been accomplished, wetting will also be prevented. It is here that the role of extractives may be sometimes very important. They may accel-erate the rate of setting,resulting in premature so l i d i f i c a t i o n , or they may retard i t , resulting in a weak joint. Different glues behave dif-ferently to similar extractives, depending on their formulation and other characteristics. Eickner (1942), in his investigation on the gluing characteristics of fifteen specdest;!concluded thatsthere";was (less difference between gluing characteristics of woods of different densities when glued with urea-formaldehyde, than when glued with casein glue. Troop and Wangaard (1950) report that casein and vegetable glues do not seem to permit as f u l l a development of the strength of white 21 oak as do the animal, urea and resorcinol resin adhesives. When comparing mahogany and white oak they found that the non-resin adhesives showed appreciably lower wood failure values in the case of white oak. Yagishita and Karasawa (1969), in their investigation of adhesion i n "f bunteenc speci es > ofuKael.i maatan, woods.,* observed-.that thei rel ati onshi p be-itw.eens bondgsttren:gth;'an,dpi.the2:app:a;rentvspecif.itt:gr.avj ty,,wastitha*- the speci es with higher specific gravity showed higher values of strength with phenol-and melamine-formaldehyde adhesives. This trend was not observed for urea-formaldehyde adhesive. Onishi and Goto (1971) studied the effects of wood extractives on the gelation time of urea-formaldehyde resin adhesive. They suggest that wood extractives have effects on the curing reaction of adhesives. The pH and buffer action of wood may affect the gelation of urea-formaldehyde resin adhesive because the rate of curing of this resin is closely depen-dent on pH. They concluded that wood extractives influence gelation time in that increased amounts of cold or boiling water extractives increased gelation time of urea-formaldehyde. The compressive strength and the ratio of plastic region to elastic of set gel decreased with increased gelation time. Yamagishi and Yoshihiro (1972) report on the mixed use of two different species in bonding some tropical woods. They found that the durability of the glue bond varied with the glue, and that specific gravity difference between two species considerably influence the glue bond durability, imoresxo with urea-formaldehyde than with resorcinol-formaldehyde adhes i ves. 22 Northcott,(1968) states that urea glues are susceptible to hy-drolysis and degrade much more rapidly under continuously wet systems than cyclic dry-wet bond degrade accelerating systems. Phenol-formaldehyde adhesives, on the other hand, are generally considered to be immune to hydrolysis and are more durable than wood, and are less degraded by continuously wet than by cyclic dry-wet bond degrade systems. He suggests that probably the degradation observed with fully cured phenol-formaldehyde adhesives is that of the wood rather than the glue. The conclusion reached in the study of influence of species and glues on bond durability is that durabilities of plywood bonded with different glues is strongly influenced by wood species. MATERIALS AND METHODS 1. Species Wood specimens from six species native to Burma were received in the form of sawn lumber with dimensions of 2 by 4 by 142 inches (5 by 10.by 3603cm)j8 There werectwelve..pieces .for.each of .the-s~peci.es. Eight of these pieces were made available for use in this study. Some of the pieces were quarter sawn and others flat sawn. For each species only heartwood was used. The wood specimens were comprised of the following species: Percentage of Burmese name Botanical name volume of com-mercial timber in Burma Pyinkado Xylia dolabriformis Benth. 21.5 Thitya Shorea obtusa Wall. f 5 5 Ingyin Pentacme siamensis Kurz. ^ ' Padauk Pterocarpussmaerocarpus.Kurz. 1.5 In Dipterocarpus tuberculatus Roxb. Kanyin Dipterocarpus sp. i 44.4 Moisture contents of the specimens were measured by taking samples one foot from the end of each board and averaging the results. The remainder of each board was then cut into two sections of 54 in. and one of 12 in. length. These boards were further sawn lengthwise into 1 by 4 by 54 in. and 1 by 4 by 12 in. dimensions. The long boards were reserved for the main experiment and the short boards for the preliminary experi-ments. 23 24 The sawn boards were then stacked with 1/2 in. spacers between boards in a 25 to 30 percent relative humidity room [dry bulb 80°F (26.7°C) and air speed approximately 200 feet per minute] to permit controlled drying of the material. Moisture contents were checked at time intervals shown in Table I using the ovendrying method (ASTM D 143-52). Prior to planing and gluing .the lumber was kept in the general laboratory condition which approximated 40 percent relative humidity and 70°F and EMC of about 8 percent. The moisture content of the wood at the time of gluing is shown in Table 1. 2. Glues Three glues were used in this study. They were supplied by Westport Chemicals Limited and the Borden Chemical Company (Canada) Limited. These glues include: 2.1 Phenol-resorcinol-formaldehyde Westport PRF 2947 Resin which, when mixed with 2947 Hardner, produces a phenol-resorcinol-formaldehyde resin adhesive that is suit-able for general gluing at temperatures of 70°F and above. It produces a waterproof and weather proof bond in accordance with CSA 0112.7-1960 (Specifications for Phenol and Resorcinol Resin Adhesives for Wood -Room and Intermediate Temperature curing). 2.2 Urea-formaldehyde Cascoresin 901-045 and 255-4812 catalyst supplied by Borden Chemical Company is a urea-resin based adhesive which cures to a water-resistant bond at temperatures of 70°F or higher in accordance with 25 CSA 0112.5-1960 (Specification for Urea Resin Adhesives for Wood - Room and High Temperature Curing). 2.3 Casein Casco-casein 402-003 supplied by Borden Chemical Company is room temperature curing and is recommended for interior use as specified in CSA 0112.3-1960 (Specification for Casein Glues for Wood). 3. Experimental Procedure 3.1 Preliminary experiments To determine the best conditions for gluing the wood specimens, 4 preliminary experiments were carried out using a 12 inch press. Only two glues, urea-formaldehyde and phenol-resorcinol resin adhesives were used. For each of the 6 species and glues, two boards each 1 by 4 by 12 inch were glued together. The lumber was planed before gluing to ensure identical thickness and smooth surface. To avoid any possible surface aging effects on gluing abil ity, the boards were glued within 4 to 6 hours after planing. Using a glue spread of 85 lb. per 1,000 sq. ft. of double glue line and recommended assembly times, the following conditions were tested: a) Gluing at room temperature (about 72°F) for 24 hours and 150 psi. The block shear strength and wood failure test indicated that the bond quality was substandard. b) Gluing at room temperature for 24 hours under 200 psi. The bond quality obtained was better than the f irst test, but far from satisfactory. c) Gluing and pressing at 100°F for 24 hours under 200 psi. The result was very satisfactory. 26 d) Using the conditions of the third preliminary experi-ment above, a different phenol-resorcinol glue provided by another adhesive manufacturer was evaluated. The results were satisfactory, but indicated that the bonding ability of the wood also differs with the adhesive used. 3.2 Main experiment gluing From the results of the preliminary experiment i t was decided to use 200 psi , 100°F and 24 hours clamp period for the main experiment for all the three glues. For each species and glue three boards were glued. The size of the press was the governing factor in determining the length of the laminations, each of which was 3/4 by 4 by 54 in. For each species and glue, six straight grain bi l lets, which were free from defects including knots, birds eye, short grain and any unusual discoloration from the shearing area, were selected. The grain direction in all the billets was parallel to the longest dimension. Just prior to gluing they were planed and assembled in pairs in such a way that each pair had approxi-mately the same specific gravity. All the surfaces of the wood remained unsanded and were free from dirt. The moisture content of the wood at the time of gluing is shown in Table 1. The range of the moisture content of pieces to be bonded in a single member did not exceed 5 percent (as determined by taking a 1 inch sample from the end and testing by the oven dry ing method), 27 3.2.1 Gluing with phenol-resorcinol-formaldehyde Glue was mixed according to the manufacturer's instructions and used within the recommended working l i fe . The spread used for phenol-resorcinol adhesive was 85 pounds per 1,000 square feet double spreading (one half of the spread was applied to each contacting surface). A grooved, rubber rolled mechanical spreader was used. The proper spread was obtained by weighing a piece of wood prior to the application of glue, then passing i t through the spreader so that glue is applied only on one side, and weighing i t again to deter-mine the amount of glue deposited on the surface. The spreader was then adjusted until just the right amount of glue was deposited on each board. For each press load, glue was applied to two pieces of each of the six species to produce one glued board per species. An open assembly time of 20 minutes and a closed assembly time of 35 minutes was used. This is within the 60 minutes total assembly time recommended by the manufac-turer. A pressure of 200 psi (which is within the range recommended for hardwoods by CS253-63 and confirmed by results of the preliminary experiments) was applied for all the six species for 24 hours. The press temperature was maintained at 100°F. The glued stock was removed after 24 hours and conditioned for 7 days before machining. Three glued stocks were made for each species. 3.2.2 Gluing with urea-formaldehyde- 2 ~ Vo . • z. Mixing of the glue was done according to manufacturer's instruc-tions. Enough glue was mixed and used before the expiration of the 28 working l i fe . Glue spread was 85 pounds per 1,000 square feet of joint area double spreading. A rubber rolled mechanical spreader was used, and steps were taken similar to those for phenol-resorcinol adhesive, to ensure the glue spread was as close as possible to the amount required. The assembly times were 5 minutes, open and 20 minutes closed. A pressure of 200 psi and a press temperature of 100°F was maintained for 24 hours. Each press load consisted of one glued stock per species. Three glued members were prepared for each species. • The members were conditioned in the laboratory before they were machined. 3.2.3 Gluing with Casco-casein Glue was mixed according to the manufacturer's instructions and used within the working l i fe . A glue spread of 85 pounds per 1,000 square feet of joint area double spread was used. Glue application procedures were similar to those for phenol-resorcinol and urea-formaldehyde resin adhesives. Assembly times were 15 minutes open and 25 minutes closed. Pressure of 200 psi and press temperature of 100°F was applied for 24 hours. Three glued members were prepared for each species. After removal from the press the members were conditioned for one week before machining. Some squeeze out was observed when pressure was applied in view of the heavy glue spreads. This is also true for the other two glues. 3.3 Specimen preparation It was intended in this study to investigate the performance of the adhesives and the species in laminated wood, as measured by resistance to shear by compression loading for all three glues; also resistance 29 to del ami nation during accelerated exposure to wetting and drying for phenol-resorcinol resin adhesive. In addition, the boil and cold soak treatment, normally used for plywood, was included to determine their effect on the glue-bond of laminated wood. 3.3.1 Block shear specimen Except for casein, from each of the glued stock thirty-six block shear specimens were cut according to CSA 0177-1965 with a slight modifi-cation, the specimen in this case being 1 1/2 by 1 3/4 by 2 in. (see Figure 1). This was because of the limited volume of material received, and the width being nominal 4 in. The 1/4 in. difference in width from the standard block shear specimen was considered to be of l i t t le significance to the validity of the test results. It has been proved by Strickler (1968) in a study on block shear specimen geometry that specimen width is of no significance to unit shear strength. Specimen length has significant effects on unit shear strength. For casein, ten shear block specimens were cut from each laminated piece for dry testing. The cut specimens were left in the air conditioned laboratory for two days before they were tested (dry). The cold soak test specimens and the boil test specimens were treated separately at periods no longer than a week from the time of cutting. The width and height of each specimen at the adhesive line was measured to the nearest 0.010 inch for determination of the shear area. For each species, ten specimens were randomly allocated from each of the three boards for dry block shear testing; ten from each board were used for the cold soak test; and ten for the boil test. For phenol-30 resorcinol resin adhesive, therefore, thirty specimens were used for each of the three-tests; 'TJii.r.tyspec-imens weredused fore case in for the dry test only. Urea-formaldehyde resin adhesive glued wood was tested in the dry condition and cold soaked only. 3.3.2 Delamination specimen In view of the size of the laminated stock, -four.--specimens, 3 inches along the grain and 4 inches across the grain were cut from each glued stock. Twelve specimens were used for each species for phenol-resorcinol resin adhesive. These were used for the vacuum-pressure cycle test (CSA 0177-1965). Only material as close as possible to flat sawn was used. 4. Testing Procedure 4.1 Dry test For block shear test for dry, cold soak and boil specimens, the block shear test method was used with the shear block area varying be-tween 2.6 and 2.8 sq.in.instead of the 3 sq. in.specified in CSA 0112.0 -1960. The standard block shearing tool containing a self aligning seat to ensure uniform distribution of the load was used (Figure 2). The load was applied by means of a Tinius 01 sen hydraulic testing machine at a continuous motion of the movable head at a rate of 0.015 in. per minute (+_ 25 percent) to failure. Shear stress at failure was recorded for each test block and shear strength at failure was calculated in pounds per square inch based 31 on the glue line area between the two laminations - measured to the nearest 0.01 sq.in.and rounded. Percent wood failure (the rupturing of wood fibers expressed as the percentage of the total area involved which shows such failure) was also recorded for each block. 4.2 Cold soak test Thirty specimens for each species for phenol-resorcinol and urea-formaldehyde resin adhesives were submerged in water at 70° to 75°F for 48 hours. Without drying, they were tested to failure as for the dry specimens. Shear strength and percent wood failure were recorded as for dry specimens. 4.3 Boil test Only specimens glued with phenol-resorcinol resin adhesive were used in this test. Thirty specimens for each species were submerged in boiling water for 4 hours, then dried in an oven at 140°F to 145°F for 20 hours, after which they were again submerged in boiling water for 4 hours, cooled in water at 70°F to 75°F and tested to failure while wet. Shear stress at failure and percent wood failure were recorded for each block. Shear strength in pounds per square inch was calculated as for the cold soak test specimens. 4.4 Vacuum-pressure cyclic test («"."/. J ' J 5 I S J J ) This test indicates the quality and durability of glued laminated timber with respect to high moisture conditions. Specimens 3 inches along the grain and 4 inches across the grain were used. Twelve specimens 32 glued with phenol-resorcinol resin adhesive were tested for each species. Specimens were placed in a vacuum-pressure tank which was then f i l led with water at room temperature. All end grain surfaces were exposed to water. A vacuum of 25 inches of mercury was drawn and held for 2 hours, after which i t was released and a pressure of 80 psi applied for two hours. The vacuum-pressure cycle was repeated. Specimens were removed from the tank and dried for a period of 88 hours in air at 80+5°F, and 25 to 30 percent relative humidity, moving at a velocity of 200 to 300 feet per minute. The entire soaking-drying cycle was repeated twice to comprise a total test period of 12 days. After drying for the last time the total length of open glue joints on the end grain surfaces of the specimens were measured to the nearest 1/16 inch. The total length of open glue joints on the two end grain surfaces of each specimen were expressed as a percentage of the entire length of glue lines exposed on these surfaces. This value was recorded as the percentage delamination of the specimen. The total delamination of the two end grain surfaces of the specimen at the con-clusion of the test should not exceed 10 percent of the total length of all glue lines on the two surfaces to pass as satisfactory (CSA 0177-65 1965). This figure is 8 percent as specified for hardwoods by U.S. Commercial Standard CS 253-63 (1963). After completion of the delamination test each block was cut into two shear block specimens (24 per species) and tested. Shear strength and percent wood failure were recorded as for the other block shear tests. 33 5. Determination of Specific Gravity and Shrinkage All specimens used for specific gravity and shrinkage tests were flat sawn. 5.1 Specific gravity Specific gravity was determined by obtaining the ovendry weight and green volume of four samples selected for each species. Only heart-wood was used for these determinations. For each species four specimens were cut, each 1 by 2 by 4 in. (instead of the standard 2 by 2 by 6 in. used in ASTM D 143-52) because of the restricted size of material received. The specimens were soaked in a tank f i l led with water for about five days at 120 psi to make sure they were saturated. Their cross-sectional and length dimensions were measured to the nearest 0.2<jnm. Green weight and volume was determined for each specimen. The specimens were then dried in an oven at 103+2°C until approxi-mately constant weight was reached. Ovendry weight and volume of each specimen was then determined. The specific gravity of each of the speci-men was calculated by dividing the ovendry weight by the green volume. This average value for the four specimens was taken as the specific gravity of the species in the consignment. Specific gravity for the green wood was also determined (^r^" ! ! ^ ^ ) « 3 vgreen volume' 5.2 Volumetric shrinkage Volumetric shrinkage was obtained from the same specimens used for specific gravity measurement. Green and ovendry volumes were computed 34 from surface measurements. The difference between green volume and ovendry volume divided by ovendry volume and multiplied by 100 gives percent volumetric shrinkage for each specimen. The average value of four specimens gives the average volumetric shrinkage of the species in the consignment from green to ovendry. 5.3 Radial and tangential shrinkage The radial and tangential shrinkage determinations were made on 1 by 4 by 1 in. specimens. Four specimens for each species were weighed green and ovendry and their dimensions measured to the nearest 0.001 in. Shrinkage values were calculated as in ASTM D 1943-52 (1972). 6. Extractive Content Determination Material used for extractive content determination was obtained from dry test samples. The broken specimens were chopped into small pieces, care being taken to avoid all material from the glune line area. No traces of glues were included in the chopped material. Airdry ground-wood was prepared by reducing the chopped material by means of a Wiley mill . The groundwood was reduced to such a size that i t could pass through a No. 40 sieve (425 ym), but be retained on a No. 60 sieve (250 urn), according to Browning (1967). Four solvents, ether, ethanol, water and acetone were used. The f irst three were used in that order over each of the groundwood samples and the last was used independently on a fresh sample for each of the 35 species. Three samples were used for each species. The procedure was as follows: A sample consisting of 10 g groundwood was placed in a cellu-lose extraction thimble which was placed in a glass Soxhlet extraction apparatus; 250 ml of ether was placed in a flask and the Soxhlet extraction apparatus was set up as in ASTM D. 1105-56. Extraction was carried out at such a rate that the solvent siphoned over at least s,ix:itri;mesispehoh:o,ur.ThTnis'was continued for 4 hours. The solvent in the flask was trans-ferred to another flask of known.weight by filtering through a small funnel fitted with f i l ter paper. Two small portions of fresh solvent were used to rinse the extraction flask and the funnel. Using a flash evaporation apparatus, the solvent in the flask was evaporated just to dryness after which the extract was dried in an oven at 105°C to constant weight. Extractives soluble in ether were calculated as a percentage of ovendry weight sample of groundwood similar to the original sample. The sample in the extraction thimble was airdrieddinntftes fumehood for 24 hours. It was then extracted using ethanol as the solvent. The same procedure as for ether was followed and the amount of extractives soluble in ethanol determined. Following ethanol extraction the groundwood in the extraction thimble was airdiitedd for 24 hours to make sure all the ethanol had evaporated. The groundwood was then added to 400 ml of 36 boiling water in a beaker and boiled for 4 hours. The water and groundwood was filtered into a flask of known weight and vacuum evaporated just to dryness after which the flask and the residue were put in an oven at 105°C and dried to constant weight. The percentage of extractives soluble in hot water was calculated as for the other two solvents. A fresh 10 g airdry groundwood sample was used for acetone extraction and the same procedure as above was used to determine percentage solubility of extractives. For each species 3 extrac-tive solubility determinations were carried out and the average value recorded. Ten grams of groundwood for each species was dried in an oven at 105°C for 24 hours to determine its moisture content and the ovendry weight which were used in the calculations. 7. pH Determination The same groundwood as that used for extractive determination was used for pH measurement as follows: Four grams of airdr'tedd groundwood was mixed with 40 cc of distil led water. After two hours the water and ground-wood was filtered and the filtrate used for pH determination using a standard Beckman pH meter. The pH was recorded after the meter had indicated a constant value. The pH average value of thrceeipsampTes forheaohcspecies was.etaken las: theap'H of the species at room temperature. RESULTS 1. Basis of Judgement The glue line shear strength and durability were judged on the basis of Commercial Standard CS 253-63 of the United States Department of Commerce (1963) which proposes separate specifications for hardwoods. Essentially the standard specifies that: a) The average block shear strength values of the test specimens shall meet the values of 90 percent maximum shear strength parallel to the grain of the wood adjusted for moisture content. b) The sheared or broken surfaces of the test samples shall develop at least an 80 percent average wood failure for wet-use adhesives with all species and for dry-use adhesives with softwoods and non-dense hardwoods. The average wood failure shall be not less than 40 percent for dry-use adhesives with dense hardwoods. c) The delamination should not exceed 8 percent for hardwood blocks (5 percent for softwood). . The minimum acceptable shear strength parallel to the grain for this experiment was obtained by carrying actual tests on solid wood shear block specimens. Ten specimens obtained randomly from three boards 37 38 of each species were tested according to ASTM D 143-52 and the average shear strength was obtained. This shear strength was regarded as the maximum value at the moisture content at test. Ninety percent of the shear strength value at the moisture content at test was taken as the minimum allowable glue line shear strength for the species (Table 2). 2. Difference Between Species Within Glue 2.1 Phenol-resorcinol-formaldehyde resin adhesive 2.1.1 Dry test shear strength and wood failure The average shear strengths for the dry test and their standard deviations are shown in Table 3. Pyinkado had the highest glue line shear strength followed by in and padauk. Kanyin ranks fourth, thitya and ingyin last. Thitya and ingyin, irrespective of their highest specific gravity, give strength values which failed to meet the minimum acceptable level (Table 2). On the basis of strength alone, the rest of the species meet the minimum required strength. The ranking of the species by Duncan's multiple range test (Walpole, 1968) is shown in Table 4 (see also Figure 3). A sound glue to wood bond should have strength equal to or greater than that of the wood itself . The higher the percent wood failure the better is the bond quality, other factors being equal. The average wood failure values, as well as the standard deviations for the dry test for the six species, are given in Table 3 and the statistical ranking in Table 5. The wood failure values for in and kanyin are highest, 39 followed by pyinkado and padauk. Ingyin and thitya have the lowest wood failure percent. The statistical ranking puts in and kanyin as not significantly different and pyinkado and padauk as not significantly different but lower than in and kanyin. Ingyin and thitya have failed to meet the minimum requirement for wood failure of 80 percent and shear strength according to CS 253-63. Padauk has passed the minimum requirement for strength, but its average wood failure value is slightly below the acceptable value of 80 percent; this difference is considered insignificant at the 5 percent level in the Duncan's test. Padauk can be considered as having passed the minimum requirement. Padauk, pyinkado, in and kanyin are therefore considered suitable for gluing with phenol-resorcinol resin adhesive. 2.1.2 Cold soak test shear strength and percent wood failure The results of the cold soak test for phenol-resorcinol resin adhesive are shown in Table 3 and Figure 4. In this test padauk shows the highest strength. The other 5 species are ranked as having strength values not significantly different (Table 4). The trend is that of reduced strength compared to the strength in the dry test, with the exception of padauk. Wood failure percentage data for the cold soak test are shown in Table 3, Figure 4 and the ranking by statistical means in Table 5. The ranking essentially follows the same pattern as that for the dry test. In andkanyin are ranked as not significantly different and having the highest value followed by padauk and pyinkado. Thitya has a low value but ingyin has the lowest. 40 2.1.3 Boil test shear strength and wood failure The results for the boil test are shown in Table 3 and Figure 5. Padauk shows the highest value followed by pyinkado. Thitya shows a much higher value than in and kanyin. Ingyin has the least strength. Statistically, padauk and pyinkado are ranked as having similar and highest strength followed by thitya. In andkanyin are ranked as not significantly different; ingyin is ranked last (Table 4). Wood failure values are shown in Table 3 and Figure 5. Statis-tical ranking is shown in Table 5. In and kanyin have the highest value and are ranked as similar followed by pyinkado. ..Padauk ranks fourth with s t i l l a high value. Thitya ranks fifth and ingyin last. 2.1.4 Vacuum-pressure cyclic test shear strength, wood failure and percent delamination Shear strength and wood failure values for the vacuum-pressure cyclic test are shown in Table 3 and Figure 6. Padauk has the highest shear strength followed by thitya and ingyin. In and kanyin have the least strength. Statistical ranking (Table 4) shows that padauk has the highest value. Ingyin, thitya and pyinkado are ranked as having strength values which are not significantly different. In and kanyin have similar strength values and are ranked last. Except for thitya and ingyin all the wood failure values are high (Table 5 and Figure 6). Kanyin and in are ranked as not significantly different (Table 5). Pyinkado and padauk are also ranked as similar. Pyinkado and in have wood failure values which are not significantly different. 41 Results for the delamination test are shown in Table 6. They show a high standard deviation for all species. Nevertheless, general trends can be observed. According to CS 253-63, pyinkado, thitya and ingyin show a delamination percentage far higher than the 8 percent allowed by this standard. Padauk, however, has 11 percent delamination, which is slightly higher than the maximum allowable. On the other hand, in the statistical comparison (Table 6) padauk is ranked with in and kanyin since their delamination values are not significantly different at the 5 percent level. Pyinkado, thitya and ingyin are ranked together, and are considered to have failed the delamination test under conditions used in this study. Further studies on these three species are therefore necessary before they can be recommended for exterior structural lamination purposes with phenol-resorcinol-formaldehyde adhesive. 2.2 Urea-formaldehyde resin adhesive 2.2.1 Dry test shear strength and wood failure This glue was included for comparison with other glues. It has a low resistance to exposure to weather or high humidity conditions at normal temperatures. However, urea-formaldehyde can be useful for bonding wood for non-structural purposes such as interior decorative panelling or furniture.) Only dry and cold soak tests were performed with this adhesive. The results for the dry test shear strength and percent wood failure for urea-formaldehyde resin adhesive are shown in Table 7 and Figure 7. Pyinkado has the highest shear strength followed by in and 42 padauk. Kanyin has almost the same value as ingyin. Thitya has the lowest value. Statistical ranking is shown in Table 8. Pyinkado is ranked highest followed by in. Ingyin,. padauk and kanyin are ranked similar. Thitya, kanyin and ingyin are ranked as having shear strength values which are not significantly different at the 5 percent level. Pyinkado, in and kanyin have strength values higher than the minimum required by CS253-63. Padauk, thitya and ingyin have shear strength values below the minimum acceptable level for hardwood gluing (Table 2). However, padauk and ingyin may be admitted as suitable for gluing i f the shear strength values for the cold test (Table 7) are taken into account. Wood failure values for all species are shown in Table 7. Kanyin and in have the highest values followed by padauk and pyinkado. Thitya and ingyin have the lowest values. Statistical ranking of the species by wood failure is shown in Table 8. In and kanyin are ranked as similar and highest. Padauk and pyinkado are also ranked si.mi lar-aind second. Thitya and ingyin are last. 2.2.2 Cold soak test shear strength and wood failure. Shear strength and wood failure values for urea-formaldehyde resin adhesive cold soak test are shown in Table 7.and Figure 8. Pyinkado and padauk have the highest shear strength values followed by ingyin. Thitya, in and kanyin have low strength values. Shear strength statistical ranking is shown in Table 8. Pyinkado, padauk and ingyin are ranked as having similar strength values. Thitya and in are also ranked as having nearly the same strength. Kanyin is ranked last and has a statistically lower value than the others. 43 All species except thitya and ingyin show very high percent wood failure (Table 7). Kanyin and padauk are ranked as similar and so are padauk and in. In and pyinkado are on the borderline of non-significant difference at the 5 percent level. Thitya has a slightly higher value than ingyin which is ranked last (Table 8). 2.3 Casein glue 2.3.1 Dry test shear strength and wood failure Casein is a popular dry use adhesive for interior structural laminating. The limitation of this adhesive is its sensitivity to mois-ture. This adhesive cannot be used i f the service environment of the product will develop moisture content in the wood greater than 12 percent. Only the dry test was carried out for this glue. The results for glue line shear strength and percent wood failure are given in Table 9 and Figure 9. Pyinkado gives the highest shear strength value followed by padauk. Pyinkado, padauk, in and kanyin have strength values which meet the minimum requirement as specified in CS 253-63 and are therefore considered suitable for gluing,with casein. Ingyin and thitya fail to meet the minimum strength requirement (Table 2). Statistical ranking of the species by shear strength is shown in Table 10. Kanyin has the highest wood failure percent followed by in and padauk. Pyinkado, thitya and ingyin show very low wood failure values (Table 9). According to the CS 253-63 standard, only 40 percent wood failure is required for hardwoods to pass the acceptable requirement for interior use. Pyinkado, thitya and ingyin failed to meet this require-ment. Kanyin, in and padauk are considered suitable for gluing with 44 casein for interior structural use. It is necessary to study further the gluing characteristics of pyinkado, thitya and ingyin before they can be recommended or rejected for gluing with casein for interior struc-tural uses. 3. Variation within Species between Glues 3.1 Pyinkado 3.1.1 Dry test shear strength and wood failure Table 11 shows a comparison of the results. Pyinkado shows equally high dry shear strength for all the three glues. The values are siijghtlyiiliowerethantthe minimumIregutfced^By SGS(253.S6-3 ('Figure 10 ) . The species shows high wood failure values for phenol-resorcinol and urea-formaldehyde resin adhesives. The values are higher than the minimum required by CS 253-63. Pyinkado, however, develops a lower wood failure than the minimum required for interior structural lamination withicasein under gluing conditions used in this study. Other gluing conditions could produce different results. 3.1.2 Cold soak test shear strength and wood failure Only phenol-resorcinol and urea-formaldehyde resin adhesives are compared (Table 11 ) . Pyinkado has a higher shear strength when glued with urea-formaldehyde resin adhesive than when glued with phenol-resorcinol resin adhesive. The cold soak test shear strength for urea-formaldehyde resin adhesive is almost the same as that for the dry test indicating that the wood may not have been properly wetted by the treatment. 45 The shear strength value for phenol-resorcinol resin adhesive is much lower than the dry test shear strength. The wood failure percent for both glues for cold soak treatment for pyinkado are high, 74 percent for phenol-resorcinol-formaldehyde and 88 percent for urea-formaldehyde. 3.2 Thitya 3.2.1 Dry test shear strength and wood failure percent Thitya showed low dry test shear strength values in all three glues and failed to meet the minimum strength requirements (CS 253-63). It had its highest shear strength value with phenol-resorcinol resin glue, followed by casein and urea-formaldehyde resin adhesive with similar values (Figure 10). The wood failure values were low in all the three glues but they were high enough to pass the minimum requirement of 40 percent with phenol-resorcinol and urea-formaldehyde resin adhesives (CS 253-63). 3.2.2 Cold soak test shear strength and wood failure percent Urea-formaldehyde resin adhesive shows higher cold soak test shear strength with thitya than phenol-resorcinol resin adhesive. The cold soak shear strength value for urea-formaldehyde resin adhesive is higher than that for the dry test. This indicates that there may have been some difficulty in wetting this wood during the cold soak treatment. Thitya had 62 percent wood failure for urea-formaldehyde resin adhesive and 47 percent for phenol-resorcinol resin adhesive in the cold soak test. These wood failure values show a slight increase in 46 wood failure with urea-formaldehyde, 10 percent higher than in the dry test, and a slight decrease from the dry test, 18 percent with phenol-resorcinol- formaldehyde. 3.3 Ingyin 3.3.1 Dry test shear strength and wood failure percent This species is very similar to thitya in its properties and behaviour. Ingyin shows low dry shear strength for all the three glues and failure to meet the minimum shear strength requirement for acceptance as gluable under conditions used in this study (CS 253-63). It developed slightly higher shear strength with phenol-resorcinol-formaldehyde and the least strength with casein (Figure 10). For all the three glues ingyin shows very low wood failure per-cent. Except for urea-formaldehyde, the values did not meet the minimum requirement of 40 percent. 3.3.2 Cold soak test shear strength and wood failure percent Ingyin shows reasonably high shear strength with urea-formalde-hyde resin adhesive but a low average value with phenol-resorcinol resin adhesive. The cold soak shear strength with urea-formaldehyde resin adhesive is much higher than the dry shear strength value indicating a great variation in strength properties and possibly a difficulty in soaking the specimen wet in the treatment. Wood failure values are low in both glues, being much lower for phenol-resorcinol resin adhesive (Table 11). 47 3.4 Padauk 3.4.1 Dry test shear strength and wood failure percent Padauk shows high shear strength values for all the three glues. It shows highest value with casein followed by phenol-resorcinol resin adhesive and lastly urea-formaldehyde resin adhesive. The urea-formaldehyde resin adhesive dry test shear strength value for this species is below ith er,.mi nifmum acogp tab" e-, for; § ui $abg J i tyj f ort h a rdwojodr.g.T m n gg'(£i;g u re -10). Wood failure percent is highest with urea-formaldehyde resin adhe-sive followed by phenol-resorcinol resin adhesive and casein. They are all within the acceptable minimum value. 3.4.2 Cold soak test shear strength and wood failure percent Cold soak shear strength values are high for both urea-formaldehyde and phenol-resorcinol resin adhesives. Wood failure values are equally high for both glues. 3.5 In 3.5.1 Dry test shear strength and wood failure percent Dry shear strength in this species is high for all the three glues and meets the minimum acceptable value. The strength is highest for phenol-resorcinol resin adhesive followed by urea-formaldehyde resin adhesive and casein glue last (Figure 10). The wood failure values are high for all the three glues and meet the minimum requirement. Though there is no significant difference be-tween the values for phenol-resorcinol and urea-formaldehyde resin adhesives, the value for casein is slightly lower than the other two. 48 3.5.2 Cold soak test shear strength and percent wood failure In shows l i t t le difference in cold soak shear strength between urea-formaldehyde and phenol-resorcinol resin adhesives. Very high wood failure values are developed by in for both urea-formaldehyde and phenol-resorcinol resin adhesives, being slightly higher in the latter adhesive. 3.6 Kanyin 3.6.1 Dry test shear strength and wood failure percent Kanyin shows similar properties to in. It develops high dry shear strength for all the three glues and passed the minimum strength requirement (CS 253-63). Kanyin has highest shear strength value with phenol-resorcinol resin adhesive followed by urea-formaldehyde resin adhesive and lastly casein. High wood failure values were shown by kanyin for all the three glues. The species passed the minimum required wood failure value for all the three glues. It showed highest wood failure value with urea-formaldehyde resin adhesive followed by phenol-resorcinol resin adhesive and casein. 3.6.2 Cold soak test shear strength and wood failure percent Kanyin developed similar shear strength values with phenol-resorcinol and urea-formaldehyde resin adhesives in the cold soak test. High wood failure values, close to 100 percent, were shown for both urea-formaldehyde and phenol-resorcinol resin adhesives in the cold soak test. 49 4. Effect of Treatment within Species The results for .effects of treatment on shear strength and wood failure percentage within species for phenol-resorcinol and urea-formalde-hyde resin adhesives are shown in Table 11. 4.1 Phenol-resorcinol-formaldehyde adhesive 4.1.1 Pyinkado: Shear strength and wood failure percentage If the dry test is considered the control treatment, pyinkado has lost shear strength from 2,887 psi for the dry test to 2,111 for the cold soak test, 2,235 psi for the vacuum-pressure cyclic test and 2,0T1 psi for the boil test. It is evident that the treatment which reduced the strength most is the boil test. The vacuum-pressure test s t i l l has higher strength value than the cold soak and boil tests. The wood failure percentage is highest in the boil test, 88 percent, followed by vacuum-pressure test at 86. Cold soak test wood failure per-centage is lower than that for the dry test, 74 and 81 percent, respec-tively. 4.1.2 Thitya: Shear strength and iwoo'd-fa-i\kire 'percentage Thitya shows a trend for decreasing shear strength from the dry test 2,437 psi to the lowest in the boil test at 1,698 psi. The vacuum-pressure treatment s t i l l has a high value (2,358 psi) which is very close to the dry test, while the cold soak treatment has a strength value (1,919 psi) intermediate between the dry test and the vacuum-pressure test. 50 Wood failure values are low in all the 4 treatments and do not vary greatly. They are 57 percent for the dry test, 47 percent for the cold soak test, 42 percent for the vacuum-pressure test and 56 percent for the boil test. 4.1.3 Ingyin: Shear strength and wood failure percentage The pattern is that of reduction in shear strength from the dry test at 2,426 psi , vacuum-pressure test at 2,305 psi , cold soak test at 1,818 psi , to less than half that of the dry test in the boil test at 1,039 psi. Wood failure values are low in all four treatments, being 21 percent in the dry test, 28 percent in the vacuum-pressure treatment, 14 percent in the boil test and 10 percent in the cold soak test. 4.1.4 Padauk: Shear strength and wood failure percentage The dry shear strength, 2,693 psi , is slightly lower than the cold soak shear strength 2,780 psi , but this could be only a variation in inherent strength of the wood itself (to be elaborated in Discussion). The vacuum-pressure test s t i l l shows high shear strength of 2,531 psi. The strength is reduced to 2,181 psi in the boil test. Wood failure percentage is about the same in all the 4 treatments, being 78 percent in the dry test, 84 percent for the cold soak test, 83 percent for the vacuum-pressure test and 80 percent for the boil test. 4.1.5 In: Shear strength and pwoodrfai:?l?uce" pecc'erffeage«*^  In is one of the lighter species in the group. There is a definite trend in reduction of shear strength from 2,719 psi for the dry test, 51 2,097 psi for the cold soak test, 1,693 psi for the vacuum-pressure test, to the lowest of 1,262 psi for the boil test shear strength, which is about half that of the dry strength. Percentage of wood failure is very high in all 4 cases, being 93 percent in the dry test, 97 percent in the cold soak test, 91 percent in the vacuum pressure test and 95 percent in the boil test. 4.1.6 Kanyin: Shear strength and ,woodaifaiMare percentage Kanyin is the lightest member of the group. It shows a trend of decreasing shear strength from 2,571 for the dry test, 1,945 psi for the cold soak test, 1,677 psi for the vacuum-pressure test to 1,264 psi for the boil test. The wood failure percent is very high in all 4 treatments. It is 94 percent for the dry test, 100 percent for the cold soak test, 93 percent for the vacuum-pressure test and 97 percent for the boil test. 4.2 Urea formaldehyde adhesive 4.2.1 Pyinkado: Shear strength and wood failure percent There is not much difference between the dry test shear strength at 2,881 psi and the cold soak test shear strength at 2,758 psi. The wood failure percent is high and nearly equal for both the dry test at 86 percent, and cold soak test at 88 percent. 4.2.2 Thitya: Shear strength and wood failure percent Shear strength values for the dry test and the cold soak test are very similar, being 2,217 psi and 2,389 psi , respectively. 52 Both treatments show low wood failure percent values. The dry test has a slightly lower value, 57 percent, than the cold soak test, 62 percent. 4.2.3 Ingyin: Shear strength and wood failure percent The dry test shows less shear strength than the cold soak test, 2,330 psi and 2,612 psi , respectively. The difference is considered to be a result of inherent variability in the wood itself . This will be further elaborated in the Discussion. Wood failure percent values are low, being 58 percent for the dry test and 47 percent for the cold soak test. 4.2.4 Padauk: Shear strength and wood failure percent Again, here, the cold soak shear strength is higher than the dry test shear strength but the difference is not very great, being 2,715 psi and 2,448 psi for the cold soak test and dry test, respectively. There is an increase in wood failure percentage from 87 for the dry test to 96 for the cold soak test. 4.2.5 In:Shear strength and wood failure percent In shows a reduction in shear strength from 2,681 psi for the dry test to 2,137 psi for the cold soak test. Percentage of wood failure is high in both tests, being 94 for the dry test and 93 for the cold soak test. 4.2.6 Kanyin: Shear strength and wood failure percent Shear strength is reduced from 2,330 psi for the dry test to 1,872 psi for the cold soak test. 53 Both treatments show very high percent wood failure. The dry test has 98 percent wood failure and the cold soak test 99 percent. 5. Within Treatment between Species Variation 5.1 Phenol-resorcinol-formaldehyde: Shear strength and wood failure percent The general trend is that for in and kanyin, which are lighter than the other species in the study there is a reduction in shear strength progressively from the dry, cold soak, vacuum-pressure to the boil test (Table 11). Ingyin lost 57 percent of the dry shear strength in the boil test, in lost 53 percent, kanyin 50 percent, padauk and pyinkado lost 30 percent each and thitya lost 19 percent of the dry shear strength in the boil test. In the vacuum-pressure test in lost 37 percent of the dry shear strength, kanyin lost 34 percent, pyinkado lost 22 percent, padauk lost 6 percent, ingyin 4 percent and thitya 3 percent. After the cold soak treatment pyinkado had lost 26 percent of the dry shear strength, ingyin 25 percent, in 24 percent, kanyin 22 percent and thitya 21 percent. Padauk on the other hand had a cold soak shear strength higher than that of the dry test. Wood failure percent vailues do not show any particular trend between species in the four treatments except for the fact that the values are highest for in and kanyin and are lowest for thitya and ingyin in all treatments. Padauk and pyinkado have high wood failure percent for all the treatments. 54 5.2 Urea-formaldehyde adhesive: Shear strength and wood failure percent The trend here is that of reduction in strength from dry to cold soak test for in , kanyin and pyinkado. For the other three species this is not observed. Pyinkado, padauk, in and kanyin show high wood failure percent, above 85 percent, for both the cold soak and the dry tests. Thitya and ingyin have low wood failure values for both treatments, equal to or below 62 percent. 6. Specific Gravity The specific gravity determination for the six Burmese woods are shown in Table 12. Thitya and ingyin have nearly the same values (0.80 and 0.81, respectively), pyinkado and padauk have nearly the same values (0.75 and 0.77)a respectively) while in has a specific gravity of 0.67 and kanyin 0.59. These values are based on ovendry weight and green volume. All the six species have green specific gravity above 1.00 (Table 12). Pyinkado and thitya have green specific gravity of 1.16, ingyin 1.15, padauk 1.14, in 1.08 and kanyin 1.06. These species would sink in water when fully saturated with water. Statistical ranking by specific gravity (ovendry wt/green volume) is shown in Table 13. Thitya, ingyin, padauk and pyinkado are ranked as not significantly different. Similarly padauk, pyinkado and in are ranked as not significantly different. In and kanyin are ranked similar. 55 On the basis of specific gravity, pyinkado and to some extent padauk can be compared to hickory [Carya glabra (Mill.) Sweet.] which has a specific gravity of 0.75. In can be compared to American beech (Fagus grandifolia Ehrh.) which has a specific gravity of 0.64. Kanyin can be compared to white oak (Quercus alba Linn.) which has a specific gravity of 0.60 7. Radial, Tangential and Volumetric Shrinkage Shrinkage values from green to ovendry weight for the six Burmese woods are shown in Table 12. In has the highest radial shrinkage in the group followed by kanyin. In shrinks radially 7.4 percent from green to ovendry, kanyin 6.7 percent, thitya 5.7 percent, ingyin 5.3 percent, pyinkado 3.7 percent and padauk, the least shrinking in the group, 2.2 percent. Kanyin has a high tangential shrinkage of 11.2 percent from green to ovendry followed by in 9.4 percent, thitya 7.3 percent, pyin-kado 6.5 percent, ingyin 6.2 percent and padauk 3.8 percent. Volumetric shrinkage from green to ovendry is highest in kanyin (17.2 percent). In has a volumetric shrinkage of 16.2, thitya 12.5, ingyin 10.8, pyinkado 10.4 and padauk 6.7 percent. Pyinkado shows a higher T/R ratio of 1.75, followed by padauk 1.72, kanyin 1.67, thitya 1.28, in 1.27, and ingyin 1.16. 56 8 pH Results of pH measurements are shown in Table 14. Extractives of all the six Burmese woods are acidic. The pH in order of increasing acidity is pyinkado 5.09, padauk 4.87, in 4.70, thitya 4.62, kanyin 4.61, and ingyin 4.47. Statistical ranking by pH for cold water extractives of the six Burmese woods is that pyinkado is ranked highest followed by padauk and ingyin which are ranked as not significantly different. In, thitya and kanyin are ranked similar, and kanyin and ingyin are also ranked similar (Table 15). 9. Extractives Table 16 gives the results of successive extraction with ether, ethanol and hot water as well as single extraction with acetone. Statis-tical ranking by percentage of extractives soluble in ether, ethanol, water and acetone is shown in Table 17. 9.1 Ether soluble extractives For ether soluble extractives pyinkado is ranked highest with 3.4 percent, padauk has 2.5 percent, while in and ingyin are ranked similar with 1.8 percent. Kanyin and thitya are ranked similar with 1.3 and 1.2 percent,respectively. 57 9.2 Ethanol soluble extractives Padauk has the highest amount of ethanol soluble extractives (9.6 percent) followed by thitya (7.3 percent), ingyin (6.4 percent), pyinkado (5.2 percent), kanyin (1.4 percent), and in ..(0.5 percent) 9.3 Hot water soluble extractives Pyinkado is ranked as having the highest amount of hot water soluble extractives (3.9 percent). Padauk has 2.1 percent, thitya 1.8 percent, and ingyin 1.1 percent. In and kanyin are ranked as not signi-ficantly different and have the least amounts (0.5 and 0.6 percent, respectively). 9.4 Acetone soluble extractives Padauk has 9.5 percent acetone soluble extractives and is ranked highest among the six Burmese woods. Pyinkado has 8.4 percent, ingyin 7.4 percent, thitya 6.9 percent, in 2.5 percent and kanyin 2.2 percent. DISCUSSION In this study the gluing conditions for each glue were care-fully controlled. For each glue all six species were treated equally in the amount of glue spread, open and closed assembly times, press temperature, pressure and clamp period. The differences in their glue bond performances, therefore, arise from the inherent characteristics of the species. 1. Influence of Specific Gravity 1.1 Phenol-resorcinol resin adhesive: Glue line dry shear strength and wood failure percent The specific gravities of the six species, in order of magnitude, are ingyin 0.81, thitya 0.80, padauk 0.77, pyinkado 0.75, in 0.67 and kanyin 0.57 (Table 2). Their dry glue line shear strength values, in the same order, are ingyin 2,426 psi , thitya 2,437 psi , padauk 2,693 psi , pyinkado 2,887 psi , in 2,719 psi and kanyin 2,571 psi (Table 3). Thitya and ingyin have specific gravities higher than the rest of the species but their glue line shear strength is lower (Figure 11). Duncan's multiple range test (Table 4) ranks pyinkado as having the highest shear strength. In and padauk are ranked as having strength of the glue bond not significantly different. Padauk and kanyin are also shown to have strength values which are not significantly different. 58 59 Kanyin, ingyin and thitya are ranked similar in shear strength. The strength of a good glue-to-wood bond should be equal to or greater than that of the wood itself . If a 'good' glue bond had been formed for all six species with phenol-resorcinol-formaldehyde adhesive the minimum acceptable strength of the glue line (Table 2) is pyinkado 2,891 psi , thitya 2,704 psi , ingyin 2,627 psi , padauk 2,736 psi , in 2,384 psi and kanyin 2,122 psi based on the actual test of wood blocks. Thitya and ingyin have values of strength 10 percent or less below the anticipated values, indicating that the bond did not develop full shear strength. Padauk is also slightly below the expected shear strength. The wood failure percent for the six species are as follows: kanyin 94 percent, in 93 percent, pyinkado 81 percent, padauk 78 percent, thitya 57 percent and ingyin 21 percent (Table 3). Duncan's multiple range test (Table 4) ranks in and kanyin as having similar wood failure percent, padauk and pyinkado also are not significantly different, thitya is ranked fifth and ingyin last. Thitya and ingyin have low wood failure percent while in and kanyin have the highest percentage. With in and kanyin, most of the bond failure occurred within the wood itself . With padauk and pyinkado some of the failure occurred in the3glue, while for thitya and ingyin most of the failure occurred in the glue rather than in the wood. This indicates that in and kanyin formed good glue bonds and thitya and ingyin formed poor glue bonds when glued with phenol-resorcinol-formaldehyde adhesi ve. On the basis of wood failure percent, shear strength and specific gravity certain observations can be made. In and kanyin have the lowest 60 specific gravities in the group, and developed very high wood failure percent and reasonably high dry shear strength. Pyinkado and padauk, the medium specific gravity species in the group developed high glue line dry shear strength and medium to high wood failure percent. On the other hand, thitya and ingyin, the group with the highest specific gravi-ties developed lowest glue line dry shear strengths and low wood failure percents (Figure 3). These results are not very much different from those obtained by other workers. Troop and Wangaard (1950) observed that glue line shear strength increases with an increase in specific gravity. They also observed a decreasing wood failure percent with increase in specific gravity. Sakuno and Goto (1970b), however, went further and established the fact that specific gravity and glue joint strength were only corre-lated up to a point. This point was found to be a specific gravity of 0.8. For all their values of specific gravity below 0.8, shear strength increased with increase in specific gravity. For values of specific gravity higher than 0.8 there was no correlation between specific gravity and shear strength. Moriya et a l . (1971) found a nearly straight line relationship between shear strength and specific gravity for values nearly in the same range as Sakuno and Goto (1970b). In this study, in view of the few species tested, no statistical correlation between specific gravity and shear strength was carried out. However, a general trend can be observed (Figure 11) and that is,that in and kanyin, the lighter species in the group, develop low shear strength and very high wood failure, pyinkado and padauk, the medium density 61 species develop high shear strength and reasonably high wood failure percent (81 and 78 percent, respectively). Thitya and ingyin, the high density species in the group, developed low shear strength and low wood failure percent when bonded with phenol-resorcinol-formaldehyde adhesive under conditions used in the study. 1.2 Urea-formaldehyde resin adhesive: Glue line dry shear strength and wood failure percent The glue line shear strength for the six species in order of decreasing specific gravity is ingyin 2,320 psi , thitya 2,217 psi , padauk 2,448 psi , in 2,681 psi and kanyin 2,330 psi (Table 7). Thitya, ingyin and padauk have shear strength values lower than pyinkado and kanyin in spite of their higher specific gravities (ingyin 0.81, thitya 0.80, padauk 0.77, pyinkado 0.75, in 0.67 and kanyin 0.59 (Figure 12). Dry shear strength statistical ranking (Table 8) shows that pyinkado has the highest shear strength followed by in. Padauk, kanyin and ingyin are ranked as having shear strength values which are not significantly different. Thitya, ingyin and kanyin are also' ranked as being similar in shear strength (see also Figure 12). Thitya and ingyin, in spite of their high specific gravities, developed lower shear strength values than the minimum allowable,2,704 psi and 2,627 psi , respectively (Table 2). Similarly, padauk developed glue line shear strength lower than the minimum acceptable value of 2,736 psi. Pyinkado had shear strength slightly below the acceptable minimum but the difference is so small as to be ignorable. In and kanyin developed strength values which are higher than the minimum acceptable (CS 253-63). 62 An examination of Table 8 shows that in and kanyin have very high wood failure percent and are ranked statistically as similar (94 and 98 percent, respectively). Padauk and pyinkado have high wood failure and are ranked as similar (87 and 86 percent, respectively). Both ingyin and thitya have low wood failure values and are statistically ranked as not significantly different. The overall picture indicates that for in and kanyin nearly all the failure occurred in the wood rather than in the glue. This indicates that the shear strength is mainly that of the wood rather than the glue. Pyinkado has high shear strength and high wood failure indicating that a good bond was formed. Padauk, on the other hand, has a low strength value for its specific gravity but a high wood failure percent, indicating that a good bond was formed and the low shear strength is a result of the inherent variability of the wood itself rather than a poor glue bond. For thitya and ingyin.it is observed that both species have low shear strength and low wood failure percentage, indicating that a poor bond was formed and that only a small part of the bond failure occurred in the wood itself . Most of the failure occurred either in the glue or in the wood glue interface. Species with lower specific gravity in the group, in and kanyin, have shown very high wood failure and moderate strength; pyinkado, a species with medium specific gravity, has shown high wood failure percent and high shear strength. Padauk, with a slightly higher specific gravity than pyinkado, has shown high wood failure percent but a slightly lower shear strength. This has been explained in terms of inherent variability of the wood itself rather than a poor glue bond. Thitya and ingyin 63 have high specific gravities but they have exhibited low glue line shear strength values and low wood failure percent, an indication that they did not glue well with urea-formaldehyde adhesive. The results are in general agreement with those observed for other tropical hardwoods by Troop and Wangaard (1950). A general trend of increasing joint strength and decreasing wood failure with increase in specific gravity was observed in their study. They also observed some uneven character of trend of wood failure and shear strength values in certain species. This is also observed in this study in the case of ingyin and thitya. Other workers such as Goto et a l . (1967), Yagishita and Karasawa (1969) also report that glue joint shear strength increases with increasing specific gravity. Sakuno and Goto (1970b) showed that the relationship between glue joint shear strength and specific gravity is only positively correlated up to specific gravity value of 0.8, for both urea-formaldehyde and phenol-formaldehyde adhesives. For values of specific gravity above 0.8, the correlation between specific gravity and glue line shear strength was not significant. For wood failure the opposite was found to be the case. In this study no statistical correlation between glue joint strength and specific gravity was done in view of the few species used. The observation for urea-formaldehyde, however, follows a similar trend to that observed by Moriya et a l . (1971) for red lauan species. The species with high specific gravity show glue line shear strength lower than that anticipated on the basis of their specific gravity. Ingyin, thitya to some extent, and padauk seem to follow the same trend (Fig. 12). 64 In and kanyin show strength properties similar to the shear strength of the wood (2,649 psi and 2,358 psi - see Table 2). Pyinkado is within the minimum acceptable shear strength. The trend, therefore, is that of increasing glue line shear strength and decreasing wood failure for kanyin, in, pyinkado and padauk. Ingyin and thitya do not follow this trend and show both low wood failure and shear strength for their specific gravity values. 1.3 Casein: Glue line dry shear strength and wood failure percent The shear strength values for the 6 species glued with casein, in order of their increasing specific gravity, are as follows: kanyin 2,274 psi , in 2,455 psi , pyinkado 2,884 psi , padauk 2,709 psi , thitya 2,251 psi and ingyin 2,118 psi. Their specific gravities in the same order are kanyin 0.59, in 0.&7. pyinkado 0.75, padauk 0.77 and ingyin 0.81 (Figure 13). Kanyin, in and pyinkado show an increasing shear strength with increase in specific gravity. Padauk shows a slight reduction in strength but this can be taken to be lying in the same general area of increasing strength with increase in specific gravity. Thitya and ingyin, on the other hand, show much lower strength values than expected for their specific gravities (Figure 13). This is similar to the observations of Sakuno and Goto (1970b) and Moriya et a l . (1971) for other hardwood species. For wood failure percent, the values are low for all species except in and kanyin which are the two with the lowest specific gravities 65 in the group. In and kanyin have 81 percent and 90 percent wood failure, respectively. The other species, ingyin 9 percent, thitya 11 percent, pyinkado 22 percent and padauk 67 percent, do not seem to have developed full glue to wood bond. Kanyin and in have good strength values for their specific gravi-ties and also high wood failure percent. They glue easily with casein under conditions used in this study. Pyinkado and padauk developed good strength properties but their wood failure percent is low (22 percent and 67 percent, respectively). On this basis they do not glue well with casein and need to be examined further under other gluing conditions. Padauk, however, meets the requirements for gluing with casein for in-terior structural use according to CS 253-63 (only 40 percent wood failure is required in addition to stipulated strength). All three glues show a trend that for kanyin, in and pyinkado there is a definite trend of increasing shear strength with increase in specific gravity. Padauk shows slightly lower values than anticipated for its specific gravity but this can be attributed to inherent vari-ability in strength of the wood itself in the case of urea-formaldehyde adhesive. In the case of phenol-resorcinol-formaldehyde adhesive and casein, the slight reduction in shear strength can be explained partly by a poor glue joint and partly by inherent variability in strength in the wood itself . Ingyin and thitya show low strength values with all the three glues in spite of their high specific gravity which is presumed to be directly a result of the poor wood glue bond. 66 The general trend for the percent wood failure is that of in-creasing wood failure with a decrease in specific gravity for all the species when glued with phenol-resorcinol-formaldehyde adhesive and urea-formaldehyde adhesive. This trend was not observed in the case of casein. Padauk and pyinkado showed abnormally low wood failure per-cent, indicating that the species do not glue well with casein. Thitya and ingyin also had very low wood failure percent. A similar observation was made by Eickner (1942) who reports that there is less difference between gluing characteristics of woods of different densities when glued with urea-formaldehyde adhesive than when glued with casein. In the present study four out of six species glued well with urea-formaldehyde resin adhesive whereas only two out of the six species glued well with casein. The results of the wood failure percent are similar to the obser-vations of Troop and Wangaard (1950). They report that casein and vege-table glues do not seem to permit as full a development of the strength of white oak as do the animal, urea and resorcinol resin adhesives. When comparing mahogany (Swietenia macrophylla King.) and white oak they found that the non-resin adhesives showed appreciably lower wood failure values in the case of white oak. In this study phenol-resorcinol-for-maldehyde and urea-formaldehyde adhesives showed reasonably high wood failure values for four out of the six species whereas casein showed high wood failure values for two species, both of which are of lower specific gravity than the rest. The influence of specific gravity on the gluing characteristics of wood, therefore, varies from glue to glue. The results of this study show that the lighter species of the group, 67 in and kanyin, glue well with all the three glues. Padauk and pyinkado glue reasonably well with phenol-resorcinol-formaldehyde and urea-for-maldehyde adhesives but poorly (when both dry shear strength and percent wood failure are considered) with casein. The species with the high specific gravity, ingyin and thitya, do not glue well with any of the three glues. 2. Influence of Extractives Extractives are extraneous components of wood which can be ex-tracted by cold or hot water, or one or more neutral solvents such as ether, alcohols, benzene, petroleum ether, or acetone. Extractives may formfrom about ltto-25 percent of the ovendryy weight of wood. Sap-wood and heartwood have different extractive contents. Wood rays have higher extractive content than fibrous cells (Isenberg et a l . 1957). Extractives comprise an extraordinary diversity of compounds. The pro-portions exhibit wide variation and some of these components are found in significant quantities in only a few species or genera. Extractives characterise a specific wood in that they affect the colour, odour, taste and resistance to fungal or insect attack. Extractives influence the gluing of wood. Their influence on the gluability of wood was demonstrated by Rapp (1948), Narayanamurti (1957), Huffman (1961), Narayanamurti et a l . (1962), Hancock (1964), Goto et a l . (1967), Imamura et a l . (1970), Cherv (1970), Chow (1971), and Scharfetter (1971) to name but a few. 68 Removal of extractives by water or other solvents has been shown to improve the gluing of lignum vitae by applying 10 percent caustic soda solution, allowing the solution to remain on the surface for 10 minutes, washing the surface with water, drying the wood and gluing it with a resorcinol type of glue. There was a great improvement in the gluability of the wood which is otherwise regarded as one of the most diff icult to glue. Extractives affect the formation of a glue bond either physically or chemically or both physically and chemically. Scharfetter (1971) reports that the gluability of wood is impaired by small quantities of certain substances even though these substances do not react chemically with the wood or the adhesive. Extractives affect the penetration of glue into the wood by acting as a barrier. Extractives may act as deterrents to adequate penetration of the fluid adhesive and may retard the dissi-pation of water or other solvents from the glue line. Migration of fatty acids to the surface of Douglas f i r veneer as a result of high temperature treatment reduces the wettability of the veneer and affects the pene-tration rate and depth of penetration of the glue (Hancock, 1964). Thus extractives act as mechanical barrier between the glue and the wood; they reduce penetration of glue into the wood, as well as retarding the dissipation of water and other solvents from the glue line. The last factor is very important for the curing of the glue. Wood adhesives cure either by cooling and/or by loss of water, by chemical action or by a combination of two or more of these. If the glue curing is inter-fered with, water or another solvent cannot be removed from the glue 69 line because of the extractive barrier, a weak glue joint results. Curing of the glue before adequate penetration is achieved also.results in a poor glue joint. Extractives affect the gluing of wood because of their chemical behaviour. They interfere with both the properties of the wood surface and the properties of the glue. It has been shown that extractives lower the surface tension of the wood surface and reduce wettability (Chugg and Gray, 1965). If the surface tension of the wood surface is sufficiently lowered the glue may display a definite receding angle of contact on some parts of the wood surface so that these remain com-pletely free from glue. This is likely to occur with glues possessing a high surface tension such as cold setting phenol-formaldehyde and urea-formaldehyde resin adhesives. The lowering of the wood surface tension results in the wood surface not being wetted by the glue, which in turn leads to a poor glue bond. The removal of extractives improves wettability and increases the pH of the wood surface (Chen, 1970). Extractives may also interfere with the bonding between the polar groups of the adhesive and the hydroxy! groups of the cellulose, that is interfere with specific adhesion. If wood has more affinity for extractive substances than the adhesive, then bonds will be formed be-tween wood and extractives rather than wood and glue. This leads to a poor glue bond. The influence of extractives on adhesion has been studied by Sanderman and Dietrichs (1957), Narayanamurti (1957), Narayanamurti et a l . (1962), Imamura et al . (1970). Extractives influence rheological properties of adhesives (Narayanamurti, 1957). 70 Sanderman and Dietrichs (1957) report that woods with a high tannin content tend to retard or inhibit the drying of oil varnishes. They also report that in iroko (Chlorophora excelsa Welw.), chlorophrin is responsible for retardation of the drying process. For Dalbergia  nigra Fr. All em. i t was found that phenolic constituents retard varnish drying. Effect of extractives of teak and cutch (Acacia catechu) have been discussed by Narayanamurti et a l . (1962). By their chemical nature extractives may accelerate the rate of setting of glue, resulting in premature solidification. They may on the other hand retard i t , resulting in a weak joint. Extractions from teak and Acacia catechu have been shown to have a pronounced effect on the setting and mechanical properties of animal glue and urea-formaldehyde resin adhesive. Extractives from afrTormosiai inhibit the setting of most adhesives, particularly animal glue, and set phenol-formaldehyde resin adhesive and polyvinyl acetate (Chugg and Gray, 1965). Imamura et a l . (1970) showed that only certain portions of extrac-tives affect gluing and varnish curing. They showed that other extractives in the wood had no inhibiting action. The effects of extractives vary from glue to glue as different glues have different pH. Most adhesives possess a degree of detergent action so that small quantities of extrac-tives are taken up by the adhesive. This is particularly true of adhe-sives such as casein or many phenol-formaldehyde plywood glues which contain sodium hydroxide (Chugg and Gray, 1965). Goto et a l . (1967) pointed out that percentage extract in the species studied had less important effect on the glue joint strength 71 than specific gravity and wettability. It is not the quantity of ex-tractives present that is very important but their chemical composition. Extractives also affect the shrinkage behaviour of wood by acting as bulking agents. This is important in gluing when the glued wood is to be exposed to varying weather conditions. 2.1 Successive extraction 2.1.1 Ether soTubT-e:extractives Pyinkado has the highest ether soluble extractives in the group; 3.4 percent followed by padauk at 2.5 percent. Thitya and kanyin have 1.2 percent and 1.3 percent, respectively. In and ingyin have similar amounts of 1.8 percent each. The species which did not glue well, thitya and ingyin have similar amounts of extractive content as the species which have glued very easily. Pyinkado and padauk, which glued moderately well, have higher amounts of ether soluble extractives (Table 17). The shear strengths for wood bonded with phenol-resorcinol-for-maldehyde resin adhesive in order of decreasing ether extract percent are pyinkado 2,887 psi , padauk 2,693 psi , in 2,719 psi , ingyin 2,426 psi, thitya 2,437 psi and kanyin 2,571 psi. There is no trend of increased shear strength of the glue joint with a decrease in percent ether extrac-tive content of the species. The same is true for wood glued with urea-formaldehyde resin adhesive and casein. This seems to point out the contention that quantity of ether soluble extractives may not be as important as the quality or type of extractives in the gluing of tropical hardwoods. Species with high ether soluble extractive content, pyinkado 72 and padauk, have shown better gluing characteristics than species with low percentage of ether extracts, ingyin and thitya. 2.1.2 Ethanol soluble extractives The amounts of ethanol soluble extractives for each of the six species are shown in Table 16 and 17. Padauk has 9.6 percent, thitya 7.3 percent, ingyin 6.4 percent, pyinkado 5.2 percent, kanyin 1.4 percent and in 0.5 percent ethanol soluble extractives as obtained by successive extraction. The amount of ethanol soluble extractives from each species could be higher i f only a single extraction were used. In and kanyin glued very easily with all the three glues. They have lower amounts of ethanol soluble extractives than the rest of the species. Padauk has the highest quantity of ethanol soluble extractives (9.6 percent) in the group but shows high glue joint strength with all the glues. It shows, however, low wood failure values in the case of casein, indicating that extractives may have affected the setting of this glue. This may have been caused by the presence of sufficient quantities of ethanol soluble extractives (this could be true for ether extracts too). Thitya and ingyin have relatively high amounts of ethanol soluble extractives, 7.3 and 6.3 percent, respectively. There is a possibility that these high amounts of ethanol extracts may be respon-sible for the poor gluing properties of these two species for all three glues. Pyinkado has 5.2 percent ethanol soluble extractives (higher than ether soluble extractives). These may explain the poor performance of the species when glued with casein. In and kanyin have low amounts of ethanol soluble extractives and they glue very easily with all three 73 glues. There is no definite trend of increased ease of gluability with decrease in ethanol extract except for the fact that in and kanyin, the species with low amounts of ethanol soluble extractives, glued very well with all three glues. 2.1.3 Hot water soluble extractives Amounts of hot water soluble extractives for all the species are shown in Tables 16 and 17. In and kanyin have the lowest amounts at 0.5 and 0.4 percent. Ingyin has 1.1 percent, thitya has 1.8 percent, padauk 2.1 percent and pyinkado has the highest amount, 3.9 percent, of hot water soluble extrac-tives. No apparent relationship between gluability and hot water extrac-tive content is shown except for the fact that species with the least hot water soluble extractive content, in and kanyin, glued well. 2.1.4 Total extractives The amounts of total extractive content for ether, ethanol and hot water are shown in Table 16. Padauk has the highest amount of ex-tractives, 14.2 percent, pyinkado 12.5 percent, thitya 10.3 percent, ingyin 9.3 percent, kanyin 3.1 percent andin has 2.8 percent. In has nearly five times less extractives than padauk and three times less than thitya and ingyin. In and kanyin, presumably because of their low extractive content, glued well with phenol-resorcinol-formaldehyde resin adhesive, urea-formaldehyde resin adhesive and casein. Pyinkado and padauk glued reasonably well with phenol-resorcinol-formaldehyde and urea-formaldehyde adhesives in spite of their high 74 extractive content. It is likely that the extractives of these species do not interfere adversely with the gluing abilities of urea-formaldehyde and phenol-resorcinol formaldehyde. Pyinkado and padauk glued poorly with casein. The extractives may have affected glue setting properties, wood wetting properties of casein, penetration of the glue into the wood, or form a mechanical barrier between the wood surface and the glue, but none of these factors have been investigated in this study. Any of these factors may have contributed to the poor performance of pyinkado and padauk when glued with casein. Thitya and ingyin,which are high density woods, show a high amount of extractive content (10.3 percent and 9.3 percent, respectively). They both performed poorly with all the three glues. Their extractives may have interfered with the wood surface characteristics, or the glues, or both. Since the glues have different properties, the most likely way the extractives could affect all of them equally is by acting as a mechanical barrier, thus denying the glues the possibility of forming bonds with the wood. The extractives may have interfered with the rate and depth of penetration of the glues into the wood. Some penetration must have taken place as evidenced by a range of wood failure values from 9 percent for casein, 21 percent for phenol-resorcinol resin adhesive to 58 percent for urea-formaldehyde resin adhesive for ingyin. Thitya had 11 percent wood failure for casein, 57 percent for urea-formaldehyde and phenol-resorcinol-formaldehyde resin adhesives. In this respect, thitya shows slightly better gluing ability than ingyin though they are both poor in gluability. 75 The extractives of thitya and ingyin may also have a wide variety of properties so that they affect the glues adversely irrespective of the type of glue. They may affect the glue's rheological properties, viscosity and the rate of curing. They may retard or speed up the curing process resulting in a prematurely cured bond or an uncured bond both of which are poor but none of these factors have been studied herein. In and kanyin have low extractive contents. It is evident that these extractives have l i t t l e , i f any, adverse effect either on the wood surface properties or the adhesivestused to glue the woods. They show good glue joint strength and wood failure percent with phenol-resorcinol resin adhesive, urea-formaldehyde resin adhesive and casein, indicating that the wood was properly wetted by the glue, there was adequate pene-tration of the glue into the wood and the glue cured properly forming a strong bond. The bond was stronger than the glue as evidenced by high wood failure percent. 2.2 Acetone soluble extractives The results of the acetone single extraction are shown in Table 16 and statistical ranking in Table 17. Similar results were obtained in this extraction to the total amount of extractives in the successive extraction except the amount of extractives is slightly lower than the total. Padauk has the highest amount of acetone soluble extractives at 9.5 percent, fol lowed by pyinkado at 8.4 percent. Ingyin has 7.4 per-cent, thitya 6.9 percent, in 2.5 percent and kanyin 2.2 percent. 76 In and kanyin have the lowest amounts of acetone soluble extrac-tives in the group and glued very e a s i l y , better than any other member of the group. Padauk and pyinkado, in spite of th e i r high extractive content, glued reasonably well with phenol-resorcinol-formaldehyde and urea-formaldehyde resin adhesives but not with casein. Ingyin and thitya have high amounts of acetone soluble extractives and did not glue well with any of the three glues. The explanation offered for the gluing behaviour of the s i x species with respect to t h e i r extractive content in the successive ex-traction holds true for the acetone extracts, too. From observation of the results obtained in this study, there i s no d i r e c t relationship between the percentage of extractives of the various solvents and the g l u a b i l i t y of the species. The species with low amounts of extractives were shown to be e a s i l y glued whereas the species with high extractive contents show a variation in t h e i r gluing behaviour. They are, on the whole, more d i f f i c u l t to glue than the species with low extractive content. The species with low s p e c i f i c gravity have low wood extractive contents and those with high s p e c i f i c gravity have high percentages of extractives, based on the oven-dry weights of the unextracted wood. The e f f e c t of s p e c i f i c gravity on g l u a b i l i t y i s more important and consistent than that of extractives as indicated by the results of this study. 77 3. Influence of pH All the six species are acidic (Table 14). Pyinkado has a pH value of 5.09, padauk 4.87, in 4.70, thitya 4.62, kanyin 4.61 and ingyin 4.47. Statistical ranking by pH is shown in Table 15. If the effect of specific gravity is ignored, i t is observed that the species with the highest pH value, that is pyinkado, has the highest glue joint dry shear strength with all the three glues, urea-formaldehyde, phenol-resorcinol-formaldehyde and casein. In, on the other hand, which is ranked similar to padauk, shows a higher glue joint shear strength than padauk for urea-formaldehyde and phenol-resorcinol-formaldehyde. Thitya and kanyin are ranked similar in their pH values and have nearly the same glue joint shear strength with thitya having slightly lower strength than kanyin. Ingyin has a pH value not significantly different from that of kanyin. The glue joint strength of ingyin for all the three glues is not very different from that of thitya, both having low glue joint strength values. Stamm (1964) states that pH affects solubility and glue transfer and that a reduction in pH causes a reduction in solubility and glue transfer, which in turn affects the glue joint strength. Acidity or alkalinity of the wood have an effect in the setting properties of phenol-resorcinol and urea-formaldehyde resin adhesives (Chugg and Gray, 1965). Joints made with resorcinol resin adhesives display a marked change in strength with change of pH. For example, the strength of particle-board can be correlated with the pH of wood chips and in some cases acid in the wood is sufficient to cure the urea-resin adhesive (Kitahara and 78 Mizumo, 1961). Onishi and Goto (1971) report that there is a certain relationship between the pH of the wood and the gelation time of urea-formaldehyde resin adhesive. The gelation time decreases with the de-creasing hydrogen ion concentration of each wood. In addition to affecting glue setting properties, the pH of the wood affects the wettability of the wood. Sakuno and Goto (1970b) showed that pH had significant influence on the wettability of woods below 0.8 specific gravity but no effect on the wettability of woods above 0.8 specific gravity. Any factor which interferes with the wett-ability of the wood in the gluing process will affect the strength of the glue joint formed. On the other hand, Goto et a l . (1967) report that the relationship between glue joint strength and pH is not significant and pH has a less important effect on joint strength than specific gravity, wettability and extractive percent. The results of this study indicate that i f the influence of specific gravity is ignored, the trend is that of increase in glue joint shear strength with increase in pH of the wood. This is shown for phenol-resorcinol-formaldehyde, urea-formaldehyde resin adhesives and casein (Figures 14, 15 and 16). It is evident up to this point that specific gravity and pH of the wood have more effect on the gluing of wood than extractive content which so far has given inconsistent results. 79 4. Influence of Shrinkage on Shear Strength and Delamination The cyclic vacuum-pressure treatment is designed to have a devastating effect on the glue to wood bond. High moisture content imposes compressive stresses on the glue line and at the same time water may act as a plasticizer for the glue, making i t more flexible and more able to resist stress (Chugg and Gray, 1965). When the wood is dried below the moisture content at which the joint was made, the glue line tends to 'open up1 and, at the same time, the glue becomes brittle and tends to shrink, so that delamination is enhanced. Cycles of high and low moisture content have more serious degrading effect on the glue line than either condition applied alone for a long time. The extent of delamination and loss of strength of the glue bond wi l l , to a large extent, depend on the gluing characteristics of the species and its movement. 4.1 Shrinkage Tangential shrinkage in wood is more important than radial or longitudinal. Wood shrinks nearly twice as much in the tangential direc-tion than in the radial direction and therefore more stresses are intro-duced in the glue line from tangential shrinkage than from radial shrinkage i f the tangential surface forms the glue line. Longitudinal shrinkage of wood is so small as to be negligible. Tangential shrinkage values for the six species are shown in Table 12. The percentage of tangential shrinkage for the species in order of increasing magnitude is as follows: padauk 3.8 percent, ingyin 80 6.2 percent, pyinkado 6.5 percent, thitya 7.3 percent, in 9.4 percent and kanyin 11.2 percent (Table 12). Shear strengths for the vacuum-pressure cycle test in the same order is padauk 2,531 psi , ingyin 2,305 psi , pyinkado 2,235 psi , thitya 2,358 psi , in 1,693 psi and kanyin 1,677 psi. Statistical ranking by shear strength values (Table 4) puts padauk as having the highest shear strength followed by thitya, ingyin and pyinkado which are ranked similar. In and kanyin are ranked last and not significantly different. The general trend, therefore, is that species with a higher tangential shrinkage (in and kanyin) have less shear strength after the vacuum-pressure treatment than species with a low tangential shrinkage, for example padauk. Ingyin, pyinkado and thitya have intermediate tangential shrinkage and have nearly equal shear strength left after the cyclic vacuum-pressure treatment. For radial shrinkage, in shrinks 7.4 percent, kanyin 6.7 percent, thitya 5.7 percent, ingyin 5.3 percent, pyinkado 3.7 percent and padauk 2.2 percent. In spite of the small magnitude of shrinkage, the results follow essentially the same trend exhibitedbby tangential shrinkage. This is true also for the total volumetric shrinkage, but here the per-centage of shrinkage, as expected, is much higher than in tangential and radial. In and kanyin show the highest shrinkage values and therefore have more stresses introduced in the glue line as a result of changes in the moisture content of the wood in the vacuum-pressure test. They therefore have less residual shear strength than padauk which has the least shrinkage values in the group. Padauk has the highest residual 81 shear strength. Pyinkado, thitya and ingyin have intermediate shrinkage values and have nearly equal shear strength left after the cyclic vacuum^ pressure test because the glue line is subjected to less stresses than for in and kanyin (Table 4). A comparison of the dry glue line shear strength (Table 3) with shear strength of specimens which received the vacuum^pressure cyclic test shows there is a trend of reduced shear strength from dry to cyclic vacuum-pressure test. In and kanyin have their strength reduced from 2,719 psi to 1,693 and 2,571 psi to 1,677 psi , respectively. Pyinkado has its glue line strength reduced from 2,887 to 2,235 psi , padauk from 2,693 psi to 2,531 psi , ingyin from 2,426 psi to 2,305 psi and thitya from 2,437 psi to 2,358 psi. In and kanyin, with higher shrinkage per-cent, have lost more strength than the rest of the species, 1,126 and 894 psi , respectively. In and kanyin, the species with least amounts of extractives have high shrinkage percent from green to ovendiryy and also higher losses in strength in the cyclic vacuums-pressure test. Padauk has the highest percent of extractives and the least shrinkage. Part of the space in the fine structure of the cell wall is occupied by extractives and therefore these substances are, in effect, natural bulking agents which reduce the shrinkage on water removal or conversely, the degree of swelling on exposure to water or liquid. Their presence in cell walls means that the space available for water absorp-tion is reduced. (It must be pointed out, however, that not all extrac-tives are within the cell walls.) Padauk has shown in this study to 82 have a moderately low reduction in strength when comparing the dry test and the vacuum-pressure cyclic test. Pyinkado has medium tangential shrinkage (6.5percent) but has a high loss of strength from the dry test to the cyclic vacuum-pressure test (652 psi). On the other hand, thitya has nearly the same tangential shrinkage percent (6.2), but a low loss of shear strength (75 psi). Ingyin with 7.3 percent tangential shrinkage lost 121 psi. It is apparent, therefore, that the 'lighter' species studied show a high shrinkage percent and a high reduction in shear strength in the cyclic vacuum-pressure test. A high shrinkage percent may indicate a high loss in strength in the cyclic vacuum-pressure test. The picture is not very clear for the other four species which are high density species. The glue bond on all the six species s t i l l follows the same pattern as that for the dry test bond (Table 3). In and kanyin s t i l l show high wood failure percent, 91 and 93 percent, respectively. Pyinkado has 86 percent wood failure, padauk 86 percent, thitya 42 percent and ingyin 28 percent. This indicates that the reduction in strength has occurred mainly in the wood itself rather than the glue in in , kanyin, padauk and pyinkado as exhibited by high wood failure values. For thitya and ingyin the treatment seems to have had l i t t l e effect on the wood or the glue. 4.2 Delamination percent Results of the delamination test for phenol-resorcinol-formaldehyde resin adhesive are shown in Table 6. Statistical ranking of delamination 83 percent is as follows: kanyin, in and padauk are ranked similar and having less delamination than pyinkado, thitya and ingyin, which are also ranked not significantly different. In trying to relate percent tangential shrinkage and percent delamination i t is evident that there is no direct relationship between the two. The 'lighter' species, in and kanyin, with high tangential shrinkage (9.4 and 11.2 percent), have low delamination percent (8 percent). Padauk with tangential shrinkage of 3.8 percent has 11 percent delamination and is ranked as similar to in and kanyin. Pyinkado, ingyin and thitya having similar tangential shrinkage percent (6.5, 6.2 and 7.3 percent), are ranked as having delamination percent not significantly different (22, 24 and 28 percent). A general observation can be made that the 'lighter' species which glued very easily with phenol-resorcinol-resin adhesive, showed less delamination than the medium and high density hardwoods in spite of the fact that the 'lighter' species showed higher shrinkage from green to ovend-ryy. This observation is similar to the theory put forward by Perry (1953). He stated that wood of high specific gravity tends to decrease the l i fe of glue bonds at a faster rate than wood of lower specific gravity, because the expansive force under wetting increases with density. As a consequence, the maximum allowable thickness in high density species is distinctly less than in low density species. Northcott (1964) presented evidence in support of the theory and stated that other things being equal, wood of high specific gravity tends to degrade bonds in service faster than wood of low specific gravity. Higher delamination percent for pyinkado, ingyin and thitya (22, 24 and 28 percent, respectively) and low delamination percent for 84 in, kanyin and padauk, therefore, are reasonably close to the theory put forward by Perry (1953) and evidenced by Northcott (1958). Padauk, in and kanyin meet the requirements for the delamination test for structural laminating according to CS 253-63. Pyinkado, thitya and ingyin failed to meet the requirements. However, in view of the limited number of samples used in this study, it is recommended that further tests be carried out for these species before they are used for structural lamination, or rejected. 5. Comparison of Treatments If the dry shear strength developed with phenol-resorcinol-resin adhesive is taken as the control value, i t is observed that the boil test has markedly reduced the strengths of all the species moresoo than the vacuum-pressure cyclic and the cold soak test. Ingyin lost 53 percent of the dry strength, in 53 percent, kanyin 50 percent, pyinkado and thitya 30 percent each and padauk 19 percent. The loss in strength for the vacuum-pressure cyclic test is higher in the lower specific gravity species. In lost 37 percent of the dry shear strength, kanyin 34 percent, pyinkado 22 percent, padauk 6 percent, ingyin 4 percent, and thitya 3 percent. For the cold soak test the loss in strength is 26 percent for pyinkado, 25 percent for ingyin, 24 percent for kanyin, 22 percent for in and 21 percent for thitya. Padauk had slightly higher strength in the cold soak test than in the dry test. The wood failure values for all four treatments are high for all the species except for thitya and ingyin, indicating that there has not been a marked deterioration of the glue bond. The reduction in strength is therefore mainly a result of the wood itself . A moderate reduction in the cold soak is mainly due to an increase in moisture content of the wood at the time of testing. In the case of the vacuum-pressure treatment, the reduction in strength can be a result of the stresses induced in the glue line due to dimensional changes of the wood caused by changes in moisture content of the wood. These changes seem to have affected in and kanyin more than padauk, ingyin, thitya and pyinkado. In and kanyin show higher shrinkage than the other species with changes in moisture content. The reduction in strength during the.boil test is assumed to be a result of the decrease in strength of the wood, because of high moisture content at the time of testing and as a result of damage to the fibers because of hydrolysis. For ingyin and thitya, though there is a great reduction in shear strength, the wood failure percent is s t i l l low for both species. This indicates that the bond degradation took place either in the glue or at the wood-glue interface. The boil test and the cold soak test are normally used to evaluate performance of plywood bonds. Though the boil test was observed to be effective in the laminated material for this study, i t was observed that the cold soak test specimens sometimes had not been saturated com-pletely even after 48 hours. This was not observed for the lighter members of the group (in and kanyin). Probably this accounts for the 86 higher cold soak test shear strength value for padauk with phenol-resorcinol resin adhesive and also for ingyin, thitya and padauk when glued with urea-formaldehyde resin adhesive. The boil test gave the most drastic reduction in shear strength, partly because of the hydrolysis effect of boiling water on the wood fibers and also because of the high moisture content of the wood at the time of testing. The vacuum pressure test reduced the shear strength of in and kanyin, more than the other four species. The cold soak test reduced the strength of all species except padauk with phenol-resorcinol resin adhesive and showed higher values than the dry test for urea-formaldehyde resin adhesive for thitya, ingyin and padauk. This test was not designed for laminated members and its results are therefore unreliable. CONCLUSION 1. This study was conducted on a limited number of wood samples from Burma and the results therefore do not necessarily reflect the gluing characteristics of the species as a whole. General conclusions, however, can be made from the observations of the behaviour of wood samples tested. Kanyin is a medium density wood while the rest are high density woods. 2. In and kanyin (specific gravity 0.67 and 0.59, respectively) showed very high wood failure values (above 90 percent) and good shear strength properties with all three glues. Pyinkado (specific gravity 0.75) showed high shear strength with all the three glues. Pyinkado developed high wood failure percent with phenol-resorcinol-formaldehyde (81 percent) and urea-formaldehyde (86 percent), but very low wood failure with casein (22 percent). Padauk (specific gravity 0.77), developed high shear strength with casein (2,709 psi) but slightly lower shear strength values for its specific gravity with phenol-resorcinol-formaldehyde (2,693 psi) and urea-formaldehyde (2,448 psi). It developed high wood failure percent with both phenol-resorcinol-formaldehyde and urea-for-maldehyde (78 percent and 87 percent, respectively), but low wood failure with casein (67 percent). Ingyin and thitya (specific gravity 0.81 and 0.80, respectively), showed low shear strength and low wood failure per-cent with all three glues. 87 88 3. In and kanyin, which in practice are classed together as ex-cellent structural timbers, glued easily with phenol-resorcinol-formaldehyde, urea-formaldehyde and casein. Pyinkado and padauk glued moderately well with phenol-resorcinol-formaldehyde and urea-formaldehyde but poorly, i f both shear strength and wood failure are taken into account, with casein. Thitya and ingyin, which are classed together in the timber trade, glue poorly with phenol-resorcinol-formaldehyde, urea-formaldehyde and casein. It is recommended, therefore, that further studies be carried out for the gluing of thitya and ingyin with all three glues, and padauk and pyinkado with casein glue. 4. The influence of specific gravity on the gluability of the species varied with the glue used. Generally, the 'lighter' species (in and kanyin) glued easily with all three glues. The 'medium density' species (pyinkado and padauk) glued well with urea-formaldehyde and phenol-resorcinol-formalde-hyde but poorly with casein. The high density species (thitya and ingyin) glued poorly with all three glues. 5. In and kanyin showed low percentage of extractives soluble in ether (1.8 and 1.3 percent, respectively), ethanol (0.5 and 1.4 percent, respectively), hot water (0.5 and 0.4 percent, respectively), and acetone (2.5 and 2.2 percent, respectively). The species with 'medium' to high specific gravity had high percentages of extractives. Apart from the fact that in and kanyin had the least amounts of extractive content, and glued well with all three glues, there is no direct relationship between the amount of extractives and the gluability of the species. In general, the species with high extractive content have shown a 89 variation in their gluing characteristics, from moderately easy to glue for pyinkado and padauk, to difficult to glue for thitya and ingyin. 6. If the influence of specific gravity is excluded, a trend of increase in glue joint strength with increase in the pH of the wood was observed. The pH of the wood was from 4.47 to 5.09. 7. The species with high tangential shrinkage, in and kanyin (9.4 and 11.2 percent, respectively), showed a higher reduction in strength than the species with low tangential shrinkage. In, kanyin and pyinkado lost shear strength more (37 percent, 34 percent and 22 percent) than padauk (6 percent), ingyin (4 percent) and thitya (3 percent). The reduction in strength was a result of stresses induced by shrinkage due to moisture content changes. 8. High delamination percent was observed for pyinkado (22 percent), thitya (28 percent) and ingyin (34 percent), and low delamination for in (8 percent), kanyin (8 percent), and padauk (11 percent). This was in agreement with the theory that wood of high specific gravity tends to decrease the l i fe of glue bonds at a faster rate than wood of low specific gravity, because the expansive force under wetting increases with density. 9. Of the four treatments, the boil test was observed to reduce the shear strength of the wood most for all the species. The reduction in strength was 57 percent for ingyin, 53 percent for in , 50 percent for kanyin, 30 percent for pyinkado and thitya and 19 percent for padauk. The wood failure percent was high for all the species except thitya and ingyin. The reduction in strength was due to a high moisture content 90 of the wood at the time of testing and possibly a reduction in the strength of the fibers of the wood as a result of hydrolysis. 10. Padauk, in and kanyin met the requirements for exterior structural lamination with phenol-resorcinol-formaldehyde. Ingyin, thitya and pyinkado failed to meet the requirement. Further study is required before they can be recommended for exterior structural lamination with phenol-resorcinol-formaldehyde. 11. Padauk, pyinkado, ingyin and kanyin met the minimum requirements for interior structural lamination with urea-formaldehyde. Kanyin, in and padauk are considered suitable for interior structural lamination with casein. Further study is required for pyinkado, ingyin and thitya for gluing with casein. Thitya requires to be studied further before i t is used for gluing with urea-formaldehyde for interior structural lamination. REFERENCES American Society for Testing and Materials. 1972. Annual book of ASTM standards Part 16. Standard methods of testing small clear speci-mens of timber. ASTM D 143-52. Philadelphia, Pa. pp.64-69. . Standard method for preparation of extractive free wood. ASTM D 1105-56. Philadelphia, Pa. pp. 371-372. Bodig, J . 1962. Wettability related to gluabilities of five Philippine mahoganies. Forest Prod. J . 12(6):265-270. British Forest Products Research Laboratory. 1956. A handbook of hardwoods. H.M.S. Stationery Office. London. 269 pp. British Standards Institution. 1955. Nomenclature of commercial timbers including sources of supply. British Standard BS 881 and 589. British standards house, London. 143pp. Browning, B. L. 1967. Methods of wood chemistry. Vol. I. Interscience Publisher, a division of John Wiley and Sons. New York. pp. 75-89. Bryant, B. S. 1968. Studies in wood adhesion. Interaction of wood surface and adhesive variables. Forest Prod. J . 18(6):57-62. Canadian Standards Association. 1960. Definitions and standard test methods for wood adhesives. CSA 112.0-1960. Ottawa, Ont. 15 pp. . Specification for casein glues for wood. CSA 0112.3-1960. Ottawa, Ont. 8 pp. . Specification for urea resin adhesives for wood. Room and high temperature curing. CSA 0112.5-1960. Ottawa, Ont. 9 pp. . . Specification for phenol and resorcinol resin adhesives for wood. (Room and intermediate temperature curing.) CSA 0112.7-1960. Ottawa, Ont. 10 pp. . 1965. Qualification code for manufacturers of structural glued laminated timber. CSA 0177-1965. Ottawa, Ont. 22 pp. 91 92 Carstensen, J . P. 1961. Gluing characteristics of softwood veneers and secondary hardwoods. Forest Prod. J . 9(7):313-315. Chen, Chia-Ming. 1970. Effect of extractive removal on adhesion and wettability of some tropical woods. Forest Prod. J . 20(1):36-41. Chow, S-Z. 1971. Infrared spectral characteristics and surface inter-action of wood at high temperatures. Wood Science and Technology 5(1971):27-31. Chugg, W. A. and V. R. Gray. 1965. The effect of wood properties on strength of glued joints. Timber Research and Development Asso-ciation. United Kingdom. 25 pp. Dost, A. W. and C. Maxey. 1964. Gluing characteristics of some Cali-fornia hardwoods: Black oak, chinkapin, madrone and tanoak. Uni-versity of California School of Forestry, Forest Prod. Lab. s Berkeley, California. No. 36. 5 pp. Freeman, H. 1959. Relation between physical and chemical properties of wood and adhesion. Forest Prod. J . 9_(12) :451-458. . and F. F. Wangaard. 1960. Effect of wettability on glue line behaviour of two urea resins. Forest Prod. J . 10(6):311-315. Gary, V. R. 1962. The wettability of wood. Forest Prod. J . 12(9): 452-461. Goto, T . , T. Sakuno and H. Onishi. 1967. Studies on the wood gluing I. On the gluability of tropical woods part I. Shimane Agr. Coll. Matsue, Japan. Bull. No. 15(A):53-60. Hancock, W. V. 1964. The influence of native fatty acids on the for-mation of glue bonds with heat treated wood. Ph.D. thesis. Faculty of Forestry, University of British Columbia, Vancouver, B. C. 176 pp. Herczeg, A. 1965. Wettability of wood. Forest Prod. J . 1_5(11) :499-505. Hse, C. Y. 1972. Wettability of southern pine veneer by phenol-formaldehyde wood adhesives. Forest Prod. J . 22(1) :37-56. Huffman, J . B. 1961. Proceedings of the conference on theory of wood adhesion. Edited by A. A. Mara. Department of Wood Science and Technology. University of Michigan, pp 2:2:4:7. Imamura, H., T. Takahashi, M. Yasue, M. Yagishita, H. Karasawa and J . Kawamura. 1970. Effect of wood extractives on gluing and coating of kapur wood. Govt. For. Expt. Sta., Meguro, Tokyo, Japan. Bull, no. 232(1):65-96. 93 Isenberg, H. I., M. A. Buchanan and L. E. Wise. 1957. Extraneous com-ponents of American pulpwoods part (1). The importance of extraneous compounds. The Paper Industry 38(11):945-946 Kitahara, K.. and Y. Mizumo. 1961. Relationship between tree species and properties of particleboard I. Relationship between acidic substances of wood and delamination resistance. J . Japan Wood Res. Soc. 7:239-241. Kukachka, B. F. 1962. Characteristics of some imported woods. U. S. Dept. of A g r i c , Forest Service, Forest Prod. Lab. Madison. Rep. No. 2242, 46 pp. . 1970. Properties of imported tropical woods. U. S. Dept. of A g r i c , Forest Service. Forest Prod. Lab. Madison. Res. paper 125. 67 pp. Marian, J . E. and D. A. Stumbo. 1962. Adhesion in wood II. Physical-chemical surface phenomena and the thermodynamic approach to adhesion. Holzforschung 16^ (6) :168-180. Moriya, K., M. Nishihara and M. Sugano. 1968. Gluing faculties of laminated wood made of fourteen species of Kalimantan woods. Govt. For. Expt. Sta., Meguro, Tokyo, Japan. Bull no. 218:215-236. , M. Sugano and Y. Chiba. 1969. Properties of tropical woods 15. Gluing faculties of laminated wood made of keruing lumber grown in Malaya. Govt. For. Expt. Sta., Meguro, Tokyo, Jpan. Bull. no. 231: 45-53. . 1971. Gluing faculties of laminated wood made of red lauan sawn boards from the Philippines. Govt. For. Expt. Sta., Meguro, Tokyo, Japan. Bull. no. 234:94-104. Narayanamurti, D. 1957. The role of extractives in wood. Holz als Roh-unwerkstoff Bd 15.(1957) Heft 95, 370-380. , R .C- Gupta and G. M. Verma. 1962. Influence of extractives on the setting of adhesives. Holzforschung und Holzverwetung 14 (1962)5/6:85-88. Northcott, P. L. 1958. Specific gravity influences wood bond durability. Adhesives Age 17;October 34-36. . 1968. Wood species and glues influence plywood bond durability. Can. Dep. Forest. Rural Devel. Forest Prod. Lab., Vancouver. Infor. Rep. VP-X-42. 24 pp. Onishi, H. and T. Goto. 1971. Studies on wood gluing VIII. The effects of wood extractives on the gelation time of urea-formaldehyde resin adhesive. Shimane Agr. Coll. Matsue, Japan. Bull. no. 5:61-65. 94 Patton, T. C. 1970. A simplified review of adhesion theory based on surface energetics. Tappi 53:3:421-429. Perry, T. D. 1953. Does water proof glue always mean water proof ply-wood. Wood and Wood Prod. 58(4):32-34. Pringle, S. L. 1969. World supply and demand of hardwoods. Proceedings of conference on tropical hardwoods. State University College of Forestry, Syracuse University. Syracuse, N. Y. 46 pp. Rodger, A. 1936. A handbook of the forest products of Burma. Govt. Printing and Stationery, Rangoon, Burma. 166 pp. Sakuno, T. and T. Goto. 1970a. Studies on the wood gluing VI. On the wettability of tropical woods. Faculty of Agriculture, Shi mane University. Matsue, Japan. Bull. no. 4:97-102. . . 1970b. Studies on the wood gluing VII. On the gluability of tropical woods part (2). Faculty of Agriculture, Shimane Univer-sity. Matsue, Japan. Bull. no. 4:103-109. Sanderman, W. and H. H. Dietrichs. 1957. "Foreign substances" cause the specific nature of timbers. Die umschan in wissenschaff u Tech-nique 57(7): 197-200. Scharfetter, H. 1971. The effectoof contamination on the gluing of wood. South Africa Council for Scientific and Industrial Research Unit. Special report. Hont 37. 33 pp. Stadelman, R. C. 1966. Forests of South East Asia. Wimmer Brothers. Memphis, Tennessee. 245 pp. . 1969. Hardwood timber supply- Southeast Asia. Conference on tropical hardwoods. State University College of Forestry at Syracuse University, Syracuse, N. Y. 21 pp. Stamm, A. J . 1964. Wood and cellulose science. The Ronald Press Co., New York. 549 pp. Strickler, M. D. 1968. Adhesive durability: specimen designs for accel-erated tests. Forest Prod. J . 18(9):84-90. Timber Development Association. Siamese Timbers. Timber Information. London. No. 3. 3 pp. Titmuss, F. H. 1965. Commercial timbers of the world. 3rd edition. The Technical Press. London. 277 pp. Troop, B. S. and F. F. Wangaard. 1950. The gluing properties of certain Tropical American woods. Office of Naval Research, Yale University School of Forestry, New Haven, Connecticut. Techn. Rep. no. 4, 10 pp. 95 U.S. Dep. Commerce. 1963. Structural glued-laminated timber. Commer-cial Standard CS 253-63, Washington, D. C. 22 pp. Walpole, R. E. 1968. Introduction to Statistics. Edited by Carl B. Allendoerfer. Collier-Macmillan Ltd., London, pp. 301-302. Wangaard, F. F. 1966. Choice of species in the tropics in relation to world trends. Proceedings of Sixth World Forestry Congress. Madrid, Spain. 3_:3121-3126. Wood Handbook. 1955. U.S. Forest Prod. Lab. Handbook No. 72. Madison, Wis. 528 pp. Yagishita, M. and H. Karasawa. 1969. Adhesion faculty in veneers of fourteen species of Kalimantan woods. Govt. For. Expt. Sta., Meguro, Tokyo, Japan. No. 218:273-285. Yamagishi, Y. and H. Yoshihiro. 1972. Study on the gluability of tropical woods I. Mixed use of two different species in bonding some tropical woods. Japan Wood Ind., Wood Techn. Assoc. 27-11:542-546. TABLES 96 TABLE 1. Moisture content change of wood during storage Moisture Content (%) at Date of Measurement No. Species Oct. 11/72 Oct. 23/72 Nov. 3/72 Nov. 10/72 s i Pyinkado 26.4 13.3 13.2 9.5 S2 Thitya 29.1 16.2 12.4 10.5 S3 Ingyin 19.8 18.0 17.2 10.7 S4 Padauk 17.0 11.9 11.0 9.3 S5 In 22.3 14.2 11.2 10.0 S6 Kanyi n 23.5 13.7 11.5 8.3 Each value was determined by use of four specimens of 1 by 4 by 1 in. wood block with ovendryy weight method. ** Moisture content at the time of gluing. TABLE 2. S p e c i f i c g r a v i t y , wood s h e a r s t r e n g t h as t e s t e d and minimum a c c e p t a b l e g l u e l i n e s h e a r s t r e n g t h a t 10 p e r c e n t m o i s t u r e c o n t e n t * f o r the s i x Burmese woods Species No. S p e c i e s S p e c i f i c g r a v i t y OD wt/ green v o l . Bulk d e n s i t y l b / c u f t Shear s t r e n g t h p a r a l l e l to g r a i n p s i Minimum a c c e p t a b l e g l u e l i n e s t r e n g t h p s i s l P y i n kado 0.75 46.8 3,212 2,891 S 2 T h i t y a 0.80 49.9 3,005 2,704 S3 I n g y i n 0.81 50.5 2,919 2,627 S 4 Padauk 0.77 48.0 3,040 2,736 S5 In 0.67 41.8 2,649 2,384 S6 Kanyin 0.59 36.8 2,358 2,122 Average m o i s t u r e c o n t e n t o f wood a t the time o f t e s t i n g . Each v a l u e i s an average o f ten b l o c k s h e a r t e s t specimens; a d j u s t e d t o 10 p e r c e n t m o i s t u r e c o n t e n t . 98 TABLE 3. Shear strength and wood failure of wood bonded with phenol-resorcinol-formaldehyde adhesive Method: Dry test Strength (psi) Wood failure (%) No. S l S2 s 3 S 4 S 5 S ^6 Species Average Standard Dev. Average Standard Pyinkado 2,887 405 81 16 Thitya 2,437 295 57 21 Ingyin 2,426 308 21 24 Padauk 2,693 314 78 11 In 2,719 200 93 10 Kanyi n 2,571 303 94 10 Method: Cold soak Strength (psi) Wood failure (%) No. Species Average Standard Dev. Average Standard Dev. S, Pyinkado 2,111 1,083 74 25 sl Thitya 1,919 741 47 28 S^ Ingyin 1,818 812 10 16 S^ Padauk 2,780 416 84 15 Sj In 2,097 177 97 9 S^ Kanyin 1,945 269 100 1 Method: Boil Strength (psi) Wood failure (%) No. Species Average Standard Dev. Average Standard Dev S l Pyinkado 2,011 361 88 13 S1 Thitya 1,698 579 56 25 s Ingyin 1,039 478 14 17 S J Padauk 2,181 443 80 10 S 4 In 1,262 161 95 9 s 5 ^6 Kanyin 1,264 151 97 8 continued 99 TABLE 3 (continued) Method: Cyclic delamination Strength (psi) Wood failure {%) No. Species Average Standard Dev. Average Standard Dev. s l s? Pyinkado 2,235 227 86 12 Thitya 2,358 195 42 22 Ingyin 2,305 322 28 21 S d Padauk 2,531 330 83 11 s 4 In 1,693 207 91 13 s 5 Kanyin 1,677 336 93 9 1 Kg/cm2 = 14.25 psi 100 TABLE 4. Duncan's multiple range test for shear strength of laminated wood bonded with phenol-resorcinol-formaldehyde adhesive METHODS RANKING (psi) Dry test S l S5 S4 S6 S2 S3 2,887 2,719 2,693 2,571 2,437 2,426 Cold soak test S4 S l S5 S6 S2 S3 2,780 2,111 2,097 1,945 1,919 1,818 Boil test S4 S l S2 S6 S5 S3 2,181 2,011 1,698 1,264 1,262 1,039 Cyclic delamination S 4 S2 S3 S l S5 S6 2,531 2,358 2,305 2,235 1,693 1,677 Means not underscored by the same line are judged as different at the 5% level. 101 TABLE 5. Duncan's multiple range test for wood failure percent of wood bonded with phenol resorcinol-formaldehyde adhesive METHODS RANKING Dry test S6 S5 S l S4 S2 S o 94 93 81 78 57 2? Cold soak test S6 S5 S4 S l S2 S3 100 97 84 74 47 10 Boil test S6 S5 S"4 S2 S3 97 95 88 80 56 l i Cyclic delamination S6 S5 S l S4 S2 S3 93 91 86 83 42 28 Means not underscored by the same line are judged as different at the 5% level. 102 TABLE 6. Percent delamination for wood bonded with phenol-resorcinol-formaldehyde adhesive and statistical comparison Delamination Species % Standard Dev. s l Pyinkado 22 13.8 S2 Thitya 28 37.4 S3 Ingyin 34 26.3 S4 Padauk 11 10.3 S5 In 8 11.4 S6 Kanyin 8 13.2 * Statistical Comparison S6 S5 S4 S l S2 S3 8 8 11 22 28 34 * Means not underscored by the same line are judged as different at the 5% level. 103 TABLE 7. Shear strength and wood failure of wood bonded with urea-formaldehyde adhesive Method: Dry Strength (psi) ..Wood failure (%) No. Species Average Standard Dev. Average Standard S l Pyinkado 2,881 413 86 14 S2 Thitya 2,217 543 57 21 S3 Ingyin 2,320 322 58 28 S4 Padauk 2,448 257 87 11 S5 In 2,681 275 94 10 * Kanyin 2,330 280 98 7 Method : Cold Soak Strength (psi) Wood failure (%) No. Species Average Standard Dev. Average Standard Dev S l Pyinkado 2,758 349 88 10 S2 Thitya 2,389 916 62 27 S3 Ingyin 2,612 825 47 28 S4 Padauk 2,715 350 96 8 S5 In 2,137 248 93 9 S6 Kanyin 1,872 303 99 5 104 TABLE 8. Statistical ranking by Duncan's test for shear strength and wood failure percent for wood bonded with urea-formaldehyde adhesive Strength (psi) METHODS RANKING* Dry test S-j S^ S^ Sg S^ 2,881 2,681 2,448 2,330 2,320 2,217 Wood failure (%) METHODS Dry test Cold soak test Cold soak test S^  S^ Sg S^ Sg 2,758 2,715 2,612 2,389 2,137 1,872 S6 S5 S4 V S3 S2 98 94 87 86 58 57 S6 S4 S5 S l S2 S3 99 96 93 88 62 47 * Means not underscored by the same line are judged as different at the 5% level. Non-significant difference at 5% level On the border line of non-significant difference at 5% level. 105 TABLE 9. Shear strength and wood failure percent for wood bonded with casein glue Method: Dry Strength (psi) Wood failure (%) No. Species Average Standard Dev. Average Standard Dev. S l Pyinkado 2,884 348 22 28 S2 Thitya 2,251 251 11 13 S3 Ingyin 2,118 483 9 19 S4 Padauk 2,709 307 67 34 S5 In 2,455 484 81 13 S6 Kanyin 2,274 242 90 10 106 TABLE 10. Statistical ranking by Duncan's test for dry shear strength and wood failure percent of wood bonded with casein glue RANKING S l S4 S5 S6 S2 S3 2,884 2,709 2,455 2,274 2,251 2,118 Dry (wood failure) S g S 5 S^ S^  S 2 Sg 90 81 67 22 11 9 METHODS Dry (strength) Means not underscored by the same line are judged as different at the 5% level. Non-significant difference at 5% level. On the border of non-significant difference at 5% level. 107 TABLE 11. Comparison of shear strength and wood failure percent for four treatments and three adhesives (urea-formaldehyde, phenol-resorcinol-formaldehyde and casein) for the six Burmese woods Shear strength (psi) Test S l S2 S3 S4 S5 S6 Phenol- Dry 2,887 2,437 2,426 2,693 2,719 2,571 resorcinol- Cold soak 2,111 1,919 1,818 2,780 2,097 1,945 formaldehyde Vacuum-pressure 2,235 2,358 2,305 2,531 1,693 1,677 Boil 2,011 1,698 1,039 2,181 1,262 1,264 Urea- . Dry 2,881 2,217 2,320 2,448 2,681 2,330 formaldehyde Cold soak 2,758 2,389 2,612 2,715 2,137 1,872 Casein Dry 2,884 2,251 2,118 2,709 2,455 2,274 Wood failure percent Phenol- S l S2 S3 S4 S5 S6 resorcinol- Dry 81 57 21 78 93 94 formaldehyde Cold soak 74 47 10 84 97 100 Vacuum pressure 86 42 28 83 91 93 Boil 88 56 14 80 95 97 Urea- Dry 86 57 58 87 94 98 formaldehyde Cold soak 88 62 47 96 93 99 Casein Dry 22 11 9 67 81 90 108 TABLE 12. Specific gravity, shrinkage percent and green moisture content, as determined in the study of the six Burmese woods Specific Specific • Shrinkage (%) from Green to OD gravity* gravity** Species 1 2 Radial Tangential Vol. T/R MC(%) Pyinkado 0.75 1.16 3.7 6.5 10.4 1.75 53.9 Thitya 0.80 1.16 5.7 7.3 12.5 1.28 44.1 Ingyin 0.81 1.15 5.3 6.2 10.8 1.16 41.2 Padauk 0.77 1.14 2.2 3.8 6.7 1.72 47.7 In 0^ 67 1.08 7.4 9.4 16.2 1.27 62.2 Kanyin 0.59 1.06 6.7 11.2 17.2 1.67 76.0 Specific gravity based on ovendryy weight/green volume of unextracted wood. Specific gravity based on green weight/green volume. Moisture content taken after the wood had been submerged in water for five days at 120 psi. TABLE 13. Statistical ranking of specific gravity of the six Burmese woods RANKING Specific S3 S2 S4 S l S5 S6 gravity .81 .80 .77 .75 .67 .59 no TABLE 14. Average pH value for cold water extractions of the six Burmese woods Species Pyinkado Thitya Ingyin Padauk In Kanyi n DH 5.09 4.62 4.47 4.87 4.70 4.61 Each value is an average of three determinations. Ill TABLE 15. Statistical ranking by pH for cold water extractions of the six Burmese woods RANKING* S l S4 S5 S2 S6 S3 pH 5,09 4.87 4.70 4.62 4.61 4.47 Any two means not underscored by the same line are significantly different at 5 percent level. TABLE 16. Percentage of extractives soluble in ether, ethanol, hot water and acetone for the six Burmese woods Extracti ves Species Specific gravity OD/wt green volume Ether* soluble (%) * Ethanol soluble (%) * Hot water soluble (%) Total* Acetone soluble (%) Pyinkado 0.75 3.4 5.2 3.9 12.5 8.4 Thitya 0.80 1.2 7.3 1.8 10.3 6.9 Ingyin 0.81 1.8 6.4 1.1 9.3 7.4 Padauk 0.77 2.5 9.6 2.1 14.2 9.5 In 0.<6:7 1.8 0.5 0.5 2.8 2.5 Kanyin 0.59 1.3 1.4 0.4 3.1 2.2 Each value is an average of three successive extractions. ** Each value is an average of three single extractions. TABLE 17. Statistical ranking by percentage of extractives soluble in ether, ethanol, water and acetone RANKING* S l S4 S5 S3 S6 S2 Ether soluble extractives 3.4 2.5 1.8 1.8 1.3 S4 2 3 1 6 Ethanol soluble extractives 9.6 7.3 6.4 5.2 1.4 0 S l S4 S2 S3 S5 S Hot water soluble extractives 3.9 2.1 1.8 1.1 0.5 0 S4 S l S3 S2 S5 S Acetone soluble extractives 9.5 8.4 7.4 6.9 2.5 2 Any two means not underscored by the same line are considered different at 5 percent level. FIGURES 114 Figure 1. Form and dimension of block shear test specimen Figure 2. Shearing tool r- 100 E3 Pyinkado F i g u r e 3. Thi t y a I n g y i n Padauk In Kanyin Dry t e s t s h e a r s t r e n g t h and p e r c e n t wood f a i l u r e f o r wood bonded w i t h p h e n o l - r e s o r c i n o l - f o r m a l d e h y d e r e s i n a d h e s i v e - 80 01 i 60 = •o o o _ 40 . 20 Shear S t r e n g t h Wood F a i l u r e 3000-1 2 0 0 0 -AC -t-> O) c 0) s_ +J oo XI E o CQ 1 0 0 0 -Pyinkado Thitya Ingyin Padauk r loo 80 Shear Strength Wood Failure -60 CD ~3 .40 o o -20 In Kanyin Figure 4. Cold soak shear strength and percent wood failure for wood glued with phenol-resorcinol-formaldehyde resin adhesive 3000 -r 2000 in Q. + J CT) O) S-+J oo O CQ 1 0 0 0 -_ 80 _ 60 • 100 Shear S t r e n g t h Wood F a i l u r e 40 £ o o 20 Pyinkado Figure 5. Thitya Ingyin Padauk In Kanyi n Boil test shear strength and percent wood failure for wood bonded with phenol-resorcinol-formaldehyde' resin adhesive' CO 3000 200O. c/i +-> CO CD S_ +-> oo -o c: o CO 1000-Pyinkado T h i t y a I n g y i n Padauk In Kanyin 1-100 - 80 60 ^ 5 CD 40 1 20 a m S h e a r S t r e n g Wood F a i l u r F i g u r e 6. Vacuum-pressure c y c l i c t e s t s h e a r s t r e n g t h and p e r c e n t wood f a i l u r e f o r wood g l u e d w i t h p h e n o l - r e s o r c i n o l -formaldehyde r e s i n a d h e s i v e 3000 H00 " 8 0 She a r S t r e n g t h ~ 2 0 0 0 -in -M CD C CU S-4-> 00 1 •a c o CQ - 6 0 - 4 0 &9 -a o o Wood F a i l u r e 1000 _ 20 Pyinkado T h i t y a I n g y i n Padauk In Kanyin F i g u r e 7. Dry t e s t s h e a r s t r e n g t h and wood f a i l u r e p e r c e n t f o r wood g l u e d w i t h u r e a - f o r m a l d e h y d e r e s i n a d h e s i v e ro o 3000 n 2000 to C L 4-> C D CD 5-+-> LO ft "looo CQ r-l 00' Byinkado hi t y a i n g y i n Padauk Kanyi n 80 •60 S h e a r S t r e n g t l Wood F a i l u r e CD 3 40 1 20 F i g u r e 8. C o l d soak s h e a r s t r e n g t h and wood f a i l u r e p e r c e n t f o r wood g l u e d w i t h u r e a - f o r m a l d e h y d e r e s i n a d h e s i v e , r o 3000 T -100 "2000-to p-+-> CO c: OJ %-+-> C O -o o C Q 1000. Pyinkado hi.tya I n g y i n Padauk Tn Kanyin 80 - 60 3 - 40 o o 20 Shear S t r e n g t h Wood F a i l u r e F i g u r e 9. Dry t e s t s h e a r s t r e n g t h and p e r c e n t wood f a i l u r e f o r wood bonded with' c a s e i n g l u e ro ro 3000 ~ PR UF Cas 2000 -C L •)-> CD CJ i-OO T 3 E O CQ 1000 PR P^  up PR PR UF Cas •4-4 UF Cas UF PR Cas UF Cas CAS - Casein glue joint strength UF - Urea-formal dehyde resin adhesive glue joint strength PR - Phenol-resorcinol-formaldehyde resin adhesive glue joint strength Pyinkado Thitya Ingyin Padauk In Kanyi n Figure 10. Phenol-resorcinol-formaldehyde, urea-formaldehyde and casein dry test shear strength ro oo 3000 - i o Pyinkado 2800 ~ In o Padauk 2600 -to D-sz +-> CD c O Kanyin £2400 T h i t y a ° Q I n g y i n ca cu </) CD ^2200 -CD 3 C5 2000 - T 1 i ; — 0.60 0.65 0.70 S p e c i f i c g r a v i t y 0.55 0T75 0.80 0.85 F i g u r e 11. R e l a t i o n s h i p between s p e c i f i c g r a v i t y and g l u e l i n e dry s h e a r s t r e n g t h for.wood g l u e d w i t h p h e n o l - r e s o r c i n o l -formaldehyde r e s i n a d h e s i v e ' ' 1 ro 3000-r 0 P y i n k a d o 2800-0 In 2600 Q Padauk 2400-Kanyi n 2200 -^ I n g y i n 0 T h i t y a 2000 0.55 0.60 "• 0.65 0.70 S p e c i f i c g r a v i t y —-3— 0.75 0.80 0.85 F i g u r e 12. R e l a t i o n s h i p between s p e c i f i c g r a v i t y and g l u e l i n e dry s h e a r s t r e n g t h f o r wood g l u e d w i t h u r e a - f o r m a l d e h y d e r e s i n a d h e s i v e O In O Pyinkado 0Padauk © Kanyin O Thitya o Ingyin T T 0.55 . 0.50 0.65 ' 0.70 Specific gravity 0.75 0.80 ' 0.85 Figure 13. Relationship between specific gravity and glue line dry shear strength for wood glued with casein glue 3000 -2800 -in g>2600 cu s_ S-n3 CU 2400 _ cu 2200 2000 Pyinkado O Padauk <? I n g y i n O Kanyin T h i t y a 4.40 4.50 4.60 4.70 4.80 4.90 —C— 5.00 5.10 PH F i g u r e 14. R e l a t i o n s h i p between pH o f wood and g l u e l i n e d ry s h e a r s t r e n g t h f o r wood g l u e d w i t h p h e n o l - r e s o r c i n o l - f o r m a l d e h y d e r e s i n a d h e s i v e ro 3000 0 P y i n k a d o 2800 -O In to a. cn E O) i_ +-> to ro CD (U E QJ 3 CJ3 2600 -Padauk 2400 2200 I n g y i n © Kanyin O T h i t y a 2000 4.40 4.50 T T 4.60 4.70 4.8.0 . PH 4.90 5.00 — i 5.10 F i g u r e 15. R e l a t i o n s h i p between pH o f wood and g l u e l i n e d r y s h e a r s t r e n g t h f o r wood g l u e d w i t h u r e a - f o r m a l d e h y d e r e s i n a d h e s i v e ro co 3000 O Pyinkado 2800 -O Padauk £ 2600 O In 2400 _ ^Kanyin O Thitya 2200 ~ O Ingyin 2000 4.40 4.50 4.60 4.70 4.80 4.90 5J0O 5.10 PH Figure 16. Relationship between pH of wood and glue line dry shear strength for wood glued with casein glue ro to APPENDIX APPENDIX I I. Species Description A. Pyinkado B. Thitya C. Ingyin D. Padauk E. In F. Kanyin 130 I. Species Description A. Pyinkado Xylia dolabriformis Benth. Family Leguminosae Pyinkado is one of the most useful of the structural woods of India and Burma (Titmuss, 1965). The timber is heavy, i t weighss from 55 to 65 lb per cu ft at 15 percent moisture.content and 72 lb per cu ft when green (44 percent moisture content) (Bri!ti!s!hoB6rofPHoddivbab's, 1956). It has a mild, non-distinctive smell, a grain that is shallowly inter-locked, with a moderately coarse but uniform texture. When freshly ' sawn the heartwood colour is of a red-brown shade, but this darkens after exposure to the air and the heartwood is then readily distinguishable from the thin brownish coloured sapwood. The planed surfaces of the wood have an oily appearance. Pyinkado is a difficult timber to work, either by hand or machine. It can be brought to a good surface and will respond satisfactorily to the normal finishing agents; some care is needed in mortising and similar operations to avoid chipping out. Durability factors are high, and pyinkado is said to be a good species for use under water, as in piling; i t is classified as extremely diff icult to treat with preservative fluids. Although i t dries slowly, pyinkado can be seasoned satisfactorily either by air or kiln methods. Uses The wood is considered unsuitable for plywood manufacture because of its weight. In Burma and India pyinkado is ranked as one of the world's 131 132 best timbers for cross ties and also for bridges.. It is suitable for heavy structural work, piles,bridge girders, decking, dockwork and flooring. It is suited to these uses because of its great strength and its resis-tance to decay and termites. Macroscopic identification Pyinkado has a diffuse porous structure. Its growth rings are normally quite visible to the naked eye. It has moderately numerous pores which are arranged in clusters and individually. Orange-coloured gummy deposits may be present in the pores. It has vessel lines that are clearly visible to the naked eye. Numerous rays which are only visible under a lens on end section are present. V,Vasicentric parenchyma is visible under a lens, while fine lines of tissue may also be present. Ripple marks are not present in the species. B. Thitya Shorea obtusa Wall. Dipterocarpaceaefamily The species is practically identical to Indian sal (Shorea robusta Gaertn.). It is one of the heavier Shorea species, weighing 50 to 60 lb per cubic foot air dry (Titmuss, 1965). The species grows very commonly in association with ingyin (Pentacme  siamensis Kurz.) and the two are used interchangeably in Burma. It has interlocked grain and a medium coarse texture. The heartwood of the species is a dark red-brown and quarter-sawn stock may show a striped figuring. The wood has a rather large number of resin canals in its structure, and these may result in splits during seasoning, an operation in which the wood also shows a tendency to warp. 133 Thitya is not an easy timber to work by hand, but does not cause trouble in machining. It is a very strong timber and is eminently suited for use for structural purposes, and also finds employment for flooring, bridging and similar work. It is also used for cross ties. Macroscopic identification The wood is diffuse porous with growth rings which are not very distinct. It has few medium-sized pores, uniformly distributed, which can be seen with the naked eye. Some of the pores are solitary but radial oblique clusters and pairs also occur. Vessel lines can be seen with the unaided eye. Its rays are moderately fine and are not always clearly distinct to the naked eye on end section. VMasicentric parenchyma is present in variable quantity, sometimes showing a tendency to become aliform. Narrow metatracheal lines of tissue may also be present. Small vertical canals with whitish deposits can be seen on all surfaces. In-distinct ripple marks may be seen sometimes. C. Ingyin Pentacme siamensis K'tirz. Dipterocarpaceae fami ly The wood of this species is hard, heavy and weighs 58 to 64 lb per cu ft air dry. The sapwood is greyish white and the heartwood is yellowish brown to red brown. The texture is medium to coarse and the grain is mostly interlocked. The wood seasons fairly well but slowly and is not liable to split and crack. It is very durable. It saws and machines with l i t t le difficulty especially in green condition. It finishes well. 134 Ingyin is used for floors, s i l l s , heavy construction, cross ties, tool handles and agricultural implements (Rodger, 1936). Macroscopic features of ingyin are similar to those of thitya. D. Padauk Pterocarpus macrocarpus Kurz. Leguminosae family Padauk is a fairly abundant tree and, next to teak, is the most valued uti l i ty timber in Burmese domestic usage. The wood weighs 50 to 60 lb per cu ft air dry. It is heavier and stronger than teak. It has very l i t t le sapwood which is greyish white. The heartwood is bright red to dark brick red with darker lines. The wood is hard and close grained. The interlocked grain produces a narrow ribbon grain figure. The growing tree is often 'girdled' as a pre-seasoning treatment, but during the drying operation surface cheeks are very likely to develop in the timber. The wood saws and works with some difficulty especially when dry, whilst the sawdust also irritates the nose and eyes. Generally i t can be clearly finished. It turns well, holds screws firmly but needs boring i f nailed. It polishes well when the grain is properly f i l led . Padauk is naturally durable for use in positions unprotected from the weather, and is especially resistant to "white ant" attack. It is diff icult to impregnate with wood preservatives. The wood is used for heavy construction and building purposes especially panelling, flooring beams, furniture. In the United States i t is used either in solid form or veneer in the manufacture of decorative panelling (Kukachka, 1962). It is said to be an important cabinet wood in Europe. 135 Padauk has a diffuse porous structure with growth rings distinct to the naked eye. Its largest vessels are clearly visible to the naked eye, but others are distinct under a lens. The vessels are solitary or in radial groups with open single perforation plates. Vessel lines are distinct to the naked eye. It has abundant metatracheal type of wood parenchyma. Paratracheal type parenchyma is also present. Ripple marks are present but are not usually visible to the naked eye; they are distinct under a hand lens. E. In Dipterocarpus tuberculatus Roxb. Dipterocarpaceaefamily In is a timber of Burmese origin but very similar woods (also produced by species of Dipterocarpus) are available from Malaya under such names as k'eruing and gurjun. It is appreciably denser than kleruing and gurjun. The wood weighs 51 to 54 lb per cu ft air dry. The moderately well defined sapwood of in is brownish-grey, while the heartwood is brownish-red; the latter may be of an extremely dark shade after prolonged exposure to the atmosphere. The grain is either straight or shallowly interlocked, the texture open but uniform, and the planed surfaces dull. It is not an easy wood to work with and resin exudation may cause some trouble in sawing or in the polishing process. The strength properties are good, the resistance to fungal infection is reasonable, but some insect damage may occur, though this is largely confined to the sapwood. In exposed conditions the wood lasts4 to 5 years. The response to preservative treatment is fairly satisfactory. It seasons 136 slowly, with some tendency to degrade, and is considered to be rather below average in respect of stability after seasoning. Air seasoning before kiln seasoning is recommended. The wood is used for heavy construction, railway wagon construction, house construction, Yarns'-, ~rafters~F'^ agricul-tural;1 i'mpTemeritsr: cultural ir:i"\n^i-s. Macroscopic identification The species is diffuse porous. The growth rings cannot be recog-nised. The pores are very large or large, quite distinct to the naked eye, evenly distributed and numerous. They are mostly solitary in arrange-ment but occasional pairs are present; some tyloses may be present. The coarse vessel lines can be seen with the naked eye. Rays are moderately few, fine and usually just visible to the naked eye on end section. Narrow wasicentric parenchyma (sometimes with a slight tendency to become aliform) and thin broken lines of metatracheal parenchyma may be seen under a hand lens. Ripple marks do not occur in the species but resin canals are present. F. Kanyin Dipterocarpus spp. The genus Dipterocarpus comprises a large number of species. The Dipterocarps are diff icult to separate both as standing trees and as sawn lumber (Stadelman, 1966). The wood weighs 40 to 49 lb per cu ft air dry. The sapwood is greyish-white and the heartwood is reddish-brown, dull with rough feel, even and coarse textured with characteristic odour. 137 It seasons moderately well but is liable to split i f left in large dimensions. Distortion, especially in the form of cupping is often considerable and slight collapse may occur. The wood is not resis-tant to termite attack or outside weathering. It is easy to impregnate with wood preservatives. It saws and machines well but does not take a good polish. It is used for railway carriage and wagon construction, bridge decking, boats, rough construction, house construction, rafters, beams and flooring. Macroscopic identification The species is diffuse porous. The growth rings cannot be recog-nised. The pores are very large or large, quite distinct to the naked eye, evenly distributed and numerous. They are mostly solitary in arrange-ment but occasional pairs are present; some tyloses may be present. The coarse vessel lines can be seen with the naked eye. Rays are moderately few, fine and usually just visible to the naked eye on end section. Narrow was:!'centric parenchyma (sometimes with a slight tendency to become aliform) and thin broken lines of metatracheal parenchyma may be seen under a hand lens. Ripple marks do not occur in the species but resin canals are present. APPENDIX II List of Abbreviations bd. ft. - board feet gr. vol. - green volume ha. - hectare hr - hour 3 lb/ft - pound per cubic foot m^  - cubic meter mc - moisture content ml - mi l l i l i ter od wt - oven dry weight psi - . pound per square inch r - roundwood' sp.gr. - specific gravity T/R - tangential/radial 138 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0075322/manifest

Comment

Related Items