Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Particle moisture content effects on the physical and mechanical properties of magnesite cement-bonded… Musokotwane, India E. O. 1982

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_1982_A6_7 M88.pdf [ 8.55MB ]
Metadata
JSON: 831-1.0095534.json
JSON-LD: 831-1.0095534-ld.json
RDF/XML (Pretty): 831-1.0095534-rdf.xml
RDF/JSON: 831-1.0095534-rdf.json
Turtle: 831-1.0095534-turtle.txt
N-Triples: 831-1.0095534-rdf-ntriples.txt
Original Record: 831-1.0095534-source.json
Full Text
831-1.0095534-fulltext.txt
Citation
831-1.0095534.ris

Full Text

PARTICLE MOISTURE CONTENT EFFECTS ON THE PHYSICAL AND MECHANICAL PROPERTIES OF MAGNESITE CEMENT-BONDED PARTICLEBOARD by INDIA E.O. MUSOKOTWANE B.Sc. (Hons.) Forestry, Makerere University, 1975 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE i n THE FACULTY OF GRADUATE STUDIES (Department of Forestry) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July 1982 © India E.O. Musokotwane-, 1982 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of FORESTRY  The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 D a t e S e P t e m b e r >15th 1982 DE-6 (3/81) i i ABSTRACT The e f f e c t s of i n i t i a l p a r t i c l e moisture content, wood-cement r a t i o and density on physical (thickness swelling and water absorption) and mechanical properties (MOE, MOR, IB and edgewise compression) were investigated. Five i n i t i a l p a r t i c l e moisture content l e v e l s - 0-6%, 8-15%, 25-30%, 40-50% and 60-80%; three wood-cement r a t i o s - 1:1, 1:1.5 and 1:2; and three density l e v e l s at each wood-cement r a t i o - 1:1 -0.472 g/cm3, 0.528 g/cm3 and 0.622 g/cm3, 1:1.5 - 0.636 g/cm3, 0.707 g/cm3 and 0.809 g/cm3; and 1:2 - 0.763 g/cm3, 0.847 g/cm3 and 0.939 g/cm were used. Combinations of the above variables gave 45 treatments. Three r e p l i c a t e boards were made for each treatment thus giving a t o t a l of 135 panels f o r the study. A t o t a l of 135 test specimens were used for each property tested. Results from the tests were compared to the German and ISO Standards for s i m i l a r boards and to the Canadian Waferboard Standard. I n i t i a l p a r t i c l e moisture content was highly s i g n i f i c a n t i n the development of physical and mechanical properties of magnesite cement-bonded particleboard. Increasing i n i t i a l p a r t i c l e moisture content from 0-6% to 60-80% resulted i n the reduction of the physical and mechanical properties of the boards. The highest i n i t i a l p a r t i c l e moisture content of (60-80%) yielded the lowest physical and mechanical properties. For manufacture of boards of favourable mechanical properties, an i n i t i a l p a r t i c l e moisture content of not more than 15% i s recommended. On the other hand, a higher i n i t i a l p a r t i c l e moisture content (>40%) i s considered desirable i f board thickness and water absorption are to be minimized. i i i A l l the mechanical properties tested consistently increased by increasing wood-cement r a t i o and density and were highest at 1:2 wood-cement r a t i o and density l e v e l 3 of each wood-cement r a t i o . Thickness swelling and water absorption were cons i s t e n t l y reduced by increasing wood-cement r a t i o and density. In both physical properties t e s t s , the 1:2 wood-cement r a t i o and density l e v e l 3 yielded the lowest values. Thirty-two of the f o r t y - f i v e treatment combinations of i n i t i a l p a r t i c l e moisture content, density and wood-cement r a t i o pass the MOE requirement of the German Standard DIN 52 362 f o r Portland cement-bonded particleboard; forty-one treatments met the minimum MOE Canadian Waferboard Standard requirements, while no treatment meet the MOR requirements f o r t h i s Standard. Eleven of the f o r t y - f i v e treatments met the minimum IB Canadian Waferboard Standard requirements. A l l the 45 treatments pass the ISO bui l d i n g board requirements i n thickness swelling, while 18 treatments pass the water absorption requirements for th i s Standard. Most of the treatment combinations compare favourably with r e s u l t s obtained i n tests conducted i n Europe for cement-bonded particleboard. i v TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS i v LIST OF TABLES v i i i LIST OF FIGURES x LIST OF APPENDICES x i i i GLOSSARY x i v ACKNOWLEDGEMENTS xv 1.0 INTRODUCTION 1 1.1 Background Information 1 1.1.1 The Paszner Process 5 1.1.2 The Study 7 1.2 Objective and Scope of Study 7 2.0 LITERATURE REVIEW .' 10 2.1 Raw Materials 10 2.1.1 Wood 10 2.1.1.1 Physical C h a r a c t e r i s t i c s of Wood .... 11 2.1.1.2 E f f e c t of Chemical C h a r a c t e r i s t i c s of Wood on Cement Setting 12 2.1.1.2.1 Wood Cellulose 14 2.1.1.2.2 L i g n i n 14 2.1.1.2.3 Wood Extractives 14 2.1.1.3 E f f e c t of Decay and Stain Fungi 16 2.1.2 Binders 18 2.1.2.1 Cements 18 2.1.2.2 Other Binders 21 V Page 2.1.3 Pretreatment of Wood 22 2.1.3.1 General 22 2.1.3.2 B i o l o g i c a l Treatments 22 2.1.3.3 Hot Water E x t r a c t i o n 23 2.1.3.4 Chemical Pretreatments 24 3.0 MATERIALS AND METHODS 26 3.1 Experimental Design 26 3.2 Preparation of Materials 27 3.2.1 Wood P a r t i c l e s Description 27 3.2.2 Magnesite Cement Description 27 3.2.3 Ammonium Polyphosphate (Reactant) Description .. 28 3.2.4 Mold 28 3.2.5 Press 28 3.2.6 Manufacture of Magnesite Cement-Bonded P a r t i c l e -board 29 3.2.6.1 Wood P a r t i c l e Moisture Content Adjustment 29 3.2.6.2 Mixing of Wood P a r t i c l e s with Reactant and Cement 30 3.2.6.2.1 The Reaction C h a r a c t e r i s t i c s 30 3.2.6.3 Pressing Time and Control of Temperature 31 3.2.6.4 Pressure 31 3.2.6.5 Board Conditioning 31 v i Page 3.2.7 Test Specimen Preparation 32 3.2.7.1 Cutting of Specimens 32 3.2.7.2 Cutting Plan for Test Specimens 32 3.2.8 Board Testing Procedures 33 3.2.9 Board Characterization 33 3.2.9.1 Moisture Content and Density 33 3.3 S t a t i s t i c a l Analysis 34 3.4 Scanning E l e c t r o n Microscopy (SEM) Study 34 4.0 RESULTS AND DISCUSSION 36 4.1 Mechanical Properties 36 4.1.1 Modulus of E l a s t i c i t y (MOE) 36 4.1.2 Modulus of Rupture (MOR) 38 4.1.3 Internal Bond Strength (IB) 39 4.1.4 Compressive Strength P a r a l l e l to Surface (Edgewise Compression) 41 4.2 Physical Properties 43 4.2.1 Moisture Content and Density of the Boards at Test 43 4.2.2 Thickness Swelling 44 4.2.2.1 Thickness Swelling from 50% R.H. to 2 h Cold Soaking 44 4.2.2.2 Thickness Swelling from 50% R.H. to 24 h Cold Soaking 45 4.2.3 Water Absorption 47 4.2.3.1 Water Absorption from 50% R.H. to 2 h Cold Soaking 47 v i i Page 4.2.3.2 Water Absorption from 50% R.H. to 24 h Cold Soaking 48 4.3 Scanning Electron Microscopy (SEM) Observations 49 4.4 Probable Factors Accounting f o r Trends i n Treatments 51 4.4.1 Mechanical Properties 51 4.4.2 Physical Properties 55 4.5 Comparison of Study Results With Some National Standards and Other Tests 57 4.5.1 German Standard for Cement-Bonded Wood Particleboard 57 4.5.2 Canadian Standard for Waferboard 59 4.5.3 International Organization for Standards (ISO) 60 4.5.4 General 61 5.0 SUMMARY AND SUGGESTIONS FOR FURTHER STUDY 63 5.1 Summary 63 5.2 Suggestions for Further Study 65 6.0 CONCLUSION 66 7.0 REFERENCES 67 v i i i LIST OF TABLES Table Page 1 Some dis t i n g u i s h i n g features of mineral-bonded wood composites. 72 2 Comparison of raw material, processing features, and mechanical properties of Portland cement and magnesite-bonded particleboards. 73 3. E f f e c t of va r i a t i o n s i n ambient temperature on onset of g e l l i n g and development of a temperature maximum for 75:25 mixture of s i l i c a sand and dead-burnt magnesite cement. 74 4 Wood raw materials for production of wood-wool-cement boards. • 75 5 Chemical components of the c e l l wall substance i n normal wood. 76 6 P r i n c i p a l compounds i n Portland cement. 77 7 The behaviour of p r i n c i p a l components which occur i n Portland cement. 78 8 Int e r a c t i o n of p a r t i c l e moisture content, density and wood-cement r a t i o on modulus of e l a s t i c i t y . 79 9 Interaction of p a r t i c l e moisture content, density, and wood-cement r a t i o on modulus of rupture. 80 10 Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on i n t e r n a l bond strength. 81 11 Interaction of p a r t i c l e moisture content, density, and wood-cement r a t i o on compressive strength p a r a l l e l to surface. 82 12 Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on board moisture content at te s t . 83 13 Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on thickness swelling a f t e r 2 h cold soaking. 84 14 Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on thickness swelling a f t e r 24 h cold soaking. 85 ix Table Page 15 Interaction of p a r t i c l e moisture content, density, and wood-cement r a t i o on water absorption after 2 h cold soaking. 86 16 Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on water absorption a f t e r 24 h cold soaking. 87 17 Analysis of variance for testing the e f f e c t s of p a r t i c l e moisture content, density and wood-cement r a t i o on MOR and MOE i n cement-bonded p a r t i c l e board. 88 18 Analysis of variance for testing the e f f e c t s of p a r t i c l e moisture content, density and wood-cement r a t i o on i n t e r n a l bond strength and compressive strength p a r a l l e l to surface i n cement-bonded particleboard. 89 19 Analysis of variance for testing the e f f e c t s of p a r t i c l e moisture content, density and wood-cement r a t i o on thickness swelling a f t e r 2 h and 24 h cold soaking. 90 20 Analysis of variance f o r t e s t i n g the e f f e c t s of p a r t i c l e moisture content, density and wood-cement r a t i o on water absorption a f t e r 2 h and 24 h cold soaking. 91 21 Interaction of p a r t i c l e moisture content, density l e v e l and wood-cement r a t i o on board density at t e s t . 92 22 Deta i l s of curve f i t t i n g of MOE and MOR. 93 23 D e t a i l s of curve f i t t i n g of edgewise compression and IB strength. 94 X LIST OF FIGURES Figure Page 1 Exotherm c h a r a c t e r i s t i c s of ammonium polyphosphate activated magnesium oxide cements. 95 2 Parts of the mold. 96 3. Dimension of mold. 97 4 Single opening press. 98 5 L e v e l l e r . 99 6 Dependence of modulus of rupture on density and wood-cement r a t i o i n t e r a c t i o n . 100 7 Dependence of modulus of rupture on p a r t i c l e moisture content and wood-cement i n t e r a c t i o n . 101 8 Dependence of modulus of e l a s t i c i t y on p a r t i c u l e moisture content and density i n t e r a c t i o n (at 1:1 wood-cement r a t i o ) . (Wood cement r a t i o s 1:1.5 and 1:2 102 are held constant.) 9 Dependence of modulus of e l a s t i c i t y on p a r t i c l e moisture content and density i n t e r a c t i o n (at 1:1.5 wood-cement r a t i o ) . (Wood-cement r a t i o s 1:1 and 1:2 are held constant). 103 10 Dependence of modulus of e l a s t i c i t y on p a r t i c l e moisture content and density i n t e r a c t i o n (at 1:2 wood cement r a t i o ) . (Wood-cement r a t i o s 1:1 and 1:1.5 are held constant). 104 11 Dependence of t e n s i l e strength perpendicular to surface IB on density and wood-cement r a t i o i n t e r a c t i o n . 105 12 Dependence of t e n s i l e strength perpendicular to surface on p a r t i c l e moisture content and wood-cement r a t i o i n t e r a c t i o n . 106 13 Dependence of compressive strength p a r a l l e l to surface on p a r t i c l e moisture content and wood-cement r a t i o i n t e r a c t i o n . 107 14 Dependence of compressive strength p a r a l l e l to surface on density and wood-cement r a t i o i n t e r a c t i o n . 108 x i Figure Page 15 Dependence of thickness swelling ( a f t e r 2 h cold soaking) on wood-cement r a t i o and p a r t i c l e moisture content i n t e r a c t i o n . 109 16 Dependence of thickness swelling ( a f t e r 24 h cold soaking) on p a r t i c l e moisture content and wood-cement r a t i o i n t e r a c t i o n . 110 17 Dependence of water absorption ( a f t e r 2 h cold soaking) on density and wood cement r a t i o i n t e r a c t i o n . I l l 18 Dependence of water absorption (a f t e r 2 h cold soaking) on p a r t i c l e moisture content and wood-cement r a t i o 112 i n t e r a c t i o n . 19 Dependence of water absorption ( a f t e r 24 h cold soaking) on density and wood-cement r a t i o i n t e r a c t i o n . 113 20 Dependence of water absorption ( a f t e r 24 h cold soaking) on p a r t i c l e moisture content and wood-cement r a t i o i n t e r a c t i o n . 114 21 El e c t r o n micrographs of c r y s t a l orf pure magnesium oxide powder. 0 115 22 Electron micrographs of c r y s t a l s of dead burnt magnesite cement powder. 116 23 Electron micrographs of c r y s t a l s of ammonium polyphosphate (reactant). 117 24 Electron micrographs of magnesite cement boards manufactured using; 0-6% p a r t i c l e moisture content, 1:1 wood-cement r a t i o and density l e v e l 1. 118 25 Electron micrographs of magnesite cement boards manufactured using; 8-15% p a r t i c l e moisture content, 1:1 wood-cement r a t i o and density l e v e l 1. 119 26 Electron micrographs of magnesite cement boards manufactured using; 25-30% p a r t i c l e moisture content, 1:1 wood-cement r a t i o , and density l e v e l 1. 120 27 Electron micrographs of magnesite cement boards manufactured using; 40-50% p a r t i c l e moisture content, 1:1 wood-cement r a t i o , and density l e v e l 1. 121 x i i Figure P a S e 28 Electron micrographs of magnesite cement boards manufactured using; 60-80% p a r t i c l e moisture content, 1:1 wood-cement r a t i o , and density l e v e l 1. 122 29 Relationship between MOE and density at d i f f e r e n t i n i t i a l p a r t i c l e moisture contents. 123 30 Relationship between MOR and density at d i f f e r e n t i n i t i a l p a r t i c l e moisture contents. 124 31 Relationship between IB strength and density at d i f f e r e n t i n i t i a l p a r t i c l e moisture contents. 125 32 Relationship between edgewise compression strength and density at d i f f e r e n t i n i t i a l p a r t i c l e moisture contents. •126 x i i i LIST OF APPENDICES Page 1. Compound composition and the sp e c i a l uses of f i v e Portland cement types. 127 2. B r i t i s h Standard 1811: 1952. Part 2. Method of test for wood chip boards, wood waste boards and sim i l a r boards. B.S.I. 128 3. ASTM 1980: Part 22. D1037 - 34.22. Method of test for compression p a r a l l e l to surface. 132 4. ASTM 1980: Part 22. D1037 - 28. Method of test for t e n s i l e strength perpendicular to surface. 134 5. ASTM 1980: Part 22. D1037 - 100. Method of test for water absorption and thickness swelling. 135 6. ASTM 1980: Part 22. D1037 - 126. Method of testing for moisture content and density. 137 x i v GLOSSARY limestone containing from 35 to 46 percent magnesium carbonate (MgC03). the mineral consisting p r i m a r i l y of f u l l y hydrated calcium sulphate, CaSo^^I^O or calcium sulphate dehydrate. dolomitic limestone that has been heated with or without additives to a temperature s u f f i c i e n t l y high and for a long enough time to decompose the carbonate structure so as to form magnesium oxide. Magnesite cement contains not less than 85 percent magnesium oxide. Magnesite, sometimes referred to as dead-burnt cement i s r e s i s t a n t to subsequent hydration and recombination with carbon dioxide. a hydraulic cement produced by pul v e r i z i n g c l i n k e r consisting e s s e n t i a l l y of hydraulic calcium s i l i c a t e s , and usually containing one or more forms, of calcium sulphate as an interground addition. substances which are not i n themselves cementitious but which react with Ca(0H) 2 i n the presence of water at ordinary temperature and thereby act as cements. Examples include c e r t a i n n a t u r a l l y occurring materials of volcanic o r i g i n . a magnesium oxychloride based cement. It i s obtained by treating a mixture of MgO and aggregate with a concentrated so l u t i o n of MgCl2» XV ACKNOWLEDGEMENTS I wish to express my gratitude to Dr. L. Paszner who was my thesis supervisor. I am g r a t e f u l for his enthusiastic d i r e c t i o n and guidance throughout the course of t h i s study and e s p e c i a l l y during the prepara-ti o n of the manuscript. Dr. R.W. Kennedy and Mr. L. Valg read the thesis and offered valuable advice. I would l i k e to thank Greg Bohnenkamp and Ed Burke who greatly helped me i n securing materials and assembly of the equipment required for the study. The patience of my ch i l d r e n Kayumba and Chuma and my s i s t e r - i n - l a w Wago and the encouragement and support of my lo v e l y wife Dorothy i s g r a t e f u l l y acknowledged. Thanks are expressed to Dr. A. Kozak and Simeon Chiyenda (Graduate Student) for s t a t i s t i c a l help and Barry Wong fo r helping with computer programming. I would l i k e to extend thanks to my fellow students e s p e c i a l l y Joao Manhaes, Laszlo Orbay, N i k i l Chandra Behera, Sitwala Imenda, Winston Mathu and K.C. Yang for the l i v e l y discussions we had i n the course of writing t h i s t h e s i s . F i n a l l y , I would l i k e to thank the Canadian Commonwealth Scholar-ship and Fellowship Committee i n co-operation with the ever progressive Zambian Government for the f i n a n c i a l support that made th i s study possible. 1 1.0 INTRODUCTION 1.1 Background Information I t i s estimated that before the end of the century an addi t i o n a l two b i l l i o n people w i l l need to be housed (4). This apparent high demand for housing w i l l r e s u l t i n increased demand for bu i l d i n g materials and wood f i b r e products. Rational and e f f i c i e n t u t i l i z a t i o n of the a v a i l a b l e growing forest reserve's w i l l be required i n order to s a t i s f y a l l demands. In addition, optimum and complete forest u t i l i z a -t i o n programmes w i l l have to be developed. An important aspect of such programmes w i l l be to f i n d uses for tree species s t i l l considered as waste wood and to develop processes leading to the manufacture of useful products from f o r e s t , m i l l and a g r i c u l t u r a l residues. Kollman (27) has estimated that more than 50% of the wood harvested ends up as waste as a r e s u l t of processing. The urgent need for providing housing, e s p e c i a l l y i n developing countries, w i l l c a l l for erection of q u i c k - b u i l t , low-cost housing u n i t s . The main feature of such housing systems are mass production and on-site erection from prefabricated engineered building materials (14). Plywood, fibreboards and particleboards have been extensively used for many years and have given excellent performance as s t r u c t u r a l materials. However, the excellent p o t e n t i a l of these wood-based panels can, under adverse conditions be negated by the attack and d e t e r i o r a t i o n by fungi and insects and degradation due to f i r e and weather. There-fore, i f these panels are to be e f f e c t i v e l y used i n housing construc-t i o n , practices that minimize these hazards must be employed; treatments 2 with preservatives and f i r e - r e t a r d a n t s , building techniques that prevent undue wetting of panel elements or the a p p l i c a t i o n of wood fi n i s h e s are usually implemented. The high cost of these protective measures, coupled with the r e l a t i v e l y high production cost of these panels, have tended to preclude t h e i r use i n housing e s p e c i a l l y i n developing countries. Wood-cement products are unique as building elements since they have the p o t e n t i a l to overcome the disadvantages that plywood and other wood-based panels have as s t r u c t u r a l elements i n housing. Dinwoodie (15) has reported that wood-cement products rate very highly i n resistance against f i r e , insect and fungi. Wood-wool slabs are reported by Chittenden et^ al_. (9) to have a Class 1 spread of flame rating and a f i r e propagation index (I) not exceeding 12.0 and a sub-index ( i ) not exceeding 6.0 when tested according to B r i t i s h Standard (B.S. 476). Deppe (14) reported that the thickness swelling of cement-bonded particleboard i s considerably lower than that of the synthetic r e s i n -bonded boards. The presently manufactured wood-cement products include: the wood-wool slabs, f i r s t made i n Austria i n 1914 using a mixture of magnesium chloride and magnesium oxide as a binder; and the cement-bonded particleboard o r i g i n a l l y developed and patented i n USA i n 1965 and 1966 by Elmendorf. The success and extensive use of wood-wool slabs for p a r t i t i o n i n g , cladding, roofing f l o o r base and as permanent shutter i n Europe and Japan are l a r g e l y due to th e i r i n s u l a t i n g properties and high degree of f i r e resistance (9). 3 The presently employed wood-cement systems, whose properties and c l a s s i f i c a t i o n are shown i n Table 1, s u f f e r from numerous disadvantages which l i m i t the world-wide growth of the industry. The basic raw material f o r the presently made wood cement systems comprise l i g n o c e l l u l o s i c fragments as f i l l e r materials and employ as binding agents some form of mineral cement. T y p i c a l cements have included Portland and other hydraulic cements, pozzolans and magnesium cement such as Sorel cement. The problem of forming products of adequate strength and with a wide density range arises because of the i n f e r i o r junction bond strengths, that i s , inadequate adherence of the mineral mass to the wood fragments (36). Substances which were shown to be the major cause for poor bonding include sugars, starches, proteins, polyphenolics, waxes, f a t t y substances, gums and various sequestered minerals (5, 12, 24). Because of i n f e r i o r bonding due to the above substances, the number of wood species for wood-cement system manufacture i s l i m i t e d . At present cement-bonded compositions are l a r g e l y r e s t r i c t e d to a few species such as spruces (Picea spp.), true f i r s (Abies spp.), poplars (Populus spp.) and some pines (Pinus spp.). Other suitably tested species require pretreatments i n the form of "mineralization" with mineral additives, (such as waterglass, calcium or magnesium chlo r i d e , borax), to either remove i n h i b i t i n g substances, to seal fragment surfaces or to convert near-surface contaminants to innocuous residues (34, 45). S t i l l i n g e r et a l . (48) has described t y p i c a l treatments to eliminate r e t a r d a t i o n or i n h i b i t i o n i n g e l l i n g and hydration of Portland cements to include: 4 (a) impregnating the fragment surfaces with soluble metal s a l t s such as chlorides or soluble sulphates of calcium or magnesium, which hasten the setting of hydraulic cement s l u r r y adjacent to the fragment, (b) digesting the extractable substances at and near fragment surfaces by treatment with baths of lime or caustic soda with or without further s t a b i l i z a t i o n by a pozzolan or a polyvalent metal s a l t , and (c) loading the surfaces of fragments with a mineral gel e.g. sodium s i l i c a t e . S t i l l i n g e r (48) further noted that such c o s t l y pretreatments necessitate at least an a d d i t i o n a l drying step for wood - Portland cement systems and thus further complicate the manufacturing process. A further processing d i f f i c u l t y with Portland cement-bonded systems such as the DURIPANEL, i s the long pressing time, 8 to 24 h, required to obtain a s o l i d product. Elmendorf (16) reported a press cycle of 7 to 9 h for cement-bonded particleboard, while Chittenden et a l . (9) noted that due to the long press dwell time coupled with a high degree of automation (e.g. i n DURIPANEL), the wood-cement systems are u n l i k e l y to prove economically viable anywhere other than a highly i n d u s t r i a l i z e d country. E f f o r t s to shorten the press dwell time and expand the l i s t of suitable wood species for wood-cement system production, led to the use of magnesite cements as binding agents. The bonding mechanism of the magnesite s e t t i n g , as described by Simatupang et^ a l . (44) i s e s s e n t i a l l y a high-temperature hydration of magnesium oxide, with wood fragments 5 treated with magnesium chloride or magnesium sulphate. A pressing time of 15 minutes was achieved with t h i s process and the i n h i b i t i n g e f f e c t s of wood extractives was minimized (44). A comparison of properties of Portland cement-bonded and magnesite-bonded particleboard i s shown i n Table 2. However, the extremely large amount of energy required to heat the panels during the i n i t i a l 10 minutes pressing time and subsequent conditioning stages, seems to be a big disadvantage with the magnesite-cement process. According to Simatupang et^ aJL. (46), the required hydration temperature i s i n the range of 120°C to 200°C. I t i s estimated that 500,000 kc a l {= 600 kWh) of heating energy and between 140-290 kWh are required to produce 1 mJ of magnesite cement-bonded particleboard. The equivalent d o l l a r requirement for energy alone at the current p r i c e of Cdn$0.04/kWh, i s Cdn$35.60/m3. It i s reported (36) that the production of magnesite-bonded particleboards by HERAKLITH i n Radentenheim, Au s t r i a was halted due to the high energy cost. 1.1.1 The Paszner Process Paszner (36) invented and patented (37) an improved process for the rapid bonding or combining l i g n o c e l l u l o s i c fragments with mineral i n -organic binder materials and forming the admixture into a s t r u c t u r a l product. The invention i s based on the discovery that when a major volume proportion of l i g n o c e l l u l o s i c fragment mass i s admixed with either aqueous solutions of phosphoric acids, or solutions of neutral phosphate s a l t s , p a r t i c u l a r l y aqueous solutions of ammonium s a l t s of polyphos-phoric acids i n amounts so as to p a r t i a l l y impregnate the fragment, 6 and the mixture i s then dusted with a minor volume proportion of magnesium or calcium oxides, hydroxides, or carbonates, a bond i s developed at the fragment surfaces which sets r a p i d l y from a gel phase to a strongly adhered, r i g i d concrete i n t e g r a l with the fragments. The presence of sugars, phenolics or extractables in the l i g n o c e l l u l o s i c fragments has no i n h i b i t o r y e f f e c t on the bond formation (36, 37). The products molded by the process according to the invention, may comprise any of the common s t r u c t u r a l shapes such as boards, panels, slabs, beams, and blocks and may comprise frames, poles, trusses and v i r t u a l l y any castable or sprayable configuration. Such products, a f t e r curing, are reported (36) to be f i r e , decay, insect, fungi and weather r e s i s t a n t . The Paszner process can u t i l i z e a major volume proportion of ligneous plant fragments such as softwoods and hardwoods, sugarcane r i n d shreds, cereal and f i b r e plant s t a l k s . The process b a s i c a l l y involves the following steps: l i g n o c e l l u l o s i c fragments, having thicknesses ranging from 0.3 mm to 8 mm including chips, shavings, s t r i p s , strands, s l i v e r s , f i b r e bundles, f i b r e s and peeled or sawn sheets, are mixed with so l u t i o n of ammonium polyphosphate supplying from 0.15 to 0.4 parts P2°5 a s phosphate ion per part of fragment by weight. To these wetted fragments, p a r t i c u l a t e cement s o l i d s comprising of dead burnt MgO, or CaO, or Mg(0H)2» or Ca(0H)2» or MgC03, or CaCO-j ranging from 0.55 to 2.5 part per part of fragment and from 0.01 to 0.80 parts of i n e r t f i l l e r p a r t i c l e s (such as s i l i c a sand) are added with rapid mixing. The mixture i s molded and held under predetermined compaction pressure u n t i l the product has 7 s o l i d i f i e d i n about 3-10 minutes. Best r e s u l t s occur when the phosphate s o l u t i o n i s mainly absorbed into the fragment surfaces before the cement s o l i d s are dusted onto the wet surfaces (36). S a t i s f a c t o r y products may also be fabricated i f the phosphoric so l u t i o n i s applied to the cement -dusted dry p a r t i c l e s (37). Depending on the magnesium oxide source and purity, the curing time of the mixture can be regulated within r e l a t i v e l y wide l i m i t s , i . e . almost instantaneous to s u f f i c i e n t l y delayed g e l l i n g (36). The cement setting i n the Paszner process, i s l a r g e l y exothermic and does not require i n i t i a t i o n by heat as indicated by Figure 1. However, Paszner (36) reported that v a r i a t i o n s i n ambient temperature have a pronounced e f f e c t on the onset of the exothermic setting reaction, as indicated i n Table 3. Products made by the process a t t a i n at l e a s t 50% of t h e i r ultimate strength within the f i r s t 15 minutes from the time of mixing, up to 75% within 30 minutes and f u l l strength i n 7 days. 1.1.2 The Study The present study employs the Paszner process i n the manufacture of cement-bonded particleboards. 1.2 Objective and Scope of Study The main objective of t h i s study i s to investigate the ef f e c t s of i n i t i a l wood p a r t i c l e moisture content on the physical and the mechani-c a l properties of the magnesite cement-bonded particleboard. This i n v e s t i g a t i o n was prompted by the pu b l i c a t i o n (FID-II/21) of the Food and A g r i c u l t u r a l Organization of the United Nations (FAO, 1978) 8 i n which Paszner (36) indicated that regardless of type or length of cure and p a r t i c l e s i z e , wood p a r t i c l e moisture content could have an adverse e f f e c t on the ultimate crushing strength of dead-burnt magnesite cement bonded products. These preliminary findings by Paszner were considered to be of paramount importance i n the manufacture of magnesite cement bonded products. Since the above-mentioned study by Paszner was exploratory i n nature, i t was decided to conduct an extensive i n v e s t i g a t i o n on the e f f e c t of a wide range of p a r t i c l e moisture content on physical and mechanical properties. The physical and mechanical properties chosen for the study are those considered by the USA F.P.L. technical report No. 10, to be important with respect to the p r a c t i c a l a p p l i c a t i o n of the products as s t r u c t u r a l elements. To study the p a r t i c l e moisture content e f f e c t s , magnesite cement-bonded particleboards were made using mixed softwoods as wood raw material. Physical and mechanical strength tests were conducted accord-ing to established standard procedures. The following board production variables were used: 1. Five l e v e l s of i n i t i a l wood p a r t i c l e moisture content, 2. Three l e v e l s of wood-cement r a t i o , 3. Three l e v e l s of density for each wood-cement r a t i o . Results from t h i s study could serve to ind i c a t e , i n addition to p a r t i c l e moisture e f f e c t s , the treatment combinations of p a r t i c l e moisture, density and wood-cement r a t i o which give the most favourable physical and mechanical properties. I f boards of high strength and 9 favourable physical properties can be manufactured from r e l a t i v e l y wet wood p a r t i c l e s , the study w i l l have an impact on energy saving, i n that the wood p a r t i c l e drying stage during manufacture, as required i n r e s i n -bonded particleboard, could be omitted. In addition, a scanning electron microscopy (SEM) study was con-ducted i n order to obtain preliminary i n d i c a t i o n s on the p a r t i c l e moisture content e f f e c t on bonding between the magnesite-cement and wood. 10 2.0 LITERATURE REVIEW This l i t e r a t u r e review focuses on the physical and chemical charac-t e r i s t i c s of the main raw materials (wood and binders) i n the presently manufactured wood - Portland cement systems. A further section of the review covers the pretreatments of wood raw material employed i n order to improve the wood-cement bond q u a l i t y . 2.1 Raw Materials B a s i c a l l y three groups of raw materials can be i d e n t i f i e d i n the manufacture of wood-cement boards, these are: 1. Wood (including other l i g n o c e l l u l o s i c m a t erials). 2. Binders (either of cement, gypsum, magnesite or r e s i n glue). 3. Pretreatment chemicals (minerals). 2.1.1 Wood Wood i s the most important of the raw materials. Owing to i t s i n t r i n s i c physical and chemical c h a r a c t e r i s t i c s i t may determine success or f a i l u r e of the f i n a l product. In the presently employed wood-cement systems, the form i n which the wood raw material i s procured, depends to a large extent, on the board material desired, i . e . , the technology available for wood p a r t i c l e preparation and the available raw material. Wood may be obtained as round logs from thinnings (24), logging and wood processing residues (24, 49) and secondary wood species. Manufacturing wastes such as slabs, edgings, and trimmings (24, 35) from m i l l s and furnit u r e 11 f a c t o r i e s , have been extensively used i n the past. Table 4 summarizes the form of wood raw material before and aft e r preparation, and the f i n a l product obtained. 2.1.1.1 Physical C h a r a c t e r i s t i c s of Wood The physical c h a r a c t e r i s t i c s of wood are considered here i n r e l a t i o n to the manufacture of wood-wool cement boards. Variations noted (18) i n hammermilling or crushing d i f f e r e n t wood species would probably be related to higher energy consumption. Crushing of wood i s a r e l a t i v e l y simple process compared with wood wool manufacture i n which i n d i v i d u a l wood sizes are determined and maintained by controlled production. For wood wool production, r e l a t i v e l y s t r a i g h t logs are s p l i t and squared for e f f i c i e n t gripping by chucks (18, 20, 23, 51). Machines are designed to handle 10-12 cm diameter logs i n 50 cm lengths. S i m i l a r l y , machines are also available for reduction of edgings, slabs and trim-mings into wood wool. Wacker (51) reported that crooked logs jam machines, reduce maximum length of i n d i v i d u a l wood ribbons ( f i b r e s ) , and tend to make ribbons b r i t t l e . Hubbard (23) noted that knots constitute p o t e n t i a l hazards for knife damage, while sound defects l i k e wane, s p i r a l grain and reaction wood reduce volume output per unit of log input. Weatherwax (53) reported that decayed wood lowers the strength of cement boards and that i t releases chemicals i n h i b i t o r y to cement se t t i n g . Bark, which i s high i n phenolic compounds, i s reported (52) to be also i n h i b i t o r y to the setting of cement and i s usually removed immediately from the fresh wood. 12 2.1.1.2 E f f e c t of Chemical C h a r a c t e r i s t i c s of Wood on Cement Setting The gross structure and complementary properties of mature wood arises from the nature and organization of i t s chemical constituents (33). The chemical components of the c e l l wall substance i n normal wood are given i n Table 5. Weatherwax (52) observed that both major and minor chemical com-ponents of wood had marked e f f e c t s on the curing (setting) of cement. Owing to the great v a r i a b i l i t y i n chemical c o n s t i t u t i o n between i n d i v i d u a l trees, a great v a r i a t i o n i s to be expected i n the s u i t a b i l i t y of woods for wood-cement boards. Such v a r i a t i o n i s most pronounced between hardwoods and softwoods (2, 13, 52, 53). Christensen et a l . (10) found that hardwood hemicellulose had a pronounced i n h i b i t o r y e f f e c t on the s e t t i n g of cement, and Sanderman (39) reported that starches tannins, sugars and c e r t a i n phenols were most i n h i b i t o r y . Wood i s an organic material, and when used as an aggregate i n cement-water mixtures may undergo c e r t a i n unfavorable reactions which a f f e c t s the s e t t i n g of cement mixture (7, 11, 17, 52, 53). Sanderman e_t a l . (38) pointed out that while small concentrations of c e r t a i n sugars may have a p o s i t i v e retarding e f f e c t , others may be without e f f e c t . Hemicelluloses hydrolyse r e a d i l y i n d i l u t e acids and a l k a l i to form free reducing sugars and sugar acids. Where hemicelluloses form a substan-t i a l f r a c t i o n of the wood substance, as i n many hardwoods, they were found to be troublesome i n cement setting (13, 32, 52, 53). Sanderman et a l . (38) reported three d i f f e r e n t groups of carbohydrates that can be i d e n t i f i e d i n wood-cement mixtures: 13 ( i ) water-soluble sugars ( i i ) water-soluble higher carbohydrates and ( i i i ) water soluble constituents formed from water insoluble carbohydrates due to the action of the a l k a l i n e cement. Not a l l of these carbohydrates hinder cement s e t t i n g . It was reported (38) that sugar additions of about 0.125%, increased the strength of cement appreciably while an addition of 0.25% of most sugars to cement sludge causes complete s e t t i n g i n h i b i t i o n . The e f f e c t of two p a r t i c u l a r sugars, fructose and r a f f i n o s e was reported. An addition of 0.5% of fructose to the s l u r r y gave a favourable r e s u l t , whereas the same concentration of r a f f i n o s e i n h i b i t e d the c r y s t a l l i z a t i o n process and caused crumbling of concrete. Sanderman e^ t a l . (38) attributed the non-inhibitory e f f e c t of fructose to the fact that i t forms a sparingly soluble compound [C6 Hi2°6 + C a 0 + 6 H 2 ° 1 w i t h l i m e w a t e r * Since Ca(0H) 2 i s a v a i l a b l e during preparation of the cement sludge, t h i s compound might be formed. In another study, Sanderman (39) stated that starches, tannins, sugars and c e r t a i n polyphenols had i n h i b i t o r y properties. Of a l l the organic constituents of wood, carbohydrates appear to have the greatest e f f e c t on the setting of cement (5). Of the major carbohydrates, the order of i n h i b i t i o n was: glucose, followed by carboxymethyl c e l l u l o s e and l a s t l y c e l l u l o s e . It has been reported that i n softwoods, the c e l l o b i o s e of decayed wood i s more i n h i b i t o r y than glucose (53). 14 2.1.1.2.1 Wood Cellulose Wood c e l l u l o s e and most other c e l l u l o s i c materials are not obtain-able i n pure form. Cellulose always contains small amounts of simple sugar compounds c l o s e l y linked with i t . These sugars, according to Wise et a l . (54) and Won-Yung et a l . (56), are mainly mannose, galactose, glucose, fructose, xylose and arabinose. On drying of wood raw material i n p a r t i c l e form, migration of these simple sugars to the surface of the p a r t i c l e takes place (19). Many such sugars act as i n h i b i t o r s of cement hydration, r e s u l t i n g i n poor wood-cement bonds. Generally, i t i s believed that i n h i b i t i o n i s the r e s u l t of interference by such sugars of c r y s t a l formation i n cement, thereby a f f e c t i n g possible mechanical i n t e r l o c k i n g between wood and cement and within the cement i t s e l f (56). Apart from i n h i b i t o r y e f f e c t s of the simple sugars, i t has been postulated by Weatherwax (53) that the r e l a t i v e l y high i n h i b i t o r y e f f e c t s of c e l l u l o s e might also be due to the hydrolysis products of c e l l u l o s e , mainly c e l l o b i o s e and c e l l o t r i o s e . 2.1.1.2.2 L i g n i n L i g n i n has been reported to have no e f f e c t on cement setting (2, 16). I t s low hygroscopicity and i n s o l u b i l i t y i n acids might explain t h i s feature of l i g n i n . 2.1.1.2.3 Wood Extractives Among the important wood extractives are polyphenols and r e s i n s . The l a t e r are the source of the steam d i s t i l l e d turpentine, t a l l o i l and r o s i n , while the former include a large number of important chemicals 15 such as tannins, anthocyanins, flavones, catechins, kinos and lignans (38). These occur both i n angiosperms and gymnosperms. Irrespective of quantity, the presence of extractives generally have profound e f f e c t s on the physico-chemical properties of wood. When the extractives are coloured, they enhance the usefulness of wood for wood wool, for pack-aging glass wares for display and they reduce the cost of dyeing the s t u f f a r t i f i c i a l l y (39). P u t r i d smell of c e r t a i n extractives d i s q u a l i f y the wood for use i n wood wool building slabs (18, 39, 49). Toxic extractives are desirable for long service l i f e of wood since they r e s i s t fungal decay and insect damage (27). Tannins, e s p e c i a l l y the hydrolyzable ones, were found to have a s i g n i f i c a n t i n h i b i t o r y e f f e c t on cement s e t t i n g (5). Terpenes were found to have no ef f e c t on cement setting (11), whereas several organic acids of higher molecular weight, such as a l i p h a t i c acids, were found to be i n h i b i t o r y . Lignosulphonic acid and hydroxylated carboxylic acids were reported as being used commercially as retarders of cement se t t i n g (22). The type and percentage of i n h i b i t o r y organic substances i n wood vary between sapwood and heartwood as well as between earlywood and latewood. Wood zone and season of cut, therefore a f f e c t setting of wood-cement mixtures (13, 53). Ground heartwood mixtures require longer se t t i n g time than those of sapwood, because heartwood contains more phenolics than sapwood. According to Kamil (24), the maximum allowable sugars and tannins content i s 1 and 2 percent respectively. T r o p i c a l hardwoods i n general contain more extractives and there-fore are less suitable for wood-cement board manufacture. 2.1.1.3 E f f e c t of Decay and Stain Fungi Wood decayed by brown-rot fungi (Peniophora spp.) may be severely i n h i b i t o r y (53). It has been shown that decayed wood can have i n h i b i -tory indices ranging from 2,000 to 11,000, while bark averages 200 and heartwood 70. The i n h i b i t o r y index i s calculated from the formula: = 100 9 - 8s 0s where: I i s the i n h i b i t o r y index, 9 i s the setting time of i n h i b i t e d cement and 9s i s the setting time of uninhibited cement (52, 53). Very few wood species s a t i s f y t h i s equation due to the very high sensi-t i v i t y of cement setting to extractives which are almost i n v a r i a b l y present i n wood. For th i s reason, i f hardwoods (which are reputed for being highly i n h i b i t o r y because of the high hemicellulose and phenolic contents), were to be used for manufacture of wood-Portland cement boards, pretreatment would be i n e v i t a b l e . Davis (13) studied the e f f e c t s of blue s t a i n on the setting of CaCl2 mineralized, excelsior-cement mixtures. The r e s u l t s of his study indicated that blue-stained wood promoted the setting of wood-concrete mixtures containing CaCl2' Mixtures containing blue-stained wood showed an e a r l i e r r i s e i n temperature and an e a r l i e r s e tting than did mixtures containing non-blue-stained wood. This finding by Davis (13) i s converse to that observed i n p a r t i a l l y decayed wood which strongly i n h i b i t s cement setting (53). However, i t should be noted that Weatherwax e_t a l . (53) did not include C aC^ i n t h e i r decayed wood-concrete mixtures. 17 Davis (13) a t t r i b u t e d the d i f f e r e n t e f f e c t s of blue-stained and p a r t i a l l y decayed wood on concrete setting as a r e f l e c t i o n of the difference i n the physiology of these two groups of organisms. Decay fungi attack the s t r u c t u r a l components of the wood and consequently lower the degree of polymerization of the wpod constituents, thereby increasing the amounts of carbohydrates soluble i n d i l u t e a l k a l i . At the same time the metabolic products of the fungi are accumulating i n the wood. Stain organisms, on the other hand, u t i l i z e soluble c o n s t i t u -ents with l i t t l e or no e f f e c t on the s t r u c t u r a l components (13). Elimination of decayed wood material i n wood-cement board manufac-ture i s therefore an important step towards trouble free cement se t t i n g . It has been suggested (51) that short storage periods of not more than one year should be i n s t i t u t e d i n order to prevent decay and subsequent production of soluble i n h i b i t o r s . K i l n drying, for both preventing decay and achieving low i n h i b i t i o n index, I, was found very e f f e c t i v e (52, 53). Weatherwax et^ al. (52) constructed curves from which tolerance lev e l s of wood-concrete slabs for various mixes of highly decayed wood can be read. Reasonable amounts of decayed wood i n wood-concrete slab mixes did not seriously reduce the strength of the f i n a l product. I t was suggested (52) that wood mixes with an i n h i b i t i o n index, I, of 40, w i l l just survive handling when the moulds are removed. Based on t h i s f i n d i n g , i t was recommended that only 12% of severely decayed wood, and either 33% of bark or 60% of heartwood could be tolerated i n wood-cement board products (52). 2.1.2 Binders From the above b r i e f discussion i t can be seen that the curing of cement i s very c r i t i c a l l y affected by the nature of the wood raw material. The type of wood being used i s also of great s i g n i f i c a n c e . However, work done so far also indicates that the type of binder and the wood pretreatment chemical used, are also important. O r i g i n a l l y , i n Germany, only magnesite was used as a bonding agent i n mineral bonded slabs. Since i t s f i r s t a p p l i c a t i o n i n 1928 i n the production of wood wool boards, Portland cement has increased i n use (14). By 1936, 39% of t o t a l Germany's output of wood wool slabs was manufactured with Portland cement, 35% with magnesite and 26% with plaster as binding agent (50). Today more than 80% of wood-wool boards are manufactured using cement (mainly Portland cement) as the binding agent (14). P l a s t e r s , r e s i n etc. are used to a lesser degree (3). For t h i s reason only Portland cement w i l l be reviewed here i n some d e t a i l , while others w i l l only be mentioned. 2.1.2.1 Cements The cements that have been used i n wood-cement boards manufacture over the years included magnesium oxysulphate, magnesium oxychloride and Portland cement. Because of abundance, Portland cement i s the most important (1, 21, 24, 50). Portland cement i s made from a mixture of lime and clay-bearing materials which are calcined to form a c l i n k e r consisting e s s e n t i a l l y several forms of hydrated calcium s i l i c a t e s . This c l i n k e r i s then pulverized to a fineness which permits nearly a l l of i t to pass through a sieve with 40,000 openings to the square inch. Gypsum can be added to 19 control the rate of set and other materials such as grinding aids and a i r entraining agents can also be added i n small amounts (31). Portland cement i s defined i n the ASTM C 150-41 as .... the product obtained by p u l v e r i z i n g c l i n k e r which consists e s s e n t i a l l y of hydraulic calcium s i l i c a t e s . I t i s s p e c i f i e d that no additions should be made to the s i l i c a t e s subsequent to c a l c i n a t i o n other than water and/or untreated calcium sulphate. According to Larson (31) Portland cement i s composed mainly of three oxides: s i l i c a (Si0 2)» lime (CaO), and alumina ( A 1 2 0 3 ) . It also contains small quantities of magnesium oxide (MgO), sulphur t r i o x i d e (SO-j) and f e r r i c oxide (Fe20.j). These oxides occur i n various combinations, the mixture of which forms the cement. The p r i n c i p a l compounds of Portland cement together with some of t h e i r c h a r a c t e r i s t i c s are shown i n Table 6. Appendix I contains the compound composition of Portland cement and gives the s p e c i a l use conditions for the f i v e types covered by the ASTM s p e c i f i c a t i o n s . Cement sets by a process of hydration which i s exothermic i n nature. The process steps include gelation as well as c r y s t a l l i z a t i o n . Both the gel and c r y s t a l s are c l o s e l y packed, f i l l the empty spaces between p a r t i c l e s and hence r e s u l t i n high hardness products (24). Hydration depends more s p e c i f i c a l l y on the surface exposed to the action of water than on the chemical c o n s t i t u t i o n of the cement (55). This has been recognized since as early as 1918 when Duff Abrahams introduced the water-cement r a t i o law. The law states that "With given concrete materials and conditions of t e s t , the quantity of mixing water used per bag of cement determines the strength of the concrete so long as the mix 20 i s of workable p l a s t i c i t y . " Photomicrographs have shown Portland cement types I and II to contain more unhydrated p a r t i c l e s a f t e r 28 days of hydration than type I I I (31). In the bonding of wood, cement i s very vulnerable to toxic organic substances as stated e a r l i e r . Sugar and other compounds prevent the hydration process of cement. It should, however, be noted that since hydration depends on the amount of cement p a r t i c l e surface i n contact with water, the wood p a r t i c l e s might l i m i t hydration by displacing water from the reaction s i t e s , thereby reducing the amount of water needed for hydration to occur. In massive construction such as large concrete dams, Portland cement l i b e r a t e s large quantities of heat (about 100 c a l o r i e s per gram of cement) (31). To avoid damage due to excessive heat, the composition of cement i s usually altered so that the high heat-producing compounds, C3S and C3A, are present i n lesser amounts. This r e s u l t s i n 60 to 70 c a l o r i e s of heat per gram of cement. On the contrary, i n the produc-t i o n of wood-wool slabs, rapid set or ea r l y attainment of maximum strength i s desirable. For t h i s reason high heat-producing cements are preferred (24, 29). The amount of hydration temperature and the time taken to reach maximum hydration temperature have been used as a measure of the s u i t a b i l i t y of woods for wood-cement board manufacture (24, 41, 52, 53). In essence, the rate at which wood i s able to diminish the a b i l i t y of high heat Portland cements to l i b e r a t e heat, can be used as an index of i t s n o n - s u i t a b i l i t y for board manufacture. Type II Portland cement described i n Table 7, which i s r i c h i n t r i c a l c i u m s i l i c a t e , and 21 alumino f e r r i t e , has been highly favoured i n wood-wool cement board manufacture. . 2.1.2.2 Other Binders Gypsum and magnesium cement are p o t e n t i a l substitutes to Portland cement i n the manufacture of wood-cement products. Magnesium cement, often c a l l e d Sorel cement, a f t e r i t s inventor, has mainly been used as a binder for the manufacture of the so-called HERAKLITH board, a wood-wool board bonded with magnesium oxysulphate (27). In some cases calcinated dolomite can s u c c e s s f u l l y replace the more expensive magnesium oxide i n magnesite cements (37). As dolomite i s a mineral with wide occurrence (25), the p o s s i b i l i t y for using t h i s binder i n the manufacture of particleboards has good p o t e n t i a l . Gypsum has been used i n the manufacture of gypsum wallboards. These boards are composed of gypsum core and a k r a f t paper wrapping and are widely used i n indoor applications (46). Simatupang et_ al^. (46), reported that FAMA International of Germany have since 1976, s u c c e s s f u l l y introduced on the market a gypsum-bonded fibreboard using c e l l u l o s e f i b r e s of recycled newspaper as raw material. Gypsum, i n the form of a hemihydrate (CaS04•1/2H20) also c a l l e d "Plaster of P a r i s " i s used as a binder i n the FAMA Process. In order to regulate c r y s t a l l i z a t i o n , retarding additives are reportedly added (46). Gypsum boards are generally weaker and more susceptible to v a r i a -tions i n atmospheric humidity than cement boards (29). They are therefore not recommended for use under continuously humid and wet conditions. The increased use of gypsum as a binder i n wood composites 22 w i l l depend, among other things, on the development of an e f f i c i e n t method for increasing i t s moisture resistance. 2.1.3 Pretreatment of Wood 2.1.3.1 General In order to increase the number of suitable wood species i n wood-cement board manufacture, preventive, b i o l o g i c a l , and chemical treatments of wood aimed at improving cement setting and strength c h a r a c t e r i s t i c s of the boards have been proposed and applied i n many cases. In addition to the treatments mentioned above, c e r t a i n practices for reinforced bonding, which tend to counterbalance bond i n h i b i t o r y e f f e c t s have been reported. Such practices as the use of wire netting, addition of i n e r t aggregates such as sand (2) were suggested (2, 30, 32, 47). Higher than normal cement r a t i o s i n the mix have been favoured i n order to impart the required strength to panels. 2.1.3.2 B i o l o g i c a l Treatments Treatment of wood with blue-stain fungi (Ceratocystis p i l i f e r a ) has been proposed by Davis (13). He claimed that t h i s group of fungi when used to i n f e c t southern pine (Pinus spp.) wood during a period of more than 4 months, decreased setting time of the cement. B i b l i s eit a l . (5) also concluded from t h e i r i nvestigations, that blue-stain fungi probably use the wood sugars that otherwise i n h i b i t cement s e t t i n g . The f e a s i b i l i t y of b i o l o g i c a l control i s , however, s t i l l remote as i t i s d i f f i c u l t to avoid contamination of the blue-stain culture from 23 wood decay fungi. If the action of blue-stain fungi were f a s t e r , close c o n t r o l over a r e l a t i v e l y short time might be f e a s i b l e . Since the required treatment periods are generally long, p r o h i b i t i v e l y large storage f a c i l i t i e s would be required for i n d u s t r i a l wood procurement. Another disadvantage of s t a i n fungi i s that they can be used only to tr e a t the l i v i n g sapwood which i s generally a r e l a t i v e l y small portion of most trees. 2.1.3.3 Hot Water Ex t r a c t i o n Probably the simplest of pretreatments of wood i n preparation for cement bonding i s hot-water extraction. It has been reported (53) that hot-water treatment considerably reduces setting time of wood-cement s l u r r y . The same workers (53) observed that the treatment shows a more pronounced setting-time reduction on cement mixed with heartwood than mixed with sapwood. Parker (35) used the hot-water pretreatment process i n his studies on sawdust-cement boards i n 1947. His treatment sequence was as follows: 1. B o i l the sawdust i n water. 2. Drain and wash with water. 3. B o i l i n a s o l u t i o n of ferrous sulphate i n water. 4. Drain and rewash. The use of ferrous sulphate was to p r e c i p i t a t e the tannins as f e r r i c tannate by oxidation. Sawdust of appreciable tannin content turned black when treated with ferrous sulphate. This stage can be omitted when sawdust of n e g l i g i b l e tannin content i s used. According to Parker (35), the hot-water treatment extracts soluble carbohydrates and 24 tannins, which otherwise dissolve r e a d i l y i n the s l i g h t l y a l k a l i n e mixing water of the cement paste. I t was observed that the treatment seemed to remove o i l y ( f a t t y ) material by the solution either i n the steam or hot-water. 2.1.3.4 Chemical Pretreatments Chemical treatments have been widely used to improve the bonding of wood to cement and hence improve the ultimate strength of the wood-cement panels. These treatments are applied with or without hot-water extraction. Chemicals, commonly c a l l e d "additives", which are i n d u s t r i a l l y used at present include the following; chlorides of calcium and magnesium, s i l i c a t e s of sodium or potassium (waterglass) and a mixture of aluminum sulphate and lime water. Sanderman (42) also reported that calcium formate and calcium acetate are used as ad d i t i v e s . When using chloride-containing additives, t h e i r metal corroding properties should be taken into account. Calcium chloride has been used to reduce setting time of wood cement s l u r r i e s . It was reported (5) that among various southern pine wood-cement mixtures, 1.0% CaCl2 reduced setting time s i g n i f i c a n t l y . Christensen (10) found that d i l u t e d solutions of 1.0 to 3.0% CaCl2 were required to neutralize the e f f e c t s of 0.1% sugars on setting time. He also reported that up to 10% solutions of hydrated lime did not decrease the setting time i n a wood-cement mixture to which 0.1% sugar was added. K l e i n l o g e l (26), reported that incorporation of more than 4% CaCl2 into the wood-cement mixture did, on the other hand, reduce the 25 strength of the f i n a l product. This could be a t t r i b u t e d to the fact that the higher additive concentration r e s u l t s i n an accelerated curing time, r e s u l t i n g i n very high-hydration temperatures which are d e t r i -mental to the wood fur n i s h . The corrosion e f f e c t (hydrolysis) could also be the cause of reduced strength when high CaCl2 concentrations are used. German wood-cement board producers disagree on preference of CaC^ to sodium s i l i c a t e (Na2SiC>3) as additives (50). Sodium s i l i c a t e was preferred because of i t s e f f i c i e n c y , but CaCl2 was preferred, i n some cases, because of i t s a v a i l a b i l i t y . Aluminum sulphate has also been used as an a d d i t i v e . Schmidt (43) suggested the use of 4 to 5% aqueous aluminum sulphate to minimize the i n h i b i t o r y e f f e c t of sugars. According to Simatupang et_ a l . (46), the e f f e c t of c e r t a i n additives i s influenced by the chemical composition of the cement. He suggested further investigations to e s t a b l i s h the p r i n c i p a l influence of various additives on cement of d i f f e r e n t chemical composition. The chemical treatment of wood i s commonly car r i e d out i n two ways. In the wood-wool industry, the wood-wool i s soaked i n the additive s o l u t i o n and excess water i s removed. In the other a p p l i c a t i o n the sol u t i o n of additives i s sprayed onto the wood. According to Simatupang et^ j i l . (46) and Sanderman (42), dipping i s not favoured due to the fact that the i n h i b i t i n g compounds may leach out from the wood into the main sol u t i o n . 26 3.0 MATERIALS AND METHODS 3.1 Experimental Design The design of the experiment included three v a r i a b l e s . There were f i v e wood p a r t i c l e moisture content l e v e l s , three wood-cement r a t i o s and three board density l e v e l s for each wood-cement r a t i o . The experimental design was as follows: Five wood p a r t i c l e moisture content l e v e l s - 0-6%, 8-15%, 25-30%, 40-50% and 60-80% Three wood-cement r a t i o s - 1:1, 1:1.5 and 1:2 Three board density l e v e l s and resultant densities for each wood-cement r a t i o as follows: Wood-cement Density r a t i o l e v e l Wood Cement Density (gm) (gm) (gm/cnr1) 1:1 1 300 300 0.472 2 350 350 0.528 3 400 400 0.622 1:1.5 1 300 450 0.636 2 350 525 0.707 3 400 600 0.809 1:2 1 300 600 0.763 2 350 700 0.847 3 400 800 0.939 A l l combinations of the above variables gave 45 treatments. Three r e p l i c a t e panels (boards) were made for each treatment, thus giving a t o t a l of 135 panels for the study. 27 Experimental Design - Layout Test Wood-cement r a t i o Density l e v e l P a r t i c l e moisture content % 0-6 8-15 25-30 40-50 60-80 1 3* 3* 3* 3* 3* i ) S t a t i c bending 1:1 2 3* 3* 3* 3* 3* i i ) Compress.//surface 3 3* 3* 3* 3* 3* 1 3* 3* 3* 3* 3* i i i ) I nternal bond 1:L5 2 3* 3* 3* 3* 3* i v ) Swelling i n 3 3* 3* 3* 3* 3* thickness + 1 3* 3* 3* 3* 3* water absorp- 1:2 2 3* 3* 3* 3* 3* t i o n 3 3* 3* 3* 3* 3* N.B. 3* - Represents number of r e p l i c a t i o n s . 3.2 Preparation of Materials 3.2.1 Wood P a r t i c l e s Description The wood p a r t i c l e s used i n the manufacture of the boards i n this study were a mixture of White Spruce (Picea glauca) and Jack Pine (Pinus  banksiana) obtained from a hammermill i n Prince Albert, Saskatchewan. The average p a r t i c l e size was 2 to 15 mm long, 0.4 to 2.5 mm wide and 0.3 to 1.55 mm i n thickness. 3.2.2 Magnesite Cement Description The cement used i n t h i s study was 100 percent commercial dead burnt magnesium oxide (MgO) - based cement supplied by Kaiser Refractories, Oakland, C a l i f o r n i a . There were no i n e r t f i l l e r p a r t i c l e s added to the 28 cement. The cement so l i d s were of grain s i z e passing 100 mesh and retained on 200 mesh screen, 70 percent passing 150 mesh. 3.2.3 Ammonium Polyphosphate (Reactant) Description The reactant used for th i s study was an aqueous sol u t i o n of commercial ammonium polyphosphate of analysis 10:34 (10% nitrogen as ammonia and 34% P2O5 a s phosphate ion) having a s p e c i f i c gravity of 1.4 and a s o l i d content of 40% by weight (37). Based on 100 grams of solution, the r e l a t i v e weight proportions of the P2O5 contents of the orthophosphate and polyphosphate compounds of the commercial ammonium polyphosphate sol u t i o n were: Orthophosphate 7.20 parts Pyrophosphate 10.10 parts Tripolyphosphate 0.61 parts Tetra and higher polyphosphate 0.20 parts 3.2.4 Mold The mold used for the manufacture of the boards was made out of 19 mm l u c i t e or p l e x i g l a s s . The mold, shown i n Figure 2, i s comprised of three parts: the plunger, the frame and the base. The dimensions of the various parts of the mold are shown i n Figure 3. The mold was designed for manufacture of 405 mm x 176 mm (16" x 7") boards of any desired thickness. 3.2.5. Press The press used i n the manufacture of boards i n t h i s study was a laboratory s i z e , single opening, hydraulic press with 900 square 29 centimetre platen area. It has a force capacity of 22,720 kg (50,000 l b s ) . The press model i s 25-12-2TMB and manufactured by Wabash Metal Products Inc. of Indiana. The press i s shown i n Figure 4. 3.2.6 Manufacture of Magnesite Cement-Bonded Particleboard 3.2.6.1 Wood P a r t i c l e Moisture Content Adjustment A sample of wood p a r t i c l e s was taken from each of the 5 bags received, for moisture content (oven-dry basis) determination. The procedure for determination of p a r t i c l e moisture content was i n accordance with ASTM standard - D143-52 of 1980. Using the formula: 100W  D 100+M where: D = wood p a r t i c l e green weight; W = weight of wood p a r t i c l e s at known moisture content; M = wood p a r t i c l e moisture content to be attained, the amount of water needed to rai s e the moisture content of a ce r t a i n weight of wood p a r t i c l e s (W) at known moisture content, was calculated. The calculated amount of water was added to a known weight of wood p a r t i c l e s (W) contained i n a p l a s t i c bag. The contents were then thoroughly mixed. The bag and i t s contents was then sealed and stored i n a cold room maintained at 35°F for a period of one month. The bags were shaken three times every week to ensure even d i s t r i b u t i o n of the water added. The average i n i t i a l moisture content of the wood p a r t i c l e s was found to be 15 percent. The lower p a r t i c l e moisture content of 0-6 per-cent required for the study was attained by conditioning the p a r t i c l e s 30 i n a c o n t r o l l e d temperature and humidity (C.T.H.) chamber at 21% RH and 38°C. Aft e r one month the f i n a l p a r t i c l e moisture content was determined before board manufacture, to ensure that the p a r t i c l e s had attained the desired moisture content l e v e l . 3.2.6.2 Mixing of Wood P a r t i c l e s with Reactant and Cement According to the treatment combination desired (see Section 3.1) a known weight of wood p a r t i c l e s was put i n a 788 mm x 450 mm p l a s t i c bag. The reactant (an aqueous sol u t i o n of ammonium polyphosphate) equivalent to 80 percent the weight of the cement, was added to the wood p a r t i c l e s and the contents mixed thoroughly and allowed to stand for 3 minutes, so as to allow some of the ca t a l y s t to be absorbed into the wood p a r t i c l e surfaces while retain i n g a surface wetting f i l m . The magnesite cement s o l i d s were then added as an adherent coating on the said wetting f i l m o and the contents again mixed thoroughly for 1 min and then quickly poured into the mold, l e v e l l e d (using a l e v e l l e r shown i n Figure 5) and put under cold press. 3.2.6.2.1 The Reaction C h a r a c t e r i s t i c s Upon applying the magnesite cement so l i d s to the wetting f i l m of ammonium polyphosphate on the wood p a r t i c l e s , an exothermic reaction follows. S t r u c t u r a l diagrams i l l u s t r a t i v e of the formation of oxyphosphate reaction products i n aqueous solutions of ammonium orthophosphate and ammonium pyrophosphate (the two main components of the reactant), respectively, with magnesium oxide, as reported by 31 Paszner (37), are schematically indicated below: 2 HNH, - 0 - P - 0 - HNH, + 3MgO — t 0 - P - O - M g - O - P — 0 -HNH, 3 4 6NH 3 + 3H 20 ammonium orthophosphate magnesium orthophosphate ( N H 4 ) 3 . ( P 0 4 ) 2 ( M g O ) 3 . P 2 0 5 . 4 H 2 0 HNH, - O - ^ - O - ^ - O - HNH, 4 2 M g O ~ * - 0 - ^ - 0 - ^ - 0 +4NH, 4 2H„0 I I 3 I I | I I t 9 9 ' O - M g - 0 HNK3 HNH 3 \ _ M g ' anmoniutn pyrophosphate magnesium pyrophosphate ( H H 4 ) 4 P 2 0 ? ( M g O ) 2 . P 2 0 5 . 3 H 2 0 3.2.6.3 Pressing Time and Control of Temperature A press dwell time of 10 min was used i n the manufacture of boards for t h i s study. A l l boards were manufactured at room temperature of approximately 20 to 24°C. No heat was applied since the reaction i s exothermic. 3.2.6.4 Pressure The molded mass was held under compaction pressure developed by pressing the plunger and base against the molded mass. The molded mass was compacted so as to produce a board of 19 mm (3/4") thickness. The pressure was varied from 0.3 to 14 kg/cm . 3.2.6.5 Board Conditioning After manufacture, the boards were stored for seven days i n a fume cupboard to allow the ammonia gas that i s generated during the reaction, 32 to escape. The boards were f i n a l l y stored i n the C.T.H. room, main-tained at a temperature of 70°F + 3.5°F, and a r e l a t i v e humidity of 50% + 2%, for a period of 60 days before preparation of test specimens.' 3.2.7 Test Specimen Preparation 3.2.7.1 Cutting of Specimens The specimens were cut using a c i r c u l a r bench rip-saw and a cross-cut saw. The use of these saws did not imply a grain d i r e c t i o n i n the boards, but was for convenience. Both these saws were f i t t e d with carbide sawteeth. No problems were encountered i n sawing the boards using these saws. Aft e r cutting 120 running metres (360 f t ) on the r i p saw and 90 running metres (270 f t ) on the cross-cut, no sharpening of the teeth was required. 3.2.7.2 Cutting Plan for Test Specimens 405 mm (16") 176 mm (7") IB IB MC+D 33 R = S t a t i c bending 200 mm x 100 mm (8" x 4") C = Compression//surface 25 mm x 100 mm (1" x 4") IB = Internal bond 50 mm x 50 mm (2" x 2") S = Thickness swelling + water absorption 150 mm x 150 mm (6" x 6") MC+D = Moisture content and density 3.2.8 Board Testing Procedures The following board properties were tested: Modulus of e l a s t i c i t y - tested i n accordance with BS 1811:1952 (see Appendix 2). Modulus of rupture - tested i n accordance with BS 1811:1952 (see Appendix 2). Compression p a r a l l e l to surface - tested i n accordance with ASTM 1980:D1037-34.22 (see Appendix 3). Tensile strength perpendicular to the surface ( i n t e r n a l bond) -tested i n accordance with ASTM 1980:D1037-28 (see Appendix 4). Thickness swelling and water absorption - tested i n accordance . with ASTM 1980-.D1037-100 (see Appendix 5). 3.2.9 Board Characterization 3.2.9.1 Moisture Content and Density The moisture content and the density of the boards at test were determined i n accordance with ASTM standards D1037-126 (see Appendix 6). 34 3.3 S t a t i s t i c a l Analysis F a c t o r i a l analysis of variance^ was performed i n order to f a c i l i t a t e the i n t e r p r e t a t i o n of the main and i n t e r a c t i n g e f f e c t s that could emerge, and to show the comparative performance of treatments. Treatments were analyzed with respect to compression, modulus of e l a s t i c i t y (MOE), modulus of rupture (MOR), i n t e r n a l bond strength (IB), thickness swelling and water absorption t e s t s . A step-wise regression was also performed with respect to edgewise compression, MOE, MOR and IB, i n order to show main ef f e c t s a f f e c t i n g these mechanical properties as well as t h e i r r e l a t i v e importance. The regression analysis was performed using the following power exponential f i t equation: f(R,M,D) - a ( R U l c L R M ° 2 c 2 M D" 3 c 3°) where; f = Mechanical property (MOE, MOR, IB or compression) R = Wood-cement r a t i o D = Density M = I n i t i a l p a r t i c l e moisture content a = Constant b's = Exponents of the power term c's = Bases of exponential term. Equations for p r e d i c t i o n of edgewise compression, MOE, MOR, and IB were generated. Analysis of variance for f a c t o r i a l experiment. 35 3.4 Scanning El e c t r o n Microscopy (SEM) Study SEM studies were undertaken on tested i n t e r n a l bond specimens i n order to obtain a preliminary i n d i c a t i o n on the e f f e c t of wood p a r t i c l e moisture content on the bonding between cement and wood i n the magnesite cement particleboard. The nature of the f a i l u r e i n the i n t e r n a l bond test specimens was also observed. SEM studies of main raw materials used i n board manufacture (magnesite cement and ammonium polyphosphate) were also done i n order to compare th e i r c r y s t a l formations to those of the f i n a l reaction products. Electron micrographs of pure magnesium oxide (the main compound i n the magnesite cement used) were also taken for comparison purposes. The scanning electron microscope used for the study was an Autoscan model No. 26 manufactured by the Etec Corporation. 36 4.0 RESULTS AND DISCUSSION 4.1 Mechanical Properties 4.1.1 Modulus of E l a s t i c i t y (MOE) Table 8 summarizes the average MOE values for 45 treatment combina-tions of i n i t i a l p a r t i c l e moisture content, density and wood-cement r a t i o . Boards made from 0-6% p a r t i c l e moisture content, wood-cement r a t i o 1:2 and density l e v e l 3 gave the highest MOE of 317.02 x 10 kg/cm 2. There i s a general reduction i n MOE values as the i n i t i a l p a r t i c l e moisture content increases from 0-6% to 60-80%. The MOE values increase with increasing density and wood-cement r a t i o . The f a c t o r i a l a n a l y s i s 1 (Table 17) for MOE shows that i n i t i a l p a r t i c l e moisture content A, density B and wood-cement r a t i o C and interactions A-B, B-C, A-C and A-B-C are a l l highly s i g n i f i c a n t at 0.01 l e v e l . The s i g n i f i c a n t i n t e r a c t i o n between the main ef f e c t s A-B-C means that we cannot d i r e c t l y say which main e f f e c t has the greatest and least influence. However, t h i s i n t e r a c t i o n may be more c l e a r l y understood by r e f e r r i n g to Figures 8, 9, and 10. Figure 8 shows that MOE i s lowest at i n i t i a l p a r t i c l e moisture content of 60-80% and density 1. The highest MOE value of 81.19 x 10 J kg/cm i s produced at density l e v e l 3 and 0-6% i n i t i a l p a r t i c l e moisture content. Density l e v e l 3 consistently gives the highest MOE values followed by density l e v e l 2, i n d i c a t i n g that density has a p o s i t i v e e f f e c t on MOE. At i n i t i a l p a r t i c l e moisture contents of 0-6%, 8-15%, and 60-80%, the difference i n MOE between density l e v e l 1 and 2 i s small. Generally, there i s an increase i n MOE Analysis of variance for f a c t o r i a l experiment. 37 with increasing density and a reduction i n MOE values with an increase i n the i n i t i a l p a r t i c l e moisture content at a l l density l e v e l s . As shown i n Figure 9, at wood-cement r a t i o 1:1.5, density l e v e l 3 and an i n i t i a l p a r t i c l e moisture content of 0-6% gave the highest MOE of 120.98 x 10 kg/cm . Although density l e v e l 3 consistently gave the highest MOE values at a l l le v e l s of i n i t i a l p a r t i c l e moisture content, the difference i n MOE between density l e v e l 2 and 3 i s consistently about 1.2 times, with no s i g n i f i c a n t difference at 40-50% i n i t i a l p a r t i c l e moisture content. As i n Figure 8, an increase i n the i n i t i a l p a r t i c l e moisture content resulted i n a reduction i n MOE. Figure 10 shows that at 1:2 wood-cement r a t i o , the highest MOE of 317.02 x 10^ kg/cm2 i s produced at density l e v e l 3 and 0-6% i n i t i a l p a r t i c l e moisture content. Apart from a sharp decrease i n MOE as i n i t i a l p a r t i c l e moisture content i s increased from 0-6% to 8-15% at density l e v e l 3, the rate of decrease i n MOE at the three density l e v e l s i s not s i g n i f i c a n t l y , d i f f e r e n t as confirmed by the nearly p a r a l l e l l i n e s . The step-wise regression analysis of MOE as a function of density and i n i t i a l p a r t i c l e moisture content, together with regression c o e f f i c i e n t s and standard errors of estimate are summarized i n Table 22. Figure 29 shows MOE vs density curves predicted to best f i t the data at d i f f e r e n t p a r t i c l e moisture content l e v e l s . Actual data points are shown. From Table 22 i t i s shown that density with a regression c o e f f i -cient of 2.6252, i s the main factor a f f e c t i n g MOE. Figure 29 shows that at a l l l e v e l s of p a r t i c l e moisture content MOE increases with increasing 38 density. However, the rate of increase i n MOE i s slowest at p a r t i c l e moisture l e v e l 5 and highest at p a r t i c l e moisture l e v e l 1. The res u l t s shown i n Figure 29 indicate that higher MOE i s achieved as density i s increased. 4.1.2 Modulus of Rupture (MOR) The average MOR values for the 45 treatment combinations of i n i t i a l p a r t i c l e moisture content, density and wood-cement r a t i o are summarized i n Table 9. In t h i s t e s t , again boards manufactured using 0-6% i n i t i a l p a r t i c l e moisture content, density l e v e l 3 and wood-cement r a t i o of 1:2, gave the highest MOR of 52.10 kg/cm2. From Table 9, i t i s noted that an increase i n the i n i t i a l p a r t i c l e moisture content had a general reduction e f f e c t on MOR and an increase i n density resulted i n an increase i n MOR. The r e s u l t s of the f a c t o r i a l analysis shown i n Table 17, indicate that at 0.01 p r o b a b i l i t y l e v e l , i n i t i a l p a r t i c l e moisture content A, density B and wood-cement r a t i o C are a l l highly s i g n i f i c a n t factors i n the development of bending strength i n magnesite bonded wood-cement boards. The interactions A-C and B-C are also highly s i g n i f i c a n t at 0.01 l e v e l . These interactions are gr a p h i c a l l y shown i n Figures 6 and 7. From Figure 6, i t i s observed that the highest MOR of 36.60 kg/cm2 i s produced at a wood-cement r a t i o of 1:2 and density l e v e l 3. The 1:2 wood-cement r a t i o c o n s i s t e n t l y gave the higher MOR values at a l l density l e v e l s followed by the 1:1.5 wood-cement r a t i o . However, the difference i n MOR between the 1:2 and 1:1.5 wood-cement r a t i o s , e s p e c i a l l y at density l e v e l 2, i s not very pronounced. 39 Figure 7 shows that the highest MOR of 41.90 kg/cm2 occurs at 0-6% i n i t i a l p a r t i c l e moisture content, and 1:2 wood-cement r a t i o . The 1:2 wood-cement r a t i o consistently gives the highest MOR values at a l l p a r t i c l e moisture content l e v e l s . As i n MOE, there i s a decrease i n MOR as i n i t i a l p a r t i c l e moisture content i s increased from 0-6% to 60-80%. For t h i s t e s t , the most favourable MOR (52.0 kg/cm2) i s given by the treatment combinations of 1:2 wood-cement r a t i o , 0-6% p a r t i c l e moisture content and density l e v e l 3 (Table 9). The re s u l t s of the regression analysis for MOR are summarized i n Table 22 and the predicted best f i t curves are shown i n Figure 30. From Table 22, i t i s shown that wood-cement r a t i o followed by density are the two main factors a f f e c t i n g MOR. Wood-cement r a t i o has a regression c o e f f i c i e n t of 3.3976 and density has a c o e f f i c i e n t of 3.2399. Figure 30 shows that MOR increases with increasing density. This trend i s true for a l l the i n i t i a l p a r t i c l e moisture content l e v e l s . As with MOE, the increase i n MOR as density increases i s higher i n p a r t i c l e moisture l e v e l 1 than l e v e l 5. However, the difference i n MOR increase with density i s small between 0-6% and 8-15% p a r t i c l e moisture contents, i n d i c a t i n g that the change i n MOR as i n i t i a l p a r t i c l e moisture content i s increased from 0-6% and 8-15% i s not very pronounced. 4.1.3 Internal Bond Strength (IB) Table 10 gives the average IB for the 45 treatment combinations of wood-cement r a t i o , i n i t i a l p a r t i c l e moisture content and density. Boards made with 1:2 wood-cement r a t i o at density l e v e l 3 and p a r t i c l e moisture content 0-6% produced the highest IB strength of 6.94 40 kg/cm2. I B generally increased with an increase i n density but was reduced as the i n i t i a l p a r t i c l e moisture content increased. Results of the f a c t o r i a l analysis of IB are summarized i n Table 18. The r e s u l t s show that i n i t i a l p a r t i c l e moisture content A, density B and wood-cement r a t i o C, s i g n i f i c a n t l y a f f e c t the IB strength i n magnesite cement-bonded boards. S i m i l a r l y , the interactions A-C and B-C are s i g n i f i c a n t at 0.01 l e v e l . The i n t e r a c t i o n A-B i s s i g n i f i c a n t at 0.05 l e v e l . From Figure 11, i t i s evident that wood-cement r a t i o 1:2 gives the highest IB strength of 4.82 kg/cnr at density l e v e l 3. Generally, at-a wood-cement r a t i o of 1:2 highest IB strength at a l l density l e v e l s i s obtained. There i s an increase i n i n t e r n a l bond strength with an increase i n density. Figure 12 shows that wood-cement r a t i o 1:2 con s i s t e n t l y gives higher IB strength at a l l i n i t i a l p a r t i c l e moisture content l e v e l s . As the p a r t i c l e moisture content increases, there i s a tendency for IB strength to decrease. The difference i n IB strength between the three wood-cement r a t i o s gets smaller as the i n i t i a l p a r t i c l e moisture content increases. This trend indicates that the influence of wood-cement r a t i o on IB strength diminishes as i n i t i a l p a r t i c l e moisture content increases, yet i t i s the most s i g n i f i c a n t factor i n determining IB. The regression analysis for IB strength i s shown i n Table 23. Curves that best f i t the data for the f i v e p a r t i c l e moisture content l e v e l s are shown i n Figure 31. From Table 23, i t i s evident that density and i n i t i a l p a r t i c l e moisture content are the two main factors a f f e c t i n g IB strength development i n magnesite cement-bonded boards. 41 From Figure 31, i t i s noted that at p a r t i c l e moisture content l e v e l 1, IB strength r a p i d l y increases as density increases. The rate of increase i n IB strength diminishes as the i n i t i a l p a r t i c l e moisture content i s increased from l e v e l 1 to 5. Figure 31 also shows that the difference i n the rate of increase between p a r t i c l e moisture content l e v e l s 3 and 4 and between 4 and 5 i s small. This trend indicates that as the p a r t i c l e moisture content i s increased from l e v e l 3 (25-30%) to l e v e l 5 (60-80%), the increase i n IB strength as density increases becomes lower. 4.1.4 Compressive Strength P a r a l l e l to Surface (Edgewise Compression) Table 11 summarizes the average edgewise compression r e s u l t s for the 45 treatment combinations of wood-cement r a t i o , density and i n i t i a l p a r t i c l e moisture content. In t h i s t e s t , the highest edgewise compression strength was obtained at 0-6% i n i t i a l p a r t i c l e moisture content, 1:2 wood-cement r a t i o and density l e v e l 3. As i n the other mechanical properties tested, boards made from high i n i t i a l p a r t i c l e moisture content showed low edgewise compression strength. An increase i n density, at any p a r t i c l e moisture content l e v e l , resulted i n an increase i n edgewise compression strength. F a c t o r i a l analysis of edgewise compression (Table 18) showed that the main e f f e c t s , i . e . i n i t i a l p a r t i c l e moisture content A, density B and wood-cement r a t i o C s i g n i f i c a n t l y a f f e c t edgewise compression at 0.01 l e v e l . S i m i l a r l y , the interactions B-C and A-C are highly s i g n i f i -cant at 0.01 l e v e l . 42 Figure 13 shows that the wood-cement r a t i o of 1:2 gives the highest edgewise compression strength of 22.80 kg/cm 2 at 0-6% i n i t i a l p a r t i c l e moisture content. The 1:1 wood-cement r a t i o gives the lowest edgewise compression strength of 2.80 kg/cm2 at 60-80% i n i t i a l p a r t i c l e moisture content. There i s a general decrease i n edgewise compression strength as the i n i t i a l p a r t i c l e moisture content i s increased. The 1:2 wood-cement r a t i o shows a sharp rate of change i n edgewise compression as i n i t i a l p a r t i c l e moisture content i s increased while the rate of change i n samples made with 1:1 and 1:1.5 wood r a t i o i s smallest as the i n i t i a l p a r t i c l e moisture content i s increased from 0-6% to 8-15%. This trend indicates that at 1:1 and 1:1.5 wood-cement r a t i o s , increasing the i n i t i a l p a r t i c l e moisture content from 0-6% to 8-15% has no s i g n i f i c a n t e f f e c t on edgewise compression strength. Figure 14 shows that there i s an increase i n edgewise compression strength as the density l e v e l increases. The rate of increase i n edgewise compression strength i n the 1:1 wood-cement r a t i o exhibits a slow change from density l e v e l 1 to 2 and then a sharp change from density l e v e l 2 to 3. The 1:2 wood-cement r a t i o shows no big change between density l e v e l s 1 and 2, and 2 and 3. In the 1:1.5 wood-cement r a t i o the rate of increase i n edgewise compression strength from density l e v e l 1 to 3 was found to be r e l a t i v e l y constant. This behaviour indicates that at wood- cement r a t i o 1:1, the increase i n density from l e v e l 1 to 2 has no major e f f e c t on edgewise compression strength while at 1:1.5 and 1:2 wood cement r a t i o s , edgewise compression strength i s more se n s i t i v e to an increase i n density. 43 In t h i s t e s t , the highest edgewise compression strength of 23.70 kg/cm i s attained at 1:2 wood-cement r a t i o and density l e v e l 3. Density l e v e l 1 and 1:1 wood-cement r a t i o produced the lowest edgewise compression of 3.10 kg/cm . Table 23 summarizes the r e s u l t s of the step-wise regression on edgewise compression strength. Figure 32 shows the best f i t curves at di f f e r e n t p a r t i c l e moisture content l e v e l s . Table 23 shows that density and wood-cement r a t i o are the main factors a f f e c t i n g edgewise compression strength. Figure 32 shows that samples made with 60-80% ( l e v e l 5) i n i t i a l p a r t i c l e moisture content have the least increase i n edgewise compres-sion strength as density i s increased, while 0-6% ( l e v e l 1) i n i t i a l p a r t i c l e moisture content shows the sharpest rate of increase. The increase i n i n i t i a l p a r t i c l e moisture content from l e v e l 3 to 5 (25-30% to 60-80%) does not seem to have a s i g n i f i c a n t e f f e c t on the rate of increase i n edgewise compression strength with an increase i n density. 4.2 Physical Properties 4.2.1 Moisture Content and Density of the Boards at Test Table 12 gives the average moisture content of the boards at test ( a f t e r 60 days i n the CTH room) f o r 45 treatment combinations of density, wood-cement r a t i o and i n i t i a l p a r t i c l e moisture content. Wood p a r t i c l e s (mixture of White Spruce and Jack Pine) stored under i d e n t i c a l conditions attained an equilibrium moisture content of 8.50 percent. From Table 12, i t i s noted that boards made from higher i n i t i a l p a r t i c l e moisture content furnish, tended to have a higher moisture 44 content. The v a r i a t i o n i n board moisture content between density le v e l s within a wood-cement r a t i o i s not s t a t i s t i c a l l y s i g n i f i c a n t . This trend applies to a l l p a r t i c l e moisture l e v e l s . Table 21 shows the average density of the boards at test (after 60 days i n the CTH room) f o r 45 treatment combinations of p a r t i c l e moisture content, density and wood-cement r a t i o . I t w i l l be noted from Table 21 that board density v a r i a t i o n between wood-cement r a t i o s and between p a r t i c l e moisture l e v e l s was s i g n i f i c a n t l y high. 4.2.2 Thickness Swelling 4.2.2.1 Thickness Swelling from 50% R.H. to 2 h Cold Soaking Table 13 shows the average thickness swelling a f t e r 2 h cold soaking for 45 treatment combinations of p a r t i c l e moisture content, density and wood-cement r a t i o . In t h i s test, the highest thickness swelling of 16.19% a f t e r 2 h cold soaking was obtained on the samples made with p a r t i c l e s having t h e i r i n i t i a l moisture content set at 0-6%, and formulated to a density l e v e l 3 and using a wood-cement r a t i o of 1:1. Generally an increase i n the i n i t i a l p a r t i c l e moisture content resulted i n reduced thickness swelling, while an increase i n density l e v e l had no s i g n i f i c a n t e f f e c t on thickness swelling. F a c t o r i a l analysis r e s u l t s of thickness swelling a f t e r 2 h cold soaking (Table 19) indicate that i n i t i a l p a r t i c l e moisture content A, wood-cement r a t i o C and the i n t e r a c t i o n A-C are highly s i g n i f i c a n t at 0.01 l e v e l . From Figure 15 the highest thickness swelling of 15.170% af t e r 2 h of cold soaking i n water i s produced by a treatment 45 combination of 1:1 wood-cement r a t i o and 0-6% i n i t i a l p a r t i c l e moisture content. The lowest thickness swelling of 1.00% i s given by 1:2 wood-cement r a t i o and 60-80% p a r t i c l e moisture content. The 1:1 wood-cement r a t i o shows a sharp decrease i n thickness swelling as the i n i t i a l p a r t i c l e moisture content increased. S i m i l a r l y , the 1:1.5 and 1:2 r a t i o s exhibit a decrease i n thickness swelling with increasing i n i t i a l p a r t i c l e moisture content but only up to 40-50% p a r t i c l e moisture con-tent l e v e l . Thereafter the 1:1.5 r a t i o shows a s i g n i f i c a n t increase i n thickness swelling whereas the 1:2 r a t i o shows no s i g n i f i c a n t increase i n thickness swelling. It i s thought that t h i s trend i s due to the v a r i a b i l i t y i n thickness swelling within and between the cement boards. Wood-cement r a t i o s 1:1.5 and 1:2 show (Figure 15) a small d i f f e r -ence i n thickness swelling at an i n i t i a l p a r t i c l e moisture content of 25-30% and 40-50%. Generally, the difference i n thickness swelling between the three wood cement r a t i o s gets smaller with increasing i n i t i a l p a r t i c l e moisture content. 4.2.2.2 Thickness Swelling from 50% R.H. to 24 h Cold Soaking The average thickness swelling for 45 treatment combinations of i n i t i a l p a r t i c l e moisture content, density and wood-cement r a t i o are shown i n Table 14. As i n the 2 h t e s t , boards made with 1:1 wood-cement r a t i o , 0-6% i n i t i a l p a r t i c l e moisture content and density l e v e l 3 gave the highest thickness swelling of 18.80%. Generally an increase i n the i n i t i a l p a r t i c l e moisture content resulted i n a decrease i n thickness swelling whereas, an increase i n density resulted i n a s l i g h t increase i n 46 thickness swelling. Surp r i s i n g l y , extending the swelling time to 24 h resulted i n l i t t l e further swelling from that achieved i n 2 h cold soaking. This observation indicates that 2 h cold soaking i s adequate i n the determination of thickness swelling i n magnesite cement boards. In Table 19, the f a c t o r i a l analysis r e s u l t s show that the main e f f e c t s of i n i t i a l p a r t i c l e moisture content A and wood-cement r a t i o C, are highly s i g n i f i c a n t at 0.01 l e v e l . S i m i l a r l y , the i n t e r a c t i o n between i n i t i a l p a r t i c l e moisture content A, and wood-cement r a t i o C i s highly s i g n i f i c a n t . Figure 16 shows a s i m i l a r trend to Figure 15. The treatment combination of 1:1 wood-cement r a t i o and i n i t i a l p a r t i c l e moisture content set at 0-6% gives the highest thickness swelling value of 16.66%. The 1:2 wood-cement r a t i o and i n i t i a l p a r t i c l e moisture content of 60-80% produced the lowest thickness swelling. Figure 16 shows that a l l the wood-cement r a t i o s e x h ibit a decrease i n thickness swelling with increasing i n i t i a l p a r t i c l e moisture content up to 40-50%. Thereafter the 1:1.5 wood-cement r a t i o shows a sharp increase i n thickness swelling whereas wood-cement r a t i o s 1:1 and 1:2 show a slow increasing trend i n thickness swelling. As observed i n the 2 h cold soaking t e s t , the increase i n thickness swelling from 40-50% to 60-80% i n i t i a l p a r t i c l e moisture, i s thought to be due to v a r i a b i l i t y between and within the boards. With boards made of p a r t i c l e s having i n i t i a l moisture content 25-30%, the difference i n thickness swelling between the 1:1.5 and 1:2 wood-cement r a t i o s i s very small (Figure 16). 47 4.2.3 Water Absorption 4.2.3.1 Water Absorption from 50% R.H. to 2 h Cold Soaking Table 15 summarizes the average water absorption r e s u l t s for the 45 treatment combinations of p a r t i c l e moisture content, density and wood-cement r a t i o . Boards made from the 1:1 wood-cement r a t i o , density l e v e l 1 and i n i t i a l p a r t i c l e moisture content of 0-6% produced the highest water absorption value of 70.20%. Generally, an increase i n the i n i t i a l p a r t i c l e moisture content resulted i n a decrease i n water absorption. It w i l l be noted that the treatment combination producing the highest water absorption values i n the 2 h cold soaking test did not correspondingly produce the highest thickness swelling. This trend i s thought to be due to the l o s s of the excess polyphosphate solids which dissolve i n water during the soaking. The f a c t o r i a l analysis r e s u l t s of water absorption a f t e r 2 h cold soaking test are summarized i n Table 20. The r e s u l t s indicate that the main e f f e c t s namely, i n i t i a l p a r t i c l e moisture content A, density B, and wood-cement r a t i o C and the interactions B-C and A-C are highly s i g n i f i -cant at 0.01 l e v e l . From the A-C i n t e r a c t i o n s , which are shown graphi-c a l l y i n Figure 18, the 1:1 wood-cement r a t i o gives the highest water absorption of 64.70% at 0-6% i n i t i a l p a r t i c l e moisture content. The 1:2 wood-cement r a t i o gives the lowest water absorption value of 19.99% at 60-80% i n i t i a l p a r t i c l e moisture content. The 1:1 wood-cement r a t i o c o n s i s t e n t l y gives the highest water absorption percentage at a l l l e v e l s 48 of i n i t i a l p a r t i c l e moisture content, followed by the 1:1.5 wood-cement r a t i o . Generally, as the i n i t i a l p a r t i c l e moisture content increases the water absorption rate decreases. This trend i s true for a l l wood-cement r a t i o s . Figure 17 shows that the 1:1 wood-cement r a t i o gives the highest water absorption value of 60.74% at density l e v e l 1. The 1:2 wood-cement r a t i o gives the lowest water absorption value of 21.26% at density l e v e l 3. There i s a general decrease i n water absorption with an increase i n density (Figure 17). The 1:1 wood-cement r a t i o consis-t e n t l y gives the highest water absorption percentage at a l l density l e v e l s . 4.2.3.2 Water Absorption from 50% R.H. to 24 h Cold Soaking Table 16 shows the average water absorption values for 45 treatment combinations of wood-cement r a t i o , density and i n i t i a l p a r t i c l e moisture content. In t h i s t e s t , the highest water absorption of 95.20% was obtained on the boards made of f u r n i s h having an i n i t i a l p a r t i c l e moisture con-tent of 0-6%, a 1:1 wood-cement r a t i o and density l e v e l 1. As i n the 2 h cold soaking t e s t , there i s a decrease i n water absorption with increasing i n i t i a l p a r t i c l e moisture content. S i m i l a r l y there i s a decrease i n water absorption with both increasing wood-cement r a t i o and density l e v e l . Results of the f a c t o r i a l analysis of water absorption a f t e r 24 h immersion i n water (Table 20) indicate that the main e f f e c t s , p a r t i c l e 49 moisture content A, density B, wood-cement r a t i o C and the interactions A-C and B-C are highly s i g n i f i c a n t at 0.01 l e v e l . The i n t e r a c t i o n A-B i s s i g n i f i c a n t at 0.05 l e v e l . Figure 20 shows that the highest water absorption of 85.67% occurs at 1:1 wood-cement r a t i o and 0-6% i n i t i a l p a r t i c l e moisture content. The lowest water absorption of 25.39% i s given by boards made with the 1:2 wood-cement r a t i o at 60-80% i n i t i a l p a r t i c l e moisture content. The 1:1 wood-cement r a t i o c o n s i s t e n t l y r e s u l t s i n higher water absorption at a l l l e v e l s of i n i t i a l p a r t i c l e moisture followed by the 1:1.5 wood-cement r a t i o . A l l the wood-cement r a t i o s exhibit a drop i n water absorption as the i n i t i a l p a r t i c l e moisture content i s raised from 0-6% to 60-80%. Conversely, samples prepared with p a r t i c l e s of high i n i t i a l moisture content absorbed less water even on prolonged soaking. Figure 19 shows that water absorption decreases with increasing density. The 1:2 wood-cement r a t i o shows a constant decrease i n water absorption with increase i n board density, while samples made from the 1:1 and 1:1.5 wood-cement r a t i o s exhibit a slow i n i t i a l decrease followed by a sharp decrease from density l e v e l 2 to 3. The highest water absorption of 79.40% for the B-C i n t e r a c t i o n i s produced at 1:1 wood-cement r a t i o and density l e v e l 1. 4.3 Scanning E l e c t r o n Microscopy (SEM) Observations The r e s u l t s of the SEM study are shown on the electron micrographs in Figures 21 to 28. Figures 21, 22 and 23 i l l u s t r a t e c r y s t a l formations i n the pure magnesium oxide powder, dead burnt magnesite cement powder and dried 50 ammonium polyphosphate r e s p e c t i v e l y . Pure magnesium oxide powder exhibits well-defined oval-shaped c r y s t a l s (Figure 21), whereas c r y s t a l s are s c a l e - l i k e i n the dead burnt magnesite cement powder (Figure 22). Ammonium polyphosphate (Figure 23) i s characterized by well defined rod-l i k e c r y s t a l s . Figures 24 to 28 serve to i l l u s t r a t e the e f f e c t of i n i t i a l p a r t i c l e moisture content on c r y s t a l formations i n dead burnt magnesite cement-bonded particleboard manufactured using a 1:1 wood-cement r a t i o and at density l e v e l 1. Figure 24 shows electron micrographs of boards made from furnish with an i n i t i a l p a r t i c l e moisture content of 0-6%. Well defined p l a t e - l i k e , near rectangular c r y s t a l formations are observed i n Figure 24(i) and the great amount of wood f a i l u r e i s observed i n Figure 2 4 ( i i ) . Boards manufactured from p a r t i c l e s having 8-15% i n i t i a l moisture content (Figure 25), show s i m i l a r c r y s t a l formations to those observed i n Figure 24. However, the c r y s t a l formations i n Figure 25, though p l a t e - l i k e i n appearance, are c i r c u l a r and c l o s e l y packed together. Figure 26 shows a well-defined spade-like c r y s t a l formation i n boards manufactured using 25-30% i n i t i a l p a r t i c l e moisture content. Figure 27 and 28 represent electron micrographs for boards manufactured using 40-50% and 60-80% i n i t i a l p a r t i c l e moisture content respectively. In Figures 27 and 28 there i s no c r y s t a l formations observed. From Figures 24 to 28, i t has been shown that the oxyphosphate, which i s the reaction product of dead burnt magnesite cement and ammonium polyphosphate has c r y s t a l formations which are d i s s i m i l a r to the c r y s t a l s of the reacting compounds shown i n Figures 21 to 23. These micrographs indicate that successful high strength bond formation between cement and wood i s strongly moisture s e n s i t i v e . Figures 24 to 26 indicate that the f u l l y developed and well defined c r y s t a l formations appear to r e s u l t i n successful bond formation between cement and wood furnish having a r e l a t i v e l y low i n i t i a l moisture content. On the other hand, Figures 27 and 28 shows no c r y s t a l formation, i n d i c a t i n g that the high i n i t i a l p a r t i c l e moisture content of 40-50% and 60-80% respectively, i n t e r f e r e d with the successful bonding between the wood and the cement. The poor bond formation observed i n Figures 27 and 28, can be at t r i b u t e d to the premature p r e c i p i t a t i o n of either the phosphate or oxyphosphate complex due to the excess water i n the wood furnish. From Table 10, i t i s observed that there i s a decrease i n i n t e r n a l bond strength as the i n i t i a l p a r t i c l e moisture content increases from 0-6% to 60-80%. Treatment combinations represented by Figures 24, 25 and 26 show a higher i n t e r n a l bond strength than those represented by Figures 27 and 28. This observation indicates that f u l l y developed and well defined c r y s t a l formation r e s u l t s i n excellent bond strength. The high i n i t i a l p a r t i c l e moisture content i n Figures 27 and 28 in t e r f e r e d with the c r y s t a l l i z a t i o n process which i s evidently associated with high strength development. 4.4 Probable Factors Accounting f o r the Trends i n Treatment 4.4.1 Mechanical Properties Results obtained and presented i n Tables 8 to 11 and those of the analysis of variance (Tables 17 and 18) and the step-wise regression (Tables 22 and 23), show beyond doubt that the main e f f e c t s of i n i t i a l 52 p a r t i c l e moisture content, density and wood-cement r a t i o and t h e i r i n t e r a c t i o n s , have a s i g n i f i c a n t e f f e c t on the development of mechanical strength i n magnesite cement-bonded boards. In a l l cases, increase i n the i n i t i a l p a r t i c l e moisture content resulted i n a decrease i n the mechanical properties of the boards (Figures 6, 7, 8, 9, 10, 12, 13). It i s evident that moisture i n t e r -feres either with the bonding between the wood p a r t i c l e s and magnesite cement or se r i o u s l y a f f e c t the strength and structure of the cement. The scanning electron microscopy study undertaken (Section 4.3) confirms the adverse e f f e c t of the high i n i t i a l p a r t i c l e moisture content on the development of the mechanical properties of magnesite cement bonded boards. Won-Yung et_ a l . (56) have studied the bond formation i n wood -Portland cement systems using scanning electron microscopy. They concluded that the bonding between the wood fragments and Portland cement was achieved by a mechanical i n t e r l o c k i n g mechanism. They at t r i b u t e d the mechanical i n t e r l o c k i n g phenomenon to the c r y s t a l s that develop during the hydration of Portland cement, once water i s added to the wood-cement mix. Parameswaran ejt a l . (34) also studied the i n t e r a c t i o n between Portland cement with wood p a r t i c l e s i n wood-Portland cement composites, using scanning electron microscopy and X-ray microanalysis. They observed that the inorganic substances of the Portland cement af f e c t the woody tissue i n that small p a r t i c l e s penetrate into the c e l l w all. They noted that the impregnation with a l k a l i n e Portland cement solutions appeared to r e s u l t i n fin e s t r u c t u r a l changes of the c e l l w all, which 53 also included the c e l l u l o s e component. I t i s thought that t h i s i n t e r -action between the wood fragment and mineral binder i s responsible for enhanced d u r a b i l i t y of wood-cement composites against microorganisms, and noncombustibility of the wood. In the board forming process employed i n t h i s study, the bonding between the wood fragments and the magnesite cement binder, as described by Paszner (37), d i f f e r s s l i g h t l y to that described by Won-Yung et a l . (56) and Parameswaran et^ a l . (34). Paszner (37) reported that the ammonium polyphosphate aqueous solutions penetrate the wood fragment, and subsequent drying r e s u l t s i n the retention of a c r y s t a l l i n e residue s a l t deposit within lumina and pores. These deposits were found to greatly enhance strength and flame retardant properties of the boards, the improvements generally increasing with loading up to the physical l i m i t . Paszner (37) also observed that when an adhered mineral deposit of metal oxyphosphate i s attached to the wood surface as a continuous anchored layer, the impairment of strength r e s u l t i n g from f i s s u r i n g and cracking produced by cutting and crushing processes of forming wood fragments i s not only o f f s e t , but the wood fragment gains i n t r i n s i c bending and compressive strengths, exceeding by about 100% (aspen) the a i r - d r y values for clear whole-wood specimens. The gain i n mechanical properties i s thought to be due to the presence of a s h e l l of mineral material at the fragment surface, through which the generated oxyphos-phate compounds have migrated for distances 30 to 150 microns p r i o r to the setting of the compound (37). In the Paszner process, the formation of the s h e l l of mineral-vegetable layer at the wood fragment surface i s followed by the c r y s t a l -54 l i z a t i o n of the c o l l o i d a l - metal oxyphosphates which have migrated into minute surface openings. C o l l o i d a l migration appears to be effected both by c a p i l l a r y transport of the suspending l i q u i d or by mechanically-induced pressure gradients therein and by e l e c t r o s t a t i c forces within the c a t a l y s t s o l u t i o n i . e . by Brownian movement (37). The c r y s t a l formation was confirmed by the res u l t s of the SEM study (Section 4.3). It i s therefore evident from Paszner's (37) observation above, that wood fragments, whose c e l l lumina are saturated with water, w i l l not be available for impregnation by the ammonium polyphosphate (reactant). I f some c a t a l y s t does p a r t i a l l y impregnate a nearly saturated c e l l lumen, the d i l u t i o n e f f e c t of the c e l l lumen water w i l l tend to wash away the s a l t s of the reactant and make these unavailable for bonding. This w i l l have an adverse e f f e c t on the amount of c r y s t a l l i n e s a l t deposi-t i o n . S i m i l a r l y , i t i s suggested that when the reactant i s applied to the surface of a wood fragment of high moisture content, followed by deposition of powdered magnesite cement, a high f l u i d i t y of the l i q u i d / s o l i d mixture w i l l r e s u l t p r i o r to gelation. This high f l u i d i t y , i t i s suggested, w i l l tend to make the cladding thickness uneven due to dripping, flowing and squeeze-out into i n t e r s t i c a l c a v i t i e s , thereby also decreasing the cement layer thickness between the p a r t i c l e s . The reduction of cladding thickness and decreased cement penetra-t i o n into the wood structure and the absence of properly c r y s t a l l i z e d , high-strength s a l t deposits on the f i b r e surface having had high i n i t i a l moisture content, are thought to be the two most important factors responsible for the observed reduction i n mechanical properties of boards made of wood p a r t i c l e s having high i n i t i a l moisture content. 55 It i s thus obvious that for highest mechanical strength boards to be made, the i n i t i a l p a r t i c l e moisture content must be as low as possible. The p a r t i c l e moisture content e f f e c t appears to be more pronounced with high wood-cement r a t i o s (Figure 7). There also appears to be no break on the p a r t i c l e moisture content/MOR curves and hence no minimum i n i t i a l p a r t i c l e moisture content above 15% can be recommended. The r e s u l t s appear to be highly consistent and the interactions highly s i g n i f i c a n t . 4.4.2 P h y s i c a l Properties Results presented i n Tables 13 and 14 show that there i s an increase i n thickness swelling from 2 to 24 h cold soaking of magnesite cement-bonded boards previously conditioned to 50% R.H. In Table 19 i t i s shown that i n i t i a l p a r t i c l e moisture content A, wood-cement r a t i o C and the interactions of A and C a l l have a highly s i g n i f i c a n t e f f e c t on the board thickness swelling following 2 to 24 h cold soaking t e s t , respectively. On the other hand, density B and the i n t e r a c t i o n of A-B and B-C do not appear to a f f e c t s i g n i f i c a n t l y the thickness swelling of the boards following 2 to 24 h cold soaking t e s t s . Broker £t jal. (8) investigated the dimensional changes (thickness swelling and shrinkage) of Portland cement-bonded boards. The r e s u l t s of t h e i r i n v e s t i g a t i o n showed that the shrinkage and swelling of the Portland cement-bonded boards was independent of the wood component, but was strongly influenced by the shrinkage and swelling of the porous hydrated Portland cement. 56 However, r e s u l t s of the present study show that both the wood-cement r a t i o and the wood fragment have an e f f e c t on the thickness swelling of the magnesite cement-bonded boards. Figures 15 and 16, show that at a l l l e v e l s of wood-cement r a t i o , there i s a decrease i n thick-ness swelling as i n i t i a l p a r t i c l e moisture content increased. Table 12 shows that a f t e r conditioning the boards at 50% R.H. f o r 60 days, the equ i l i b r a t e d moisture content of the boards showed an increasing trend as the p a r t i c l e moisture content increased from 0-6% to 60-80%. Further, an increase i n the equilibrium moisture content of the boards was noted as the wood-cement was increased from 1:1 to 1:2. It i s therefore evident that since the wood fragments were p a r t i a l l y f i l l e d with water, at the time of t h e i r encasement i n the r i g i d oxyphosphate s h e l l , t h e i r capacity to absorb more water and swell further i s li m i t e d , r e s u l t i n g i n l i t t l e to no thickness swelling occurring. Regarding the e f f e c t of the wood-cement r a t i o , i t i s suggested that a high cement f r a c t i o n r e s u l t s i n the formation of a superior gel mass which i s impermeable to water. It i s quite apparent that moisture absorption of the boards i s cont r o l l e d by the e f f e c t s of both the i n i t i a l p a r t i c l e moisture content and the wood-cement r a t i o . Since water absorption r e s u l t s i n swelling, boards made of p a r t i c l e s bonded i n t h e i r maximum swollen condition w i l l show a lesser tendency to pick up water. Unfortunately, t h i s better dimensional s t a b i l i t y property of the magnesite cement boards made of high moisture content f i b r e s does not coincide with tendencies of maximum mechanical strength development. Thus, a compromise must be reached as to which parameter should be maximized i n the product. 57 Obviously, the decision w i l l depend on the most c r i t i c a l requirement i . e . strength versus dimensional s t a b i l i t y . 4.5 Comparison of Study Results with some National Standards and  Other Tests 4.5.1 German Standard f or Cement-bonded Wood Particleboard The product s p e c i f i c a t i o n s of the German Standard DIN 52 361 and DIN 52 362 for cement-bonded particleboard are shown i n the following table: A. German Standard f o r cement-bonded wood particleboard. Wood p a r t i c l e Binder type Density M.O.E 3 2 3 gm/cm kg/cm xlO (DIN 52 361) (DIN 52 362) Flakes/shavings Magnesium cement 0.9-1.50 70-130 Flakes/shavings Portland cement 1.0-1.35 60-150 Wood Pulp Gypsum < 1.0 40-70 Treatment combinations from t h i s study that meet the above exterior grade cement-bonded particleboard requirements for MOE are summarized i n the following table: 58 B. Test MOE r e s u l t s that meet German Standard DIN 52 362 f o r magnesite cement-bonded particleboard. Wood/ Density P a r t i c l e moisture content % cement l e v e l r a t i o 0-6 8-15 25-30 40-50 60-80 MOE, kg/cm2 x 10 3 1:1.0 2 _ _ _ _ _ 3 81.187 72.803 67.437 53.563 39.923 1 77.553 47.610 42.237 1:1.5 2 110.393 99.800 73.950 61.960 42.917 3 120.980 111.883 86.853 67.670 55.253 1 136.410 117.017 93.377 41.623 1:2.0 2 156.777 140.917 126.523 88.647 55.633 3 317.017 187.333 163.873 119.680 96.169 As noted from the comparison above, thirty-two of the f o r t y - f i v e treatments meet the s p e c i f i c a t i o n requirements f o r the German Standard for modulus of e l a s t i c i t y . It i s evident from the above comparison that boards with a higher density and wood-cement r a t i o gave the highest MOE and hence met the standard requirement for cement bonded wood particleboard. It i s in t e r e s t i n g to note that denser boards made from very wet p a r t i c l e s met the s p e c i f i c a t i o n requirements. The implication of t h i s observation i s that i t may not be necessary to dry wood p a r t i c l e s p r i o r to manufacture of magnesite cement-bonded particleboard i n order to meet the German 59 s p e c i f i c a t i o n for M.O.E. This can r e s u l t i n considerable heat energy savings. 4.5.2 Canadian Standard f o r Waferboard No s p e c i f i e d standard ex i s t s f o r cement-bonded particleboard i n Canada and the USA. In t h i s section an attempt w i l l be made to compare the test r e s u l t s of the present study to waferboard standard s p e c i f i c a -t i o n s . Waferboard has been chosen for comparison because of i t s favour-able growth i n the e x t e r i o r grade panel market. The cement-bonded particleboard i s also intended for use as an exterior grade panel. The properties s p e c i f i c a t i o n s of the Canadian Standard Can 3 -0188.0-M78 for waferboard are as follows: C. Canadian Waferboard Standard. Property kg/cm' MOE 27,530.00 MOR 143.00 Tensile strength perpendicular to surface (IB) 2.85 Treatment combinations from t h i s study, that meet the above requirements are as follows: 1. MOE Forty-one out of f o r t y - f i v e treatment combinations meet the minimum requirements s p e c i f i e d by Canadian Waferboard Standard 60 for MOE. This implies that for use categories where MOE i s of prime consideration the magnesite cement-bonded boards can favourably compete with waferboard. The treatment combinations that meet the MOE Canadian Waferboard Standard are indicated i n Table 8. From Table 9, there i s no treatment combination that meets t h i s Standard. 3. Internal Bond Strength From the comparison i n Table 10, eleven out of the 45 t r e a t -ments combinations from the study, meet the Canadian Waferboard Standard requirements for IB strength. 4.5.3 International Organization f o r Standards (ISO) Water absorption and thickness swelling of building boards (wood-based panel boards), when calculated according to I.S.O. P u b l i c a t i o n 769 s h a l l not exceed the following values: D. ISO requirement for water absorption and thickness swelling i n b u i l d i n g boards. 2. MOR Max water absorbed %1 Max swelling i n thickness % Building board 40.00 20.00 Based on immersion i n 20°C water for 24 h + 15 min. 61 Treatment combination from the study that meet the above requirement are as follows: 1. Thickness swelling a f t e r 24 h cold soaking (Table 14) A l l the 45 treatment combinations meet the I.S.O. bu i l d i n g board requirements i n thickness swelling. 2. Water absorption a f t e r 24 h cold soaking. Treatment combinations from the study that meet I.S.O. requirements are indicated i n Table 16. Out of the 45 treatments, 18 treatment combinations meet the requirements. As noted i n Figures 19 and 20 a decrease i n water absorption r e s u l t s as the board density and p a r t i c l e moisture content increases. Thus, f o r manufacture of boards with acceptable water absorption and thickneses swelling the higher wood-cement r a t i o s and higher density are favourable for t h i s purpose. The i n i t i a l p a r t i c l e moisture content should be as high as possible. 4.5.4 General In tests c a r r i e d out i n Europe, Deppe (14) reported that s i n g l e -layer cement-bonded wood chipboard (type DURIPANEL), i n comparison with resin-bonded panel boards, are weaker i n modulus of rupture, modulus of e l a s t i c i t y and i n t e r n a l bond strength. On the other hand, he reported that thickness swelling was much lower i n the single-layered cement-bonded boards than i n resin-bonded boards. For example, tests by 62 E.M.P.A. of Zurich-Dubendorf and the Universtat Karlsruhe and res u l t s reported by Deppe (14) showed the following properties for Portland cement-bonded and resin-bonded particleboard (resin-bonded boards i n brackets): (a) MOE 20,000 (32,000) kg/cm2 (b) MOR 150 (220 + 12) kg/cm2 (c) I.B. strength 5.0 (4.0) kg/cm2. (d) Thickness swelling % a f t e r 24 h 2.7 (11.0) (e) Shrinkage ( a i r dry to normal) % 0.15 (0.25 to 0.50) (f ) Board density 1100 (750 + 25) kg/cm3 (g) Board thickness (mm) 11.5 + 4 (13.0 + 0.3). From the above comparison, no MOR treatment combination from the study met the test r e s u l t s reported. Forty-one, f i v e , and forty-one treatment combinations met the test r e s u l t s reported above for MOE, IB and thickness swelling, r e s p e c t i v e l y . 63 5.0 SUMMARY AND SUGGESTIONS FOR FURTHER STUDY 5.1 Summary The information l i s t e d below contributes to the knowledge of the e f f e c t s of i n i t i a l p a r t i c l e moisture content on the physical and mechanical properties of magnesite cement-bonded boards. 1. The analysis of variance and regression analysis performed have established that the i n i t i a l p a r t i c l e moisture content has a pronounced e f f e c t on the physical and mechanical properties of magnesite cement bonded boards. An increase i n i n i t i a l p a r t i -c l e moisture content resulted i n a decrease i n the physical and strength parameters tested. However, high i n i t i a l p a r t i c l e moisture content i s desirable i f boards of low water absorption capacity are to be manufactured. The difference i n most physical and mechanical properties tested has tended to be small between the 0-6% and 8-15% i n i t i a l p a r t i c l e moisture content, thus no p a r t i c l e moisture content of more than 15% i s recommended. 2. S i m i l a r l y , the e f f e c t of wood-cement r a t i o (a major factor i n manufacturing cost) i s found to be highly s i g n i f i c a n t i n a l l the ANOVAs c a r r i e d out. The e f f e c t of density i s found highly s i g n i f i c a n t i n 6 of the 8 ANOVAs. However, the e f f e c t of density on thickness swelling from 50% r e l a t i v e humidity to 2 h and 24 h cold soaking was overshadowed by the s i g n i f i c a n t e f f e c t of the i n i t i a l moisture content of the f i b r e f u r n i s h . 64 This e f f e c t c e r t a i n l y would warrant closer anatomical and microscopic i n v e s t i g a t i o n . 3. The study has demonstrated that boards with high mechanical properties r e s u l t as the wood-cement r a t i o and density increase. Wood cement r a t i o 1:2 at density l e v e l 3 gave the highest mechanical properties followed by 1:1.5 wood-cement r a t i o . However, magnesium oxyphosphate bonded boards did not reach the MOR values reported for DURISOL boards. This i s possibly due to the lower density of the boards at equivalent wood-cement r a t i o s . 4. Thirty-two out of the 45 treatments meet the requirements of the German Standard for the 3 categories of cement-bonded boards. However, only 21 of the treatment combinations pass the German requirements of the magnesite-bonded particleboards. It should be noted that at 1:2 wood cement r a t i o , boards made from wood p a r t i c l e s with 40-50% and 60-80% moisture content pass the German Standard, while at the lower wood-cement r a t i o (1:1) only those boards made from lower p a r t i c l e moisture content (0-6% and 8-15%) pass the standard for cement-bonded boards. 5. The physical properties tested (thickness swelling and water absorption) are found to worsen with increasing wood-cement r a t i o , density and decrease i n p a r t i c l e moisture content. It i s found that 18 out of the 45 treatment combinations pass the 65 board water absorption of the I.S.O. standard while a l l the 45 treatments pass the thickness swelling I.S.O. standard requirement. 6. I t i s found that the boards manufactured i n t h i s study, e x h i b i t , i n comparison to resin-bonded boards: (a) s u b s t a n t i a l l y lower MOR and comparable MOE (b) but possess higher i n t e r n a l bond strength, lower thickness swelling and water absorption (a favourable occurrence). 5.2 Suggestions f o r Further Study Conclusive evidence has been presented for the e f f e c t of the i n i t i a l p a r t i c l e moisture content on the physical and mechanical proper-t i e s of magnesium oxyphosphate cement-bonded particleboard. The study has also shown that depending on treatment combinations of p a r t i c l e moisture content, density and wood-cement r a t i o , boards that meet the I.S.O. and German standard s p e c i f i c a t i o n s for s i m i l a r boards can be manufactured. An immediately suggested area of study i s as follows: The cost of dead burnt magnesium oxide-based cement powder i s i n the region of US$0.17/lb. This cost i s considerably high. However, i t has been shown i n i n i t i a l tests by Paszner (37) that an i n e r t material such as raw powdered dolomite can successfully replace a major part of the more expensive magnesium oxide. It i s suggested that a study should be c a r r i e d out to determine the l e v e l of magnesium oxide replacement with dolomite at which undersirable loss i n mechanical and physical properties occurs. Such a study would have a far-reaching impact on the economics of the board manufactured by t h i s technology. 66 6.0 CONCLUSION The findings of t h i s study are considered to be of i n t e r e s t to the would-be manufacturers of the cement-bonded particleboard using phos-phate activated dead burnt magnesite cement as the binder. There i s no doubt that the short curing time, coupled with the favourable physical and mechanical properties of the boards made by t h i s process w i l l be of great i n t e r e s t i n both developed and developing countries. The boards manufactured i n t h i s study show properties which meet the European products codes and compare favourably with products made by the conventional Portland cement technology. It i s , however, noted that due to the inertness of the cement bonding process to wood sugars and phenolics, a l l wood species can be used for the process. The rapid room temperature curing c h a r a c t e r i s t i c s of the phosphate activated magnesite process provides further advantages i n product processing and reduced material handling within the plant. In order to manufacture ammonium polyphosphate activated magnesite cement boards of favourable mechanical properties, an i n i t i a l p a r t i c l e moisture content of not more than 15% i s recommended, together with 1:2 wood-cement r a t i o at density l e v e l 3. Favourable dimensional s t a b i l i t y properties (thickness swelling) are achieved when an i n i t i a l p a r t i c l e moisture of more than 25% and 1:2 wood-cement r a t i o with density l e v e l 3 are used. 67 7.0 REFERENCES 1. Aisher, O.A., G.E. Bessey, W.K. Blenkinsopp, M.J. Preedlander, and E. Morris. 1945. The German wood-wool building slab industry. F i n a l Rep. B r i . I n t e l l . Obj. Sub-comm. No. 956, Item No. 22. 44 pp + f i g s . 2. Anon. 1943. Sawdust-cement. Notes from the Information Bureau of the Building Research Station. J. Royal Inst, of A r c h i t e c t s . Sept. p. 259-263. 3. Anon. 1968. The manufacture of resin-bonded wood-wool products. Inf. B u l l . l/IB/13. Timber Res. Dev. Ass., Hughenden Valley, High Wycombe, Bucks. Feb. 24 pp. 4. Anon. 1971. Production of Prefab Wooden Houses. ID/61 (ID/WG 49/5, Rev. 1). UNIDO, New York. 94 pp. 5. B i b l i s , E.J. and Lo Chen-Fan. 1968. Sugars and other wood extrac-ti v e s e f f e c t on the setting of southern pine-cement mixtures. For. Prod. J . 18(8):28-34. 6. Bison-Werke. 1977. Cement-bonded particleboard. F.A.O. P o r t f o l i o of small-scale wood-based panel plants. 24 pp. 7. Brauner, S. and L.B. Copeland. 1964. The chemistry of concrete. Sc. Amer. 210(4):81-92. 8. Brooker, F.W. and M.H. Simatupang. 1974. Causes of dimensional changes of wood-cement materials. Holz-als-Roh Werkstoffe. 32(4):150-155. 9. Chittenden, A.E., A.J. Hawkes, and H.R. Hamilton. 1974. Wood-cement systems. F.A.O. FO/WCWBP/75. Doc. No. 99. New Delhi, India. 20 pp. 10. Christensen, L.E. 1948. The e f f e c t of sugars and wood extractives on the properties of Portland cement mixtures. B.S. thesis, U n i v e r s i t y of Wisconsin. 11. Clare, M.E. and P.T. Cherwood. 1956. Further studies on the ef f e c t of organic matter on the setting of soil-cement mixtures. J. Appl. Chem. 6:317-324. 12. Dass, A. and J . George. 1969. S u i t a b i l i t y of some Indian timbers for Portland cement-bonded woodwool boards. J. Timb. Developm. Ass. India 15(2):21-26. 13. Davis, T.C. 1966. E f f e c t of blue s t a i n on setting of e x c e l s i o r -cement mixtures. For. Prod. J . 16(6):49-50. 68 14. Deppe, H.J. 1975. On the production and a p p l i c a t i o n of cement-bonded wood chipboards. Proceedings of the 8th Particleboard Symposium, Washington State Univ., Pullman, Wash, pp. 267-287. 15. Dinwoodie, J.M. 1978. Wood-cement particleboard. B.R.E. Infor-mation Report. Princes Risborough Lab. IS 2/78. 2 pp. 16. Elmendorf, A. and T.W. Vaughan. 1963. Newly developed mineral-bonded panel products. F.A.O.PPP CONS/PAPER 3.3. 18 pp. 17. Ernsberger, F.M. and W.G. France. 1945. Portland cement disper-sion by absorption of calcium lignosulphonate Industr. Engin: Chem. 37(6):593-600. 18. F.A.O. 1963. Plywood and other wood-based panel products. F.A.O. Rep. Int. Consult., Rome. 8-19 July. 223 pp. 19. Fischer, F., 0. Weinhaus, M. Ryssel, and J . Olbrecht. 1974. The water soluble carbohydrates of wood and t h e i r influence on the production of l i g h t weight wood-cement boards. Holztechnologie 15(1):12-19. 20. Garland, H. 1949. Aspen for e x c e l s i o r . St. Paul Lake State For. Experim. Station. Aspen Report No. 16. 10 pp. 21. Geleff, S. 1948. Wood-wool concrete. Indian Concrete J . , Bombay, India 22(12):271-274. 22. Hobbs, C. 1964. Concrete additives from the viewpoint of the large contractor. Chem. and Industry Mar. 28:526-535. 23. Hubbard, E. 1920. U t i l i z a t i o n of wood waste.' Third Revised E d i t i o n . H.B. Stocks Ed. (translated from German from the second revised and enlarged e d i t i o n ) . Scott Green Wood and Son, London, England, p. 218-228. 24. Kamil, N. 1970. F e a s i b i l i t y of establishing wood-wool board industry i n Indonesia. Lembaga-Lembaga, Penelitan Kehutan, Pengumunan No. 95, Bangor Indonesia. 56 pp + 23 pp f i g s , (translated at UBC). 25. K e i l , F. 1972. Zement, Herstellung und Eigenschaften. Springer Varlag, B e r l i n , Heidelberge, N. York. pp. 1-10. 26. Kleinogel, A. 1950. Influences of concrete. (Translated by F.S. Morgantroth), Fredrick Unger Publish. Co., New York. 279 pp. ( O r i g i n a l not seen). 69 27. Kollman, F.F.P. 1951. Wood-Technology and the technology of Wood-Based M a t e r i a l s . Part I. Springer Verlag, B e r l i n and Munchen, p. 416 and 913. 28. Kollman, F.F.P. 1963. Processes and equipment used i n wood-wool boards, small and large units. Fixed investment, working c a p i t a l and operating data. Tech. Pap. I n t e r n a t l . Cons. Plywood and other Wood-Based Panel Products. F.A.O./PPP Cons./Paper 3.21, F.A.O., Rome. 6 pp. 29. Kollman, F.F.P. 1963. Basic properties of wood-wool board. I n t e r n a t l . Cons. Plywood and other Wood-Based Panel Products. F.A.O./PPP Cons./Paper 5.18. F.A.O., Rome. 15 pp. 30. Korneyv, V.A. 1969. An e f f e c t i v e building material made from logging waste. USDA Forest Service, Wash., D.C. FPL-702. Translated from Russian by J o i n t Public. Res. Services, Dept. of Comm. 6 pp. 31. Larson, T.D. 1963. Portland cement and Asphalt concretes. McGraw-Hill Book Co. Inc., N. York. 32. Neubauer, L.W. and S.A. Witzel. 1940. Sawdust concrete test r e s u l t s . Agric. Engin. 21(9):363-366. 33. Panshin, A.J. and C. de Zeeuw. 1970. Textbook of Wood Technology. Vol. 1, t h i r d ed. McGraw-Hill Book Co. Inc. 705 pp. 34. Parameswaran, N., F.W. Broker, and M.H. Simatupang. 1977. Zur Mikrotechnologie mineraligebundener Holzwerkstoffe. Holzforsch, 31(6):173-178. 35. Parker, T.W. 1947. Sawdust-cement and other sawdust bui l d i n g products. Chemistry and Industry Sept. 27:593-596. 36. Paszner, L. 1978. Novel development i n non-restricted bonding of l i g n o c e l l u l o s i c s by mineral binders. Special Paper, 8th World Forestry Congress. F.A.O. FID-II/21, Jarkata, Indonesia. 10 pp. 37. Paszner, L. 1980. Mineral-clad ligneous bodies and method of adhering a mineral deposit i n wood fragment surfaces. Canadian Patent No. 1081718. 38. Sanderman, W. and M. Brendel. 1956. The "cement poisoning" of wood extractives and i t s dependence on chemical c o n s t i t u t i o n . (Translated by WFPL, Vancouver, B.C., Canada). Holz-Roh Werkstoff 14(8):307-313. 39. Sanderman, W., H.J. Preusser, and W. Schweers. 1960. Studies on mineral-bonded wood materials: The e f f e c t of wood extractives on the setting of cement-bonded wood materials. Holzforsch 14(3):70-77. 70 40. Sanderman, W. and R. Kohler. 1964. A short s u i t a b i l i t y test on wood species f or cement-bonded wood-based materials. Holzforsch 18(1,2):53-59. 41. Sanderman, W. 1969. Research r e s u l t s on wood-cement systems. T r o p i c a l Products. Inst., London, England. (Translated from Chemie und chemische technik. Mineral-gerbundener Holz-werkstoff Internationale Arbeitstagung). 16 pp. 42. Sanderman, W. 1970. Technical processes for the manufacture of wood-wool cement boards and t h e i r adaptation for u t i l i z a t i o n of a g r i c u l t u r a l wastes. UNIDO, ID/WG 83/4. 43. Schmidt, L. 1958. Selection of the mineralizer and of the method of m i n e r a l i z a t i o n i n making cement f i b r o l i t e . S t r o i t e l Materialy 4(12):20-22. Chem. Astr . 8, 575-576, 1959. 44. Simatupang, M.H. and H.G. Schwartz. 1975. Magnesit als Bindermittel fur Spanplatten. Holz Z e n t r a l b l a t t , Stuttgart 87:1129-1130. 45. Simatupang, M.H. 1975. Zur Eignung der verschiedenen Holzarten zur Herstellung von zementgebundenen Holzwerkstoffen. Holz-Z e n t r a l b l a t t , Stuttgart, Nr 31, p. 415. 46. Simatupang, M.H., G.H. Schwartz, and F.W. Broker. 1978. Small scale plants for the manufacture of mineral-bonded wood composites. F.A.O. FID-II/21-3. 8th World Forestry Congress, Jarkata, Indonesia. 20 pp. 47. Spraul, J.R. 1941. Acid r e s i s t a n t concrete coatings. A g r i c . Engin. June:209-210. 48. S t i l l i n g e r , J.R., and I.W. Wentworth. 1977. Product process and economics of producing s t r u c t u r a l wood-cement panels by Bison-Werke system. Proceedings of the 11th Particleboard Symposium, Wash. State Univ., Pullman, Wash. pp. 383-410. 49. Strable, J.R. 1943. Manufacture of wood-wool slabs for building purposes. Gyproc Products Ltd., B r i t i s h Patent No. 566, 890, June 1. B r i . Abs. B l , 1945 (130). 50. Thorogood, A.L. 1946. Supplementary report on the German wood-wool slab industry. B r i t . I n t e l l . Obj. Sub. Comm. No. 1489, Item No. 22. pp. 45. 51. Wacker, H. 1960. Wood-wool and machines for making i t . Res. Inf. Serv. t r a n s l a t i o n No. 411, US FPL Madison, pp. 16. Trans-lated from Holz Roh Werkstoff 18(4):142-152. 52. Weatherwax, R.W. 1964. E f f e c t of wood on setting of Portland cement. For. Prod. J . 14(12):567-570. 71 53. Weatherwax, R.W. and H. Tarkow. 1967. E f f e c t of wood on the set t i n g of Portland cement: Decayed wood as an i n h i b i t o r . For. Prod. J . 17(7):30-32. 54. Wise, L.E. and K.H. Lauer. 1962. Cel l u l o s e and the hemicel-l u l o s e s . Pulp and Paper Sc. and Techn. V o l . 1. McGraw- H i l l Book Co. Inc., N. York. pp. 54-73. 55. Witt, J.C. 1966. Portland cement technology. Chemical Publishing Co. Inc., N. York. pp. 99-185. 56. Won-Yung, A. and A.A. Moslem!. 1980. On the S.E.M. examination of wood - Portland cement bonds. Wood Sc. 13(2):77-82. 72 TABLE 1. Some distinguishing features of mineral-bonded wood composites* Less compressed mineral-bonded wood composites Higher compressed mineral-bonded wood composites Structure of material Wood p a r t i c l e / f i b r e coarse-porous wood-wool porous flakes, shavings, sawdust compact surface flakes, shavings flakes, shavings wood pulp P a r t i c l e dimensions length width in mm thickness Actual binders Density in g/cm3 by DIN Bending strength In N/mm2 by DIN Fire test by DIN 4102 A2 • non-combustible Bl - d i f f i c u l t Inflammable Names Products Information 500 3-6 0.2 - 0.5 Portland cement and magnesium cement 0.36 1101 0.57 0.4 - 1.7 1101 d i f f i c u l t inflammable Bl wood-wood l i g h t -weight building boards acc. to DIN 1101 and 1104 Boards 4-60 2-15 to nearly 6 Portland cement 0.45 - 0.80 1101 0.4 - 2.0 1101 d i f f i c u l t inflammable non-combustible* Bl DURIS0L-IS0-Span-ISO-Tex-, VEL0X-building elements Boards, mouldings, building elements such as resin bonded pa r t i c l e board <0.001 Bundesverband Lelchtbauplatten Beethovenstr. 8 D-8000 Hunchen 15 10-30 1-3 nearly 0.3 ( p a r t i a l l y fine particle surface) Portland cement magnesium cement gypsum 1.0 - 1.35 52 361 6.0 - 15 52 362 d i f f i c u l t inflammable non-combustible* A2-B1 BER-building elements* Duripanel Fama panel Isopanel boards, mouldings boards 0.9 - 1.15 52 361 >7.0 - 13.0 52 362 d i f f i c u l t inflammable A2-B1 Homogen Pyroverth* Informatloosverbund Masslvbau-DSmmstelne Postfach 1348 D-8190 Uolfratshauaen <1.0 32 361 4.0 - 7.0 52 362 non-combustible A2 Fermacell boards •Source: Simatupang (46). TABLE 2. Comparison of raw materials, processing features and mechanical properties of cement- and magnesite-bonded wood p a r t i c l e boards* Cement-bonded Magnesite-bonded Pyroverth-olok''" Homogen - M' Wood p a r t i c l e s free from cement i n h i b i t o r s wood and a g r i c u l t u r a l residues Binder Portland cement calcined magnesite Additives A 1 2 ( S 0 4 ) 3 + Ca(0H) 2 MgCl 2 or MgS04 Manufacturing process Setting time 8 - 2 4 hrs approx. 10 min. Equipment special equipment due to long setting time conventional wood p a r t i c l e s board machinery with s l i g h t modifications Ratio binder/wood 2.5:1 1 - 1.5:1 approx. 1.0:1 Board properties S p e c i f i c gravity ( a i r dry) 1.25 0.87 1.15 Bending strength (N/mm2) 10.0 - 13.0 7.0 13.0 Compression strength (N/mm2) 15.0 not a v a i l a b l e 110.0 Fungal resistance good not known good Termite resistance good not known susceptible - good^ F i r e resistance (DIN 4102) Bl A2 Bl Thickness swelling after 2 h water soaking 0.8 - 1.2% 4 4 A2 = non combustible 1) Joachim H i l l e , D-2000 Hamburg 36, Jungfernstieg 36, West Germany. Bl = d i f f i c u l t inflammable 2) Osterreichische Homogenholzges. mbH, A-8775 Kalwang, Aust r i a . 3) Depends on termite species. *Source: Sumatupang (46). TABLE 3. E f f e c t of v a r i a t i o n of ambient temperature on onset of g e l l i n g and development of a temperature maximum for 75:25 mixture of s i l i c a sand and dead-burnt magnesite cement* Curing parameter Ambient temperature, T, °C +32 +27 +22 +15 -1 G e l l i n g Time, T , min crxt • 3 5 7 25 150 Cure Time, T , min 18 20 23 56 280 ' max.' Exotherm Max. Temp. °C 68 63 56 40 17 *Source: Paszner (36). TABLE 4. Wood raw material for production of wood-cement boards C e l l u l o s i c materials P a r t i c l e type F i n a l product wood/cement combination Round wood (logs) slabs, edgings, trimmings wood wool wood-wool cement boards Round wood (logs) slabs, edgings trimmings crushed wood* cement-wood boards Sawdust, shavings and chips f i n e or coarse p a r t i c l e s sawdust, wood chip or wood shaving concrete board Fibres, f l a x , straw and bagasse f i b r e or crushed p a r t i c l e s bagasse - or f l a x -fibro-cement board *Source: Korneyev (30) TABLE 5. Chemical components of the c e l l wall substance i n normal wood* I. PRIMARY COMPONENTS A. Total polysaccharide f r a c t i o n s expressed as holocellulose ... ... ... ... 60-70 percent 1. Cellulose 40-50 percent (Long chain polymer with low s o l u b i l i t y ) 2. Hemicellulose 15-20 percent Noncellulosic polysaccharides; these are r e a d i l y soluble i n d i l u t e a l k a l i and hydrolyzable by d i l u t e acids to component sugars and uronic acids. B. L i g n i n ... ... ... ... ... 15-35 percent I I . SECONDARY COMPONENTS A. Tannins B. V o l a t i l e o i l s and resins C. Gums, latex, a l k a l o i d s , and other complex organic compounds including dyes and c o l o r i n g materials. D. Ash ... ... ... ... ... ... l e s s than 1 percent *Source: Panshin et a l . (33) TABLE 6. P r i n c i p a l compounds i n Portland cement* Name of compound Oxide composition Abbreviation Tricalcium s i l i c a t e 3CaO.Si02 C 3 S Dicalcium s i l i c a t e 2CaO.Si02 C 2S Tricalcium aluminate SCaO.A^O^ C 3 A Tetracalcium aluminoferrite 4CaO.AI2O3.Fe203 C^AF *Source: Larson (31). TABLE 7. The behaviour of p r i n c i p a l components which occur i n Portland cement* Relative behaviour of each compound** Property C 3S C 2S C 3A C 4AF Rate of reaction Medium Slow Fast Slow Heat l i b e r a t e d , per unit of compound Medium Small Large Small Cementing value per unit of compound: Early Ultimate Good Good Poor Poor Good Poor Poor Poor * Source: Larson (31). * C 3 S i s t r i c a l c i u m s i l i c a t e ; C 2 S , Dicalcium s i l i c a t e ; C 3 A ; T r i c a l c i u m aluminate; and C 4 A F , Tetracalcium aluminoferrite. 79 TABLE 8. Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on modulus of e l a s t i c i t y Wood- Density P a r t i c l e moisture content % cement l e v e l r a t i o 0-6 8-15 25-30 40-50 60-80 Modulus of E l a s t i c i t y , kg/cm zxlQ : } 1 32.947* 28.610* 13.487 13.370 11.727 1:1 2 36.650* 36.537* 34.167* 28.583* 14.317 3 81.187* 72.803* 67.437* 53.563* 39.923* 1 77.553* 47.610* 42.237* 39.700* 29.743* 1:1.5 2 110.393* 99.800* 73.950* 61.960* 42.917* 3 120.980* 111.883* 86.853* 67.670* 55.253* 1 136.410* 117.017* 93.377* 41.623* 29.313* 2 156.777* 140.917* 126.523* 88.647* 55.633* 3 317.017* 187.333* 163.873* 119.680* 96.164* *Indicates treatment combination r e s u l t s that meet Canadian Waferboard Standard requirements for MOE. 80 TABLE 9. Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on modulus of rupture Wood- Density cement l e v e l r a t i o 0-6 P a r t i c l e moisture content % 8-15 25-30 40-50 60-80 Modulus of Rupture, kg/cm^ 1 8.390 7.573 3.070 2.253 1.800 1:1 2 13.100 11.463 8.187 6.313 5.117 3 25.173 16.547 13.097 9.467 7.523 1 20.673 13.917 12.400 9.620 3.477 1:1.5 2 38.273 27.427 16.987 13.550 7.873 3 41.957 37.723 30.227 23.123 16.167 1 35.407 25.170 18.010 9.007 5.113 1:2 2 38.067 29.173 22.823 15.713 10.297 3 52.103 44.310 36.313 27.387 22.873 81 TABLE 10. Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on i n t e r n a l bond strength Wood- Density cement l e v e l r a t i o 0-6 P a r t i c l e moisture content % 8-15 25-30 40-50 60-80 Internal Bond, kg/cmz 1 0.916 0.833 0.639 0.499 0.339 1:1 2 1.610 1.288 1.198 0.722 0.639 3 2.499 1.846 1.666 0.916 0.876 1 1.860 1.541 1.332 0.985 0.639 1:1.5 2 3.402* 2.374 2.138 1.693 1.166 3 4.332* 2.707 2.499 1.943 1.527 1 3.776* 3.320* 2.332 1.249 0.694 1:2 2 5.636* 4.235* 2.916* 1.926 1.470 3 6.942* 5.803* 4.915* 3.776* 2.666 *Indicates treatment combination r e s u l t s that meet the Canadian Waferboard Standard requirements f o r IB. 82 TABLE 11. Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on compressive strength p a r a l l e l to surface Wood-cement Density P a r t i c l e moisture content % r a t i o l e v e l Q _ 6 8 _ 1 5 2 5 _ 3 Q 4 Q _ 5 0 6 Q _ 8 0 Compressive strength p a r a l l e l to surface, kg/cm2 1 4.420 4.070 2.737 2.527 1.757 1:1.0 2 7.370 5.403 3.510 2.490 2.107 3 13.123 11.540 8.910 7.440 4.423 1 9.683 8.943 5.613 4.777 2.753 1:1.5 2 14.243 12.807 10.703 8.527 6.320 3 19.053 18.650 15.617 11.687 9.473 1 17.893 17.020 5.790 4.833 4.130 1:2.0 2 22.107 19.647 16.837 9.477 7.190 3 28.377 26.843 24.887 21.230 17.017 83 TABLE 12. Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on board moisture content at test Wood- Density cement l e v e l r a t i o 0-6 P a r t i c l e moisture content % 8-15 25-30 40-50 60-80 Board moisture content % 1 10.133 11.333 13.633 14.867 15.467 1:1 2 10.833 11.600 14.333 15.767 16.267 3 11.033 12.517 13.233 16.000 16.900 1 11.207 12.067 15.067 15.167 16.267 1:1.5 2 11.800 12.833 15.067 16.733 16.867 3 12.267 13.667 14.667 17.167 15.600 1 13.033 14.067 16.067 16.300 16.500 1:2 2 13.400 15.133 16.500 15.767 16.533 3 13.700 15.300 16.633 17.633 17.830 *EMC of wood furnish (White Spruce and Jack Pine mixture) under i d e n t i c a l conditions was 8.50%. 84 TABLE 13. Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on thickness swelling a f t e r 2 h cold soaking Wood- Density P a r t i c l e moisture content % cement l e v e l — " r a t i o 0-6 8-15 25-30 40-50 60-80 Thickness swelling % 1 15.407 8.897 2.737 2.633 2.187 1:1 2 13.917 12.090 3.917 2.793 1.293 3 16.187 9.930 3.760 1.683 1.443 1 6.797 4.513 2.057 1.290 1.253 1:1.5 2 7.880 3.930 2.770 1.810 2.783 3 8.187 3.820 2.203 0.647 1.717 3.610 3.563 4.157 2.623 2.583 2.863 2.007 1.650 2.190 1.507 0.680 0.487 1.090 1.347 0.553 85 TABLE 14. Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on thickness swelling a f t e r 24 h cold soaking Wood- Density P a r t i c l e moisture content % cement l e v e l r a t i o 0-6 8-15 25-30 40-50 60-80 Thickness swelling % 1 16.313 10.627 3.663 3.100 3.850 1:1 2 14.880 13.500 4.227 3.737 3.367 3 18.800 11.307 4.500 2.220 3.947 1 7.520 5.313 2.273 2.047 1.697 1:1.5 2 9.333 6.200 3.830 2.043 5.200 3 9.107 4.847 2.637 1.930 4.487 1 4.610 3.303 2.730 1.957 1.790 1:2 2 4.140 2.967 2.040 1.043 1.630 3 5.193 3.577 2.627 1.190 2.270 86 TABLE 15. Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on water absorption a f t e r 2 h cold soaking Wood- Density P a r t i c l e moisture content % cement l e v e l r a t i o 0-6 8-15 25-30 40-50 60-80 Water absorption % 1 70.223 63.563 61.210 61.930 46.760 1:1 2 66.543 57.637 59.863 54.330 47.063 3 57.263 51.770 48.970 50.020 37.807 1 45.113 39.217 39.803 37.060 31.643 1:1.5 2 41.870 36.767 35.873 32.853 28.480 3 35.333 30.247 27.637 32.043 26.033 25.857 26.583 25.067 24.187 24.650 21.703 26.137 23.270 21.523 24.557 22.333 19.673 21.180 20.313 18.340 87 TABLE 16. Interaction of p a r t i c l e moisture content, density and wood-cement r a t i o on water absorption a f t e r 24 h cold soaking Wood- Density cement l e v e l r a t i o 0-6 P a r t i c l e moisture content % 8-15 25-30 40-50 60-80 Water absorption % 1 95.190 82.347 77.223 75.440 66.697 1:1 2 91.050 72.330 77.607 69.883 62.647 3 70.773 62.667 61.287 61.093 52.547 1 60.917 56.617 53.307 48.200 42.913 1:1.5 2 60.760 48.470 48.287 42.107 41.350 3 45.683 36.393* 34.610* 40.883 33.653* 1 37.087* 34.960* 33.960* 30.373* 27.593* 1:2 2 34.193* 31.690* 28.727* 26.920* 25.890* 3 32.213* 28.233* 25.497* 24.123* 22.687* *Indicates treatment combination r e s u l t s that meet the ISO Standard requirements for water absorption. TABLE 17. Analysis of variance for testing the eff e c t s of p a r t i c l e moisture content, density and wood-cement r a t i o on MOR and MOE i n cement-bonded particleboard Src. Source D.F. Sum of squares Mean square F value F. Prob. No. ANALYSIS OF VARIANCE FOR VARIABLE MODULUS OF RUPTURE 1 Moisture (A) 4 7837.431074 1959.357666 164.0359 0.0000** 2 Density (B) 2 5290.605228 2645.302490 221.4627 0.0000** 3 AB 8 140.735446 17.591919 1.4728 0.1776 4 W/C r a t i o (C) 2 6692.250833 3346.125244 280.1350 0.0000** 5 AC 8 1046.745375 130.843170 10.9541 0.0000** 6 BC 4 442.820172 110.705032 9.2681 0.0000** 7 ABC 16 314.158665 19.634903 1.6438 0.0732 8 Error 90 1075.021933 11.944688 9 Tot a l 134 22839.768726 ANALYSIS OF VARIANCE FOR VARIABLE MODULUS OF ELASTICITY 1 Moisture (A) 4 98652.096388 24663.023438 61.7286 0.0000** 2 Density (B) 2 78992.633024 39496.316406 98.8546 0.0000** 3 AB 8 10096.653132 1262.081543 3.1588 0.0035** 4 W/C r a t i o (C) 2 172152.270135 86076.125000 215.4384 0.0000** 5 AC 8 39513.250532 4939.156250 12.3621 0.0000** 6 BC 4 16544.635163 4136.156250 10.3523 0.0000** 7 ABC 16 14935.878215 933.492188 2.3364 0.0062** 8 Error 90 35958.552000 399.539307 9 Total 134 466845.968588 * * S i g n i f i c a n t at 0.01 l e v e l . 89 TABLE 18. Analysis of variance f o r testing the e f f e c t s of p a r t i c l e moisture content, density and wood-cement r a t i o on i n t e r n a l bond strength and compressive strength p a r a l l e l to surface Src. Source D.F. Sum of squares Mean square F value F. Prob. No. ANALYSIS OF VARIANCE FOR VARIABLE INTERNAL BOND STRENGTH 1 Moisture (A) 4 89.095596 22.273895 77.2174 0.0000** 2 Density (B) 2 59.039160 29.519577 102.3362 0.0000** 3 AB 8 5.036775 6.295969E-01 2.1826 0.0359* 4 W/C r a t i o (C) 2 127.065196 63.532593 220.2498 0.0000** 5 AC 8 24.270048 3.033755 10.5172 0.0000** 6 BC 4 13.506140 3.376534 11.7055 0.0000** 7 ABC 16 1.763792 1.102369E-01 0.3822 0.9833 8 Error 90 25.961123 2.884569E-01 9 Tot a l 134 345.737830 ANALYSIS OF VARIANCE FOR VARIABLE COMPRESSION//SURFACE 1 Moisture (A) 4 1551.253144 387.813232 68.5518 0.0000** 2 Density (B) 2 2044.713761 1022.356689 180.7170 0.0000** 3 AB 8 55.658127 6.957266 1.2298 0.2906 4 W/C r a t i o (C) 2 2608.437228 1304.218506 230.5403 0.0000** 5 AC 8 242.171550 30.271439 5.3509 0.0000** 6 BC 4 254.139941 63.534973 11.2308 0.0000** 7 ABC 16 126.354793 7.897174 1.3959 0.1615 8 Error 90 509.150400 5.657227 9 Tot a l 134 7391.878944 * S i g n i f i c a n t at 0.05 l e v e l . * * S i g n i f i c a n t at 0.01 l e v e l . TABLE 19. Analysis of variance for testing the e f f e c t s of p a r t i c l e moisture content, density and wood-cement r a t i o on thickness swelling a f t e r 2 h and 24 h cold soaking Src. Source D.F. Sum of squares Mean square F value F. Prob. No. ANALYSIS OF VARIANCE FOR VARIABLE THICKNESS SWELLING: 50% R.H. to 2 h cold soaking 1 Moisture (A) 4 1102.298941 275.574707 206.3927 0.0000** 2 Density (B) 2 2.061453 1.030726 0.7720 0.4690 3 AB 8 14.600450 1.825056 1.3669 0.2212 4 W/C r a t i o (C) 2 485.138564 242.569275 181.6731 0.0000** 5 AC 8 441.349828 55.168716 41.3188 0.0000** 6 BC 4 3.207049 8.017622E-01 0.6005 0.6663 7 ABC 16 24.619381 1.538711 1.1524 0.3215 8 Error 90 120.167667 1.335196 9 Tota l 134 2193.443333 ANALYSIS OF VARIANCE FOR VARIABLE THICKNESS SWELLING: 50% R.H. to 24 h cold soaking 1 Moisture (A) 4 1168.691551 292.172852 159.9616 0.0000** 2 Density (B) 2 .7.717213 3.858606 2.1125 0.0000** 3 AB 8 23.359698 2.919962 1.5986 0.1246 4 W/C r a t i o (C) 2 608.858258 304.428955 166.6717 0.1356 5 AC 8 464.813720 58.101700 31.8101 0.0000** 6 BC 4 16.797502 4.199375 2.2991 0.0000** 7 ABC 16 32.812364 2.050773 1.1228 0.3468 8 Error 90 164.386733 1.826519 9 Tota l 134 2487.437040 * * S i g n i f i c a n t at 0.01 l e v e l . 91 TABLE 20. Analysis of variance for testing the ef f e c t s of p a r t i c l e moisture content, density and wood-cement r a t i o on water absorption a f t e r 2 h and 24 h cold soaking Src. Source D.F. Sum of squares Mean square F value F. Prob. No. ANALYSIS OF VARIANCE FOR VARIABLE WATER ABSORPTION: 50% R.H. to 2 h cold soaking 1 Moisture (A) 4 2310 .570341 577.642578 64.7348 0.0000** 2 Density (B) 2 1363 .540013 681.769775 76.4041 0.0000** 3 AB 8 73 .731201 9.216400 1.0329 0.4178 4 W/C r a t i o (C) 2 24625 .742443 12312.871094 1379.8704 0.0000** 5 AC 8 548 .409628 68.551193 7.6823 0.0000** 6 BC 4 286 .547973 71.636978 8.0282 0.0000** 7 ABC 16 106 .662856 6.666428 0.7471 0.7395 8 Error 90 803 .088667 8.923207 9 Total 134 30118 .293133 ANALYSIS OF VARIANCE FOR VARIABLE WATER ABSORPTION : 50% R.H. to 24 h cold soaking 1 Moisture (A) 4 4073.753249 1018.438232 70.0652 0.0000** 2 Density (B) 2 3828.178601 1914.089111 131.6831 0.0000** 3 AB 8 301.358147 37.669754 2.5916 0.0136* 4 W/C r a t i o (C) 2 40998.176313 20499.085939 1410.2700 0.0000 5 AC 8 627.592480 78.449051 5.3970 0.0000** 6 BC 4 592.671256 148.167801 10.1935 0.0000** 7 ABC 16 243.455284 15.215955 1.0468 0.4174 8 Error 90 164.386733 1.826519 9 Tota l 134 51973.387197 * S i g n i f i c a n t at 0.05 l e v e l . * * S i g n i f i c a n t at 0.01 l e v e l . 92 TABLE 21. Interaction of p a r t i c l e moisture content, density l e v e l and wood-cement r a t i o on board density at test Wood- Density P a r t i c l e moisture content % cement l e v e l -r a t i o 0-6 .8-15 25-30 40-50 60-80 Board density gm/cnr 1 0.515 0.545 0.423 0.447 0.429 1:1 2 0.592 0.569 0.498 0.528 0.452 3 0.655 0.687 0.606 0.602 0.558 1 0.711 0.694 0.636 0.605 0.534 1:1.5 2 0.782 0.762 0.678 0.707 0.604 3 0.840 0.882 0.821 0.773 0.731 1 0.812 0.839 0.761 0.715 0.688 1:2 2 0.895 0.868 0.845 0.883 0.746 3 0.933 1.028 0.905 0.924 0.904 93 TABLE 22. D e t a i l s of curve f i t t i n g of MOE and MOR Variable P a r t i a l C o e f f i c i e n t „ J 1 Std. error T-stat S i g n i f . MOE CONSTANT 5.3183=b0 0.53413-1 99.570 0.0000 LNDNS 0.89477 2.6252=^ 0.11403 23.022 0.0000 MOISFRAC -0.49338 -0.72820=b2 0.11174 -6.5170 0.0000 MOR CONSTANT 8.3061=b0 1.3582 6.1154 0.0000 LNDNS 0.60263 3.2399=^ 0.21476 15.086 0.0000 MOISFRAC -0.60263 -1.0874=b2 0.12629 -8.6102 0.0000 WCRATIO -0.24423 -3.9492=b3 1.3753 -2.8715 0.0048 LNWC -0.28398 3.3976=b4 1.0061 3.3769 0.0010 N.B. LNDNS - Log density MOISFRAC - Moisture f r a c t i o n LNWC - Log wood-cement r a t i o WCRATIO - Wood-cement r a t i o * * S i g n i f i c a n t at 0.01 l e v e l . ^"Standard error of regression parameters. 94 TABLE 23. De t a i l s of curve f i t t i n g of edgewise compression and IB strength A ** Variable P a r t i a l C o e f f i c i e n t Std. e r r o r 1 T-stat S i g n i f . Edgewise compression strength CONSTANT LNDNS LNMF LNWC 0.77787 -0.38745 0.23682 3.2971=b0 3.2619=^ -0.1405=b2 0.51232=b-: 0.17667 0.23024 0.29217-1 0.118363 18.662 14.167 -4.8103 2.7899 0.0000 0.0000 0.0000 0.0000 CONSTANT LNDNS LNMF 0.81528 -0.49970 IB strength -2.4939=b0 0.16064 -15.524 0.0000 3.7323=bx 0.23073 16.176 0.0000 -0.21695b2 0.32733-1 -6.6280 0.0000 N.B. LNDNS - Log density LNWC - Log wood-cement r a t i o LNMF - Log moisture f r a c t i o n * * S i g n i f i c a n t at 0.01 l e v e l . •'•Standard error of regression parameters. Mixing 'crit. *max. Reaction Time (Min) Fig. 1 Exotherm Characteristics of Ammonium Polyphosphate Activated Magnesium Ox Cements. Source: Paszner (36) Figure 2. P a r t s of Mold 97 SCALE 1 inch = 17.6 cm (all measurement in cm) F i gu re 3. Dimensions of mo 98 Figure 4. S i n g l e opening press Figure 5. L e v e l l e r 100 4 0 , Density Level Fig. 6 Dependence of Modulus of Rupture on Density and Wood-Cement ratio interaction. 101 ' 1 1 1 1 0-6% 8-15% 25-30% 40-50% 60-80% Particle Moisture Content % Fig. 7 Dependence of Modulus of Rupture on Particle Moisture Content and Wood-Cement ratio interaction. 102 1 1 — — " 1 1 1 0-6% 8-15% 25-30% 40-50% 60-80% Particle Moisture Content % Fig. 8 Dependence of Modulus of Elasticity on Particle Moisture Content and Density interaction (at 1:1 Wood-Cement Ratio). (Wood-Cement ratios: 1:1.5 and 1:2 held constant.) 103 Particle Moisture Content % Fig. 9 Dependence of Modulus of Elasticity on Particle Moisture Content and Density at 1:1.5 Wood-Cement «ar/o.(Wood-Cement ratios: 1:1 and 1:2 are held constant.) 104 Particle Moisture Content % Fig. 10 Dependence of Modulus of Elasticity on Particle Moisture Content and Density a Wood-Cement Ratio. (Wood-Cement ratios 1:1.5 and 1:1 held constant.) 105 106 * i 1 • 0-6% 8-15% 25-30% 40-50% 60-80% Particle Moisture Content % Fig. 12 Dependence of Tensile Strength 1 surface on Particle Moisture Content and Wood-Cement ratio interaction. 107 i , , . 1 • 0-6% 8-15% 25-30% 40-50% 60-80% Particle Moisture Content % Fig. 13 Dependence of Compressive strength / / surface on Particle Moisture Content and Wood-Cement ratio interaction. 108 Density Level Fig. 14 Dependence of Compressive strength / / surface on Density and Wood-Cement ratio in-teraction. 109 Particle Moisture Content % Fig. 15 Dependence of Thickness Swelling(after 2h cold soaking)on Wood-Cement ratio and par-ticle moisture content interaction. 110 — , , — — , . 1 0-6% 8-15% 25-30% 40-50% 60-80% Particle Moisture Content Fig. 16 Dependence of Thickness Swelling (after 24h cold soaking) on Particle Moisture Content and Wood-Cement ratio interaction. I l l I : , , r  1 2 3 Density Level Fig. 17 Dependence of Water Absorption (after 2h cold soaking) on Density and Wood-Cemen ratio interaction. 112 Fig. 18 Dependence of Water Absorption (after 2h cold soaking) on Particle Moisture Content and Wood-Cement ratio interaction. 113 t r Density Level Fig. 19 Dependence of Water Absorption (after 24h ratio interaction. cold soaking) on Density and Wood-Cement 114 Fig.20. Dependence of Water Absorption (after 24h cold soaking) on Particle Moisture Content and Wood-Cement ratio interaction. I Figure 21. Electron micrographs of c r y s t a l s of pure magnesium oxide powder 116 Figure 22. E l e c t r o n micrographs of c r y s t a l s of dead burnt magnesite cement powder Figure 23. Electron micrographs of c r y s t a l s of Ammonium Polyphosphate (reactant) 118 Figure 24. Electron micrographs of magnesite cement boards manufactured using; 0-6% i n i t i a l p a r t i c l e moisture content, 1:1 wood-cement r a t i o and density l e v e l 1 119 4DOOX Figure 25. Electron micrographs of magnesite cement boards manufactured using; 8-15% i n i t i a l p a r t i c l e moisture content, 1:1 wood-cement r a t i o and density l e v e l 1 Figure 26. Electron micrographs of magnesite cement boards manufactured using; 25-30% i n i t i a l p a r t i c l e moisture content, 1:1 wood-cement r a t i o and density l e v e l 1 Figure 27. Electron micrographs of magnesite cement boards manufactured using; 40-50% i n i t i a l p a r t i c l e moisture content, 1:1 wood-cement r a t i o and density l e v e l 1 123 Fig.29 Relationship between MOE and density at d i f f e r e n t i n i t i a l p a r t i c l e moisture contents. 1 1 1 1 1 1 0.2 0.4 0.6 0.B 1.0, 1.2 D E N S I T Y . g / c m 3 Fig.30 Relationship between MOR and density at d i f f e r e n t i n i t i a l p a r t i c l e moisture contents. 125 Fig.31 Relationship between IB strength and density at d i f f e r e n t i n i t i a l p a r t i c l e moisture contents. 126 1 1 1 1 1 r~ 0.2 0.4 0.6 0.8 1.0 , 1.2 D E N S I T Y , g * m 3 Fig.32 Relationship between edgewise compression strength and density at d i f f e r e n t i n i t i a l p a r t i c l e moisture contents APPENDIX V4- Portlond Camant Concrttti dncing compounds, C t S and C 5A, arc present in lesser Amounts, re-nilting in CO to 70 rnl per g heat liberation. The American Society for Testing mid Materials given a specifi-r.-itinn (ASTM V I50-G11 covering live type* of portland cement* which an' intended to cover llie prinripnl nrcas where special proper-ties are neiiliil. A further modification of the properties of |>ortlnnd cement concrete can l>c iiehieved by the use of ndinixtures. Tliis topic will lie discussed in :i later section. Table 1-3 shows typical compound composition for the five type* of Portland cement covered by ASTM Spi-cifirnlion*, -.and n brief description of each follows. TOBLA I '3 COMPOUND COMPETITION OL PORTLAND CTMTNTI T \ | i r OF RRINRNT I. NORMAL II . MOHIFI^L 111. HIGH RARLY ATRRNCTH IN', L/im HINT V . SIILFAIR RESISTANT Typa I This is onhnary portland cement for use where special properties arc not required. When ordinary portland cement is to be exposed to severe frost action, then Type IA may_ be specified. This is similar to Type 1 except that an air-entraining agent ha* been added. Typt II This type is a moderate variation of Type I and is used where some sulfate action is indicated or where a somewhat lower heat of hydration is nrcdcil. Type IIA should l>e used for combined frost nnd sulfate attack. Type III Type III cement is designed for uso where high early" strength is needed liocausc of a particular^on8lfuctiorr~Hruatiou. PcvcrnI factors canTbnlrfliutc to-tins Tiigh early strength. First, enrly strength can lie improved chemically by using n higher percentage of _<J".iS. Second, the hydration and hnrdening can be improved by phys-ically grinding the cement finch The surface area of the cement exposed to the action of water and therefore the rate of hydration partially depend on the fineness of the individual cement particles. For • given weight of cement, the surface area, measured in aqunrc ccnti-COMPOUND COMPOSITION, % c s 15 <4 S3 20 38 0,A C . A F 27 I I , 10 | at I « i I I 7 in R, * 8 13 7 V U . 8 Iwltoducllan »a Portland Caianl Co««»»l»i IS meters per gram, will be greatest for the finer material. Not only will the hydration proceed more rapidly for the fine cement, but it will also be more complete in it given period of time. The limit to fineness occurs when the particle* are so smnll that minute amount* of moisture will prehydrntc nnd thereby drMroy the rrnirtit during handling nnd storage. Again, Typo 111 A i* nvnilnhlc tor situations where fast hnrdening and resistance to froil action are required. Type IV Type IV cement is for use where the heat of hydration must be kept to an absolute minimum. As bus already been mrn-tinned, this is accomplished chemically by using minimum amounts of tricnlciuin silientc and tricalcium illuminate. While the cement enn be modified to reduce the hent liberated, other STEP- may alto be necessary to control temperatures in mass concrete. Among the step* taken in heat control arc the following: controlling placement tem-peratures of the constituent mntcrinl* in the concrete mul use of n cooling solution circulated through pipes embedded in the concrete. Typ« V Type V cement is specified for use where there is exten-sive exposure to sulfates. This condition occurs most often in hy-draulic structures carrying water* with n high alkali control. Sulfate resistance can be improved chemically by reducing the tricalcium aluminatc content. Of the five types available Type I is the most widely used and is best for the normal situation. Chemical or physirnl alteration of this type of cement can improve certain specific properties, but at the same time other properties may be ndversely nffected. l'nr example, strength is reduced somcwhnt for Type IV and V. Table 1-4 com-pares the strengths of the various types of cement with Type I at three different moist curing interval*. Tabla I i Approsimata Rilollv« Stfangthi of Con<t«t« oi Aflaclad by Typa Camant TYPO OF PORTLAND RRINRNT CUMPRRASIVR STRRNRLLI, r, NF NNRMNL PORTLAND RRINRNT RONRRRTE 3 DNYI 28 HNYA 3 MONTHS 1. KORNINL 100 100 ion I I . MODIFIED 80 85 1110 III . ILIIILI EARLY STRENGTH ITIN 130 115 I V . IX>W LICNT 5(1 05 (XI V . HTDFALC RRAIATANT ns US 85 APPENDIX II BRITISH S T A N D A R D SPECIFICATION METHODS OF TEST FOR WOOD CHIP BOARDS WOOD WASTE BOARDS & SIMILAR BOARDS B.S. 1811 :1952 Price 2}- net, post free BRITISH STANDARDS INSTITUTION Incorporated by Royal Charter T e l e g r a m s : Standards, Sowest, London Telephone : Abbey 3333 Sales Branch Telephone : Victoria 0522 24 V I C T O R I A S T R E E T , L O N D O N , . S . W . I B.S. 1811 : 1952 If precise daU are required as to the condition of the lest pieces at the time of testing, a small portion shall be cut from each of the test pieces and weighed immediately after the test has been carried out. The volatile content of each of these portions (hall be determined in accordance with Test 6, Part 2, and reported with the results of the particular test to which they refer. " r P A R T 2. M E T H O D S O F T E S T ( ^ 1. T E S T F O R F L E X U R A L S T R E N G T H ( T R A N S V E R S E ) Six test pieces, each 8 in. X 4 in . shall be prepared and conditioned as specified in C a u s e s 2 and 3. Each test piece shall be simply supported on parallel rollers having a radius o f % in . to in . , spaced at 6 in. centre-to-centre and free to rotate in ball or roller bearings. A load 6hall then be applied at the centre o f the span along a line parallel w i th lhe end supports by means of a bar roundedi to a radius o f between % in. and j£ in. (See F i g . 1.) . / i / T h e j o a d shall be j ipp l i e^ at_an even rate, or at a rate such as w i l l ^ ^ ^ ^ c r ^ * 4 A* ' produce an even rate o f increase in strain, and shall be so adjusted that \ the test will be completed in a period pfjiolJess than half a minute and not ^ ^lr'-^-S more than 4 minutes^ T h e load which each test piece fails to support shall — /ipce'/****?' be reported, together with the mean ultimate failing load. The effective modulus o f rupture^ f, shall be calculated as follows : — If \ V = mean ultimate failing load in pounds to the nearest pound o r to within per cent o f the load measured span between centres o f supports (6 in.) mean width o f test pieces in inches, determined in accordance with Test 5, Part 2 a b d — mean thickness o f test pieces in inches, determined in accordance with Test 5, Part 2 NOTE 1. If the faces of the test pieces have different finishes, they shall be tested with the load applied to the fair face. NOTE 2. If a board has a distinct grain h wilt be necessary to repeal the strength tests and record separate results for lest pieces cut parallel with, and at right angles to, the grain. NOTE 3. Iftheflexnralstrength when wet is required, the test pieces shall be immersed in water for 24 hours and the water absorption determined in accordance with Test T. Part 2 of this standard. The test pieces shall then be tested immediately for fiexnral strength and the effective modulus of rupture determined as above. NOTE 4. This test is designed for determining the breaking load of boards up to about \yt in. in thickness. When testing materials of greater thicknesses an appropriate depth/span ratio should be chosen. os In S3 H in. to H in. radius 3 in. Bin. ^ l ^ ' l l / > ' " ' t 0 X f l i n ~ r a d l u s 3in. 6 in.span. < 3 in. to in. radius Fig. 1. Typical diagram showing application of test for flcxurnl strength and deflection o 131 B.S. 1811 : 19S2 2. T E S T F O R D E F L E C T I O N A N D E F F E C T I V E M O D U L U S O F E L A S T I C I T Y Six test pieces, each 8 in. x 4 in . , shall be prepared and conditioned as specified in Clauses 2 and 3. Each test piece shall be simply supported on parallel rollers having a radius o f % in . to M in. spaced at 6 in. centres and free to rotate in ball or roller bearings. A load shall then be applied at the centre o f the span-along a line parallel with the end supports by means o f a bar rounded to a radius of between % in. and in . (See F i g . 1.) A n initial load shall be applied equivalent to 10 per cent o f the ultimate failing load* and the deflection under this load shall be treated as the zero condition. The load shall then be increased to 33)4 per cent o f the ultimate failing load and after a lapse o f 30 seconds the deflection at the centre shall be measured to an accuracy o f ± 0-005 in. T h e increase in deflection for each of the test pieces shall be reported together with the mean increase. T h e effective modulus o f elasticity, E , in bending shall be calculated as follows : — I f W = increase in load in pounds s — measured span between centres o f supports (6 in.) A •= mean increase in deflection, in inches b = mean width o f test pieces in inches, determined in accordance with Test 5, Part 2 d = mean thickness o f test pieces in inches, determined in accordance with Test 5, Part 2 WJ* t n e n E - H^nT , b / " l - i n ' 3. T E S T F O R D E F L E C T I O N U N D E R S U S T A I N E D L O A D Six test pieces each (36</ + 4) in. long by 6 in. wide, where d is the nominal thickness o f the board, shall be prepared and conditioned as speciSed in Clauses 2 and 3. E a c h test piece shall be simply supported on horizontal parallel rollers having a radius o f % in. to H in. , spaced at a distance centre to centre o f 36 times the nominal thickness o f the board. T h e test load shall be applied at the mid-point o f the span along a line parallel to the end supports by means o f a bar rounded to a radius o f ft in. to K in. and carrying a stirrup, from which a weight shall be suspended so that the total weight including the bar and stirrup shall be 2 lb. The position * For the purposes of this lest, if the ultimate railing load has not already been determined in accordance with Test 1, Part 2, it will be permissible to adopt the mean of the results of at least two tests, carried out in accordance with Test 1, Part 2, on test pieces cut from the same, or similar, boards as those being tested for deflection. 9 APPENDIX I I I 4&!t) 01037 each specimen as ^ i f . c d in Sections 9 . 126 and 127. T E N S I L E S T R E N G T H P E R P E N D I C U L A R T O S U R F A C E 28. Scope 28.1 The test for tensile strength perpendic-ular to the surface shall be made on specimens in the dry condition to determined cohesion of the fiberboard in the direction perpendicu-lar to the plane of the board. NOTE 13—Thu test is included because of tbe increased use of fiberboards and particle boards where wood, plywood, or other materials are glued to the board, or where the internal bond strength of the board is an important property. Tests in tbe soaked condition shall be made if tbe material is to be used under severe conditions. of load shall pass through the cor. specimen. of the 31 . Speed of Testing 31.1 App ly the load continuously through-out the test at a uniform rate of motion of the movable crosshead of the testing machine of 0.08 in . / in. (cm/cm) of thickness per min. NOTE 15—It is not intended that the testing machine speed shall be varied for small differences in fiberboard thickness, but rather that it shall not vary more than ± 5 0 percent from that specified above (see Note 9). 32. Test Data and Report 32.1 Obtain maximum loads from which calculate the stress at failure. Calculate strength values in pounds per square inch (or kilograms per square millimeter), for which the measured dimensions of the specimen shall be used. Include the location of the line of failure in the report. 29 . Test Specimen 29.1 The test specimen shall be 2 i n . (50 mm) square and tbe thickness shall be that o f the finished board. Loading blocks of steel or aluminum alloy 2 in . square and 1 i n . (25 mm) in thickness shall be effectively bonded with a suitable adhesive (Note 14) to the 2-in. square faces of the specimen as shown in F ig . 5, which is a detail o f the specimen and loading fixtures. Cross-sectional dimensions of the specimen shall be measured to an COMPRESSION S T R E N G T H P A R A L L E L TO accuracy of not less than ± 0 . 3 percent T h * S U R F A C E e maximum distance from the center of the universal joint or self-aligning head to the glued surface of the specimen shall be 3 in . (76 mm). NOTE 14—Any suitable adhesive that provides an adequate bond may be used for bonding the fiberboard specimen to the loading blocks. Epoxy resins are recommended as a satisfactory bonding agent. Tbe pressure required to bond the blocks to the specimen will depend on tbe density of the board and the adhesive used, and should not be so great as to measurably damage the fiberboard. The resulting bond shall be at feast as strong as the cohesive strength of the material perpendicular to the plane of the fiberboard. 30. Procedure 30.1 Engage the loading fixtures, such as are shown in F ig . 5, attached to the heads of the testing machine, with the blocks attached to the specimen. Stress the specimen by sepa-ration of the heads of the testing machine until failure occurs. T h e direction of loading shall be as nearly perpendicular to the faces of the fiberboard as possible, and the center 3 3 . Moisture Content 33.1 Determine the moisture content of each specimen on a separate sample prepared from the same material, as specified in 126.2 and Section 127. 34. Scope 34.1 The test for compression strength par-allel to the surface shall be made on specimens both in the dry and in the soaked condition. Tests shall be made of specimens both with the load applied parallel and perpendicular to the long dimension of the board to determine whether or not the material has directional properties. 34.2 Because of the large variation in char-acter of wood-base fiber and particle panel materials and the differences in manufactured thicknesses, one procedure is not applicable for all materials. One of the three procedures detailed as follows shall be used depending on the character and thickness of the board being evaluated. 34.2.1 Procedure A (Laminated Specimen) shall be used for materials ' /§ in . (10 mm) or more but less than 1 in . (25 mm) in nominal thickness, particularly when modulus of elas-ticity and stress at proportional limit are re-304 # D1 0 3 7 quired, in this procedure when materials less than 1 in. in thickness are evaluated, two or three thicknesses shall be laminated to pro-vide a nominal thickness of at least 1 in. but no amount more than that amount than nec-essary. The nominal size of the specimen shall be 1 by 4 in. (25 by 101 mm) (with the 4-in. dimension parallel to the applied force) by the thickness as laminated. 34.2.2 Procedure B (Lateral Support) shall be used for materials less than ' /« in. in thickness, particularly when modulus of elas-ticity and stress at proportional limit are re-quired. Specimens shall be 1 by 4 in. by the thickness, as manufactured and evaluations made in a suitable lateral support device. The 4-in. long dimension shall be parallel to the applied force. 34.2.3 Procedure C (Short Column) shall be used when maximum crushing strength only is required or where the thickness of the board material is 1 in. or more and either maximum crushing strength modulus of elas-ticity, and stress at proportional limit or only maximum crushing strength is required. When the material being evaluated is 1 in. or less in thickness, the width of the specimen shall be 1 in . , the thickness shall be as manufactured, and the length (height as tested) shall be four times the thickness. When the material being evaluated is more than 1 in. in thickness, the width shall be equal to the nominal thickness and the length (height as loaded) shall be four times the nominal thickness. 35. Test Specimen 35.1 The test specimens shall be carefully sawed with surfaces smooth and planes at right angles to the faces of the boards as manufactured. For the laminated specimens (Procedure A ) , pieces of board at least 1 in. (25 mm) larger in length and width than the finished size of specimen shall be laminated using thin spreads of epoxy resin (Note 16) and pressures not exceeding 50 psi (3.5 kg/ cm 2 ) . Specimens shall be sawed from the laminated pieces after at least 8 h of curing of the resin al room temperature. The width and thickness shall be measured to at least the nearest 0.001 in. (0.025 mm). These two dimensions shall be used to calculate net cross-sectional area for modulus of elasticity, and stress at proportional limit and maximum load. NOTE 1 6 - T h e epoxy resin is required because it contains no water or other swelling agent that might produce initial stresses adjacent to the glue lines. 36. Specimens Soaked Before Test 36.1 Specimens to be tested in the soaked condition shall be prepared in accordance with Section 13. 37. Procedure 37.1 Load the specimens through a spheri-cal loading block, preferably of the suspended self-aligning type. Center them carefully in the testing machine in a vertical plane as shown in Figs. 18 (unsupported 4-in. (101-mm specimen) and 19 (laterally supported pack device). 4 Apply loading at a uniform rate of head travel of the testing machine of 0.005 in. (0.12 mm)/in. of length/min. NOTE 17 —Speed of test therefore for the 4-in. specimen of Methods A and fi shall be 0.020 in. per minute (see Note 9 for permitted variation in testing speed). 38. Load-Deformation Curves 38.1 When required, obtain load-deforma-tion curves for the full duration of each test. Figure 18 shows a Lamb's Roller Compres-someter on an unsupported specimen. Figure 19 shows a Marten's Mirror Compressometer on a laterally supported specimen. Use these or equally accurate instruments for measuring deformation. Choose increments in loading so that not less than 12 and preferably at least 15 readings are obtained before proportional limit. Read deformation to the nearest 0.0001 in . (0.002 mm). Attach compressometers over the central portion of the length; points of attachment (gage points) shall be at least 1 in. (25 mm) from the ends of specimens. 39 . Moisture Content and Specific Gravity 39.1 Use the entire compression parallel to surface specimen for moisture content deter-mination except when the capacity of the drying oven is too small for convenient drying 4 The lateral support device is detailed in Fig. 2 of ASTM Methods D 805, Testing Veneer, Plywood, and Other Glued Veneer Construct ions, Animal Book of ASTM Standards, Part 22. 134 APPENDIX TV each specimen as specified in Sections 9, 126 and 127. TENSILE STRENGTH PERPENDICULAR TO SURFACE 28. Scope 28.1 The test for tensile strength perpendic-ular to the surface shall be made on specimens in the dry condition to determined cohesion of the fiberboard in the direction perpendicu-lar to the plane of the board. NOTE 13—This test is included because of the increased use of fiberboards and particle boards where wood, plywood, or other materials are glued to the board, or where the internal bond strength of the board is an important property. Tests in the soaked condition shall be made if the material is to be used under severe conditions. 29. Test Specimen 29.1 The test specimen shall be 2 in. (SO mm) square and the thickness shall be that of the finished board. Loading blocks of steel or aluminum alloy 2 in. square and 1 in. (25 mm) in thickness shall be effectively bonded with a suitable adhesive (Note 14) to the 2-in. square faces of the specimen as shown in Fig. 5, which is a detail of the specimen and loading fixtures. Cross-sectional dimensions of the specimen shall be measured to an accuracy of not less than ± 0 . 3 percent. The maximum distance from the center of the universal joint or self-aligning head to the glued surface of the specimen shall be 3 in. (76 mm). NOTE 14 —Any suitable adhesive that provides an adequate bond may be used for bonding the fiberboard specimen to the loading blocks. Epoxy resins are recommended as a satisfactory bonding agent. Tbe pressure required to bond the blocks to the specimen will depend on the density of the board and the adhesive used, and should not be so great as to measurably damage the fiberboard. The resulting bond shall be at feast as strong as the cohesive strength of the material perpendicular to the plane of the fiberboard. 30. Procedure 30.1 Engage the loading fixtures, such as are shown in Fig. 5, attached to the heads of the testing machine, with the blocks attached to the specimen. Stress the specimen by sepa-ration of the heads of the testing machine until failure occurs. The direction of loading shall be as nearly perpendicular to the faces of the fiberboard as possible, and the center D 1037 of load shall pass through the center of the specimen. 31. Speed of Testing 31.1 Apply the load continuously through-out the test at a uniform rate of motion of the movable crosshead of the testing machine of 0.08 in./in. (cm/cm) of thickness per min. NOTE I S - I t is not intended that the testing machine speed shall be varied for small differences in fiberboard thickness, but rather that it shall not vary more than ± 5 0 percent from that specified above (see Note 9). 32. Test Data and Report 32.1 Obtain maximum loads from which calculate the stress at failure. Calculate strength values in pounds per square inch (or kilograms per square millimeter), for which the measured dimensions of the specimen shall be used. Include the location of the line of failure in the report. 33. Moisture Content 33.1 Determine the moisture content of each specimen on a separate sample prepared from the same material, as specified in 126.2 and Section 127. 135 APPENDIX V. materials with the materials most commonly used alternately to them.' 97. Test Specimen 97.1 The area of the test specimen to be abraded shall be 2 by 3 in. (50 by 76 mm), and the specimen shall be fabricated from a piece of the board 2 by 4 in. (50 by 101 mm) by the thickness of the material (Note 32) as shown in Fig. 16. The specimens shall be conditioned before test (see Section 4) and the test made in the same conditioned atmos-phere. The actual dimensions of the abrading area of the specimen shall be measured to the nearest 0.01 in. (0.2 mm). The thickness of the test specimen shall be measured to at least the nearest 0.001 in. (0.02 mm) near each corner and the center. NOTE 32 - When the board tested is less than '/j in. (12 mm) thick, either sufficient thicknesses shall be laminated together to provide the '/j-in. thick-ness or the specimen shall be backed by a thickness of wood or plywood sufficient or provide the ' / i-in. total thickness of specimen required. 98. Procedure 98.1 Conduct the test on the Navy-type abrasion machine 7 as shown in F ig . 17, using as the abrading medium new N o . 80 grit aluminum oxide, or equivalent. Apply the grit continuously (Note 33) to the 14-in. (355-mm) diameter steel disk, which serves as a platform supporting the specimen and rotates at the rale of 23'/J rpm. Rotate the specimen in the same direction as the steel disk at the rate of 32Vi rpm. Superimpose a load of 10 lb (4.5 kg) on the test specimen. The machine is designed so that twice each revolution the specimen is raised Vi» in. (1.6 mm) above the steel disk and immediately lowered. Deter-mine the decrease in the thickness of the specimen at the end of each 100 revolutions of the steel disk by measuring the thickness of the specimen to the nearest 0.001 in. (0.02 mm) near each corner and at the center, after brushing to remove any dust or abrading material adhering to the surface of the speci-men. The mean of the five recordings shall be taken as the loss in thickness. Repeat this procedure until the specimen has 500 revolu-tions of wear or as required (Note 34). NOTE 33-The Navy wear tester is so designed that there is an excess of grit on the abrading disk D 1037 at all times. During all parts of the abrading action, except when the specimen is in the raised position, the specimen is pushing a small amount of grit a head of it. NOTE 34-When values of accumulated wear are plotted as ordinates against revolutions, the slope of the curve is a straight line for wear through uniform materials. When the rate of wear per 100 revolutions of the abrading disk is not uniform after the first 200 revolutions, it is probably due to a change in abrasion resistance with depth from the original surface of the material being tested. 99. Report 99.1 The report shall include the follow-ing: 99.1.1 Loss in thickness in inches per 100 revolutions of wear if uniform, and 99.1.2 If the amount of wear changes with depth from the original, surface values for each 100 revolutions. W A T E R ABSORPTION A N D THICKNESS r SWELLING 100. Scope 100.1 A test shall be made to determine the water-absorption characteristics of build-ing boards. 101. Test Specimen 10] .1 The test specimen shall be 12 by 12 in. (304 by 304 mm) in size, or 6 by 6 in. (152 by 152 mm) in size with all four edges smoothly and squarely trimmed. 102. Conditioning Prior to Test 102.1 The test specimen shall be condi-tioned as nearly as deemed practical to con-stant weight and moisture content in a condi-tioning chamber maintained at a relative hu-midity of 65 ± 1 percent and a temperature of 20 ± 3 C (68 ± 6 F) . The moisture content after conditioning shall be reported. 103. Weight, Thickness, and Volume or Test Specimen 103.1 After conditioning, weigh the speci-men to an accuracy of not less than ± 0 . 2 • U. S. Forest Products Laboratory Report R 1732, "The Abrasion Resistance of Wood as Determined with the U S Navy Wear Test Machine." ' The Navy-type wear tester may be constructed from drawings obtainable from the U.S. Navy or the Forest Products Laboratory. It is manufactured commercially by the Tinius Olsen Testing Machine Co., Willow Grove, Pa. till) percent and measure the width, length, and thickness to an accuracy of not less than ± 0 . 3 percent. Compute the volume of the specimen from these measurements. Measure the thick-ness to an accuracy of ± 0 . 3 percent at four points midway along each side 1 in. (25 mm) in from the edge of the specimen and average for the thickness swelling determination. NOTE 35—Where a common practice or special consideration requires edge thickness determina-tions at the edge or another distance from the edge !as the present practice for particle board is 'h in. 12 mm)), the edge distance used shall be given. 104. Submersion in Water 104.1 Submerge the specimen horizontally under 1 in. (25 mm) of distilled water main-tained at a temperature of 20 ± 1 C (68 ± 2 F) . A s an alternative to the above method of submersion, specimens may be submerged vertically (Note 36). After a 2-h submersion, suspend the specimen to drain for 10 min, at the end of which time remove the excess surface water and immediately weigh the specimen and determine the thickness. Sub-merge the specimen for an additional period of 22 h and report the above weighing and measuring procedure (Note 37). NOTE 36 —The amounts of water absorbed for tests of this duration are not the same for the two methods of submersion. Specimens suspended ver-tically will absorb considerably more water than those suspended horizontally. Therefore, values obtained from the two methods are not comparable. NOTE 37 —Soluble materials in some of the board products may influence water absorption values if the water is reused; therefore fresh water shall be used for each test. When tap water has been proven sufficiently pure so that results of test are not affected, it may be used as an alternative to distilled water. 105. Drying After Submersion 105.1 After submersion dry the specimen in an oven at 103 ± 2 C as outlined in Sections 126 and 127, and calculate the mois-ture content (based on oven-dry weight) from the weights after conditioning and after 2 and 24-h submersion. 106. Calculation and Report 106.1 Calculate the amount of water ab-sorbed from the increase in weight of the specimen during the submersion, and express the water absorption both as the percentage 0 1037 by volume and by weight based on the volume and the weight, respectively, after condition-ing. Assume the specific gravity of the water to be 1.00 for this purpose. Express the thickness swelling as a percentage of the orig-inal thickness. When any other size of speci-men than the 12-in. (304-mm) square one i s used, the report shall include the size used. In addition, give the method of submersion i f other than horizontal. L I N E A R V A R I A T I O N WITH C H A N G E IN MOISTURE CONTENT 107. Scope 107.1 Tests of linear variation with change in moisture content shall be made to measure the dimensional stability of a fiberboard w i t h change in moisture content. • 108. Test Specimen 108.1 The test specimens, when possible, shall be 3 in. (76 mm) in width and at least 12 in. (304 mm) in length. Two specimens shall be provided, one cut parallel with the long dimension of each board and one from t h e same board cut at right angles to the long dimension. When a board does not permit obtaining a 12-in. (304-mm) specimen, t h e maximum length possible shall be used, bui i t shall be at least 6 in. (152 mm). 109. Procedure 109.1 Follow the following or any equall\ or more accurate method for measuring spec-imens: Condition to practical equilibrium (Note 38) specimens carefully sawed square and smooth at a relative humidity of 50 ± 2 percent and a temperature of 20 ± 3 C (68 ± 6 F) . Measure the length of each specimen to the nearest 0.001 in. (0.02 mm) in a comparator like or similar to the one shown in Figs. 13 and 14 using a standard bar of the same nominal length as the specimen for reference. For each measurement orient the specimen in the same way in the comparator, for example, numbered surface up with num-bers reading from the side toward the opera-tor. Then condition the specimens to practical equilibrium (Note 39) at a relative humidit) of 90 ± 5 percent and a temperature of 20 ± 3 C (68 ± 6 F ) , place in the comparator 314 APPENDIX VI <Sl» to Ike surface. Include the sire of specimen and necessary details regarding method* of •tenure menu in the report. MOtfTURE CONTENT AND SPtCltlC GiAvmr IM. Pnwadwc 126.1 Delermine the specific gravity (or density) of each Malic bending specimen al lime of mi from the dimcmiom. weight, and moisture content, ai provided in Seciiom 9, 19. and 127. When specific gravity determi-nation! are required on specimens of prism form from other lens, like procedure may be used. The moeuure content of a specimen may be determined from a coupon cut from the lest specimen m in the sialic bending lest, or the entire letl ipecimen may be used for the morslure determination. 126.2 When for any reason additional de-terminations of morilure content and specific gravity (or density) are required, prepare separate samplei for Ihrt determination horn the lame material at is used m preparing the teal specimens. These moislure content and specific gravity specimens shad be the full thickness of the material and 3 m. (76 mm) •ide and 6 in. (152 mm) long. Smaller speci-mens may he used when deemed necessary. Condition Ihe specimens in accordance with the provisions of Section S. 126.3 Measure Ihe dimensions to an accu-racy of art len than ±0.3 percent, and the •eight to an accuracy of not lets lhan ±11 2 percent. Obtain ihe overt-dry arcighl after drying Ihe ipecimen m an oven al 103 * 2 C until approitmately constant weight it at-tained. in. Cahamlliai 127.1 The moislure content shall he careu-hled at follows: *f - laofiir - n/f| where: M • moisture content, percent. W m initial weight, and F * final weight when oven-dry. 127.2 Calculate Ihe specific gravity a* f(»|. lows (Note 45): 01037 where: ¥ - final weight when oven dry, g. L M length ol coupiin, in. (or mm), w • wnlih tit coupon, m. (or mm), f thickness of coupon, in. (or mm), and K - \, when melrk units of weight and measurement are used; or Html, when metric units of weight and U.S. customary units of measurement are used. Nora 45-The specific gravity as demraimdhy this equation n based on volume al lesl and weight when oven dry. 127.3 When desired, the density based on weight and volume of Ihe specimens after conditioning may he calculated. INTERLAMINAR SHEAR IM. 8cage 128.1 Tests in interlaminar shear (shear in Ihe plane of the hoard) shall he made on specimens bonded between two steel loading plates loaded in compression to obtain strength and deformation properties of wood-base panel materials. One half of the lest specimens shall be prepared with the long dimension parallel and Ihe other half with the long dimension perpendicular to Ihe long di-mension of ihe board in order lo evaluate any directional properties. 129. Slgrdflcawes 129.1 Shear properties In Ihe plane of the board (interlaminar shear) duplicate the kind of stressing in shear encountered in such glued structural assemblies as structural sandwiches and adjacent to gluelines between flanges and webs in bos and l-hcams and gusset plates in Irusset. The procedure used follows chisely the requirements of Method C 27.1. While it apparently yields values in ihe same plane as Ihe "block shear" lesl ol Sections H7 lo 911. values obtained are not comparable because of effects of friction in the tool and the fact that failure in Ihe Mock shear lest can only occur in a Vt-in. (3-mm) thick area in the middle of Ihe hoard. Ihe interlaminar sheai sltenglh lesl oilers the additional advantage thai shear deformation data can he obtained when ilesnable. 13d. lesl Specimen 110 I I he inlrihnnin.il shear tests sh.ill he €0 made on specimens 2 in. (Stl mm) wide and 6 in. (ISO mm) long by Ihe thickness of Ihe material, when Ihe board material it Vi in. (12.5 mm) in thickness or less, For materials more lhan Vi in. (12.5 mm) thick, the speci-men shall have a width of al least twice Ihe thickness hut not less lhan 2 in. (Stl mm) and a length al least 12 timet Ihe thickness (Note 46). The edges of Ihe ipecimen shall be tawed square and smooth. The length, width, and thickness shall he measured to an accuracy of at least ±0.3 percent. 1311.2 Steel loading platet Va in. (about 20 mm) thick, having a width equal lo the speci-men width and a length equal to Ihe specimen length plus 'It in. (7 mm) shall he bonded lo each face of Ihe specimen as shown in Fig. 20 using suitable adhesive (Note 47). The load-ing ends uf Ihe plates shall protect 'It in. (7 mm) beyond Ihe end of the specimen and they shall be beveled al 45 deg and cn rented at shown. Eilrrme care shall he used in applying adhesive to minimum spreads are used to prevent it infusing into Ihe specimen and thus reinforcing the hoard. Nora 46-A length ratio of 12:1 ta pmuiUd as a minimum to thai secondary normal sircssct arc mini in al. Nora 47-Set Note 14 131. L-aaHwg Pitxiwart 131.1 The load shall be applied through notched fillings such that Ihe line of action of the direct compressive force shall pass through the diagonally opposite corners of Ihe specimen as shown in fig 211. The tower fitting shall he placed on a spherical hearing block as shown in Fig. 21 so thai Ihe load is uniformly dislrihuted across the width of Ihe specimen. I». Speed of Teatkag 132.1 The load shall he applied continu-ously throughout ihe test al a uniform rate of million of the movable crnsshead of Ihe test-ing machine equal to 11.1102 limes ihe length (inch or centimeter) per minute (see Note 9). 133. laud-Orfonaialioa IHIa 133.1 When shcaiing modulus is required, data lor plotting loud dclotmalion curves can he obtained by usiny die dial gate :uiane.e-rrtent shown in lig. J|. »lmli ineasuics ihe 01037 displacement of one plate with respect to the other. The interlaminar shearing modulus it defined at Ihe slope of Ihe siraight-lmc por-tion of Ihe stress-strain curve. A secant mod -ulus may be calculated for data which do not have an initial siiaight-lme relationship. If shearing modulut values are included in Ihe report, Ihe method used lo delermine them shall also be reported. 134). fn*wlaTavs Cna4atsf ami SpceMc faavvtay 1.34.1 II moislure content or specific grav-ity, or both, are required, a separate ipecimen taken from Ihe hoard adjacent lo the shear ipecimen shall be used and the determinations made as detailed in Sections 126 and 127. I3S. CatcatstftMs ami Regaarf 135.1 The interlaminar shearing strength shall be calculated for each specimen by the following equation, and ihe values deter-mined shall be included Hi Ihe report: /. - F/l* 135.2 The interlaminar shearing mudufua shall he calculated for each specimen by Ihe following equation (Note 4R), and the values determined shall be inducted in Ihe report: G - rV/IAr where: b - width of specimen, m. (or cm), tf " thickness of specimen, in. (or cm), f, m interlaminar shearing strength, ptl (or kg/cm'). G " interlaminar shearing modulus, pti (or kg/cm'). L " length of specimen, in. (or cm), P m maaimum load, lh (or kg). P, m load at proportional limit or point where secant intersects load-deforma-tion curve, lb (or kg), and r • dial reading or displacement of one plate with respect to the other at load f i , in. (or cm). . Nota 4 H - T V equation In 12? 2 above a s s a m e t l h a l i h e s t r a i n s i n Ihe l o a d i n g p l a l e s and i n Ihe b o n d b e t w e e n t h e p l a t e s a m i the s p e c i m e n a r e n e g l i g i b l e . EDGEWISE SHEAR r— 136. Scope 1.16.1 Icsls in edgewise shear (shear nor-

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.831.1-0095534/manifest

Comment

Related Items