MSU RETURNING MATERIALS: P1ace in book drop to LIBRARJES remove this checkout from w. your record. FINES will be charged if book is returned after the date stamped be1ow. L * 3 5 " Nov-122001 DEVELOPMENT OF PHYSICAL AND MECHANICAL PROPERTIES OF PARTICLEBOARD AND MEDIUM DENSITY FIBERBOARD WITH PARTICULAR EMPHASIS ON INTERNAL BOND STRENGTH BY M. Mehdi Faezipour A DISSERTATION Submitted to Michigan State University In partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Forestry 1988 ABSTRACT DEVELOPMENT OF PHYSICAL AND MECHANICAL PROPERTIES OF PARTICLEBOARD AND MEDIUM DENSITY FIBERBOARD WITH PARTICULAR EMPHASIS ON INTERNAL BOND STRENGTH BY M. Mehdi Faezipour The purpose of this study was first to discuss how different factors and process variables influence the physical and mechanical properties of particleboard and medium density fiberboard (MDF) with emphasis on internal bond (18) strength property, second to evaluate the relationship between IB and density of two commercial particleboards and medium density fiberboards. Included in this paper are evaluations of some important related properties and discussion and evaluation of different IB test methods. There were strong correlations between modulus of elasticity and density of both MDF and particleboard. There was a very slight correlation between core density and IB of MDF althougth over a wider range of density 18 was a function of density. However, there was a close correlation between core density and 18 of particleboard. It was concluded that there is a possibility of reducing overall density of MDF and saving costs without sacrificing 13. For many application, the resulting reduction of MOE would have little practical significance. ACKNOWLEDGMENTS Appreciations and special thanks are due Dr. Otto Suchsland for his valuable guidance, criticisms and counsel in the preparation of this study. I also like to thank the other three committee members, Dr. Alan Sliker, Dr. Henry A. Huber and Dr. Paul A. Rubin. Finally I like to thank Mr. Ivan Borton for his help with laboratory work and Mr. Hong Xu, Ph.D. candidate, for his help with computer analysis. ii TABLE OF CONTENTS Page ACWOWLEDGEMNTOOOOOOOOOOOO ....... O 00000000000000 ii LIST OF TABLES.......................... ...... ... v LIST OF FIGURES................. ............ ..... vi LIST OF APPENDICES....................... ........ Xi Chapter I. INTRODUCTION................................ 1 Classification of wood composite boards... 3 Historic development...................... 7 Wet-process fiberboard.................. 10 Platen-processed particleboard .......... 12 Extruded particleboard.................. 14 Dry-process hardboard..... .............. 15 Medium density fiberboard.... ........... 18 Total capacity............................ 18 General particleboard and MDF processing systems.........OOOOOO......... ...... 0.... 22 Control of properties....... ...... ... ..... 30 II. omEmIVESOOOOI......OOOOOOOOCOOO0.0. ........ 35 III. DEVELOPMENT OF PHYSICAL AND MECHANICAL PROPERTIES OF PARTICLEBOARD AND MDF ......... 38 The structure of particleboard and fiberboard...................... .......... 38 Major factors developing physical and mechanical properties..................... 40 SpeCieSOO......OOOOOOOOOO...... ..... O... 40 ReSinbondOOOOOOOOOOOOOOOOO0............ 42 Particle geometry....................... 49 Densification of boards.......... ....... 52 DenSitYOOCOOOOOOOO00.00.00.000...O..0... 52 Density variations in boards.. ....... ... 53 Multi-layer boards........................ 55 The mechanics, process techniques and control of three-layer boards........... 56 Control of density gradient............. 60 Some special feature of MDF.......... ..... 64 The measurement of density variation ...... 70 iii Chapter Page IV. IB OF PARTICLEBOARD AND MDF ............. .... 72 The internal bond test methods.... ........ 73 Relationship between 18 and density ....... 82 The internal bond of three—layer boards... 84 Inconsistancy in literatures regarding IB and core density relationship.... ...... ... 86 V. EXPERIMENTS : PART I - DETERMINATION OF RELATIONSHIPS BETWEEN DENSITY AND IB OF TWO COMMERCIAL PARTICLEBOARDS AND MDF........... 89 Purpose of experiment.......... ........ ... 89 Preparation of specimens.. ....... ......... 89 Density profile testing procedure ..... .... 90 MOE testing procedure.............. ....... 97 IB testing procedure...................... 108 Results (part I)................ .......... 115 Density and density profile....... ...... 115 Modulus of elasticity.... ............... 120 Internal bond........ ................... 125 Limitation of experiment......... ...... ... 129 EXPERIMENTS : PART II - EXTENSION OF IB TEST T0 LARGER RANGE OF DENSITY ....... ... ........ 129 Purpose of experiment ..................... 129 Preparation of specimens....... ........... 130 Density profile testing procedure... ...... 130 I8 layer testing procedure ................ 130 Results (part 11) ......... ............... 139 Layer IB strengh. ....................... 139 Limitation of experiment .................. 141 V1. SUMMARY AND DISCUSSION.. ..... ..... ......... 143 VII. CONSEQUENCES OF REDUCING CORE DENSITY ...... 146 V111. CONCLUSION.. ........... . ................... 149 APPENDICES................ ....................... 152 LIST OF REFERENCES............ ................... 161 iv LIST OF TABLES Page Particleboard and MDF production by region, 1976-19860000000...-0............OOOOOOOOO 20 Historical profile : Shipments and production of U.S. particleboard and MDF, 1959-86..... 21 Property requirements of mat-formed wood particleboard.............................. 33 Property requirements for MDF .............. 34 The overall design and number of test specimens. .................. 91 Summary of test results : properties of particleboard ................ 116 Summary of test results : properties of fiberboard. .................. 118 Summary of test results : ID of MDF and particleboard as function of distance from board surface........ ..... . .................. .... 140 Examples of computing density and MOE while reducing core density and maintaining TB of MDF............ .......... 148 Figure 1. 9. 10. 11. 12. 13. 14. LIST OF FIGURES Classification of wood composition boards by particle size, density and process type Manufacturing diagram of typical composite boardOOIOOOOOOO ..... ......OOOOOOOOOOOOIOOO Summary schematic of fiberboard processing Simplified schematic of wet and dry fiberboard processes ...................... Principal particleboard manufacturing processeSOOOOOOOOOOOOO0.0... OOOOOOOOOOOOOO Historical profile, U.S. particleboard shipments........................... ...... Typical flow chart of particleboard manufacturing plant................. ...... Principle of radio frequency board press.. Typical process flow chart of MDF ......... Arrangement of particles in the manufacturing of plywood and particleboard Example of bending strength of flakeboard affected by both species and board density Resin efficiency in particleboard as function of resin droplet size..... ....... Example of MOR of particleboard as influenced by resin content ............... Example of dependence of some strength properties of solid wood and particleboard on moisture content........ ..... .... ...... vi Page 16 19 23 28 31 39 41 44 45 48 Figure Page 15. Most common forms of wood composite elementSOOOOOOIOOOOOOO. ............ 0...... 50 16. Example of MOE of three-layer flakeboard as affected by raw material and process variables................. ........ . ....... 59 17. Example of dependence of IE and shear strength on core density of three-layer flakeboardOOOOOO.......OOOOOOOOOOOOOOOOOOO 61 18. Example of dependence of layer density of particleboard on press closing time ....... 63 19. Example of the influence of short closing time on the shape of density profile of waferboard........ ........................ 65 20. Example of the influence of long press closing time on the shape of density profile of waferboard... ..... . ............ 65 21. Example of relationship between pressure or closing time and density profile of MDF 68 22. Example of relationship between press closing time and MOE for two press platen temperatures of MDF....... ................ 69 23. Correlation between standard 18 test and modified standard IB test of particleboard 77 24. Correlation between standard IB test and push-out test of particleboard ..... ....... 77 25. Example of relation between layer strength and layer density of particleboard. ....... 79 26. Example of relationship between standard IB test and shear strength of particleboard.. 81 27. Example of relationship between compaction ratio and IB strength of flakeboard ....... 85 28. Example of non linear correlation between IB and density of MDF... .......... 88 29. Diagram for cutting test specimens from 4 by 4 squares ft. boards ........ . ........ 92 vii Figure 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. Density profile for particleboard # PB-E (gravimetric method) ...................... Density profile for MDF # FB-C (gravimetric method). ........... Equipment schematic of gamma radiation method of density profile determination... Specimen stage of gamma radiation method.. Electrical components of gamma radiation methOd. ...... ......OOOOOOOOO OOOOOOOOOOOOOO Density profile for particleboard # PB-E (gamma radiation method using 0.020" interval). ................... Density profile for particleboard # PB-E (gamma radiation method using 0.0050" interval). .................. Density profile for MDF # FB-C (gamma radiation method using 0.020" interval) ............. Density profile for MDF # FB-C (gamma radiation method using 0.0050" interval) ............ Apparatus for measuring static MOE.. ...... Cross section (A) and dimensions (B) of MOE test specimen.............. ........ Ultrasonic timer for measuring dynamic MOE test...... .................... Dimensions of 18 test specimen used in part I of experiment..... Compression shear test specimen undergoing test .................. viii Page 95 96 98 99 100 101 102 103 104 106 107 109 110 111 Figure 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. Carbide tipped milling cutter mounted on a milling machine for removing thin slices of 18 test specimen....................... Special jig for measuring length reduction of IE test specimen ...... Regressions of MOE on densities of boards........................ ............ Regressions of MOE on densities of particleboard (two directions).. .......... Regressions of MOE on densities of MDF (two directions) ...................... Regressions of static MOE on dynamic MOE Of boards.....OOOOOOOOOO0.0...O. ...... Regressions of IE on core densities of MDF and particleboard .................. MDF and particleboard IB test specimens showing 1 in square area on planed surface created by grooves to which aluminum block will be glued....................... MDF 18 test specimen after gluing aluminum blocks to face and back ( ready for test). MDF IB test specimen undergoing test ...... MDF and particleboard IB test specimens showing failure areas at 0.040 in from the surface................ ...... ...... ....... MDF and particleboard IB test specimens showing failure areas at 0.160 in from the surface....................... ..... ....... MDF and particleboard IB test specimens showing failure areas at 0.280 in from the surfaCeOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOO MDF and particleboard IB test specimens showing failure areas at 0.360 in from the surface.........OOOOOOCOOI......OOOOOO...O ix Page 113 114 121 122 123 126 128 132 133 134 135 136 137 138 Figure Page 58. Regressions of layer IE on layer densities of MDF and particleboard.. ...... 142 59. Relations between overall density and core density of MDF when maintaining face density and IB....... ................ 147 60. Measurment of length reduction of IE test specimen......... ................. 155 LIST OF APPENDICES Appendix Page 1. Board specifications......... ............. 152 2. Determination of machined length of IE test specimen............... ........... 154 3. Experimental data for layer IB and layer density relationships ............... 157 xi Chapter 1 INTRODUCTION The composition board industry, including hardboard, insulation board, particleboard, strandboard, and medium density fiberboard is of very recent origin. These boards are distinct from solid wood in that they are composed of wooden elements of varying sizes held together by an adhesive. Commercial composition board products have some very desirable characteristics such as availability in large sheets,smooth surfaces, uniformity of properties from sheet to sheet, and freedom from localized defects (44). A wood particleboard is composed of essentially small dry wood particles that are randomly distributed or oriented, arranged in layers and glued together with a resin binder and pressed into sheet form. The particle- board industry in the U.S. had its start in the 1950’s and it was originally established to utilize wood waste. It has grown to be one of the major forest products industries in the U.S. Particleboard has been developed, not solely as a utilizer of wood waste, but as a new product for new uses (32 ). Particles for the boards can be made from almost any type of wood, whether whole logs or wood residues such as trimmings and shavings from lumber or plywood manufacture. Particleboard is used by the furniture, kitchen cabinet and store fixture industries, usually in the form of cores in veneered or otherwise overlayed panel material. Particleboard also is suitable for many other applications in furniture, building and other wood using industries. Oriented strand board (OSB) is a type of particleboard which is used in structural applications. Growing pulp mill demand for low-grade wood waste and increasing utilization of wood waste as fuel are causing some changes in the particleboard industry. Medium density fiberboard (MDF) is a more recent product in direct competition with particleboard. It can be manufactured from wood waste and low grade woods unsuitable for pulping. Medium density fiberboard is produced by reducing raw material to fibers in pressurized disk refiners. These fibers are then bonded together with low viscosity, low tack, synthetic resin and formed into a mat. The process offers an opportunity for profitably utilizing wood once considered waste because it tolerates wide variation in raw material such as species composition, geometric configuration and bark inclusions (91, 127). MDF, according to the common usage of the term, refers to the thick ( 3/8 to 1 in) medium density fiberboard that is generally sold in the industerial core stock market. Its properties, such as bending strength, modulus of elasticity, internal bond, machinability, and scew holding power, meet the levels required for these applications. Just what these requirement are, in terms of the above properties, is not precisely defined. The furniture manufacturer, in many cases, simply knows from experience that a certain core material is suitable in a given application. The commertial standard simply identifies those materials that can be successfully used for spesific purpose (91). A unique combination of moderate overall density, suitable density profile, and resin content results in excellent edge machining and edge finishing characteristics, qualities essential to superior furniture stocks. The first MDF plant was built in the U.S. in 1965. MDF-dry successfully competes with particleboard, a lower cost core material, on the strength of its more uniform and solid edges, which allow direct finishing. It is expected that MDF-dry will make further in- roads into the core stock market in the future (91). Classification of wood composite boards A general classification of wood composition boards by particle size, density, and process type is shown in figure 1. In this classification, plywood is also included in wood composition boards in which the particles are veneer A Hm .wom v .mmau mmoooua oaowuumm an mpuoon :o .H.ousmfim 62.106.334.035: .cuz<.lulo¢ at- O; . a >h_>(¢0 Quanta n.- sheets of regular dimensions that can easily and systematically be assembled and laminated without densification. Particleboard is based on particles that are very small compared with veneer sheets, but many times larger than the wood cell. Waferboard and strandboard are particleboards made from rather large particles and intended for structural applications. Strand board or oriented strand board (OSB) is a particleboard of oriented particles, with the use of particles deliberately manufactured with optimum particle-geometry which can yield materials equalling or surpassing the structural capability and reliability of sawn lumber and plywood. The term wood particleboard usually refers to a particleboard of random particles (5,44,91). Fiberboards are made from a furnish consisting of elements with dimensions of the same order of magnitude as those of the wood cells. Insulation board, MDF-dry, MDF wet, hardboard-dry and hardboard-wet fall into the general category of fiberboard. Paper is not a fiberboard but made out of fibers and its process is similar to fiberboard technologies (44,91). A manufacturing diagram of typical composition board is shown in figure 2. MATERIAL . STORAGE OVENS Figure 2. Manufacturing diagram of typical composite board. ( Ref. 44 ) Historic development Particleboards and fiberboards are of very recent historic origin. A tremendous amount of residual material resulting from the production of lumber and plywood has been used for fuel over the centuries. There is no doubt that people over the centuries have visualized taking the waste material and producing some type of man-made board. However, records show that such developments are of very recent origin. Maloney (44) summarized a brief history of particleboard and fiberboard which is used as a reference for most parts of the historic section of this paper. The following paragraphes will cover a brief historic development of wet-process fiberboard, platen-pressed particleboard, extruded particleboard, dry-process hardboard and medium density fiberboard. Although fiberboards including insulation board and hardboard wet and dry processes will not be included in this paper, a brief history of these boards will be given for a relatively complete backround of wood composite development and because of the similarity of these processes to the MDF process. See figures 3, 4 for summary schematics of the fiberboard process. 3: § §§ lumuuo_ -33.- ¥ WIT! WEED MIC i‘ .. m" 42:... r ‘E E F I umhuuvuumo «aromas -.....— _ I 2 l. , ' 3% im“*’1~év~:« L——-:!szs HM! M MD VET E; \\—- _:__ —t Figure 3. Summary schematic of fiberboard processing. ( Ref. 91 ) ’6 8 E 3‘. I l k WOOD ‘41 '3 BOARD I—r ‘fi‘ PULP LI——H WATER WATER WET PROCESS SHEEI' ‘ F ORA“ 7' ION [1]£”fl"”' luunmm 8 k ‘4‘ . W l It ADHESIVE ,3 q ‘ I \ D "...—4 {sh ‘ as om Pnocess o k Figure 4. Simplified schematic of wet and dry fiberboard processes. ( Ref. 91 ) 10 Wet-process fiberboard Lyman obtained a United State Patent in 1858 on fiberboard. In 1914, C. G. Muench made the world ’s first insulation board out of wood fibers. His idea was a partial adaptation of the process used in making paper. Wood fibers were carried in water from which the fiber slurry was metered out onto a continuously moving screen. The water was drained away through the screen leaving a mat of interlaced fibers. This material was then oven dried, yielding the sheets of board. Although insulation board has been refined over the years, the key point is that it is a wet process system based on papermaking technology. This type of board ranges in density from 10 to 25 lbs/cu.ft., which is lower than wood and other raw materials from which it is usually made. It is widely used for sheathing, interior paneling, rigid roof insulation, and as a siding. It is a relatively low cost material which requires little, if any, binding agent. It is possible also to densify the insulation board after drying by consolidation in a heated press. This results in higher-density hardboard products (44). William H. Mason discovered the process of producing wet-process hardboard in 1924 and his discovery established the Masonite Process. To prepare the fiber, chips are subjected to high-temperature steam in a 11 pressure vessel called a gun. The chips are subjected to approximately 600 psi steam pressure for one minute, and then the steam pressure is quickly raised to more than 1000 psi for five seconds. The pressure is suddenly released and the chips are exhausted from the gun. The presure built up within the chips literally explodes them into a coarse mass of fibers. This mass can be further reduced into individual fibers or fiber bundles by means of an attrition mill. Lignin, in general terms, is nature’s glue for binding the wood fibers together. In the steaming operation the lignin bonds between the cellulose fibers are softened, which permits a more improved disintegration of the chips into fiber bundles as compared to grinding unsteamed chips in an attrition mill. After the fluffy masses of fiber are refined, the fiber is washed and the fines and hemicellulose solubles are removed. The remaining stock then goes to a former, somewhat similar to the one for producing insulation board. This type of mat is pressed into its final density in a hot press. A screen is used on the bottom of the mat to allow the moisture to escape during the pressing operation. This results in a smooth— one side (SlS) board. If a smooth-two-side (S28) board is desired, the wet mat must first be oven dried before it is put into the hot press. Otherwise, the high moisture content of the mat will cause an excessive amount of steam to be generated, which will blow the board apart 12 during or immediately after pressing. In comparison to insulation board, hardboard must have a density of at least 31 lbs/cu.ft. (44). Platen-pressed particleboard Ernst Hubbard in 1887 published the early concept of particleboard process under the title of : Utilization of wood waste. He proposed to manufacture artificial wood from sawdust and blood albumin under application of pressure and heat. Krammer in 1889 obtained a German patent for a method for gluing planer shavings onto linen cloth and then laying up the cloth layers in a cross-lap construction much like that for plywood. Watson, an American, in 1905 made thin wood particleboard. This patent shows clearly a flakeboard very similar to some types of board made today. Watson could be called the inventor of the flake type of particleboard. Backman, a German, in 1918 suggested making a board with chips or wood dust in the center and surface veneers on the outside. This particular formulation is now the basis for a certain structural building panel. Freudenberg, a German in 1926 talked about utilizing planer shavings with the adhesive available at that time for making a board. He noted that the adhesive level should be between three and ten percent, which, is about 13 the range for the present-day particleboard. Nevin, an American, in 1933 recommended the mixing of coarse sawdust and waste wood shavings with an adhesive and then forming and compressing them under the application of heat. A Frenchman, Antoni, in 1933 discussed boards of a mixture of wood fibers and particles and large elements such as excelsior or even metal netting in a board that was to be bonded with phenolic or urea glues. Carson, an American was awarded a patent in 1936 for which he had applied initially in 1932, for establishing a regular production line for producing particleboard. A binding agent, which was a urea- formaldehyde-condensation product dilutable in water, was to be sprayed onto the wood particles in a rotary-drum blender. Before hot pressing, a prepressing operation was to take place, and he proposed covering the final board with a thermoplastic coating of synthetic resin. Much of what he discussed here will be found in many particleboard plants today. During 1937-1960 particleboard production gradually appeared throughout the world. Most probably, commercial particleboard was produced after world war II. About 1960 the massive expansion of the United States particleboard industry commenced. Techniques were developed to produce boards with smooth surfaces, and resins were refined to speed up curing times in the press. 14 Significant efforts are being made to bring particleboard and fiberboard further into the structural building panel market in direct competition with plywood. In the U.S., the greatest usages of structural particleboard are in the underlayment and industrial core stock. Industrial and mobile home decking are using structural particleboard as well. Structural flake board is approved for use in Canada. other particleboards are used structurally throughout the world. The development of this particular segment of the board industry has been phenomenal since world war II. Many different types of board plants have been built around the world, based not only on wood waste and roundwood cut especially for particleboard but also on other lignocellulosic material such as bagasse and flax (44). Extruded particleboard The extrusion method of producing particleboard was initially developed in Germany in the years 1947-1949. In the United States, several plants were built in the 1950 ’5 based upon the original concept from Europe as well as on developments by U.S. firms. The plants producing this type of board in the U.S. are relatively small and are captive in the sense that they produce board for use by other parts of their firm ’5 operations, 15 notably in furniture factories. Extrusion of particleboard is a different method of production of particleboard. Particles are forced by an oscillating ram through a die, consisting of two parallel heated plates. The distance between the plates is fixed, and is equal to the thickness of the board (1,44). see figure 5. The extruded board cannot meet the standard established for platen-pressed particleboard. Particles in this system are randomly oriented in a plane perpendicular to the board surface and hence the boards have very low bending strength and stiffness. They must be overlaid with some other material such as thin particleboard, veneer or a paper or plastic laminate (44). Dry-process hardboard The next development in the composition board field was that of manufacturing hardboard by means of a dry process in 1945. In this production process, the chips or other residual materials are conveyed through cooking and grinding units where the fiber is prepared. It is also now normal to apply the resin and wax to the raw material prior to grinding so that the attrition mill simultaneously prepares the fibers and blends the additives with the fibers . In initial experiments and in the early plants, the resin was added 16 A om .wom v .moumoooum mcflusuomuacme pumonoHowuumm Hmmflocqum .m ousmwm QHHUOIL IOumalhflfl mmfluoxm DNIGOhlhtt ./. 33.. 63.5 I! _— scan—m I... \ t R...“ Icons-a amused alum 17 after the fibers were prepared. The mats were then formed using air and mechanical means in a continuous system. This mat, however, did not have the fiber interlocking that is possible when forming mats by the wet process. Because of this lack of interlocking and very low, if any, adhesion that can be obtained by lignin flow and hydrogen bonding in the hot press, the dry- process hardboard is dependent upon an added binder for the development of physical properties. If the mat has a moisture content below 10 percent, it normally is pressed into a smooth-two sides board which has a very light color in contrast to boards prepared by the wet-process system. At moisture contents over 10 percent, it is necessary to use a screen on one side of a mat, much in the same way as pressing a wet-process mat. The boards manufactured by this method enter the same market as the wet-process board. Some people hold that the wet-process board is somewhat superior in properties because of the self-bonding of the fibers, the intermeshing of the fibers as developed in the wet- forming operation, and the removal of the hemicellulose. The elimination of the massive amounts of water necessary for forming the mat is an advantage of the dry hardboard. However, synthetic resins, which are expensive, must be used as binders in the dry-process board (44). 18 Medium density fiberboard In the mid-1960 's the most recent development in the board industry occured. This breakthrough has been called medium density fiberboard. Most of the plants producing this board use the pressurized refiner for generating finer fibers, which have far greater bulk than those produced by atmospheric refiners. This particular board is usually much thicker than the traditional hardboard, and goes to the furniture industry for use as core stock. Much of the recent interest in building new plants has centered on the MDF process (44). Total Capacity Particleboard production in the United State is currently nearly four billion sq.ft. per year (1987) as indicated in figure 6 and tables 1,2 (68). In the South-East region, annual production of industrial board has increased from 635 million sq. ft. in 1976 to its current level of 1.1 billion sq. ft. The West has also seen growth in industrial board production, from 900 million sq. ft. in 1976 to its current level of 1.14 billion sq.ft. However, some products have seen a sharp decline in overall production levels. In the South-East region, for instance, the production of floor underlayment has declined steadily, as have production levels of mobile decking. The West, however, continues to 19 1 mo .mmm . .mucoemwzm cumonmaofluumm .m.D .oawuoum Hmowuoumwm .o madman 00 nut. 00 «0 po 0. Oh Ch 55 Oh as '5 an «s. pk Oh no .0 so on no '0 no «0 —0 00 on i? l‘ \ [III 0580.0 0:3: 0.303 .5538... Alll\\llN, <\ :8- i... :- ...: g!!- In... 3.3.45.3 ...-(Onwaoficf. .0... 20 0s. . .mOm 02.6%.. 0.00%.. 2.0.00m 30.0: R. . .00. 0 . 0.00 .. 00. NON N00.00N mmv.0bv mm0.m . v ANo.00". 000600 000.000 0m0.0mm 08.500 30.0%.. 00.000. 092 mvmdv. .. .mv.m.o.. «400.90.. 000.000 315 . 0 Nov.00~. b . 0.005 00v.000 bmm.m00 Ommdvm bmb.80 000.000. . 000.000. . 000.N00 vvm.m00 N. . N000 >00. . 0N. 0m . .0. N. 03.00.. 00.1.00 0.0.000 000600 .28.. .2826... o o o h. ..m _N©.N o .00.m «.00 0mm mm. 050920 9::an noEooEoo m co... .03... 638.020 0.96 0. 1.2.5.3 200. 000.0... 000.9. . 02.00 003... 00100 000.9. «.0060 000.0m mo0.0m v . ...mm 00m.mv c 60400 00. ..0 o0~.00 v . v.00 0 .000 09.00 8000 6.0.0.00 o 08.00 m00.0m. mnvdm .0. .mm. 0m0.mm 02.0. . 0o0.mm 80.00. 80.00 2.0.00 0.0.0. 00000 . 3.0.0 00m.mm . 00.00 000. . v 00. .3. v .0.0. .me. 0vm.0m . mm. .0. 800 05:2.0 uooD . mo .umm . www.mm «00.0m . .0.vm 000.00 000.00 3.0.34 000.0v N0. .00 505.00 .860 00v.00 000.02 000.0. . mimmm. b0. .00. 50.3.. Nov. . b. 000.00. 00v. . Om 940.00. 000.00. 000.00. 0:108 0.302 380 .3. 48.4 5.50 4.833.: 00.000 000. .4002” 000. 00m.mom v00. 00m.m..m m00. m0m.m.m ~00. .0...0m .00. mo. .00m 000. ..0...mm 0S0. m0m<.0m 050. 4.00.63 2.0. ..00.mmm 04.0. 5&qu 3030.3 500.0% 000. .80.... 000. 000$? .00. 0mm.0m0 000. 050.03. N00. 000.000 .00. $0.80 000. 40mg... 060. 000.08 04.0. 50.00.. 2.0. 00.00.. 0.0. $010.25 Be» 305 .350: toga-om a manna. 00041990~.sglxzzggn.azgwosaypampmnux‘uzsu.paooaizufitn§~ 21 000. 0030.... 000. 800.. 0... .0. 800%. :0 0:0 .000. 0030.... 000. 20.”; 0... .o. loan-ll :0 0030 0.0 0030.. .300 000. £000.... 000. 200» :. :0..0:0o.0 .20. 0... .0 80. .0 in 0:0 .08. 0:0 08. .0 200> 0... :00200 :0..u:00.0 .0.0. 0... .0 .00. :0... $0. .0. 0:30.50 0.000 0003...“. 000. 0.0.00 «050.. 00:... 0:0 3:0...0...» .0 03.0.. 0... c. 00020:. n. 05000.0...00 0002...”. .m.:0:.0...m>.o..0:00.0 .0.0. 0... .0 .00 :0... $0. 30000.0 m. 50.20 m..:0.0 200 9050.. .00. .000. 0.... c. 00020:. a. .:0.0 0.000.303 0.....0500 0:0 .0 m.:0.:0...m\:0..0:00.0 0.; ate: 0.00m . .000 80.6! 00060. 80.0000 .00.Nv0 000.805 80. mm... 0.8 338 20...... 3.023 3.8.. 208....“ 82 new 1.2. R38 80%. 0......3. £39 8......» 3.2 .52 3.8 5.08 $32 808.... 30.3. 03.8.... 82 we: 38 a}: 802.. $0.80. ......me «8.82 82 3: .38 33.» 03...... 80.8.. 0%... «8.8.... 52 S... :8 . 83...... Exam $58. $38 5.2.... 82 ...... 0...... 3.8» $0.30 80:... 808: 8.08... 2.2 3.: 92.. 238 ...QSN 2.38.. 80...... 808.... 2.2 ...... 0...... 333 3. 08 81.8.. v8.8... 318...... 2.2 3m 0...... 83.: 8020 238.. 508.... :08... 2.2 v. .0 0 000 3.0.; $0.000 000.00? . 000600 80.n00d 000. .. 2.. 802.. «3.0.... mm. .8 838... :2 v 000 000.00 000.00. .0 000.000 0010010 000. 0000. 000. . v0 81000.0 000. 0.80 . 3.000 .NN.000.N .00. «2.. 5.8. 5...... 22 0,80 000. . 00 000.30.. 000. m .... 81:... 5.8.. 82 . 00 .0000w~ 00.600.— 08. 0.00 03.000 8.0.0.0 08. . .00 000.000 $5.000 08. 000.00. 000.50 .00. 000.0. . 0.0.000 08. 000.00. 80.08 N00. 830 52 80...N 08. 000.03 000. .5: ...... 0.34.25. 9.38.. .53.... .8553}. .28. a... a. 3.3.5: 0.8qu 230: ism .0 33> 5.0.: 34.028... 0.830%! . me .000 . «...-on .. x... .8... 22...... 08.8.... 820.32 .08.. .23 033.008.. .3. .o 80312... .38 085.20 "Gag“ 30%: N 0.009 22 produce floor underlayment at slightly higher levels than those of 1976, although_ mobile decking has dropped sharply (68). As shown in figure 6 and tables 1,2, the production of MDF in the U.S. has grown rapidly in the course of the last ten years, going from just over 215 million sq.ft. in 1975 to a 1986 total of almost 781 million sq.ft. The value of this production has increased even more rapidly, begining at 31.4 million dollars in 1975 to over 267 million dollars in 1986. General particleboard and MDF processing systems In this section, the basic particleboard platen type and MDF manufacturing process will be described briefly. Figure 7 shows a typical layout for a platen-pressed board plant. In the particleboard process, wood raw material in the form of planer shavings, plywood ends, re-cycled particleboard panels, as well as round wood chips, etc are stored in the bins. Usually, finer materials are used to produce faces of panels, and coarser particles are used in the cores of panels. Screening is an essential part of separating the particles into material to be used for face and core construction. The screening process is followed by drying. When the particles enter the storage site, their moisture 23 A we .umm . .ucmam m:wusuumwscms oumoanUauuam mo guano 3o~u acadmha .h musmwm _ guzzae oz< . . A: aa.ooo a¢¢oa muncosz: ascoc umoozmz<3 oz: m3¢m «mango o9 _ amazman much same b h 7y“: m ma‘Um ac: mandamus "u nuns 94: can .zmagoq noun“ d D;— . - -..—8m .. I amazuqn uzoo ; sun znn ununu mmszn “non - «so. . . Ergo out... 2933 8828 o w. .o o . g 5.53 62.2: 53 Side 2:... c . . 34m. ...... ¢2§Sw -IJ _ “Em q weae Jflu.. 393.5 «3.... 39:: 2.sz r h :3 sun: tr > .uuauc ( 4 I , >.. a: D J! «was... .3338 cacao: 9.8 I 2‘2. 44:33: .3. 3:. D cameo: as: .oaz. 2:22: 32 each product has a particular set of properties which are important because of the intended end use. In general, the physical properties usually evaluated for board products include static bending (which covers modulus of elasticity and modulus of rupture), internal bond, water absorption, thickness swelling, and linear expansion with change in moisture content. Since properties are usually associated with the density of the product, the density is usually determined along with the physical properties. Measurement of the physical properties of particleboard and fiberboard is generally performed according to the standard methods of evaluating the properties of wood-base fiber and particle panel material ( ASTM D 1037) (2). There are of course other accepted standard methods which are used in other countries such as the Canadian Standard (CSA) (9) and the International Standard(ISO) and so forth. Tables 3, 4 show the general properties requirements for particleboard and MDF. The quality control of particleboard or MDF is associated with testing of these product properties. Many different types of tools can be used in the quality- control effort. These can range from sophisticated instrumentation and computers to micrometers and hand rules. Most plants use statistical quality-control procedures which can provide data on process capability, 33 6003.3! 2.30.8.0 80...: .0 00.2.2.6... no ...-3 38.-0:33.“. 1.032% .80... . . 60...... a. a...) 2.3.8:...- .:..0.-0 ...-:00 ....- .o..0.... .... 03...... .0100. 0......0..a 2.0.2.0.. 203:... .80....2...’ ..Ia .0336! 5.138 .30 03-0... 3...! 010:. $830.03... V0!!- .0... ”a ...».p o 032.3%... 5...... .... ...-...: .52.... ......8...81|8..8§ ...... ...-... ...-8...: ...-333:... 3.3.... .... H . x... . .- .. 8.3.... ado :2 9n .. on 3... .0... on 3.6 98...: N... 8n.~ u .6 na}... on...” ...: .3. ad... a: an... no... no 02... 08...: 0.2 8... _ 5.2.06 3.3.002 .n .03 ....- .u ... on... .62 on... as: 8... an... a...“ 80 :0... 8963 can 8: ~ 8 ...}... on n62 30 an... on... 2. n . v.~ 08...: no. Svfi _ 5.2.0.. ...... .< . N «0»... 09¢ . 8...... ....- .. ... on... «...... ...... on... .«6 on 02... 6863 a... 8!. N 3 ...}... ..n - no... 3. On... I... 2 03.. 98.3. ...... 8o _ 5.3.0.. .5._ ..v a .. 8.3.... 0.2. co. #3. a: on... .0... 8 on..." 80.80 mi. Svfi a ... na}... on...” ...~.. 3. «.3. n2 an... no... .... v2... :86: ... .. 80.. . 5.80.. c.3902 .- .05 ....- I ... . - . an... 3... cc. 2'.“ 98...: 0.2 806 ~ on... .0 n5.... on - . .202 3.0 3.... on. :3 Sta 8963 ad. 89.« . ...-:09 ...... .< . .. .8... as... 3. ...... .... ...... .... 3.... a... ...... ...... 3.... ...... 3.. ....g. .....- ........ 3.. a}... 3.- 5.... .... ...... B.- ..t. 13.0 .0..- 5.... 0am... 0...... 53.5.5 3 5.02.80 03...... «0.3.». 5.2.5 .02.... 150.... \o \o 9.55.. 30th 243 an: A we . 00.. v 2883...... ..8... 353.....33 assess... 56%... . m 03...... 34 Table 4. Property requirements for MDF. ( Ref. 49 ) imam m M (Tm Stung!!! W W d at Want to Luna W mm mm Some.) W has Eda. inch. psi psi psi 96 I08- lbs. 1310 and , wow 3.000 300.000 90 0.30 325 275 m and above 2.000 250.000 80 0.30 300 225 ‘Form m MIMI mum at.“ incl: or Inc. the Uncut Expumon value Moll 000.35%. 35 assist in setting specifications, and help with control decisions. There are several basically different methods of production of particleboard and MDF and many variables within each method. It is not within the scope of this study to outline those methods of productions or all of the parameters. An effort will be made to investigate some of the basic variables that are always present. These basic variables influence the mechanical and physical properties of particleboard and medium density fiberboard. Chapter II OBJECTIVE The objectives of this study are two fold: The first is to describe the factors involved in the development of strength, stiffness, and density variation in particleboard and MDF. The second is to investigate the relationship between density and internal bond strength of particleboard and MDF. A number of factors affect and develop the strength properties of the boards. Among the major parameters are wood species, the raw material type, mat moisture level, the resin type and distribution, board density, particle orientation, .particle geometry, and pressing cycle variables. 36 The glue line contact and development of mechanical properties of particleboard and MDF depend significantly on densification. Almost all of the above mentioned parameters interact with each other in one way or another. A change in any one of these factors will result in a change of the effects of many of the other related factors in the board process and consequently in changes of strength properties. Thus each factor cannot be thought of as an individual entity which can be manipulated easily to control the board process. However, once it is recognized that there is an interrelationship between a number of factors and how different parameters and process variables influence the strength properties, a more complete grasp of the process can be attained and actual manipulation can be achieved for controlling and modifying the strength property of special interest, namely, internal bond strength and its relations to density. The subject of internal bond strength property of particleboard and MDF has been investigated by many researchers. However, there is an inconsistency in the results regarding the relationship between IB and density and density gradient. While many workers have found a clear relationship between IB and average density and between 18 and core density of particleboard, 37 some workers claim that those relations are not strong and might not exist (18, 70, 71, 97). It is believed that MDF is not similar to particleboard in terms of core density and IB relationship. It is interesting that, if there is no relationship between core density and IB of the board, then it may be possible to reduce the density to some extent and gain the benefits of density reductions. So, the second objective of this paper is to determine those relationships for two commercial particleboard and MDF and searching for a possibility of density reduction. However, any density reduction would require sacrifices in one or an other of several important board properties, e.g., the density reduction causes a reduction of the bending strength. In order to minimize strength properties, it would be important to know the relationship between overall density, density gradient, face and core densities and the major important strength properties. The scope of the experiment is limited to only two commercial products. The results, therefore, although they may used as valuable indicators, cannot be extended directly to other commercial particleboard and MDF. Chapter III DEVELOPMENT OF PHYSICAL AND MECHANICAL PROPERTIES OF PARTICLEBOARD AND MDF As mentioned previously, there are many factors affecting physical and mechanical properties of particleboard and MDF. Among the major factors are : type of raw material, raw material density, resin type and distribution, press cycle variables and many other factors. In the following sections, the structure of particleboard and fiberboard will be outlined first. The main factors for developing physical and mechanical properties will be discussed next. The structure of particleboard and fiberboard The characteristic feature of woody cells is the special nature of their cell wall. The layered construction of the cell wall and the organization of the fibrils in the layers have an important bearing on the strength properties of individual fibers, which in turn will be reflected in the behavior of solid wood and wood products (91). Particleboard can be viewed as a structure consisting of layers like plywood. Each layer consists of a certain amount of wood substance with void spaces interspersed (83), see figure 10. This structure has an effect on the formation of continuous glue lines. Upon compression of this mat in a 38 II. //I./ [III II//. 7. III/III I- III/I IIIII/II. - l’IIIIlI/II I 'IIIIIIIII. '/III ”III [III ’I/I/l/l/ I 'IIIII. 'III/II/ III/III I- III/III. - VIII/I. -'IIIII 'Illll/I 'III. III] VII/III ”Ii IIII. ’IIIII II; ' 'III - III III/IIII/Im -/ ’/I/I//// I-I III/III. III- Ill/m II'll/II PARTICLEBOARD Figure 10. Arrangement of particles in the manufacture of plywood and particleboard. ( Ref. 91 ) 40 hot press, however, the amount of void space is reduced and the total area of contact enlarged. The discontinuity of the layers may, therefore, be compared with the surface irregulatities of solid veneer sheets in plywood (91). Major factors developing physical and mechanical properties Species is one of the most significant factors influencing the physical and mechanical properties of the boards. Numerous wood species are currently utilized in the manufacture of particleboard and MDF throughout the world. The technology exists for using almost any wood species for the manufacture of particleboard and MDF. Particleboard manufacturers have a definite preference for softwood because the quality of a softwood board at a given board weight is considerably higher than that of a board made from heavy hardwoods. In other words, hardwood boards must be heavier, to be of the same quality as softwood boards. Aspen and other lower density hardwoods are, of course, exceptions. A wood of low strength properties (aspen) produces a particleboard with higher strength properties than another board made with a wood of high strength at the same board density, see figure 11. This is so because 41 L Legend: 6 —°— poplar --o-- pine A '-'-.'-‘ MD a 5.. 5 -~---— men \ 3 '§ " E 3 3- w a: p— (n 32' 3 l- wl l l l 1 0.4 0.5 0.6 0.7 0.8 BOARD SPECIFIC GRAVITY Figure 11. Example of bending strength of flakeboard affected by both species and board density. ( Ref. 5 ) 42 unit weight of low density wood occupies a greater volume than the same weight of high density wood. When these volumes of wood are compressed to the dimensions of a board, a higher relative contact will occur in the case of the greater volume of wood and a better glue bond between flakes results (29,82). Most often the question of species or raw material mix is settled on economic considerations rather than technological ones, although these two are not always independent (47). Much of the problem of species variation can be handled by constantly subdividing the various types of raw material recieved from various sources and mixing them back together (44). Resin bond The two principal and predominant resins used for particleboard manufacture are urea and phenol formaldehyde. The amount of resin used in board manufacture is expressed in one of two ways : (a) Percent of resin solids based on the oven dry weight of wood furnish. (b) Amount of resin solids used ( in grams or pounds ) per unit surface area of the particles. The amount used in manufacture of particleboard varies over a considerably range . Most often however, the amount of resin used ranges from 5 to 9% for urea 43 formaldehyde resin and from 3 to 6% for phenol formaldehyde. When multi-layer particleboard is intended for furniture manufacture, the resin usage in face layers may be as high as 10-12% , with 6% or higher in the core (47). Resin efficiency has been defined as the application of a minimum quantity of adhesive to wood particles, with resulting optimum physical properties in a pressed board. Theoretically, resin needs to be applied only to those portions of particles which will come into contact with other particles. The resin droplets are sufficient to act upon the entire contact area of the particle (8,13, 37,). Lehmann (37) showed how fine atomization influences strength properties. He noted that at a density of 0.65 g/cu.cm. boards of desired strength level could be obtained with 19.6% to 27.9% less resin with fine atomization than with coarse atomization, see figure 12. However, he pointed out that regardless of resin efficiency, in the density range of the experiment, static bending increases rapidly with increase in resin content from 2 percent to 4 percent. However, those values will not increase significantly with increase in resin content from 4 percent to 8 percent, see figure 13. The adhesive bonding of wood fibers involves thermodynamic and kinetic parameters and their interactions (99). 44 INCREASING PERCENTAGE or REsm INTERNAL BOND RESIN DROPLET SIZE ; Figure 12. Resin efficiency in particleboard as function of resin drOplet size. ( Ref. 100 ) 4S a; 4- & ui' C a 3. d. 3 2 IL ° 2_ kcomse g ’ , SPRAY " 4 11' 3 J O 3 .. specmc envm . o - 0.55 '. a 0.73 c 1 1 1 L ' o z 4 s a RESIN CONTENT, PERCENT Figure 13 . Example of MOR of particleboard as influenced by resin content. ( Ref. 34 ) 46 To give the adhesive polymer mobility, it is applied to wood as a solution. As soon as the adhesive solution is applied to the wood, the solvent wets and then penetrates and adsorbs into the fiber, even without pressure, leaving most of the resin solids on the fiber surface. The rate of loss of solvent depends on the moisture content of the wood fibers, roughness of the surface, adhesive concentration and inherent attraction between wood and solvent (99). Application of pressure to adhesive-laden flakes spreads and transfers the increasingly viscous adhesive and causes bulk flow of adhesive solids and solvent through the wood surface into the large pores and capillaries in the fiber wall, thus reducing the thickness of the film (99). If too much adhesive penetrates into the fiber structure, the film is discontinuous on the fiber surface (99). Cure of the adhesive involves two interrelated process: solvent loss and chemical crosslinking. A high solvent content of the adhesive retards crosslinking. Too rapid loss of solvent makes the immobile solids incapable of forming a crosslinking film (99). Heating increases both solvent loss and crosslinking, although each rate increases differently with temperature (99). Most of the gluing parameters are affected by time, 47 temperature, pressure, and fiber properties. So, it is possible to manipulate the adhesive property to form a good bond. The thermodynamic processes of wetting and adsorption are less amenable to manipulation than other gluing parameters (99). Although many wood properties influence bonding, moisture content is most important, see figure 14. When moisture content is high, wetting occurs readily, but the amount of water adsorbed is low. So, at any specific wood moisture content, the assembly time and amount of adhesive are to be adjusted for optimum bond quality (99). Density also affects gluing in several ways. The glueline may vary in thickness because dense woods are more difficult for adhesives to penetrate and are less conformable under pressure. Because most structural wood adhesives do not fill gaps, thick areas in glulines tend to be weak (99). Dense woods exert more stress on cured glulines because of changes in wood moisture content. In addition, the greater the wood density, the greater the compression required to consolidate particles and flakes into a board (35,99). The broad range of densities that may be found within the same piece of wood may create bonds of different quality in a board 99). 48 1‘” ‘5 Inn/cmz ‘\\j W \ 12% \L\ ‘2 ’\ . - \ 3' 000 \ 10 l :1 l‘ g \ ...... i g \\ \ Putnam . E noos\ “ I 1 t \ 7 \ § § '- 2 3 \ \i I i Emu Ix:\\. j: ' 5: 5 \ | ' a \\ ' ? % .\\\ fi 1 ‘ ‘i \m-——-—-—1. 200:: . ‘ z M— a A ‘ 9 Example of dependence of some strength prOperties of solid 20 n Moisture concern Figure 14. wood and particleboard on moisture content. ( Ref. 31 ) 49 Particle geometry The total range of sizes and shapes of wood elements that could possibly be considered for composition boards are shown in figure is. This is a general classification of wood elements starting with the largest element, the log. Smaller element can be defined until the molecular level is reached. Included in this classification, are the typical elements used in particle composites, namely chips, flakes or wafers, strands and particles. Flakes are particles of predetermined dimensions and are produced by the action of the knives cutting across the grain (either radially, tangentially, or at an angle between) and they are flat, thin particles. In the following, the term flake refers to a special type of particle. Particle geometry has tremendous effect on mechanical properties. When considering flake for raw material, two aspect of flake geometry are of significant importance : flake thickness and flake size. The flake thickness relates to the degree of flexibility of each flake when compressed into one layer of particle or flakeboard. The flake size controls the total surface area exposed for resin dispersion. Together, these two aspects of flake geometry influence, to a large extent, the mechanical 50 Figure 15. Most common forms of wood composite elements. ( Ref. 41 ) 51 strength of the board, the surface characteristics, the machining properties, and properties such as weatherability and dimensional stability (4). Suchsland (81) introduced another aspect of flake geometry; the flake length-width ratio. He observed that the area of definite glue line contact between flakes was a function of length-width ratio of individual flakes, and that was a significant factor in developing bending strength. The ratio of flake length to thickness is another indicator suggested by some workers (6,57). While modulus of elasticity increases with thinner or longer flakes, internal bond strength improves by using shorter and smaller sizes of flakes (6, 45, 48, 62, 59, 69). The linear expansion of flakeboard tends to increase with increasing flake thickness and to decrease with increasing flake length (44). The thickness swell of flakeboard tends to increase with increasing flake thickness except with low board densities (44). With mixture of particles having widely differing geometrical characteristics, the strength properties tend in general to be intermediate between the properties obtained with the different kinds of particles separately (44). 52 Densification of boards Low glue spreads and structural characteristics of the mat require densification. The fundamental variable affecting board properties is the degree of contact between particles. So, densification is the major step developing a good glue bond between particles and developing the strength properties of the loose particleboard mat to produce a relatively strong thin particleboard. This is done by using a hot press. The common techniques utilized to compress the loose mat into the final thickness consists of using steel bars (stops) whose thickness is equal to that desired for the consolidated boards. In this technique, the press is closed until it reaches the stops. Once the stops are reached, the pressure applied to the densified mat begins to drop. The densification of the mat involves considerable plastic deformation. Density Density is one of the most important and most easily determined characteristics of composition boards. It allows for rough estimate of other physical and mechanical properties. High insulation values, dimensional stability, and low cost are generally associated with low board densities 53 whereas, boards of higher densities generally offer higher strength properties, have higher expansion coefficient and are more costly (82). Density is a measure of compactness of the individual particles. Thus, the two most important factors controlling the average final density of a board are the raw material density and the compaction of the mat in the hot press. Any changes in one of these factors requires an adjustment of the other if the average board density is to remain constant. Either of these factors can also be changed to increase or decrease the average board density (47). At a given board density an increase in raw material density causes a decrease of particleboard strength properties (4,15,16,40,46,69,95,97). Most researches have found a positive relationship between particleboard properties and board density. An increase in board density increases values of modulus of elasticity (MOE), modulus of rupture (MGR), and internal bond strength (18) (4,24,77,94,95,97). Density variations in boards The density of a particleboard is subject to variations both within a board and between boards of a given sample. These variations of the density may be due to the limitations of the manufacturing equipment and variations in the raw material qualities, or they are 54 inherent in the process. The density variations between boards of a given sample are due to the former reason while those within a given board are due to both sources of variations (82). The density distribution in a homogeneous particleboard has a horizontal and a vertical component. The horizontal density distribution is a direct consequence of the discontinuity of flake layers and is determined by the particle geometry for a given compression ratio while the vertical density distribution is the result of interactions between most raw material and process variables. The vertical density distribution or density gradient of a board is highly dependent upon the particle configuration, moisture distribution in the mat entering the press, rate of press closing, temperature of the hot press, reactivity of the resin, and the compressive strength of the wood particles. The effect of temperature and moisture content on mat compressibility is highly responsible for the formation of a density gradient through the board thickness. As the hot press is closed against the mat, vaporized moisture in surface layers migrates toward the more cooler core. This moisture movement facilitates transfer of heat into the board (70,83). Particle geometry also, influences the rate of moisture 55 migration and therefore the range and the shape of density gradient (14). The time required to close the press to stops is directly related to the initial pressure; hence, a high initial pressure will cause the mat to reach final thickness before sufficient heat is transferred from the platens to the mat interior. With a lower initial pressure, press closure speed will be slower, thereby allowing the interior to attain a higher temperature and a lower compressive strength before the stops are reached. The slower the press closure speed , the lower will be the vertical density gradient (78,82). The vertical density distibution has a marked effect on the mechanical properties of the boards (77,82). High bending strength and stiffness are associated with high face densities. Internal bond, shear strength, screw-holding power, and related properties of particleboard are normally functions of the core density (82). Multi-layer boards Single layer board or homogenous boards were the first type of particleboard to be developed. They are almost isotropic in strength and physical properties in the plane of the board, due to the random arrangement of the particles. Their strength properties are very closely S6 allied to density (1). In commercial production, the most common pattern of layer-density is a sandwich construction, or multi-layer board, having the faces of considerably higher density than the core and frequently containing finer particles in the face layer. Because of the larger size of particles in the core, the core cannot consolidate to quite the same degree as the faces, and hence, this core is invariably of a lower density and frequently incorporates a lower proportion of resin than the outer layers. This type of construction results in a board having adequate bending strength and shearing stiffness for most purpose, but a relatively low tensile strength perpendicular to the plane of the board (1,28, 84). The mechanics , process techniques and control of three layer boards Sandwich structure in particleboard is the result of a laminating process in which the manufacture of the laminas and their assembly are combined in one process. The final properties of each individual layer are composite functions of raw material variables and variables of the laminating process. The properties of the individual layers of a three- layer flake board are functions of their densities. The 57 relations between layer density and overall density is expressed by the following equation (81) : St - Sr(l - x) Where : $1 = Layer density, g/cu.cm. St = Overall density, g/cu.cm. 51' Core density, g/cu.cm. X ll Shelling ratio, = (total thickness - core thickness) """"" ESEQI'EEIE£;;;;"""'" For a given overall density and a given overall thickness, core density, face density, and shelling ratio (a ratio of face thickness to total thickness) can be controlled only indirectly by varying the following factors: 1- Raw material variables affecting the compressibility such as species, flake geometry and initial moisture content. 2- ratio of core weight to face weight. 3- Press cycle (81). In particleboard structures made up of more than a single layer, a basic characteristic involves the differential strength properties between the face and core layers. This type of differentiation permits densification at surface layers designed to withstand higher bending stresses without undue imposition on cost 58 and product weight. To obtain the strength differentiation, one or more of these techniques are employed (47, 81) : -- A higher adhesive content in the face layers. -- different particles( e.g. smaller, thinner) in the face layers. -- Lower density wood species in the face layers. -- Processing techniques which reduce compression strength of the face particles, e.g. , using particles of higher moisture content. -- Surface particle orientation to achieve higher bending strength in the direction of orientation with other parameters ( resin content, particle size, etc.) kept constant. From the following equation, face modulus of elasticity can be derived, knowing the average MOE, core MOE, and shelling ratio (81) : MOE(t) - (1-x)3MOE(c) MOE(f) = --------------------- psi 1 - (1-x)3 Where : MOE(f) = MOE of face, psi. MOE(t) = MOE of total board, psi. MOE(c) = MOE of core, psi. x = Shelling ratio. See figure 16 for an influence of raw material and process variables on the MOE. 59 nu.” an an mm W' o o.» as o: no on m on as m Ntmvolmnm . , o l commutes? 3 ' 3 8, ’ no mmwmfl; ' i 1 . f) A anmo-m nan-x . i 'V / um um com A g, mr m'dmw. :,’ ,4; ! I a “'3 g 8 a z»? mun-mun“ cal—um Figure 16. Example of MOE of three-layer flakeboard as affected by raw material and process variables. ( Ref. 81 ) 60 The relationship between face density and face MOE and between core density and core MOE of particleboard have been studied by some researchers. Good linear relationships have been found between face density and face MOE and between core density and core MOE (18, 70, 83). As face density increases the relations become non linear (18). The relationship between core density and IB is reported by some workers to be linear (10, 11, 81), see figure 17. Control of density gradient In general, the linear relations between MOE and core or face densities, and between IB and core density depend only on the density of the species, flake geometry and glue spread. They would only change when one or more of these raw material variables is varied. They do not depend on moisture content or press cycle and can be established independently for any combination of species, flake geometry and glue spread (81). As mentioned previously, press cycle and closing time have tremendous effect on the strength properties. The basic function of pressing operation in the manufacture of particleboard is the development of an adhesive bond between individual particles and to densify the mat into the desired thickness. 61 I '°° % ! Cl\f //’ - I50 ... ,7’ - o g 0 o . l//.h ' l/ I. g m o /J’I g E «n ’,' . ” z IMIE i /’:' o o I- // ' 4 i“ /r" s ' I- ‘ a 200 #fl—-*- "if—’1‘"- - 11" —‘{--~—%—-——‘ 50 u 3 I I i 1 an an .no so .uo l“ .uo .no culnunwv «av Figure 17. Example of dependence of IE and shear strength on core density of three-layer-flakeboard. ( Ref. 81 ) 62 The necessity for minimizing press times has been of considerable important in press cycle design and selection of raw material variables although care must be taken to insure that resin is completely cured despite shorter pressing time. Prepressing and high frequency preheating are possible ways to achieve such a reduction. Factors which affect the time are : temperature required to cure resin, desired board density and wood species used (22). The most important aspect of the press cycle is closing to position. Closing time refers to the time period between initial pressure application and the moment at which the mat is compressed to the thickness of the stops. High initial pressures result in short closing time, high face densities and lower core density, see figure 18. Low initial pressure result in longer closing time and a more uniform density distribution at the same average density (56). The press closing time not only influences the range of density through the board thickness, but also the shape of the resulting density profile. With short closing time, the density profile takes on a U-shape. Density would be highest on the surface and then would decrease along a regular gradient to a minimum in the core. As press closing time increases, the density 63 ..L r 3i at» me, 30L. 'cLou-mac "°‘| I | 70% I g , I . . I 0 I0 20 so 40 t Percent Distance From Face Figure 18 . Example of dependence of layer density of particleboard on press closing time. ( Ref. 10 ) 64 gradient goes through different M-shape configuration, see figures 19-20. Reducing press closing time would lower core density just as it would increase face density (56, 70). Short closing time would improve MOE and MGR (70, 78, 83). High densified surfaces increase the bending strength of particleboard but the resultant lower density of the core region normally reduces IB strength ( 10, 11, 39, 81, 83). However, some workers reported that short closing time would increase IB strength and core density and internal bond might not have a close relationship (18, 70, 71). Some special features of MDF In the previous sections, major factors for developing physical and mechanical properties of particleboard have been disscussed. MDF has almost similar properties as particleboard in most cases. Some of the inherent advantages of MDF over conventional particleboard include : higher 18, face screw holding, edge screw holding and better edge machining. In contrast, MOE of MDF is lower than that of particleboard. Comparing the two products, MDF is more costly than particleboard (86). 65 a! Figure 19. Example of the influence of short closing time on the shape of density gradient. ( Ref. 70 ) Figure 200 Example of the influence of long press closing time on the shape of density profile of waferboard. ( Ref. 70 ) 66 The smaller fiber elements used in MDF result in a much more uniform distribution of the material in the board. This reduces the variations of its properties. MDF of a good quality can be made from a furnish that contains bark, provided resin distribution is adequate. By inclusion of bark, IB and other strength properties are reduced to some extent. These losses could be countered by altering the press schedule or by reducing the percentage of fines by screening (103). Bending strength of MDF similarly to that of particleboard is affected by various factors. Bending strength of MDF at comparable densities would be greater in boards made from raw material of low specific gravity (103). The density and density distribution directly affect the MOE of MDF (90). This effect is explained in terms of a strong correlation between the MOE and face density of MDF. However, some workers did not find any correlation between MOE and average density of MDF (86). Unlike particleboard, core density of MDF does not appear to have as strong an effect on the IB strength (86, 89, 90). The density gradient of MDF is an important variable that can be controlled to a considerable extent during the manufacturing process. An extreme density gradient of 67 MDF has the advantage of high MOE at a very moderate average density. Its disadvantage might be that the correspondingly low center density could result in low level of IB strength, edge screw holding power and similar properties which rely on the integrity of the center layer of the board (86). However, a uniform density gradient would produce high IB but poor bending properties (76, 103). This could only be achieved by eliminating the temperature gradient during the compression period, either by heating the mat uniformly throughout its thickness or by pressing the mat at room temperature. The first possibility may be realized by using high frequency heating (76, 89). The second option would be approached by using very high pressure which would bring closing times to a minimum (89). Like in the case of particleboard, high face density and low core density of MDF are achieved by short closing time, see figures 21-22. Fiber alignment would improve the bending stiffness in the aligned direction of MDF. Therefore, MDF with properties in one direction equal to those of random boards could be made at lower densities and with less raw material (104). But fiber alignment does not seem to have a significant effect on IE of MDF (74, 86). 68 DENSITY CONTRAST —O 0——- PRESSWE CLOSING TIME ——0 Figure 21. Example of relationship between pressure or closing time and density profile of MDF. ( Ref. 89 ) 69 humps MODILUS OF ELASTICITY- CLOSING TIME - SECONDS Figure 22. Example of relationship betwee press closing time and MOE of two press platen temperatures. ( Ref. 90 ) 70 The measurement of density variation The measurement of layer density has been extensively used in particleboard and fiberboard research for studying the relation of the board structure to physical and mechanical properties. Measurement of the density gradient is of use in qualitative evaluations of the manufacturing process and may be accomplished using several techniques. The normal and traditional method of determining the vertical density gradient is to accurately determine the weight and thickness of a given specimen, remove a thin layer from top and bottom surfaces and accurately determine the weight and thickness of the resultant specimen. The density of the removed layers can then be determined from the weight and volume of materials removed. This is normally referred to as the gravimetric method. Although this method is simple, it is very tedious, is time consuming, has limited resolution, and destroys the specimens. Consequently, it has not been widely used in production control. Stevens (75) in an attempt to develope a gravimetric method, introduced a slicing apparatus to remove a thin smooth layer, as thin as 0.25 mm. He described the limitation of the surface smoothness and precision of measurement. Nearn and Bassett (51) examined x-ray 71 radiography as a means of detecting density variation within a board. In their experiment , x-ray is applied to the cross section of the boards. Dark portions on the exposed film indicated areas of higher density, whereas, light portions indicated areas of lower density. This result can be used as a rough estimate of density variation and does not provide an accurate density profile. A fast and yet simple non destructive test method for determining the density gradient is the gamma radiation method. The density gradient in this method is determined by incrementally measuring the amount of radiation which passes through a sample (33). The non destructive quality of this measurement allows the correlation of mechanical properties to density gradient characteristics on the same sample and could be used for in-plant quality control, too. This method is not significantly affected by the normal ranges of resin content, changes in wood species and ambient moisture conditions. The cost of this system is relatively low, it was s 5200 in 1982. The limitations of this method are that only 3 in wide samples can be used in the specimen holder of the stage and the poor resolution in the outer layers. Winistorfer et al (101, 102) used a direct scanning densitometer based on the gamma radiation technique to measure density profile which may be used 72 for a wide range of specimen thickness. Chapter IV IB OF PARTICLEBOARD AND MDF When the press is closing in the manufacturing of particleboard or medium density fiberboard, the core-zone of the midlle layer compresses least. Therefore, its strength is also the lowest. Thus, it is worthwhile to pay attention to this strength, namely, internal bond. The internal bond (IB) or the tensile strength perpendicular to the surface of particleboard or fiberboard has been used as an important indication of board quality in both production and utilization. The internal bond test is important for all composites; it provides direct information on the quality of the adhesive bond between wood elements and shows the location of the weakest surface within the board. Thus, the IB test is generally considered the most significant for determining composite board quality. It not only reveals the quality of the glue bond, which in turn allows estimates of related properties, but it is also an important quality control tool, which, in combination with the MOE provides clues to the balance of board characteristics, which is affected, for example, by the press cycle. 73 The internal bond test methods The American standard Test for determining the internal bond strength is ASTM D 1037 (2) which is the same as the Canadian Standard (9). It requires 2 in sq. specimens glued between two metal blocks. The glued specimen-block assemblies are pulled apart at a constant rate of strain applied in the direction perpendicular to the surfaces of the board. The specimen is tested to failure and the failing load, which occurs in the weakest planes, is recorded. The 13 strength is expressed in psi. The Japanese-Industry Standard (JIS) has adapted a tensile delamination (push-off) test as a simpler method for evaluating the internal bond of particleboard (25). The Australian Forest Products Laboratory applies the torsion-shear test for routine determination of torsional properties related to internal bond (42). The American Standard Test has the advantages that it evaluates the weakest plane in the board, which does not always coincide with the center plane, and that it is equally applicable to thick and thin boards. On the other hand, the necessity of laminating each individual specimen between steel blocks and cleaning the metal block surfaces is inconvenient and time consuming. It also has technological disadvantages such as the dissimilarity of the two materials and the glue lines being subjected to high stresses. Besides, the standard 74 IB test method is not entirely satisfactory for production quality control work in which a quick result is frequently of primary concern, and it may not be economically feasible for a board user such as a small furniture maker who would have to acquire a relatively expensive testing machine. Soper and Hann (72) showed that when using the standard method of the IB test, the adhesive and bonding conditions must be carefully selected. The type of adhesive used to glue board specimens to the blocks, how this adhesive is applied, and whether the surface of the board specimen is sanded or unsanded could significantly affect the apparent internal bond strength values obtained. For example, when the surface layers of the board are weak, using an epoxy resin can cause misleadingly high internal bond strength values by reinforcing those surfaces, or the hot-melt adhesive applied at a temperature of 285 F could cause heat deterioration of particleboard, resulting in low internal bond strength. The need for a speedy, yet reliable method to control quality of particleboard and fiberboard in manufacturing has encouraged many attempts to devise alternative method for determining 18 strength. However, each alteration of the standard test might have disadvantages. Most of them either require gluing or restrict failure to a predetermined plane. 75 Most alternative IB test methods are based on determining torsion strength, shear, torsion shear, or twisting shear strength with good linear correlations between those properties and IB strength derived from standard test. Knowles (29) described a rapid method determining IB strength in particleboard manufacture. His idea was to organize and simplify the routine of plant hot testing of particleboard for IB. A sample preparation jig which quickly and accurately sizes the test pieces and thin metal tabs (instead of using 2 metal blocks) which cool rapidly provide the means for a rapid test response. It is possible to reduce the response time between sample removal from the line and data evaluation. The benefit of quickly testing IB would be that quality assurance personnel can respond quickly to changing test results by adjusting manufacturing variables. Following Knowles, Lebens and Hall (34) suggested models of electrically heated hand-held adhesive applicators to speed up gluing of internal bond test specimens. Lehmann (36) examined two different IB test methods (and compared them with standard test. In a modified ASTM test wooden blocks were substituted for metal, the other was a new push-out test. In the push-out test, specimens were prepared by cutting halfway through them from opposite sides with hole saws of two different diameters. 76 Force was applied at the bottom of each out until failure occurred in the area between the cuts. Lehmann found a good linear correlation between modified ASTM and ASTM IB test method and between push-out and ASTM test. The push- out IB test method was faster, the modified ASTM was intermediate in terms of time consumed and the standard ASTM test was slowest. However, in the push-out method, failure is forced into the center of the specimens, wheras they could occur in the zone of least strength using the standard test. The strength values obtained with steel blocks were higher than those obtained using wood blocks. These differences between the two methods can perhaps be explained partly by stress concentration induced by the action of yoke arrangement at the edges, see figures 23-24. Szabo and Gaudert (92) used the idea of Lehmann using wood end blocks instead of metal blocks with eye screw attachments for IE test. They explained that the stress field over the entire block area in an eye screw attachment is more uniform than that of a yoke arrangement. Lehmann (38) in a recent study, outlined a fast IB test method for both urea and phenol formaldehyde in composite material. This test is based on a simple shear test device and a short exposure to high temperature and humidity in a small pressure cooker. This test does not need gluing and it is a fast test and an accurate one. '77 . on .uom . .ouaonoaoauuna uo anon “sous-=6 can name an thousand coozuma couucnouuou .va «mamas .. .5! I?! 36.98. in... 09. o 4 d I a I“ d 1‘ 4 11 1 d) 1 d 4\ \ r 7 \ .. \. \ \AI .22... I A e a 3.3.2... 1 I... r L MM ., 18mm . aw . on .uoa . .ma masons 3 .32 :5! a aim-E at 8 {*444dI1 IU=J.=§ . r 034. . as: . 4 1 l J j d A. F- 3 l‘sOo.0.. p .cHMOAdqomuuan «0 bomb on mustangs tamudeI 6cm been an chansons coozaob coaDMAOuuoU 84'” m IA '01" OIMI 78 Shen et al (64,67) introduced a new method for determining IB strength. They found a good relationship between torsion shear and internal bond strength. Their technique is simple, rapid, low cost and does not not involve gluing problems. The speed of testing might be of interest to particleboard manufacturers, and the low cost of the test equipment which includes a precision torque wrench and a pair of modified wrench sockets may be of interest to particleboard users who are not equipped with a conventional testing machine. This test may give erroneous results if the weakest plane does not concide with the center plane of the boards. Besides, it is difficult to test thin specimens using this method. Shen (63) also found a very close correlation between a twisting—shear test and IB. This method could overcome the difficulties of determining IB using the torsion-shear technique on thin specimens. Again, the limitation of this metod is that failure is forced to occur in the center plane of the board. The other short coming of these methods is that they will not offer the accuracy of a testing machine which is necessary for research studies. Shen and Carroll (65) examined the correlation between layer strength and layer density. They used the same technique of torsion shear to analyse relationship between layer 18 and layer density, see figure 25. Vogt (98) used stress wave analysis to evaluate the 79 or I 3 soo- f t e s > I- 5 5 o z 40*- U ’ I 3 ’ A E 3 I C 9 ! O I e I . 3 i a 30+ x.“ ,e 7 O 9 so Is 20 Figure 25. Example of relation between layer strength and layer density of particleboard. ( Ref. 65 ) 80 IB strength. That is a non destructive test method can provide useful predictions of IB strength. Many other workers examined the correlation between torsion strength, shear strength and IB (l7,39,53,66,105). One of the alternative tests based on the well- established high correlation between shear strength in the plane of a particleboard and IB strength, and which is also used in this paper was introduced by Suchsland (85). This alternative test does satisfy the two important advantages of the standard test : evaluation of the weakest plane not necessarly in the center plane, and applicability of the test to thin boards. Suchsland demonstrated that if a particleboard specimen could be oriented so that its center plane concides with the plane of maximal shear stresses in a column, then its shear strength could be determined in a compression test. Although this method does not require any wooden or metal blocks it does need gluing . The procedure of this method will be described in more details later, see figure 26. Hall and Haygreen (20,21) developed the " Minnesota Shear Test " which in principle is similar to Suchsland's test in that it utilizes compressive and shear forces. However, by using an apparatus that orients the plane of the specimen at 45 degree to the direction of the compressive load, gluing is not required. 81 400 177') 300L- A ' ,9 , I 200 I00 Y - INTERNAL BOND - PSI a o 400 800 I200 I600 zooo ‘ x - SHEAR smeucrn - PSI Figure 26. Example of relationship between standard 13 test and shear strength of particleboard. ( Ref. 85 ) 82 Relationship between IB and density As pointed out previously, there are many factors influencing the strength development of particleboard and fiberboard. The effects of the main variables on strength properties including internal bond strength were outlined.In this section, the relationship between IB and board density will be described briefly. The internal bond strength property of composite boards is dependent on many factors such as wood species, board density, density profile, resin type, resin efficiency, resin distribution, particle geometry and press cycle. Efforts are continually being made to improve internal strength by alteration of particle geometry, by altering resin content and method of application (54, 60), by modification of press schedules, by heat treatment (87), by oil tempering (19), by preventing degradation of resins (3,52,61), and by experiments with many other factors (7). Results of many early researches have revealed that higher IB values occur with increase in board density. Vital et a1 (97) observed that IB values generally increased linearly with increased board density but these values were not as closely related to board density as were MOR and MOE. Vital et al reported a decrease in IB at constant board density as the compaction ratio 83 increased from 1.2 to 1.6. The IB decrease was attributed to the increased amount of flake damage at the high compaction. Hse (23) showed that IE was significantly lower for denser species than for other less dense species. Stewart and Lehmann (77) offered that althougth IB increased with increasing panel density, this increase was not related to species density. In their experiment, they used four different species, three different thicknesses of cross-grain flake, and various board densities. They observed that Red Oak IB values were very high. This high value in IB was due to the furnish having a higher percentage of smaller particles. Red Oak, a ring-porous species, breaks down more compeletly into smaller particles during flake and board manufacture than does a uniformly textured wood . In general, more uniform horizontal density distribution of particles and the use of smaller particles result in better contact between particles under application of heat and pressure and thus the resulting boards have better bond between particles and have fewer weak points and consequently higher 18 strength than boards made from particles not broken down to the same extent. 84 The internal bond of three-layer boards The internal bond strength of three-layer boards improve as core particle configuration changes from a long wide flake to planer shavings (6,77,80,93). Shorter and smaller size of flakes results in better IB. Hse (24) found a linear correlation between IB and compaction ratio, see figure 27. Maloney (43) in a study of resin distribution in layered particleboard examined various resin levels of each layer while keeping the total resin content in the board constant at 6 % . He found that high face resin level produced superior MOR values but very low IB properties. In contrast, with low resin level in the faces, internal bond properties were much higher, but many test failures occurred in the surface. Removing the relatively soft faces of these latter boards by planing shifted the failure location back into the core and further increased the IB strength. Maloney found the optimum percent of resin in both face and core was 6% . To improve IB of three-layer boards one could increase the core density while maintaining the shelling ratio (the ratio of the face thickness to the total thickness) at the expense of MOE (81). This is 85 400 Y ' -358 + 423: r I 0.05 300 I- § n k \ 200 I- Q ~ I00 '- o 1 A «as I9 I} to COIPACY IO N RATIO Figure 27. Example of relationship between compaction ratio and IB strength of flakeboard. ( Ref. 24 ) 86 possible by manipulating the press cycle. By changing the press closing time alone, the core density of particleboard of the same overall density could be varied so that the internal bond strength varied as much as 70 percent (10). Again high IB value could not be attained without loss of MOE. Place (55) in an attempt to make a three-layer particleboard with wood bark as core reported similar results. He found that IE was improved by using higher percentages of resin content. He mentions that in order to get acceptable values for IB, the boards needed more densification in the core at the expense of bending properties. Inconsistancy in literatures regarding IB and core density relationship As pointed out earlier, density is one of the most important factors which influence most of the board properties. Reviewing the literature shows that there are many studies on 13 strength and its correlation to density and density gradient. Althouth the readers may find many studies which resulted in a positive and strong linear relationship between IB and density and density gradient, yet, there are some reports indicating a non linear correlation betweeb IB and 87 density of particleboard and MDF, see figure 28. Some workers claim that there is no correlation between center density and TB of MDF (88). On the other hand, there is no doubt that over a wider range of density, IB is indeed a function of density. It seems that a definite conclusion with regard to the relationship between core density and IB of MDF cannot be drawn at this time and that the behavior of MDF may differ substantially from that of particleboard. Therefore, it was decided to investigate these relations in two samples of commercial particleboard and MDF. This study consists of two parts. In the first part of the experiment, the relationships between density and IB is determined. These relationships are examined over a limited range of density. Related important properties such as MOE are also considered. In the second part of experiment, those relationships are evaluated over a wider range of density for both commercial boards. 88 “or 120- e Nov-Ir... - High-Iroqueney I um} ,4 I. I 3 a 2 80 anew-3626+ 6.3" e’ O 2 3 I so.” I d 60 ,r 1‘: ' " fl 5 o‘-"" " 40 2 ; Ie-zu-Iue+o.zseo e- 1 I w.» 20 :—1 ' I 3. 44 50 neueono oeuemr Ire/"3) Figure 28. Example of non linear correlation between 13 and density of MDF. ( Ref. 76 ) Chapter V EXPERIMENTS : PART I - DETERMINATION OF RELATIONSHIPS BETWEEN DENSITY AND IB OF TWO COMMERCIAL PARTICLEBOARDS AND MDF Purpose of experiment It was the purpose of this part of the experiment to investigate the inconsistancies discussed previously and to clear up the relationships between IB and density of the core for both MDF and particleboard. Also to establish whether there are significant differences between MDF and particleboard in this regard. Preparation of specimens Rather than trying to obtain commercial boards made at various density levels, the natural variablities of overall and layer densities of boards made at one target density was used as a basis for establishing the desired relationships. Two 4 by 4 square ft., 0.75 in thick sections of boards were obtained from a southern particleboard mill and two 4 by 5 square ft., 0.75 in thick sections of boards from a southern MDF mill ( see board specification in appendix 1). 89 90 The two sections of each type are identified in the following as board A and B. The following properties were determined for each of the two board types: Vertical density distribution. MOE of total board. MOE of core. MOE of face (derived from core and overall MOE). IB ( overall). Overall density (determined on MOE specimens). Core density [determined on MOE (core) specimens]. Core density ( determined on IB specimens). Face density ( derived from overall and core densities of MOE specimens). The overall design and number of specimens used in both parts of the experiment are shown in table 5. Specimens were cut from board sections as indicated in figure 29. All test specimens were conditioned in a controlled room at 65 deg. F and 50 percent relative humidity. The moisture content of the conditioned specimens ranged between 6.9 and 7.4 percent. Density profile testing procedure For the purpose of this study, since neither board type had clearly defined face and core layers, a 91 .umdu mo: owEmcap .mddanomm umou mo Hones: can cmamoc Hamuo>o one _IIIIII:11 now owns Oman mumz mcoaaommm MOS ma 0:» mo sumo no downs u maoz 0mm 04 «a 4H m o posses cowumwcmu MEEmm cowuznauumwp auwmcmo m m m N m tongue Hmowuuo> OauuoEw>mum 6mm 603 663 mm m~ ueoacmum Susmcmc can ma 66H cm on on on abandon ouoo 0cm mH 44H ma ms 6H ma ma ma ma ma anamcoc 6coo 4 = 4 = 4 a 4 s .aufimcmp Humuo>o .moz m oumom a oumom m tumom m canon .2308 an: bHMOAOaOauumm em ”HAM-H. 92 Density Distribution [Em Bending Internal Bond I. 48 .. 4. v.‘.'oo;.o. .. Q7 I‘m“ ‘ q. I. '3‘... QT ...... . 9.14.3.3. "-- 03 '.~° -..-.' ‘.'. Q3“. .2: '.,‘.‘,. . .:0 0° 9' -.- -- 30.00 53;!-., 400' ...‘J .1. :3? 3:3. 0....‘. ..." . 0;. .. :, .‘3.0 on. :7:.‘.; ......oo:e.;- ‘6‘- W. ..- . ’0 .0 0‘0.‘ “0.; -'.'f.'.‘:.‘.':‘:"-' -- ”:‘2. -. ....-.u-' ” mo. ... ... :‘O‘... ..‘ .33. b I . . 0.... ’0' . ;.Q&o.\.$..;. ...... r00: ".‘ .... ...... -- a 0t o :. ’;?‘ .3:3. :0 ‘0. ....- .oo.~..-: ' 6.. . .0 . . ‘. ...... I0... I 6.. .0... O... '.'.0- '. 1:3” "'000 he? 32:9 3. “0.. - 9.0.0.’. .....‘ ”Inf. ' 3. " “on , .13.. ~.-“" Moo. '3: 9:... :3: O.:.~mx o. O. . ‘ . 0 ;'.?2q. 4‘. :o‘:e.. . .. ‘ ' . . O 0.0. o O o ..., 0.... .... .-. . 48 " £3.53: ”iii-'3 ' :‘a' .443"??? - ..- . o e on I ‘: z .. ’.-.-.3- :31... «no . '-'-:-" .- ‘33.". m..;;...*: a... 3......0. 0.;e..‘e:: :o‘..'e:"f .-< . 00‘ ° ’° £121.?{5-169'ud 3:03,. '.‘.°I,’.‘.":7 ,' :1 3;.:.3:'-;;-T.5.1 000.: Z .".°o°.° mm .....é.0.’..:.'"0‘ .. e .0... .00 >2".'.':::..‘”‘ ..C’::. ne'er'§:eo‘.o'o.: .0. .-.... . . .0 0.0!-ed‘900830I ”... :0 : :0 . 0‘..1 I :o‘O-I 00‘... - .0... 03.0-03w ....e 00......‘ .....’ 0.0.;‘k0000.0 22 " F- ”4......” - «om :t': .2. :‘t .: 6:9.- ‘\ .q-...oq1 0.... , o o o. o O .0 0' -:.° '-:-° ve- ... “was“: ....2.0’ .".:. .3 .~ .. . 2......c - a o o . I o o. . e;o'..e\.“'.‘:°.'?.': ... 0::‘7. .;;.01 $3.": '93-? 3 "“°' “"“""'.'.'31 ‘0‘. ~*...... C”... .:;.;“... one o o. ’9 ° . . :00. O O .. '1'. - «‘53-: 0"... '33:.“ : 2.} ~31 0 a ?....‘.. .. 0‘..‘..-.:‘ {1. ~.*.--.- ‘ .I ”to. \3==;-.’E°:: %‘°?¢eoo _.d g ‘. .0 u 0-0: :3‘;:— ' : ::... * o... 0:0; '.‘. ’.:.. °----'- . ' .0... o-.§ “°" ' ."‘.:.": 3...}: 0:00. ..:o”"-o"-=3:'- ..... . . w ‘.. .. 4 ...“ '6-.. .. .00... "05 o’.’ 5‘ - Figure 29. Diagram for cutting test specimens from 4 by 4 squares ft. boards. 93 subjective judgement was made as to their relative thicknesses, based on the vertical density distribution. This approximate defining of face and core was based on the average board density in which assuming areas above average board. density line as face and areas under average board density line as core. The intersection of average board density and density gradient lines would indicate the thickness of the face of each density profile figure. The average face thickness was 0.32 in and the average core thickness was 0.43 in for both particleboard and MDF. Two methods were used for the determination of the vertical density distribution, the gravimetric and gamma radiation method. In the gravimetric method, thin layers parallel to the surface are removed by planing. By weighing and measuring (thickness) of the specimens before and after the removal of each layer, the density of the layers can be calculated : St - Sr(l - x) 31 = ---------------- {1) $1 = Layer density, g/cu.cm. St = density of specimen before layer removal, g/cu.cm. Sr = Density of specimen after layer removal, g/cu.cm. 94 x = Shelling ratio, = total thickness - core thickness """"" ESZ;I'Z£IZIZ;;;;"""" The test specimens, 1 in wide, 22 in long, and 0.75 in thick were weighed to the nearest one hundredth of a gram and the center thickness measured to the nearest one thousandth of an inch. In a given pair of matched strips, material was planed from the top surface of one and the bottom surface of the other. These pieces were then laminated together with the planed surface as the contact area. Since the planer was preset to a given thickness, initial thickness of all laminated specimens was constant. Thin layers of uniform thickness were now removed from both sides until half the original board thickness had been planed off each side. Layer thickness was 0.050 in and the last layer included the specimen center line. Samples of results are shown in figures 30- 31. The principle used to measure density gradient using gamma radiatio is simply to pass radiation through a sample of board so oriented that the board surface is parallel to the direction of radiation. The attenuation of radiation is a function of board density. Such tests were conducted on 2 in wide, 10 in long, and 0.75 in thick specimens at the Forest Products Laboratory, 9s .Atoguoa Oauumea>uumvmlmm # oumonoAOauumm How oawuoum auwmcma .om musmwm A mmmcxoflnu «\A V .cw oommusm tumon Eouu mucouuwn 66m.o mam.o maw.o mmw.o . mam.o mag.o m~6.o maw.o o .eo.=o.\m Hmb.o u auwucoo .m>< 6IIIIIIIIerIIIIIwaI - om O O OOH .loua .:OVH °£qreuap '54: go quaazad 96 .Anonuma oauumsa>memcozmm 4 662 com mangoes Shannon .Hm musmwm A mmmcxoflnu «\H 9.:fi oommusm bacon scum muscumaa arm.o mfim.o mmN.o mmN.o me.o mfiaoo mwo.c .50.:o\\m Hom.o a >uwmcop .m>< OIIIIIIIIIVII «66.6 o oo .6 a) ooa Sow” fova 'qusuap 'Bae go nuaozaa 97 Madison, Wisconsin. Each sample had its density profile measured using 0.020 in thickness intervals and in a few cases using 0.0050 in thickness intervals. Samples of results as well as the schematic of the equipment are shown in figures 32-38. MOE testing procedure Because it was desirable to determine both face and core moduli on the same specimens, two non-destructive test methods were used, one static, the other dynamic. The static method consisted of applying two center load increments while determining the change in deflection due to the second load increment of a single supported beam. [P(2)-P(1)1L MOE = ---------------- psi {2) [Y(2)-Y(1)]4BD Where : P(l) = First load increment, lbs. P(2) = Second load increment, lbs. Y(1) = Deflection after application of first load increment, in. 2(2) Deflection after application of second load increment, in. L = Span, in. B = Width of specimen, in. 98 Figure 32. Equipment schematic of gamma radiation method of density profile determination. Equipment Schematic RADIATION SOURCE / — — SPECIMEN LVDT Positioning _ Stage DETECTOR I-—Wc-:.H VOLTAGE SOURCE] AMPLIFIER/ANALYZER RATE METER CRT Printer Plotter '1 99 Figure 33. Specimen stage of gamma radiation method. 100 Figure 34. Electrical components of gamma radiation method. m T U M m H N O 101 .Aan>uoucfl :o~o.o mcfims cosmos coaunwtnu neammvmlmm # tumonodowuunm uOm mawwoum audmcoo .mm mucosa mmuzxunzh ma :zuuzum . 8 or an a I I I I I d I d I I I o 0's. I nsz mmUZ¥UHzp UAazuobcfi =Omoo.o manna bosons codunwcnu meanmvamm * ounOAOHOquunm you oawuoum auwncoo .bm ousmwh mmfieafiszkuhzusfir_ 65— 90 an 0? SN G u n u u u u u u u u a 6:. .. 95 826.85 355 .r s 8 Lr d .Eo.20\m 4.3.5 u xbancmt .034 r... Lr ovmw no L. 00.3 V. L. A a. a . L 90.5 _d .r .3 n as..: I if a a. L. 8. m A e:- u A a a... 103 .Aam>umbca =o~o.o mcams cornea coaumscmu mesmmvoimm t was now cannons anamcmo db d \, mmuzxa=:.uc.:fififimi on 0v D L “ .hm ousmwh SN QVh. I nsz wmuszHZh U4¢tuamcoo .m>< AL L ON av 'OAV JO LNBOHEd ALIAVUO OIJIOBdS 104 .Aan>uouc« =omoo.o mean: pocuoe coaunwpnu meaanOImm # has now mawuoum abandon .mm ousmqm meZKH§¢FLa_h2mQ§¥_ OO- 00 GO 0? ON a u u u u u u u u u u o V's. I nsz mmuzxcnzh uaazud6c0© .m>< .fi m“ a .- vau m. or I. o L. 00.3 V 0:! A .0 .. on . m fl 3 : . - of a I \ .. . ~ 8 : ow. w m ..A L. cv— 105 D = Thickness of specimen, in. [ P(2) was well below the porportional limit load, which had been determined by preliminary experiments]. Figure 39 shows the set-up for the static MOE test. Specimens were 1 in wide, 22 in long and the span between supports was 18.2 in (figure 40). This test was applied first to the complete specimen( face and core) and then to the same specimen after the face layers (0.32 in) had been removed by planing. From the total MOE and core MOE, the MOE of the faces was derived as follows (81): MOE(t) - (l-x)3MOE(c) MOE(f) = --------------------- psi {3} 1-(1-x)3 Where : MOE(f) = MOE of face, psi. MOE(t) = MOE of total board, psi. MOE(c) = MOE of core, psi. x Shelling ratio. The above tests were supplemented by a dynamic method employing a stress-wave tester and using the same specimens as used in the static test. The principle of the determination of the dynamic MOE using the stress-wave technique is to measure the time required for a stress wave to travel the length of the specimen. The 106 Figure 39. Apparatus for measuring static MOE. 107 //////////////////, N l /////////////////// 11 T - Core thickness. t 8 Single face thickness. 3/4" 1~ 22»- 41 Figure 40. Cross sectiOn (A) and dimensions (B) of MOE test specimen 108 velocity of the stress wave is porportional to MOE (26,30). Figure 41 shows the dynamic MOE test apparatus. For all MOE specimen, overall density and core density were determined by weighing and measurement of dimensions. The density of face was derived using equation {1). IB testing procedure Internal bond strength in this part of the experiment was determined by the compression shear test as suggested by Suchsland (85). This test involves the lamination of shear strip specimens of 0.75 in by 0.75 in by 24 in between two boards sections and cutting the laminate to yield 12 compression shear test specimens, each 1 in wide, 5 in long and 0.75 in thick ( figure 42). The actual shear specimen is now oriented at 45 degrees relative to the direction of the applied axial compression load. The specimens were tested in compression at a crosshead speed (rate of loading) of 0.05 in/min. until they failed in shear (figure 43). The shear strength was calculated as Tmax --- {4} 2A Where : Tmax = Shear stress at failure, psi. 109 Figure 41. Ultrasonic timer for measuring dynamic MOE test. 110 2.5“ L b.‘... I . '0'": Figure 42. Dimensions of TB test specimen used in part I of experiment. 111 Figure 43. Compression shear test specimen undergoing test. 112 P = Axial load, lbs. 3’ II Cross sectional area of column, square in. The values of IB where obtained from the relation: IB = Shear stress* 0.208. This is based on statistical analysis of extensive tests by Suchsland (85). In order to relate internal bond strength to core density, the density of the failure area had to be determined carefully. This was accomplished by removing thin slices from each side of the shear failure plane and by measuring the resulting weight loss and total specimen length reduction. The thin slices (about 0.10 in thick) were removed from both specimen halves by a carbide tipped milling cutter mounted on a milling machine (figure 44). The specimene were so oriented that the saw cuts were aligned at exactly 45 degrees to the long dimension of the specimens. The length reduction of the specimen, which is equal to the thickness of the removed slices near the failure area, was determined with a special jig which eliminated the error due to posible slipping along the 45 degrees contact faces when reassemblying the specimen halves for measurement in a straight line. The setup for measuring length reduction is shown in figure 45. Explanation of the length measurement is given in appendix 2. 113 Figure 44. Carbide tipped milling cutter mounted on a milling machine for removing thin slices of IB test specimen. 114 Figure 45. Special jig for measuring length reduction of TB test specimen. 115 RESULTS (part I) Tables 6-7 show a summary of the test results as well as statistical analysis of each property. The evaluated properties of particleboard are shown in table 6 while those of MDF are shown in table 7. Density and Density Profile Values of overall density ranged from 0.757 to 0.844 g/cu.cm for particleboard and 0.761 to 0.822 g/cu.cm for MDF. The density range of particleboard was about 43% wider than the density range of MDF. Both types of boards had almost the same average density ( 0.801 and 0.799 g/cu.cm for particleboard and MDF respectively). These results were obtained from MOE test specimens. The average density obtained from the gamma radiation method were 0.811 plus or minus 0.0059 g/cu.cm for MDF and 0.804 plus or minus 0.195 g/cu.cm for particleboard which are very close to the above results. Samples of the results of density profile determination are shown in figures 30-31 and 35-38. The face density of particleboard ranged from 0.793 to 1.057 g/cu.cm and from 0.888 to 0.967 g/cu.cm for MDF. Here, the range for particleboard was twice that for MDF. The range of the core density values was from 0.686 to 0.757 g/cu.cm for particleboard while those for ”*- 116 Table 6. Summary of test results : properties of particleboard. Overall density Of bending test0.801H0.757 0.844 0.087 0.024 3.0 Q/ cu.cm. Face density . g.’cu.cm. 0.920 0.793 1.057 0.264 0.041 4.5 Core density g/Icu.cm. 0.719H0.686 0.757 0.071 0.018 2.5 Overall MOE u 492 433 563 130 37.8 7.7 1.000 psi Overall MOE 1.000 psi .1 426 353 494 141 57.0 11.9 "1xnn5nzrnbs* average 1.000 psi 459 353 563 210 55.6 12.1 Face MOE 1.000 psi n 560 491 639 148 44.3 7.9 Face MOE 1.000 psi .1 480 394 560 166 58.6 12.2 Note : MOE was tested in two perpendicular directions. The two results are arbitrarily identified as «parallel and perpendicular ("g 3.) 117 Table 6 ( cont'd. ) Coeff Mean Min. Max. Range Std. of Dev. Var.% Face MOE 520 394 639 245 65.4 12.6 1.000 psive‘age Core MOE 1.000 psi " 234 181 278 97 21.5 9.2 Core MOE 1.000 psi ‘ 216 174 247 73 23.3 10.8 Core MOE average 225 174 278 104 24.1 10.7 1.000.125; Overall dynamic MOE 572 481 670 189 55.9 9.8 1,990 psi Core dynamic . MOE 363 315 413 98 30.4 8.4 1,000 psi Internal Bond 130 88 165 77 18.8 14.5 psi acre density of IE test 0.747 0.693 0.827 0.134 0.038 5.1 g,/cu.cm. Note : MOE was tested in two perpendicular directions. The two results are arbitrarily identified as parallel and perpendicular( ". * ). Summary of test results : 118 Table 7. properties of MDF Coeff Mean Min. Max. Range Std. of Dev . Var .76 vera ensi of bending test0.799 0.761 0.822 0.061 0.016 2.0 g,/cu.cm. Face density 0.937 0.888 0.967 0.079 0.020 2.1 gI/cu.cm. Core density g/ch.cm. 0.701 0.671 0.723 0.052 0.014 2.0 Overall MOE 1.000 psi " 586 561 619 58 13.3 2.3 Overall MOE 1.000 psi. , 547 508 586 78 23.7 4.3 Overall MOE average 567 508 619 111 27.6 4.9 ....LML Face MOE 1.000 psi " 647 621 686 65 15.0 2.3 Face MOE 1.000 psi J. 603 560 648 88 26.5 4.4 Note : MOE was tested in two perpendicular directions. The two results are arbitrarily identified as parallel and perpendicular(II. * ). 119 Table 7 ( cont'd. ) Range Std. Face MOE average 625 560 686 126 30.9 4.9 1.299.231 Core MOE " 345 319 372 53 12.7 3.7 1,099 psi Core MOE 1.000Apsi Core MOE average 333 269 372 76 18.6 5.6 1,000 psi 7576:“): dynamic MOE 646 593 692 99 29.1 4.5 1,000 psi ‘ 320 269 347 51 15.5 4.8 C010 dynamic 442 409 467 53 17.0 3.9 Internal Bond 129 105 160 ‘ 55 12.0 9.3 psi “CUTI'Uifllfcy of 13 test 3.726 0.679 0.772 0.093 0.019 2.6 g/'cu.cm. Mote : MOE was tested in two perpendicular directions. The two results are arbitrarily identified as parallel and perpendicular( N, l ). 120 MDF ranged from 0.671 to .723 g/cu.cm. Again, the core density range for particleboard was 37% wider than for MDF. Despite the noted differences in average and range of density and density gradient, the two types of boards are remarkably similar in both respects. Modulus of Elasticity Values of overall MOE of particleboard ranged ,from 353,000 to 563,000 psi, while the corresponding values for MDF were 508,000 to 619,000 psi. The average values of MOE were 459,000 and 567,000 psi for particleboard and MDF respectively. Values of face MOE ranged from 394,000 to 639,000 psi with mean value of 520,000 psi for particleboard and from 560,000 to 686,000 psi with mean value of 625,000 psi for MDF. Core MOE range was from 174,000 to 278,000 psi with average of 225,000 psi for particleboard and 269,000 to 372,000 psi with average of 333,000 psi for MDF. Analyses of variance indicated that there were significant differences between the two directions of the board for MOE values (overall, face and core ) in both types of boards. The results of the relationships between static MOE (overall, face and core ) and density (overall, face and core ) and between overall MOE and face density are presented in graphical forms in figures 46-48. As Static MOE 1,000 psi. 121 700 600‘- 500'” 400.. 300-- ---IParticleboard 200 """" ’ MDF " 1. Overall MOE.-overa11 density. 2. Overall MOE and face density 3, Face MOE and face density 4. Core MOE and core density 100 ‘ : - . 0.6 0.7 0.8 0.9 1.0 Density g,/cu.cm. Figure 46. Regressions of MOE on densities of boards. 122 700 1, Overall MOE and overall density 2, Overall MOE and face density. 3; Face MOE and face density 4. Core MOE and core density 600 b ' g ”. Parallel *. Perpendicular. , % ... _ E” :5 500 - 1° 1° o .3 m 400 . 9 o .... U n U ca 300 - 200 . loo 1 e s L 0.6 0.7 ' 0.8 0.9 1.0 1.1 Density g/’cu.cm. Figure 47. Regressions of MOE on densities of particleboard (two directions). Static MOE 1,000 psi. 123 700 600» 5004 400" Parallel 3000 Perpendicular Overall MOE. overall density Overall MOE and face density Face MOE and face density Core MOE and core density 200 : .4 : 4 0.6 0.7 0.8 0.9 1.0 1.1 Density g/'cu.cm. Figure 48 e Regressions of MOE on densities of MDF (two directions). 124 can be seen in these figures, the MOE is a linear function of density. From analyses of variance for regression ( 73 ), the equations for the regression line of MOE (Y) on density (X) of particleboard were determined as follows : For overall MOE and overall density: Y = - 1,117,000 + 1,968,000X with a correlation coefficient of : r = 0.85. {5} For overall MOE and face density: Y = - 528,100 + 1,073,000X with : r = 0.79. {6) For core MOE and core density: Y = - 529,300 + 1,048,000X with : r = 0.79. {7) For face MOE and face density: Y = - 653,500 + 1,275,000X with : r = 0.80. {8) The corresponding equations for the regression of MOE (Y) on density (X) of MDF were determined as follows : For overall MOE and overall density: Y = - 735,300 + 1,630,000X with r = 0.96. {9) For overall MOE and face density: Y = - 682,600 + 1,333,000x with r = 0.95. {10) For core MOE and core density: 125 y = - 473,700 + 1,150,000x with : r = 0.89. {11) For face MOE and face density: Y = - 757,300 + 1,475,000X with : r = 0.94. {12} Values of dynamic MOE are presented in tables 6 and 7. These values are higher than static MOE values. The relationship between dynamic and static MOE is shown in the same graph for MDF and particleboard in figure 49. The regression equations of static MOE (Y) 0n dynamic MOE(X) are : For particleboard: Y = - 62,800 + 0.924.701x with r = 0.90. {13) For MDF: Y = 60,700 + 0.782,527X with r = 0.92. {14) Internal Bond The results of the first internal bond tests represent the tensile strength of the core alone, since failure occured in the core in all cases. The internal bond values ranged from 88 to 165 psi for particleboard while those values for MDF ranged from 105 to 160 psi. Although both types of boards had the same average IB strength ( 129 and 130 psi for MDF and 126 700 7 -- Particleboard f .... m5! ° ...? a “600 I. «o Io Io .3 9': 0500 ' .4 U . n U a: 400 . 300 I! L 1 J 300 400 500 600 700 Dynamic MOE 1.000 psi. Figure 49.. Regressions of static MOE on dynamic MOE of boards. 127 particleboard ), the IB range for particleboard was 40% wider than that for MDF. Besides, IB variations between boards of particleboard were significant while variations between boards of MDF were not significant. The results of core density determination at the failure plane of the compression shear test specimens are presented in tables 6 and 7. These values ranged from 0.679 to 0.772 g/cu.cm for MDF and 0.693 to 0.827 g/cu.cm for particleboard. As expected, this range was about 44% greater for particleboard than for MDF. Again, core density variations between boards of particleboard were significant. Core density determined on IE specimens was always higher than that determined on MOE specimens. The reason may be that machining inaccuracies ( rough surfaces ) would always tend to result in higher density values rather than in lower one. The relationships between IB and core density at the failure point are presented in graphical form in figure 50 for MDF and particleboard. As shown in this figure, the IB is a linear function of the core density of particleboard. From the analysis of variance for the regression, the equation for the regression of IB (Y) on core density at failure point (X) is : Y = - 172.367 + 404.316X with r = 0.81. {15} However, in case of MDF, the correlation between core density and IB was very weak ( r = 0.23 ). 128 180 — Par t ic 1eboard 1600 140” 18 psi. 100.. l L 80 3 L r 0.65 0.70 0.75 0.80 0.85 Core density g,/cu.cm. Figure 50. Regressions of IE on core densities of MDF and particleboard. 129 Limitation of experiment As mentioned previously, the length reduction measurement was the difficult part of the experiment in terms of possible error. Although a special jig was used to eliminate such error, the length reduction measurement is subject to error. Another limitation is that since the failure areas are not plane, density determination of the failure area is also subject to error because it might not be representative of the exact failure plane. Chapter V EXPERIMENTS : PART II - EXTENSION OF IB TEST TO LARGER RANGE OF DENSITY Purpose of experiment In the first part of experiment the relationship between core density and IB of two commercial particleboards and MDF were determined. The results inicated that in a limited range of core density there is a very slight correlation between IB and center density of MDF. But IB was a function of core density of particleboard. The purpose of this part of experiment was to extend those relations to a larger range of density of particleboard and MDF. 130 Preparation of specimens The samples were obtained from the same boards as used in the first part of the experiment. So, the general design and general conditions of this part were similar to first part. In order to obtain a wider density range for the IB- density relationship, advantage was taken of the vertical density distribution. The ID test was so designed that the failure was forced to occur at various distance from the board surface. Therefore, two properties were determined : Vertical density distribution and layer internal bond strength. Density profile testing procedure Only the gamma radiation test method was used in this part as described earlier. Samples were chosen so that it was possible to conduct the IB test and density profile determination on the same sample of the boards. IB layer testing procedure 18 strength was determined at 10 location of each test sample. The standard IB test specimen was modified asin the following explanation : 131 The thickness of 10 sets of board specimens was adjusted by planing in such a way that their thickness varried from full board thickness by intervals of 0.040 in. The thinnest set of specimens, thus, had a thickness after planing of board thickness - 0.400 in. The specimens were then cut to 2 by 2 in square. Grooves, 1/8 in wide and 0.040 in deep were then milled into the planed face of the specimens. These grooves were cut parallel to the four specimen edges and at a distance from the edges so that their intersections created a 1 by 1 in square in the center of the specimen, see figure 51. 2 by 2 in square aluminum blocks were glued to the back of the specimens and 1 by 1 in square aluminum blocks were glued to the 1 by 1 in square field in the center of the top face of the specimens, see figure 52. The following IB test therefore evaluated the tensile strength in 0.040 in thick layers of the board at various locations between board surface and board center, see figures 53-57. The IB of each 0.04 in layer of the board was calculated as : T IB = --- {16} A Where : IB = Internal bond, psi. T = Axial load, lbs. Figure 51. 132 MDF and particleboard IB test specimens showing 1 in square area on planed surface created by grooves to which aluminum block will be glued. Top : Particleboard Down : MDF 133 Figure 52. MDF IB test specimen after gluing aluminum blocks to face and back ( ready for test). 134 Figure 53. MDF IB test specimen undergoing test. 135 Figure 54. MDF and particleboard IB test specimens showing failure areas at 0.040 in from the surface. Top : Particleboard Down : MDF 136 Figure 55. MDF and particleboard IB test specimens showing failure areas at 0.160 in from the surface. Top : Particleboard Down : MDF 137 Figure 56. MDF and particleboard IB test specimens showing failure areas at 0.280 in from _the surface. Top : Particleboard Down : MDF 138 Figure 57. MDF and particleboard IB test specimens showing failure areas at 0.360 in from the surface. Top : Particleboard Down : MDF 139 A = Failure area, equal to 1 in square. The density of these layers was obtained by the gamma radiation method as described earlier. The experimental data for layer IB and layer density relationships of MDF and particleboard is given in appendix 3. Results (part II) Table 8 shows the summary of the test results of the internal bond strength values and densities in different depth of particleboard and medium density fiberboard. The results of density profile determination were discussed in part I. In this part, the results of layer IB test will be described. Layer IB strength IB of the layers ranged from 71 psi to 213 psi for MDF and from 107 psi to 256 psi for particleboard. The relationship between IB of the layers and layer densities is shown in figure 58 for both MDF and particleboard. From the analysis of variance for the regression, the equation for the regression of IB (y) on the density at failure point (x) of particleboard is : y = - 104.6 + 333.8x with r = 0.79 {17) However, the relationship between layer IB and layer densities of MDF was broken down into two line segments 140 Table 8. Summary of test results : IB of MDF and particleboard as function of distance from board surface. 140 a Table 8. ! Board Depth I Area 13 Layer densit in ' in2 (mean) 4 psi (mean) 4g/cm 0.041 lbyl 213.1 4 23.8 1.047 4 0.013 0.080 n " 192.0 4 16.0 0.980 4 0.012 ,4 0.126 n " 165.6 4 6.5 0.904 4 0.011 m 0.168 n " 128.6 4 6.8 0.825 4 0.011 g 0.202 " " 102.1 4 8.6 0.775 4 0.010 g 0.245 " " 83.1 4 8.7 0.728 4 0.008 3 0.284 " n 78.0 4 8.8 0.700 4 0.008 ,6 0.326 n 70.9 4 5.3 0.685 4 0.008 3 0.365 " 76.4 4 5.5 0.678 4 0.007 .2 0.404 n " 73.9 4 6.0 0.678 4 0.007 Avg. density 0.811 t 0.006 0.042 lbyl 255.8 4 41.5 0.974 4 0.040 0.084 n " 182.0 4 8.7 0.954 4 0.044 0.125 " " 196.8 4 11.2 0.874 4 0.045 '3 0.166 " " 138.2 4 15.6 0.805 4 0.039 8 0.203 n " 142.4 4 25.0 0.767 4 0.034 '8 0.244 " " 141.4 4 15.8 0.728 4 0.031 '3 0.282 " " 158.8 4 28.4 0.717 4 0.022’ 3: 0.325 .. " 107.0 4 11.2 0.710 4 0.027 g 0.363 n " 140.0 4 25.6 0.694 4 0.032 m 0.405 n " 130.8 4 20.1 0.684 4 0.029 Avg. density 0.804 4 0.020 141 as shown in figure 58. The upper line represents the correlation between layer 18 and layer density for the density range of 0.771 to 1.05 g/cu.cm. The corresponding equation for MDF within that range of density is : y = - 229.2 + 431.8x with r = 0.89 {18} The lower line represents the correlation between layer IB and layer density for the density range of 0.67 to 0.77 g/cu.cm. which is about the same range as that of the core density of MDF in the first part of the experiment. Again, within this narrow range of core density, the correlation between IB and density of MDF was found to be very weak (r = 0.23). Both particleboard and MDF show strong overall correlations between IB and layer densities. There appears to be no fundamental difference in the response of MDF compared with particleboard. Limitation of experiment Since failure of the IB layer test specimens of particleboard did not occur at the exact predetermined location, and did not have smooth failures, assigning a layer density to the layer IB of particleboard tested is subject to a certain error. In general, determination of core density and relating it to the internal bond strength at a specific 1J12 canons-03:.- e... .2! no 0:32.00 usua— co 3 no:- uo saga-300.. a.- .un 259: .IU.QO\O bus-:00 nuns.— 1». h 3— p a... . sum . FL... . NW A... v i 'I'UCUC ' o ”usedUdUHC. II II o. ”- .0“ f .8— . l .c: o no 1 ... Too“ ..\s ..u \ . r .n. u I .. ... vOmN I 1 luri "god (1 10501 143 location was one of the difficulties of this study in both parts. Chapter VI SUMMARY AND DISCUSSION In the firSt part of this paper, the important factors developing strength properties of particleboard and MDF were discussed. Resin efficiency, wood species, board density, temperatyre, particle geometry and press cycle variables are among the major important variables in this regard. However, there are interrelationships between most of the above mentioned factors. The density distribution in a particleboard or MDF has a horizontal and vertical component. The density gradient can be controlled to some extent by manipulating raw material variables and press cycle. MOE and IB seemed to have similarities in terms of responding to many factors in the manufacture of both MDF and particleboard . Both properties are sensitive to board density, species, compaction ratio, and flake geometry. Long or thin particle will improve MOE wheras short and small particles in the core region will improve IB. Resin content has a stronger influence on 18 than MOE. Press cycle and closing time have opposite affects on MOE and IB. A Short closing time improves MOE 144 at the expense of IB. In contrast, relatively long closing time results in higher 18 and lower MOE values. High surface moisture content combined with initial pressure develops significantly superior MOE and reduces the IB. High density gradients normally result in higher MOE and lower IB. Moderate density gradients or uniform density throughout the thickness of the board results in high IB and low MOE. So, manufacturers would have to decide which properties are most important for their products and make the necessary adjustment and trade off between 18 and MOE. In practice, it seems that maintaining the IB strength property is of primary interest because this property cannot be enhanced by fabricators after the board is produced. On the other hand, it is practically possible to reinforce the MOE strength property even after processing simply by adding a thin layer of veneer or other overlay material. Because of the time consuming, relatively slow and inconvenient glue assembly of IE test specimens according to the standard method, many alternative methods of testing 18 strength have been developed. However, not all alternative 18 test methods can provide the necessary accuracy and feasibility of the standard test. Again, the manufacturers are to decide which test methods are appropriate for their products in terms of accuracy, cost, speed and suitability for quality control. 145 The standard requirements of the commercial standard, of course, are based on the standard IB test. Many researchers have been working on the subject of IB strength and its relation to density and density profile. of particleboard and MDF. Because of inconsistancy regarding IB and core density relationships in the results of such studies, the experimental part of this paper includes investigations of those above mentioned relationships for two commercial particleboards and MDF and determination of related properties. This experimental part has two components. In the first part of the experiment, relationships between IB and a limited range of density (core density) of particleboard and MDF were determined . In the second part of the experiment, those relations were investigated over a wider range of density. The results of the first experimental part confirmed the virtual absence of a core density effect on the IB of MDF, while a relatively strong correlation was found between IB and core density of particleboard. The apparent lack of this correlation in MDF was probably due to the relatively narrow range of the core density. However, the extension of the experiment in the second part showed that the relationships between 18 and density does exist for MDF as well as particleboard. It may be noted that core and face materials follow the same relationship between MOE and density but at 146 different levels for particleboard and MDF and that MOE is very density sensitive (figure 46). The dynamic MOE test produces higher values than the static test and appears to be a good predictor of static MOE. This predictability may, however, be limited to individual products only. Chapter VII CONSEQUENCES OF REDUCING CORE DENSITY The results of the first experiment : A very slight correlation between IB and core density of MDF encourage speculation for the possibility of reducing the core density of MDF without suffering undue penalties in terms of properties deterioration. Some exploratory calculations in this regard were made and are discussed in the following : Suppose that, the core density of MDF is being reduced from 0.770 to 0.680 g/cu.cm. If the face density was maintained at, say, 1.0, 0.95 or at 0.90 g/cu.cm., the overal density would be determined by using equation {1}. These values would be reduced from 0.868, 0.847 or 0.826 to 0.817, 0.795, or 0.774 g/cu.cm. respectively (figure 59). Furthermore, face MOE, core MOE and overall MOE when maintaining such face densities could be calculated by using equations {12, 11, 3} as shown in table 9 example 1. The overall density reductions are 147 Figure 59. Relations between overall density and core density of MDF when maintaining face density and IB. 147a 0.90 0 o (0 0| 0.80. Overall density g,/cu.cm. 0.75 i r :w i 0.68 0.70 0.72 0.74 0.76 0.77 Core density g,/cu.cm. Figure S9. 148 Table 9. Examples of computing density and MOE while reducing core density and maintaining 13 of MDF. .N madamxm .d oHQmem 148 a .so.=0\\m as _ .amd ooo.a Am «ms.o cosumsoo magma. Hmm own Ham coo see one cosumsam manage >uamcoc Hamuo>o mo: Hamu0>o .80.=o\\m Awe .smn ooo.s Add 848.6 cosuaauo mammal mom «He can ~Hv mom Nae coaumsvo mammal auwmcoo 00mm mot ouoo .amd coo.s Am .auo ooo.~ ANA vac 60886566 mammal ohm ohm eve 666 was was coaumsem mammal m0! 009m . m0: 00mm .«mm ooo.~ Add .Eo.so\\m AA mom cosumscm mammal was.ojo~m.o¢mms.o sem.g pam.n mom.o :omuasem means. mo: muoo . — abandon Hamuo>o .Eo.so\ m .Eo.so\m ome.o Acmosowsv omo.o o6~.o ome.o oss.oroo.o ops.p Amazon. >udmcoc ouOU . auwmcoo ouoo .smd ooo.s .50.:0\\m mme Acossnssmmm. om.o om.o mm. mm.o o.~ o.~ Acoauossuum. mo: admum>o . abandon 00mm .m manna 149 5.9%, 6.1%, and 6.3% respectively. Considering figure 46, which shows that MOE values of MDF are much higher than those of particleboard and that the mean MOE value of particleboard is still higher than the standard requirement (400,000 psi), one might ask, why not reduce overall MOE of MDF to the average MOE value of particleboard (459,000 psi ). Here, again, the new face MOE, new face density and finally, new average board density would be obtained simply by using equations {11, 3, 12, 1 } while again reducing core density to 0.680 g/cu.cm. and maintaining IB ( table 9 example 2). This would be equivalent to a 5.9% total density reduction. Of course, even small amounts of density reduction mean cost savings in terms of raw material, press cycle, and transportation. At the same time, there might result other board quality improvements, as for instance, reduced thickness swelling. There are disadvantages too : Lower edge density, lower screw holding power, etc. Chapter VIII CONCLUSION From the results obtained in this study, the following conclusion may be drawn : 150 1. Density is a very important and dominating variable affecting physical and mechanical properties of particleboard and MDF. 2. Resin efficiency, densification, particle geometry and press cycle are some of the other important factors developing mechanical properties of particleboard and MDF. 3. Density gradient of MDF or particleboard is one of the important characteristics of these boards which can be manipulated by raw material variables and press cycle. 4. IB and MOE are dependent properties. Measures to increase one may adversely affect the other. 5. MDF and particleboard have different density-IE relationships. 6. IB of particleboard is strongly related to core density. 7. 18 of MDF has a very slight correlation with core density. 8. By extension of the range of densities, strong linear correlations were found between IB and density of both board types. 9. It may be possible to reduce the overall density of MDF without sacrificing the IB. 10. The greater homogeneity of MDF is reflected in smaller variations of its properties. 11. There might be other factors besides density 151 affecting the relationship between layer IB and layer density as tested in our experiment. One such factor might be the change in fiber orientation as function of distance from borad surface. APPENDICES 152 Appendix 1. Board specifications. a: SurPine particleboard specifications : Raw material : Planer shavings, plywood trim,reclaims of Pine. Resin : Urea formaldehyde, 7.5% face and 6.25% core. Wax : 0.4% . Urea : 0.4%( scavenger ). Press : 5' by 16’, with 20 openings. Pressure : 500 psi. Total cycle time 6.6 minutes including closingtime and opening. Sanded 0.04 in each side. Board density 48 lbs/cu.ft. b: Willamette MDF specifications : Raw material : 90% Pine planer shaving, 10% hardwood chips. Resin : Urea formaldehyde, about 9% . wax : 0.5% . Urea : 0.4% or less( scavenger ). Press : 5’, by 18’, with 13 openings. 153 Pressure 500 psi. Total cycle time : 7 minutes including closing time and opening. Board density : 48 lbs/cu.ft. 154 Appendix 2. Determination of machined length of IB test specimen. The length reduction of the specimen due to the removal by means of a milling cutter of material on both sides of the failure plane can be determined with the help of a special jig, figure 60. The shaded areas on the sketch are restraining steel bars mounted on a steel base plate. The distance A is so chosen that it is approximately equal to L1/2 + D/2. What is shown in the sketch is a hypothetical specimen separated without loss of material along a plane inclined by an angle of 45 degree relative to the long axis of the specimen. The dial gage measures distance Bl. If now this specimen was taken out of the jig, material removed along the 45 degree planes to the dotted lines and placed back in the jig, only part b would shift. Part a would be in exactly the same position it had occupied initially. Part b would shift downward by a distance equal to the length of the arrow which is exactly the reduction in total length (Ll-L2) of the specimen.. The dial reading is now 82. Therefore, Bl-B2=L1-L2=AL {19} 155 Figure 60. Measurement of length reduction of 18 test specimen. 155a HA «\H4 .00 musmwm N mv Q+HA W“ . . - < m «\HA «\o I e A 1 ,A. D II II 1.1 7/ . I; _ / am an m . N _@ Q+HA _ . . _ h 156 However, Bl connot be measured and must be expressed in terms of L1 and D, where L1 is the length of before testing. One obtains from the sketch : 3 Bl Bl 11L L1/2 + 0/2 - A Ll/2 + 0/2 + 5 L1 + D - A Bl - 82 L1 + D - A - 82 the specimen {20} {21} {22} {23} 157 Appendix 3. Experimental data for layer IB and layer density relationships. Layer density Layer IB Layer density Layer IB g/cu.cm. psi g/cu.cm. psi 1.051 206 0.974 221 1.051 217 0.974 213 1.051 251 0.974 221 1.051 211 0.907 150 1.051 215 0.907 165 1.054 235 0.907 172 1.054 185 0.907 158 1.054 230 0.907 187 1.054 217 0.903 130 1.054 234 0.903 172 1.047 228 0.903 175 1.047 162 0.903 178 1.047 263 0.903 139 1.047 250 0.905 155 1.047 242 0.905 182 1.036 235 0.905 173 1.036 204 0.905 169 1.036 217 0.905 168 1.036 236 0.901 160 1.036 220 0.901 168 0.987 207 0.901 170 0.987 213 0.901 174 0.987 232 0.901 167 0.987 247 0.828 135 0.987 232 0.828 126 0.988 - 170 0.828 150 0.988 160 0.828 137 0.988 152 0.828 127 0.988 169 0.822 130 0.988 118 0.822 148 0.974 188 0.822 133 0.974 161 0.822 137 0.974 170 0.822 110 0.974 175 0.831 130 0.974 159 0.831 112 0.974 211 0.831 120 0.974 221 0.831 105 Layer density Layer IB Layer density Layer IB g/cu.cm. psi g/cu.cm. psi 0.831 117 0.726 80 0.821 130 0.726 108 0.821 153 0.726 83 0.821 100 0.706 42 0.821 130 0.706 56 0.821 141 0.706 83 0.781 67 0.706 71 0.781 83 0.706 85 0.781 96 0.695 55 0.781 116 0.695 66 0.781 78 0.695 89 0.776 100 0.695 52 0.776 101 0.695 56 0.776 98 0.705 90 0.776 76 0.705 106 0.776 92 0.705 66 0.771 120 0.705 87 0.771 110 0.705 84 0.771 97 0.694 107 0.771 90 0.694 87 0.771 142 0.694 85 0.771 108 0.694 95 0.771 123 0.694 98 0.771 116 0.689 56 0.771 116 0.689 83 0.771 112 0.689 81 0.733 100 0.689 74 0.733 80 0.689 61 0.733 103 0.681 65 0.733 116 0.681 57 0.733 90 0.681 68 0.723 61 0.681 63 0.723 57 0.681 55 0.723 59 0.688 70 0.723 59 0.688 59 0.723 75 0.688 81 0.732 72 0.688 66 0.732 78 0.688 91 0.732 78 0.681 70 0.732 103 0.681 70 0.732 112 0.681 96 0.726 66 0.681 74 0.726 82 0.681 78 159 Layer density Layer IB Layer density Layer IB g/cu.cm. psi g/cu.cm. psi 0.678 60 0.678 72 0.678 70 0.678 93 0.678 67 0.670 72 0.670 91 0.670 74 0.670 70 0.670 57 0.685 81 0.685 75 0.685 88 0.685 74 0.685 57 0.679 74 0.679 87 0.679 91 0.679 95 0.679 79 0.675 75 0.675 77 0.675 63 0.675 86 0.675 92 0.677 85 0.677 81 0.677 55 0.677 76 0.677 62 0.687 59 0.687 57 0.687 50 0.687 78 0.687 88 0.671 68 0.671 86 0.671 65 0.671 86 0.671 89 160 Board : Particleboard Layer density Layer IB Layer density Layer IB g/cu.cm. psi g/cu.cm. psi 0.974 275 0.694 166 0.974 212 0.694 155 0.974 242 0.694 130 0.974 300 0.694 135 0.974 250 0.694 114 0.954 175 0.684 155 0.954 190 0.684 138 0.954 182 0.684 112 0.954 175 0.684 129 0.954 188 0.684 120 0.874 207 0.874 195 0.874 186 0.874 191 0.874 205 0.805 142 0.805 116 0.805 144 0.805 142 0.805 147 0.767 129 0.767 136 0.767 121 0.767 156 0.767 170 0.728 122 0.728 135 0.728 148 0.728 150 0.728 152 0.717 150 0.717 182 0.717 160 0.717 177 0.717 125 0.710 105 0.710 120 0.710 98 0.710 112 0.710 100 LIST OF REFERENCES 10. 11. LIST OF REFERENCES Akers, L. E. 1966. Particleboard and hardboard. Pergamon Press Inc., 44-01 21st street, Long Island City, New York. 172 pp. American Society for Testing and Material. 1979. Standard method of evaluating the properties of wood-base fiber and particle panel materials. ASTM D 1037-78. Philadelphia, PA. Anon. 1965, 1966. Centre Technique Du Bois, Paris, France. Agglos Nrs. 304 and 319. Bhagwat, S. 1971. Physical and mechanical variations in cottonwood and hickory flakeboards made from flakes of three sizes. For. Prod. J. 21 (9): 101-103. Bodig, J. and B. A. Jayne. 1982. Mechanics of wood and wood composites. Van Nostrand Reinhold company Inc. 135 west 50th street, New York, N.Y. 712 pp. Brumbaugh J. 1960. Effect of dimensions on properties of particleboards. For. Prod. J. 10(5) 243-246. Bryan, E. L. 1962. Maximum strength properties for particleboard. For. Prod. J. 12(2): 59-64. Burrows, C. H. 1961. Some factors affecting resin efficiency in flake board. For. Prod. J. 11(1): 27-34. Canadian Standards Association. 1968. CSA. Standard 0188- 1968: Mat-formed wood particleboard. Can. Stand. Assoc., 77 Spencer street, Ottawa, Canada. Carroll, M. 1963. Efficiency of urea- and phenol- formaldehyde in particleboard. For. Prod. J. 13(3): 113-120. Carroll, M. and D. Mcvey. 1962. An analysis of resin efficiency in particleboard. For. Prod. J. 12(7): 305-310. 161 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 162 Chow, P. 1976. PrOperties of medium-density, dry-formed fiberboard from seven hardwood residues and bark. For. Prod. J. 26(5): 48-55. Christensen, R. L. and P. Robitschek. 1974. Efficiency of resin-wood particle blending. For. Prod. J. 24(7): 22-25. Denisov,O. B., P. P. Anisou, and P. E. Zuban. 1975. Investigation of the permeability of particle mats. Holtechnologie. 16(1): 10-14. Food and Agricalture Organization of United Nations 1958. Fiber and particleboard. Report of an intre- national consulatation on insulation board, hard- board and particleboard. FAO. Rome. 179 pp. Foster, W. G. 1967. Species variation. In procee- dings of first symposium on particleboard. Washington State Univ. Pullman, Wash. pp. 13-20. Gaudert, P. 1974. New torsion test for particleboard. For. Prod. J. 24(2): 35-37. Geimer, R. L. , H. M. Montrey and W. F. Lehmann. 1975. Effects of layer characteristics on the properties of three-layer particleboards. For. Prod. J. 25(3): 19-29. Hall, H. and R. Gertieiansen. 1974. Improving the properties of commercial phenolic-bonded, structural-type particleboards by oil tempering. For. Prod. J. 24(3): 40-42. Hall, H. and J. G. Haygreen. 1983. The Minnesota shear test. For. Prod. J. 33(9): 29-32. Hall, H. , J. G. Haygreen and Y. Lee. 1984. Minnesota shear test reproducibility and correlation with internal bond. For. Prod. J. 34 (9): 49-52. Halligan, A. F. 1969. Recent glues and gluing research applied to particleboard. For. Prod. J. 19(1): 44-52. Hse, C. Y. 1975. Properties of flakeboards from hardwoods growing on southern pine sites. For. Prod. J. 25(3): 48-53. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 163 Hse, C. Y., P. Koch., C. W. Mcmillin., and E. W. Price. 1975. Laboratory-scale development of a structural exterior flakeboard from hardwoods growing on southern pine sites. For. Prod. J. 25(4): 42-50. JIS - 5908 - 1961. Specification for particleboard Jap. Ind. Stand. Tokyo, Japan. Jung, J. 1982. Properties of parallel laminated veneer from stress-wave-tested veneers. For. Prod. J. 32(7): 30-35. Kelly, M. W. and R. G. Pearson. 1977. Properties of veneered southern pine low density fiberboard from hardwood total-tree chips. For. Prod. J. 27 (9): 28-37. Keylwerth, R. 1958. On the mechanics of the multilayer particleboard. Holz als Roh -und Werkstoff. 16(11): 419 - 430. Knowles, L. 1981. Rapid method to determine internal bond and density variations of particleboard. For. Prod. J. 31(12): 51-53. Koch, P. and G. E. Woodson. 1968. Laminating butt-jointed, log-run southern pine veneers into long beams of uniform high strength. For. Prod. J. 18(10): 45-51. Kollmann, F. F. P. , E. W. Kuenzi, and A. J. Stamm. 1975. Principles of wood science and technology . II Wood based materials. Stringer-verlag New York Heidelberg Berlin 1975. 703 pp. Larmore, F. D. 1959. Influence of specific gravity and resin content on properties of particleboard. For. Prod. J. 9(4): 131-134. Laufenberg, T. L. 1986. Using gamma radiation to measure density gradients in reconstituted wood products. For. Prod. J. 36(2): 59-62. Lebens, G. A. and H. J. Hall. 1983. Hot melt adhesive applicator speeds gluing of internal bond test specimens. For. Prod. J. 33(7/8): 33-34. Lehmann, W. F. 1960. The effects of moisture content, board density and press temperature on the dimensional and strength properties of flat-pressed flakeboard. School of Forestry, North Carolina State Univ., Raleigh, NC. 106 pp. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 164 Lehmann, W. F. 1965. Simplified test of internal bond in particleboard. For. Prod. J. 15(5): 223-4. Lehmann, W. F. 1970. Resin efficiency in particleboard as influenced by density, atomization and resin content. For. Prod. J. 20(11): 48-54. Lehmann, W. F. 1986. outline of a fast durability test for UP and UP adhesives in composite materials. Proceedings 20th international particle- board/composite materials symposium. Washington State Univ. Pullman, Wash. 105 - 122. Liiri, O. , A. Kivisto, M. Tuominen, and M.Aho. 1980 Determination of the internal bond of particleboard and fiberboard. Holz als Roh-und Werkstoff 38(1980) 185-193. Lynam, F. C. 1959. Factors influencing the properties of wood chipboard. Wood 24: 18 - 20. Marra, G. G. 1972. The future of engineered wood materials. Quest 9(4) : 1-8. Mack, J. J. 1948. A study of the torsion test. Effect of size of specimen. Division of Forest Products, Council for Scientific and Industrial Research. Commonwealth of Australia, Melbourne. Maloney, T. M. 1970. Resin distribution in layered particleboard. For. Prod. J. 20(1): 43-52. Maloney, T. M. 1977. Modern particleboard & dry process fiberboard manufacturing. Miller Freeman publications, Inc. 500 Howard street. San Francisco, California. 672 pp. May, H. , and G. Stegmann. 1966. Measures for control of raw material yield in particleboard manufacture: part 1: Methods for determination of dimensions of wood particles and for evaluation of sifting processes. Holz als Roh-und Werkstoff 24 (7): 305 - 311. Mitchell, D. V. 1957. The suitability of five southern hardwood species for the manufacture of dry formed, flat-pressed particleboard. School of Forestry, North Carolina State Univ., Raleigh, NC. Moslemi, A. A. 1974. particleboard. Volume 1 : Materials. Southern illinois Univ. Press. 244 pp. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 165 Mottet, A. L. 1967. The particle geometry factor in particleboard manufacturing. In proceedings of first symposium on particleboard. Washington State Univ., Pullman, Wash. National Particleboard Association. 1980. Medium density fiberboard for interior use. ANSI/ A 208 2-1980. ------------------------ . 1987. Mat-formed wood particleboard. ANSI/ A 208. 1-1987. Nearn, W. T. and K. Bassett. 1967. X-ray determination and use of surface-to-surface density profile in fiberboard. For. Prod. J. 17(1): 73-74. Neusser, H. and W. Schall. 1970. Experiments concerning the hydrolytic phenomenon in particle- board. Holzforschung und Holzverwertung. 22(16): 116 - 120. Passialis, C. and G. Tsoumis. 1982. A method for testing internal bond of particleboard. Wood and fiber, 14(2), 1982; 159-161. Philippou, J. L., W. E. Johns., E. Zavarin., and T. Nguyen. 1982. Bonding of particleboard using hydrogen peroxide, lignosulfonates, and furfuryl alcohol: the effect of process parameters. For. Prod. J. 32(3): 27-32. Place, T. A. and T. M. Maloney. 1977. Internal bond and moisture response properties of three layer wood-bark boards. For. Prod. J. 27(3): 50-54. Plath, 1., and E. Schnitzer. 1974. The density profile a criterion for evaluating particleboard. Holz als Roh-und Werkstoff. 32(11): 443 - 449. Post, P. W. 1961. Mechanical and dimensional properties of flake board. For. Prod. J. 11(1): 34-37. Price, E. 1974. Analysis of southern hardwoods as furnish for a wood flake- resin composite structural material. Ph.D. diss., Univ. of Illinois Urbana- champaign. 148 pp. Rackwitz, G. 1963. Influence of chip dimensions on some properties of wood particleboard. Holz als Roh- und Werkstoff 21 (1963) 200-209. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 166 Rice, J. T. and R. H. Carey. 1978. Wood density and board composition effects on phenolic resin- bonded flake board. For. Prod. J. 28(4): 21-27. Robitschek, P. and R. L. Christensen. 1976. Degradation phenomena in urea-formaldehyde resin- bonded particleboard. For. Prod. J. 26(12): 43-46. Schwarz, F. E., R. L. Andersen, and A. G. Kageler. 1968. Resin distribution and how variations affect board quality. In proceedings of second symposium on particleboard. Washington State Univ. Pullman, Wash. Shen, K. C. 1970. Correlation between internal bond and the shear strength measured by twisting thin plates of particleboard. For. Prod. J. 20(11) 16-20. Shen, K.C. and M. N. Carroll. 1969. A new method for evaluation of internal strength of particleboard For. Prod. J. 19(8): 17-22. -------------- . 1970. Measurement of layer strength distribution in particleboard. For. Prod. J. 20(6): 53-55. Shen, K. C. , M. N. Carroll., and B. Wrangham. 1969. Compression shear strength of particleboard and its relationship to internal bond. Dep. Fish. Forest. Inform. Rep. OP-X-22. Shen, K. C. and B. Wrangham. 1971. A rapid accelerated-aging test procedure for phenolic particleboard. For. Prod. J. 21(5): 30-33. Sherrill, S. 1987. " A particle of difference " Particleboard/ MDF. A current view of these industries in North America. Crow digest special report. C. C. Crow Publications. Vol. 2 No. 3. Shuler, C. E. and R. A. Kelly. 1976. Effect of flake geometry on mechanical properties of eastern flake-type particleboard. For. Prod. J. 26(6): 24-28. Smith, D. C. 1982. Waferboard press closing strategies. For. Prod. J. 32(3): 40-45. Snyder, J. L. , J. T. Rice., and C. A. Hart. 1967. Influence of selected resin and bonding factors on flakeboard properties. For. Prod. J. 17(8): 49-57. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 167 Soper, V. R. and R. A. Hann. 1964. Variables in specimen preparation influence results of IB tests on particleboard. For. Prod. J. 14(6): 261-264. Steel, R. G. D. and J. H. Torrie. 1980. Principles and procedures of statistics. Mcgraw-Hill Book Co. New York. 633 pp. Steinmetz, P. E. and C. W. Polley. 1974. Influences of fiber alignment on stiffness and dimensional stability of high-density-dry formed hardboard. For. Prod. J. 24(5): 45-50. Stevens, R. R. 1978. Slicing apparatus aids in determination of layer-density of particleboard. For. Prod. J. 28(9): 51-52. Stevens, R. R. and G. E. Woodson. 1977. High- frequency and hot-platen curing of medium-density fiberboards. For. Prod. J. 27(1): 46-50. Stewart, H. A. and W. F. Lehmann. 1973. High quality particleboard from cross-grain, knife-planed hardwood flakes. For. Prod. J. 23(8): 52-60. Strickler, M. D. 1959. Properties of douglas-fir flakeboard. For. Prod. J. 9((7): 203-216. Suchsland, O. 1959. The strength of glue joints in wood obtained with minimum glue spread. Quart. Bull. Mich. Agr. Exp. Sta. Mich. State Univ. 41(3): 577-599. ------------ . 1959. An analysis of the particle- board process. Quat. Bull. Mich. Agr. Exp. Sta. Mich. State Univ. 43(2): 350-372. ------------ . 1960. An analysis of two-species three-layer wood flake board. Quat. Bull. Mich. Agr. Exp. Sta. Mich. State Univ. 43(2): 375-393. ------------ . 1962. The density distribution in flake board. Quat. Bull. Mich. Agr. Exp. Sta. Mich. State Univ. 45(11): 104-121. ------------ . 1967. Behavior of a particleboard mat during the press cycle. For. Prod. J. 17(2): 51-57. ' ------------ . 1972. From wood waste to particle- board. Ext. Bull. E-735. Marketing Series. Coop. Ext. Serv. Mich. State Univ. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 168 Suchsland, O. 1977. Compression shear test for determination of internal bond strength in particleboard. For. Prod. J. 27(1): 32-36. ------------ . 1978. Medium density fiberboard production possibilities in the Tennessee Valley Region. Process, market, and raw materials basis. Tennessee Valley Authority. Norris, Tennessee., technical note. B 29. Suchsland, O. and R. C. Enlow. 1968. Heat treatment of exterior particleboard. For. Prod. J. 18(8): 24-28. Suchsland, O. , D. E. Lyon, and P. E. Short. 1978. Selected properties of commercial medium-density fiberboards. For. Prod. J. 28(9): 45-49. Suchsland, O. and G. E. Woodson. 1974. Effect of press cycle variables on density gradient of medium density fiberboard. Eight Wash. State Univ. Symp. on particleboard. Pullman. Wash. P.375-396. Suchsland, O. and G. E. Woodson. 1976. Properties of medium density fiberboard produced in an oil- treated laboratory press. USDA For. Serv. res. Pap. SO-116. , South. For. Exp. Sta. New Orlean, La. 10 p. ---------------------- . 1987. Fiberboard manufacturing practices in the United States. U.S. Dep. of Agr. For. Serv, Agr. Handbook No. 640 263 p. Szabo, T. and P. C. L. Gaudert. 1978. Fast internal bond test for waferboard. For. Prod. J. 28(8): 38-40. Talbott, J. W. and T. M. Maloney. 1957. Effect of several production variables on modulus of rupture and internal bond strength of boards made of green douglas-fir planer shaving. For. Prod. J. Oct. 1957. 395-398. Tang, R. C. and E. W. Price. 1983. Analysis of the dimensional stability of wood based composites. Int. J. Modelling and simulation. 3(2): 94 - 104. Turner, H. D. 1954. Effect of particle size and shape on strength and dimensional stability of resin-bonded wood-particle panels. For. Prod. J. Oct. 1954. 210-223. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 169 U. S. Dep. of Agricalture, Forest Service. 1986. Wood particleboard and flakeboard types, grades and uses. Forest Products Laboratory. General technical Report. ( FPL-GTR-53). Vital, B. R. , W. F. Lehmann., and R. S. Boone. 1974. How species and board densities affect properties of exotic hardwood particleboards. For. Prod. J. 24(12): 37-45. Vogt, J.J. 1986. Longitudinal stress waves as predictors of internal bond strength. Proceeding 20 th international particleboard / composite materials symposium. Washington State Univ. Pullman, Wash. 385 - 402. Wellons, J. D. 1981. Bonding in wood composites. Proceedings of a conference on adhesion in cellulosic and wood-based composites. Queen ’5 University, Kingston, Ontario, Canada. Wilson, J. B. and R. L. Krahmer. 1976. Particle- board: Microscopic observations of resin distri- bution and board fracture. For. Prod. J. 26(11): 42-45. Winistorfer, P. M., W. C. Davis. and W. W. Moschler Jr. 1986. A direct scanning densitometer to measure density profile in wood composite products. For. Prod. J. 36(11/12): 82-86. Winistorfer, P. M., and W. W. Mochler. 1986. Comparison of the properties of medium density fiberboard. U. 8. Consumer Prod. Safety Comm. contr. No. CPSC - P - 85 - 1186. Woodson, G. E. 1976. Effects of bark, density profile, and resin content on medium-density fiber- boards from southern hardwoods. For. Prod. J. 26(2): 39-42. ------------- . 1977. Density profile and fiber alignment in fiberboard from three southern hard- woods. For. Prod. J. 27(8): 29-34. Ylinen, A. 1963. A coparitive study of different types of shear tests on wood. The State Institute for Technical Research, Helsinki, Finland. "71.111111117111111111'“