1 International Conference on Environmentally ...

1

Proceedings of ICECFOP1 – 1st International Conference on EnvironmentallyCompatible Forest Products Fernando Pessoa University Oporto, Portugal 22 – 24 September 2004

Edited by Fernando Caldeira Jorge

Copies available from: Edições Universidade Fernando Pessoa Praça 9 de Abril, 349 4249-004 PORTO – PORTUGAL [email protected]

ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

2

The opinions expressed in these proceedings are those of the authors and do nor necessarily represent those of the editor.


3

Message from the Director of the Faculty of Science and Technology - UFP The purpose of ICECFOP Conferences is to encourage discussion between scientists and industrialists working on the different topics of basic and applied sciences related to the research and innovation on wood products and manufacturing technologies, with low environmental impacts. These proceedings reflect the interaction between researchers and businessmen from various fields, by including a diversity of subjects. As Director of the Faculty of Science and Technology of Fernando Pessoa University (UFP), I would like to express my gratitude to CEMAS, especially to Dr. Fernando Caldeira, Coordinator of the Scientific and Organisation Commissions of ICECFOP, and to Geonúcleo (an union of students of UFP) and to Ecole Supérieure du Bois, Nantes, France, for their outstanding collaboration in this conference. To all the speakers and participants, I would like to welcome to Porto and to this conference. I wish you a nice stay in Oporto. Dr. Álvaro Monteiro Director of the Faculty of Science and Technology Fernando Pessoa University, Oporto, Portugal


4


5

Preface Nowadays, environmental issues are paramount to all industrial sectors in the developed countries and in many developing countries, and even the personal life of citizens is in one way or another conditioned by environment-protection policies. Many world-leading companies work ahead of legislation by setting environmental goals for themselves. This behaviour is not entirely altruistic as it also makes economic sense because they can save costs through reducing the amount of residues they produce and the cost of treating these wastes, plus they can promote themselves through their actions in order to their visibility in the market place. This trend is being followed by many more companies that work for and towards modern societies. The modern citizen is much more aware that environmental quality is closely related to quality of life. Therefore, more and more people tend to prefer ecological products, i.e. products which have a much lower environmental impact. Aware of this feature of modern societies, CEMAS, the Centre of Modelling and Environmental Systems Analysis, of the Faculty of Science of Technology of Fernando Pessoa University, which has staff with expertise in many areas of the environmental science and engineering, but also in wood science and forest products, identified an opportunity for a international conference that would bring together environmental science and forestry/wood science. This conference also aims to bring scientists and industrialists of the world together, so that they can present the latest research and innovation on wood products and technologies with the low environmental impacts, or even with environmental benefits, the so-called clean technologies, as well as data on environmental impacts and pollution emissions from the existing processes. And ICECFOP1 is now born. We hope that you will find it useful for your work, whether you are from business or academia, and that when you leave, on the 24th September, you will feel that it was worth the time and expense of attending. We hope too that your know-how will be enriched and that you find new opportunities for cooperation with other people and other countries. We wish to make ICECFOP a biannual conference. If you are able to spend some time in Portugal, then we wish a pleasant stay. Enjoy the city of Oporto, which is most famous for its Port Wine, but also for its historical heritage, and the peculiar costumes of its people, among other topics of interest. Enjoy our climate and the traditional hospitality of the Portuguese people, and try to visit other regions of the country. Last but not least, we are proud to announce the Keynote speaker of this first edition of ICECFOP: Dr. Roger Rowell, from United States Department of Agriculture, Forest Service, Forest Products Laboratory, and Biological Systems Engineering Department, University of Wisconsin, Madison, WI. He is a worldwide famous scientist, with a vast experience and curriculum related to the research on chemical modification of lignocellulosics for property enhanced composites, materials science of natural fibres, composites from sustainable agroresources, etc. He has received 10 honours and awards, written more than 300 papers, edited 9 books and been granted 22 patents. Enjoy the conference and your stay. Fernando Caldeira Jorge Associate Professor ([email protected]) Coordinator of the Scientific and Organisation Commissions of ICECFOP1 CEMAS - Centre for Modelling and Environmental Systems Analysis Faculty of Science and Technology - Fernando Pessoa University Oporto, Portugal ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

6


7

Table of Contents Session 1: Opening Session and Keynote Address Biobased Composites: A 21st Century Perspective Roger M. Rowell ..................................................................................................... 15

Session 2: Broad-Scope Communications Solving the Problem of Misunderstanding at the International Conferences, Dealt with Ecology and Forestry Julia V. Ivanova, Valentin A. Makuev .................................................................... 31 The Corkmark Process: Natural is Good but not Enough Maria Carolina Varela, Diogo Holstein Campilho, Manuel Holstein Campilho .... 35

Session 3: LCA of Forest Products Methodological Considerations on LCA Applied to Wooden Products Marco Fioravanti, Paolo Frankl, Linda Toracca .....................................................

47

Evaluation of the Environmental Performance of Printing and Writing Paper Using Life-Cycle Assessment Ana Cláudia Dias, Margarida Louro, Luís Arroja, Isabel Capela ........................... 57 Ecological Aspects of Forest Roads Network Development Vladimir V. Nikitin .................................................................................................

71

Session 4: Treatment of Effluents with Forest Residues Application of Pine Bark as a Sorbent for Organic Pollutants in Effluents I. Brás, L. Lemos, A. Alves, M.F.R. Pereira ........................................................... 79

Session 5: Environmental Impact of Wood Preservation Creosote Spills as a Cause of Odours in River Waters. Identification of the Source and Strategies Adopted for Minimizing its Impact Emese Borsiczky, Fabián-Ángel Sánchez-Alonso, Jordi Martin-Alonso ............... 93


8

Session 6: Heat-Treatment of Wood Hygroscopicity in Heat-Treated Wood: Effect of Extractives Marcos M. González-Peña, Martin C. Breese, Callum A. S. Hill ..........................

105

Session 7: Nonwood-Fibre Products Minimizing Environmental Burden of Oil Palm Trunk Residues Through Development of Laminated Veneer Lumber Product Kamarulzaman Nordin, Mohd Ariff Jamaludin, Mansur Ahmad, Hashim W. Samsi, Abdul Hamid Salleh and Zaihan Jalaludin ..............................

123

Manufacturing of Fiber Composite Medium Density Fiberboards (MDF) Based on Annual Plant Fiber and Urea Formaldehyde Resin Sören Halvarsson, Magnus Norgren, Håkan Edlund ..............................................

131

Manufacturing Parameters on the Properties of Cement-Bonded Boards Using Sugarcane Bagasse Dwight A. Eusebio .................................................................................................. 149

Session 8: Alternative Methods for Wood Preservation Environmental Friendly Wood Protection – Furfurylated Wood as Alternative to Traditional Wood Preservation Stig Lande, Mats Westin, Marc H. Schneider ......................................................... 163 Toxicity of Pine Resin Derivatives to Subterranean Termites (Isoptera: Rhinotermitidae) L. Nunes, T. Nobre, B. Gigante, A. M. Silva .......................................................... 177 Studies of Wood Preservation with Boron. Research on Fixation Mechanisms using Maritime Pine A. M. Ramos, F. Caldeira Jorge, C. Botelho ..........................................................

187

Preliminary Evaluation of Fungicidal and Termiticidal Activities of Hydrolysates from Biomass Slurry Fuel Production from Wood S. Nami Kartal, Noriaki Katsumata, Yuji Imamura, Fujio Tsuchiya, Katsuaki Ohsato ............................................................................ 195


9

Session 9: Wood and Lignocellulosics Recycling Session 9a: Recycling by Chemical Processing Recycling of Chromium-Copper-Arsenate (CCA) Treated Wood by Thermochemical Conversion Toshimitsu Hata, Paul Bronsveld, Tomo Kakitani, Dietrich Meier, Takeshi Kajimoto, Yuji Imamura ...........................................................................

207

New Environmentally Friendly Method to Coat Metal with Wood Distillate Hannu Kuopanportti, Esko Hotti, Mikko Selenius, Reijo Lappalainen .................. 217

Session 9b: Physical Processing Recycling of Cork Residues into Cork-Cement Panels or Light-Weight Concrete Catarina Pereira, Fernando Caldeira Jorge, Mark Irle, José M.F. Ferreira, Afonso Serra Neves, Joana Sousa Coutinho ...........................................................

225

Floating Concrete! Light-Weight Concrete from Granulated Cork Waste Mark Irle, Sukhdeo Karade, Kevin Maher .............................................................. 235 Composites of Cork and Used Beverage Cartoons Luís Gil, Paulo Cortiço ...........................................................................................

245

General Waste Handling and Recycling in Particle Board Production – A Hungarian Case Study M. Varga, T. Alpár, G. Németh ..............................................................................

255

Characterization of Urea-Formaldehyde Resins for the Production of Composites Made with Recycled and Waste Wood Ruben Araújo, Luísa Carvalho, Fernão Magalhães, Adélio Mendes ...................... 269

Session 10: Adhesives from Renewable Resources Joining Wood by Friction Welding Bernhard Stamm, Julius Natterer, Parviz Navi .......................................................

283

Improved Water Resistance of Bio-Based Adhesives for Wood Bonding Charles R. Frihart, James M. Wescott ....................................................................

293

Response of Six Indonesian Wood Species to Lignin Resorcinol Formaldehyde I.M. Sulastiningsih, Adi Santoso ............................................................................

303


10

Session 11: VOC’s and other Atmospheric Pollutants Environmental Management in Wood Processing Industries and the European Legislation on VOC Emission Control Ana Fonseca ............................................................................................................ 313 Activities to Comply with the European VOC Directive in Slovenian Furniture Industry Borut Kricej, Miro Tomazic, Matjaz Pavlic, Marko Petric ..................................... 325 Optimization of Fine Air-Borne Dust Collecting System in Woodworking Factories Using Computer Simulation and Finite Element Method (FEM) Yoshihisa Fujii, Tomoko Kawano, Yuko Fujiwara and Shogo Okumura ..............

333

Extended Abstracts from Poster Presentations The Contribution of Wood Products to Carbon Sequestration in Portugal Ana Cláudia Dias, Margarida Louro, Luís Arroja, Isabel Capela ........................... 345 The Challenge of Bonding Treated Wood Charles R. Frihart ....................................................................................................

351

Wettability and pH Value Related to Glualibility of Calcutta Bamboo Mansur Ahmad, Frederick A. Kamke .....................................................................

357

Interlaminar Shear Strength of Multi-Layered Oriented Strength Board from Plantation Species Mohd Ariff Jamaluddin, Mansur Ahmad, Kamarulzaman Nordin, Awg Roslan Awg Bosman ......................................................................................

365

Philippine Bast Fibers for Local Cottage Industry Arsenio B. Ella ........................................................................................................

371

Thermal Loading in Wood Cutting Tools Csanády Etele .......................................................................................................... 381 Improving the Dimensional Stability of the Quebracho Blanco Sawn Timber Using Tannin and Polyethyleneglycol Rolando Hipólito Martínez, Luis García Esteban, Juan Carlos Medina .................

389

Verification of the Antimicrobial Activity in Tree Species of the Chaco Forest Used in Popular Medicine and Assessment of their Potential Use in Pharmacology Corzo, A.G, Soberón, J.R., Carranza, M.E., Martínez, R.H., Vattuone, M.A. .......

397


11 Formaldehyde Emissions from Wood-Melamine-Formaldehyde Resin Compounds during the Extrusion Process Irina Rosca, Ildiko Tanczos, Harald Schmidt .........................................................

405

Introducing a Simple Method for Laboratory Testing on Subterranean, Dry-Wood and Damp-Wood Termites to Evaluate Susceptibility of Wood Paimin Sukartana, Jasni .......................................................................................... 411 Novel Non-Destructive Methods to Detect Biodegradation of Wooden Construction Yoshihisa Fujii, Yoshiyuki Yanase, Yuko Fujiwara, Masahiro Miura, Shogo Okumura, Nami Kartal, Tsuyoshi Yoshimura, Yuji Imamura ....................

417

Chemical and Anatomical Characterization of Cypress Wood for Pulp Production Sandra Esteves, António Santos, Cristina Morais, Ofélia Anjos, Rogério Simões, Helena Pereira .............................................................................

427

Commercial Exploitation of Forest Resources and Commercial Use Of Forest Land In Bangladesh Mohammed Alamgir Sharif ....................................................................................

435

A Review of Chain of Custody Requirements for Different Forest Certification Schemes Kate Wingate, Paul Mcfarlane ................................................................................

441

Choice and Formation of Harvester Park for Russian Logging Enterprises, Taking into Consideration Ecological Requirements Valentin A. Makuev ................................................................................................ 449 Estimating of the Volume of Sinked Timber in Rivers Karpachev Sergey ...................................................................................................

459

Logistic Systems in Transporting of Wood Materials S.P. Karpachev, V.V. Lozovetskiy .......................................................................

461


12


13

Session 1 Opening Session and Keynote Address


14


15

BIOBASED COMPOSITES: A 21ST CENTURY PERSPECTIVE Roger M. Rowell United States Department of Agriculture, Forest Service, Forest Products Laboratory, and Biological Systems Engineering Department, University of Wisconsin, Madison, WI

SUMMARY Biobased resources have played a major role throughout human history. Even the earliest humans learned to use these resources to make shelters, cook food, construct tools, make clothing, keep records and produce weapons. Collectively, society learned very early the great advantages of a resource that was widely distributed, multifunctional, strong, easy to work, aesthetic, biodegradable and renewable. Composites made of biobased resources are being “rediscovered” in the 21st century. Many “new” concepts are circulating in the press, at scientific meetings, in discussions and on web sites all over the world on the advantages of using a “green”, “low cost”, “biodegradable”, “renewable”, “recyclable”, and “sustainable” resource to produce composites. While these are powerful concepts, they are often misapplied, misused, and misunderstood. These are not just “buzz” concepts that are used in a grant to get funding. They are part of the reason biobased composites are getting more attention in the materials world. The largest single potential growth in the use of biobased resources is in the composites sector. More specifically, in housing components. Issues to be considered in this sector include land use, resource supply, carbon sequestration, reduced energy use, utilization of mixed resources, recycling, competing materials, durability, life cycle assessment, quality assurance, codes and standards, tradition, marketing, and appropriate use.

INTRODUCTION For most of the world, the main biobased resource used today is wood. Some countries, however, have very little wood left and have turned to utilizing other types of bioresources for products. In order to take advantage of bioresources, it is important to understand what they are, how they are isolated, and how they are processed into products. Cellulose is the most abundant natural polymer in the world. It is estimated that 830 million tons of cellulose are produced each year through photosynthesis (Krassig 1993). If the average plant (on a dry weight basis) contains 40% cellulose, the annual biobased resource would be approximately 2000 million dry tons. This compares to 225 x 109 tons which is the estimated world reserves of petroleum and natural gas. While the biobased resource is renewable, the petroleum and gas resources are not. Biobased resources are available in many different sizes and shapes depending on the intended product: solid, veneers, strands, flakes, chips, particles, fiber bundles, fibers (aspect ratio > 10), flour (aspect ratio < 10), cellulose crystallites, simple sugars, carbon (Marra 1979). The energy required to isolate a given fraction goes up as selected furnish size gets smaller (Maloney 1996). In terms of tensile strength, clear wood has a tensile strength of 12 GPa, the cell wall, 35 GPa, and the cellulose microcrystal 137 GPa (compared to Kevlar at 130 GPa. Biobased resources are ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

16 anisotropic, hydroscopic, non-abrasive, porous, viscoelastic, biodegradable, combustible, compostible, reactive, high aspect ratio, high strength to weight ratio, have good insulation properties (sound, electrical and thermal), the structure is hollow, laminated, with molecular layers and with an integrated matrix. Biobased resources have been traditionally used for textiles, pulp and paper, packaging, composites, filters, sorbents, geotextiles, fuel, and chemicals. Of these products, those keeping pace with world population growth are textiles, pulp and paper, packaging, and chemicals. Sectors growing faster than world population are energy consumption, materials consumption, and clean water. That is to say, that increased demand for energy, materials and clean water are growing faster than the population increase. The fastest growing population type in the world is the middle class and their demand for materials is growing. For example, in the United States in 1970, the average size of a single family home was 140 square meters. In 1990, this had risen to 193 square meter and it is projected that by the year 2050, the average size will be 240 square meter. The demand for energy has caused an increase in energy prices and it is projected that energy costs will continue to increase over the next decade. For this reason, reducing energy used in all bioresource processing is a major goal of the industry in the future. The demand for clean water is also on the rise due to more world wide chemical use and increased pollution.

COMPETITION OF RESOURCES Global biobased resources play a major role in the conversion of carbon dioxide to oxygen which is essential for human life. With the continuing debate over clean air, the need to retain a large portion of the biobased resource, especially forests, for oxygen generation will be one of the major arguments (along with other environmental and multiple use considerations) for not cutting trees. There will be increased competition for land use as the world population increases. In 1830, the world population was 1 billion. Over the next hundred years, it doubled to 2 billion. At the present rate of growth, the population increases by 1 billion people every eleven years. Biobased composites provide an opportunity to fill a growing need for materials, however, there will be a greater and greater need for food and feed. There is a continuing debate about the use of renewable versus non-renewable resources. The future trend will favor the use of renewables but within an economic framework. Recycling will become a much more important issue and will give the nonrenewable resources an argument in favor of "sustainable non-renewables" based on the reuse of these resources without significant loss of performance properties. These issues can be answered in a life cycle assessment which gives a valid comparison of one product versus another. In order to insure a continuous supply of biobased resources, management of the agricultural producing land should be under a proactive system of land management whose goal is both sustainable agriculture and the promotion of healthy ecosystems. Ecosystem management is not a euphemism for preservation, which might imply benign neglect. Sustainable agriculture denotes a balance between conservation and ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

17 utilization of agricultural lands to serve both social and economic needs, from local, national and global vantage points. Sustainable agriculture does not represent exploitation but rather is aimed toward meeting all the needs of the present generation without compromising the ability of future generations to meet their needs. It encompasses, in the present case, a continuous production of biobased composites utilizing the biobased resources, considerations of multi-land use, and the protection, restoration, and conservation of the total ecosystem. It has been stated that wood is renewable. This is not true. The tree the wood came from is renewable and, therefore, the need for healthy ecosystems is where the emphasis should be in discussions of renewability. DRIVERS OF TECHNOLOGY There are three major factors driving technology in the biobased material sector: cost, performance and the environment. Most biobased composites are large volume low cost and are therefore cost driven. That is to say that the cost of the product is the major driver in this industry. Performance is important but for most biobased composites it is not the primary driver of the technology. In some cases, performance is a consideration for the industry to avoid law suits and for the consumer, assurance of produce durability with more leisure time. Consumers are willing to pay more for improved performance but how much more is debatable. The environment is another consideration but just how much more consumers will pay for an “environmentally friendly” product is still a subject of debate.

RECYCLING AND ENVIRONMENTAL ISSUES Using environmentally sound technologies to make biobased materials that are cost effective and performance driven is the direction in which we need to go. At the present time, many of our current composites that are cost effective may not be the best from the environmental perspective and vice versa. Many companies talk about moving toward "green technologies" but this concept has yet to be clearly defined. In fact, many concepts in the environmental arena are not clearly defined or understood. For example, we use the term recyclable to mean the reuse of a resource. But, what does this term mean if there is no recycling program in your area? And, if there is a recycling program, how much of a given resource is, in fact, recycled? Further, if it is recycled, what products are going to be made from that resource and how have the properties of a given product changed due to the recycle content? Many, if not most, products made today from recycled resources are more costly and have reduced performance properties as compared to the original product made using virgin resources. What we call recycling is really not more than isolation and reuse. Future technologies in recycling will be much different than anything we use today. For example, different types of plastics will be sorted into chemical composition groups and either re-melted or broken down into their basic chemical components to be used in the same or different products. Metals, glass, rubbers, and other resources will be sorted by type, color, chemical composition and either reused in a pure form or broken down into more ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

18 basic components to be used again in new and different products. Perhaps, biobased resources will some day be either burned to recover energy or broken down into simple chemical units to be used in completely different ways than can now be imagined. This might include single carbon chemistry (carbon monoxide) that can be used to make a wide variety of chemicals. We also use the term "biodegradable" as one of the advantages of using biobased resources. A biobased product disposed of in a modern land fill does not undergo aerobic biodegradation due to the lack of oxygen in these sealed, air tight "tombs". So, a product can be sold as biodegradable but may never biodegrade. Even using the term "compostible" may also be misleading since it may not end up in a composting recycling program. Another interesting aspect to using biodegradability as a promotional element in advertising is that we do not want all products to biodegrade. Certainly it is not wise to advocate the use of biodegradable structural components in exterior airplane wings or critical components to a high rise buildings! So, biodegradability will have a place in future products but we must apply the concept where we can use it to our advantage. This might be termed appropriate use. There is a lot of discussion pro and con for the use of biobased resources for energy use. While this may appear to be outside the scope of this paper, we must realize that the ultimate end point of recycling biobased resources is composting, burning or land filling. Unless we come up with new technologies to control the degradation of the resource due to recycling, we must carefully consider the full life cycle of our biobased materials. It is interesting that a group of people may be very much in favor of composting and strongly opposed to burning. The opposition comes from the generation of carbon dioxide from the combustion process. What they do not understand is that approximately the same amount of carbon dioxide is produced as a result of composting as is produced as a result of burning. Burning biobased resources is not the major concern in the global issue of carbon dioxide production. Burning biobased resources can be considered a cyclic phenomenon. That is, the carbon dioxide produced by burning or composting biobased resources is equal to the carbon dioxide consumed by the biobased resource in the production of the resource. The real carbon dioxide problem is the burning of fossil fuels. It is estimated that 1.7 billion tons of carbon dioxide are emitted from power plants in the United States each year from burning fossil fuels (Herzog 1997). The potential global conversion of carbon dioxide to oxygen by photosynthetic pathways can not convert the vast amount of carbon dioxide generated by the burning of sequestered carbon (oil and coal) millions of years old. From the environmental stand point, one of the big issues in biobased composites is the nature of the adhesive used. Most of the industrial adhesives used today are petroleumbased, i.e. phenol, formaldehyde, urea, isocyanates, PVA, etc. Concerns on the use of petroleum based adhesives include volatiles released in the production of composites, volatiles released in the use of composites, toxicity of resins in production, use, recycling, and disposal of composites, and costs.


19 Research is underway to develop new adhesive systems that are not petroleum-based, that are not toxic, that are based on the use of renewable resources, and that are based on a better understand of the mechanisms of adhesion. These mechanisms require a much better understanding of interface and interphase relationships between the biobased elements and the matrix. Some of the newest research in this area involves the use of enzymes, surface activation, biotechnology, chemical modification, and cold plasma technology.

PRODUCTS There are many opportunities to expand existing markets and develop new markets for biobased materials (Rowell 1997). One past trend that exists in the biobased composite industry is that when a new product is developed, its intended market is, in many cases, to replace an existing bio-based composite. We have what might be called a biobased materials pie. We expand the market of one biobased composite at the expense of another biobased composite. For example, oriented strandboard or flakeboard may be targeted to replace part of the plywood market. This does not expand the overall size of the biobased composite pie, it only changes the size of each piece within the pie. We need to expand the whole pie! We need to start thinking about a new approach. One in which we introduce new products into markets to replace products not made from biobased resources. We need to both reclaim lost markets to plastics, metals, and other synthetics and get into markets we have never been in before. When assessing potential markets for biobased materials, there are two important aspects to consider in the decision making process: markets and products. Within each of these, there are two possible options: existing markets and new markets and existing products and new products. There are already existing products in existing markets and those will continue and, in some cases, grow. What is desired is to use an existing market to introduce a new product or go after a new market with an existing product. The greatest risk occurs when one tries to introduce a new product into a new market. This approach is much riskier but can work. As was stated before, the trend in the biobased industry is to go toward smaller and smaller elements in composites. At the fiber level, most defects in the growing plant have been eliminated so fiber-based products tend to be more homogeneous. Fibers are available from many plants other than trees so the availability of different fiber types is greatly expanded. Wood fibers come in two lengths: short and shorter. So wood fiber is limited to short fiber applications unless it is combined with longer agricultural fibers for applications in a wider array of products. There are potential products that can come from short plant fiber alone, some from long fiber bundles combined with short fibers, and some from the long fibers alone. Using all potential fibers in various combinations provides us with a new resource base to consider many new and exciting products. There are many new product potentials to be considered for future development. Markets for existing products will expand but whole new markets are possible. The following is just a partial list of new possibilities that are mainly based on using fibers.


20 They include: geotextiles, filters, structural composites, non-structural composites, molded products, packaging, and combinations with other resources.

Geotextiles Long fibers such as kenaf, jute, cotton, sisal, agave, etc. can be combined with short fibers and formed into flexible fiber mats, which can be made by physical entanglement (carding), nonwoven needling, or thermoplastic fiber melt matrix technologies (see Figure 1). In carding, the fibers are combed, mixed and physically entangled into a felted mat. These are usually of high density but can be made at almost any density. A needle punched mat is produced in a machine which passes a randomly formed machine made web through a needle board that produces a mat in which the fibers are mechanically entangled. The density of this type of mat can be controlled by the amount of fiber going through the needle board or by overlapping needled mats to give the desired density. In the thermoplastic fiber matrix, the biobased fibers are held in the mat using a thermally softened thermoplastic fiber such as polypropylene or polyethylene. Medium- to high-density fiber mats can be used in several ways. One is the use as a geotextile. Geotextiles derive their name from the two words geo and textile and, therefore, mean the use of fabrics in association with the earth. Geotextiles have a large variety of uses. These can be used for mulch around newly planted seedlings. The mats provide the benefits of natural mulch; in addition, controlled-release fertilizers, repellents, insecticides, and herbicides can be added to the mats as needed. Research results on the combination of mulch and pesticides in agronomic crops have been promising. The addition of such chemicals could be based on silvicultural prescriptions to ensure seedling survival and early development on planting sites where severe nutritional deficiencies, animal damage, insect attack, and weed problems are anticipated. Medium density fiber mats can also be used to replace dirt or sod for grass seeding around new homesites or along highway embankments. Grass or other type of seed can be incorporated in the fiber mat. Fiber mats promote seed germination and good moisture retention. Low and medium density fiber mats can be used for soil stabilization around new or existing construction sites. Steep slopes, without root stabilization, lead to erosion and loss of top soil.

Figure 1 – Air laid mat made from wood fiber (95%) and cotton fiber (5%).


21

Medium and high density fiber mats can also be used below ground in road and other types of construction as a natural separator between different materials in the layering of the back fill. It is important to restrain slippage and mixing of the different layers by placing separators between the various layers. Jute, kenaf, and flax geotextiles have been shown to work very well in these applications but the potential exists for any of the long biobased fibers.

Filters It is estimated that global spending on filtration (including dust collectors, air filtration, liquid cartridges, membranes and liquid macro-filtration) will increase from $17 billion in 1998 to $75 billion by 2020. The fastest-growing non-industrial application area for filter media is for the generation of clean water. Medium and high density fiber mats or fiber filler containers can be used for air filters. The density of the mats can be varied, depending on the size and quantity of material being filtered and the volume of air required to pass through the filter per unit of time. Air filters can be made to remove particulates and/or can be impregnated or reacted with various chemicals as an air freshener or cleanser. Medium to high density mats can also be used as filtering aids to take particulates, nutrients, oil/grease, nitrogen, phosphorus, heavy metals and toxic organics out of contaminated water.

Structural Composites A structural composite is defined as one that is required to carry a load in use. In most cases, these require a thermosetting resin matrix. In the housing industry, for example, these represent load bearing walls, roof systems, subflooring, stairs, framing components, furniture, etc. In most, if not all cases, performance requirements of these composites are spelled out in codes and/or in specifications set forth by local or national organizations. Structural composites can range widely in performance from high performance materials used in the aerospace industry down to biobased composites which have lower performance requirements. Within the biobased composites, performance varies from multi-layered plywood and laminated lumber to low cost particleboard. Structural biobased composites, intended for indoor use, are usually made with a low cost adhesive which is not stable to moisture while exterior grade composites use a thermosetting resin that is higher in cost but stable to moisture.

Non-Structural Composites As the name implies, non-structural composites are not intended to carry a load in use. These can be made using a thermoplastic matrix or a thermosetting matrix and are used for such products as doors, windows, furniture, gaskets, ceiling tiles, automotive ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

22 interior parts, molding, etc. These are generally lower in cost than structural composites and have fewer codes and specifications associated with them. Because of the aesthetic nature of biobased composites, they lend themselves to products that "surround" people like wall coverings, room dividers, and furniture.

Three-Dimensional Composites The present wood-based composite industry mainly produces two dimensional (flat) sheet products. In some cases, these flat sheets are cut into pieces and glued/fastened together to make shaped products such as drawers, boxes, and packaging. Flat sheet fiber composite products are made by making a gravity formed mat of fibers with an adhesive and then pressing. If the final shape can be produced during the pressing step, then the secondary manufacturing profits can be realized by the primary board producer. Instead of making low cost flat sheet type composites, it is possible to make complex shaped composites directly using the long fibers alone or combinations of long and short fibers. In this technology, fiber mats are made similar to the ones described for use as geotextiles except during mat formation an adhesive is added by dipping or spraying of the fiber before mat formation or added as a powder during mat formation. The mat is then shaped and densified by a thermoforming step. Within certain limits, any size, shape, thickness, and density is possible. These molded composites can be used for structural or non-structural applications as well as packaging, and can be combined with other materials to form new classes of composites. This technology will be described later.

Packaging Medium and high density biobased fiber composites can be used for small containers, for example, in the food industry and for large sea-going containers for commodity goods. These composites can be shaped to suit the product by using the molding technology described previously or made into low cost, flat sheets and made into containers. Biobased materials can and have been used for pallets where cost and weight are critical factors. Moldability has been a key factor in the development of the biobased pallet. Biobased fiber composites can also be used in returnable containers where the product is reused several times. These containers can range from simple crease-fold types to more solid, even nestable, types. Long bio-fiber fabric and mats can be overlayed with thermoplastic films such as polyethylene or polypropylene to be used to package such products as concrete, foods, chemicals, and fertilizer. Corrosive chemicals require the plastic film to make them more water resistant and reduce degradation of the biobased fiber. There are also many applications for biobased fiber as paper sheet products for packaging. These vary from simple paper wrappers to corrugated, multi-folded, multilayered packaging. ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

23

Combinations with Other Resources It is possible to make completely new types of composites by combining different resources. It is possible to combine, blend, or alloy biobased fiber with other materials such as glass, metals, plastics, and synthetics to produce new classes of materials. The objective will be to combine two or more materials in such a way that a synergism between the components results in a new material that is much better than the individual components. The concept is to focus on product performance rather than product resource. That is, use whatever resource is needed to provide the desired properties of the product and not be an advocate for any one resource. For example, there is a very thin layer of aluminum on the inside of every potato chip bag to exclude moisture and air resulting in a very long shelf life and product freshness. Similarity, a thin film of aluminum could be put on the outside of a biobased composite that would provide ultra violet protection and a moisture barrier with a low cost core providing sound and thermal insulation in a siding product. We tend, however, to be advocates (because of our training) for a single resource because it is what we understand. As a result we have aluminum siding, wood siding, brick siding, etc. Wood fiber may be combined with agricultural fiber for new composites. Wood has a higher density than annual plants so there will be more bulk when using agricultural crop fiber. There are also concerns about the seasonality of annual crops that requires considerations of harvesting, separating, drying, storing, cleaning, handling, and shipping. In the present system of using wood, storage costs can be reduced by letting the live tree stand until needed. With any annual crop, harvesting must be done at a certain time and storage/drying/cleaning/separating will be required. This will almost certainly increase costs of using agro-based resources over wood depending on land and labor costs, however, in those countries where there is little or no wood resource left or where restrictions are in place to limit the use of wood, alternate sources of fiber are needed if there is to be a natural fiber industry in those countries. The total global inventory of all bio-resources is over 4 billion metric tons, of which wood represents 1.75 billion, straw, 1.15 billion, stalks, 0.97 billion, and 0.13 billion for all other (grass, leaf, core, bagasse, etc). While these global inventory numbers are interesting they are not very useful. The potential resource available within a 5, 10 or 20 km radius from a central processing location is the kind of data that is needed before any decisions can be made on an economic, sustainable source of raw material. Biobased fiber/glass fiber composites can be made using the glass as a surface material or combined as a fiber with biobased fiber. Composites of this type can have a very high stiffness to weight ratio. The long bast fibers can also be used in place of glass fiber in resin injection molding (RIM) or used to replace, or in combination with, glass fiber in resin transfer molding (RTM) technologies. Problems of dimensional stability and compatibility with the resin must be addressed but this could also lead to new markets for property enhanced biobased resources. Metal fibers can also be combined with stabilized fiber in a matrix configuration in the same way metal fibers are added to rubber to produce wear-resistant aircraft tires. A ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

24 metal matrix offers excellent temperature resistance and improved strength properties, and the ductility of the metal lends toughness to the resulting composite. Application for metal matrix composites could be in the cooler parts of the skin of ultra-high-speed aircraft. Technology also exists for making molded products using perforated metal plates embedded in a phenolic-coated fiber mat, which is then pressed into various shaped sections. Bio-fibers can also be combined in an inorganic matrix. Such composites are dimensionally and thermally stable, and they can be used as substitutes for asbestos composites. Inorganic bonded bast fiber composites can also be made with variable densities that can be used for structural applications. One of the biggest new areas of research in the value added area is in combining natural fibers with thermoplastics. Since prices for plastics have risen sharply over the past few years, adding a natural powder or fiber to plastics provides a cost reduction to the plastic industry (and in some cases increases performance as well) but to the biobased industry, this represents an increased value for the biobased component. Blending of the plastics with the biobased fibers may require compatibilization to improve dispersion, flow and mechanical properties of the composite. Extrusion of biobased filled plastics for the automotive industry is well known and has been used for more than twenty years. Typical blending involves the plastic-filler/reinforcement to be shear mixed at temperatures above the softening point of the plastics. The heated mixture is then typically extruded into "small rods", that are then cut into short lengths to produce a conventional pellet. The pellets can then be used in typical injection or compression molding techniques. To reduce the cost of this blending process, direct injection molding of bio-fiber/plastics can be done. The direct injection molding process probably has limitations on the amount of filler/fiber that can be used in the composite, and is also likely to be limited to particulate or shorter fiber. The chemical characteristics of the surface and bulk of the bio-fibers are also important in the blending with plastics. The ability of the matrix of the lignocellulosic (hemicellulose and lignin) to soften in the presence of moisture at plastic processing temperature may give these materials unique characteristics to develop novel processing techniques. The primary advantages of using biobased fibers as fillers/reinforcements in plastics are low densities, non abrasive, high possible filling levels resulting in high stiffness properties, high specific properties, easily recyclable, will not fracture when processing over sharp curvatures, biodegradable, wide variety of fibers available throughout the world, would generate rural jobs, increases non-food agricultural/farm based economy, low energy consumption, and low cost. Material cost savings due the incorporation of the relatively low cost bio-fibers and the higher filling levels possible, coupled with the advantage of being non-abrasive to the mixing and molding equipment are benefits that are not likely to be ignored by the plastics industry for use in the automotive, building, appliance and other applications. This is just a few of the potential expanded or new product lines that can be produced from a wide variety of biobased resources. In most cases in the Western world, biobased composite production units are very large. This is based on the economy of scale and the fact that most of the existing biobased ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

25 composite industries are based on high volume sales. Future mills, in many countries will be smaller production plants with the possibility of mobile units that can be moved to a given production site. Equipment for these smaller mills will be made within that country using indigenous materials and local labor. The cost of importing new, and in some cases, used equipment from Western countries is prohibitive for most developing countries.

PROCESSING For the most part, the processing of bioresources into biomaterials has not changed much in the past twenty to fifty years. There have been improvements in sawing techniques to reduce residues and new processing lines have been set up to produce composites. But, most of these lines were developed in the area of cheap energy and abundant clean water. Energy is no longer cheap and clean water regulations have become more stringent. The biomaterials industry needs to take a new look at ways to cut down in both energy and clean water use. In some cases, the present processes to convert solid wood into composites can be tightened to reduce energy and water use. In other cases, new conversion technologies need to be developed with the specific objective to increase conversion efficiency, generating less waste, and using less water and energy.

DURABILITY We have used biobased materials for so long that we tend to accept their performance limitations in use such as swelling, shrinking, rotting, burning, ultraviolet radiation degradation, etc. By learning to live with these limitations, we have also limited our expectations of performance, which, ultimately, limits our ability to accept new concepts for improved performance and expanded markets. Biomaterials are very familiar materials that have been used for centuries by common people for low cost, medium to low performance markets. We may have limited our expectations of biomaterials to a time long gone while years of advances in chemistry and materials science research have been taking place. Biobased resources were designed, after millions of years of evolution, to perform, in nature, in a wet environment. Nature is programmed to recycle these resources, in a timely way, back to basic building blocks of carbon dioxide and water through biological, thermal, aqueous, photochemical, chemical, and mechanical degradations. In order to expand the use of biomaterials used in adverse environments, it is necessary to interfere with natures recycling chemistry. We have industries that chemically treat biomaterials to perform better in adverse environments. Wood preservatives, fire retardants, ultraviolet energy absorbers, water repellents, paints, and coatings have all been developed to help protect biomaterials from environmental degradation. All of these treatments are coming under attack from an environmental stand point and new technologies need to be developed. Many of the traditional broad-spectra toxic wood preservatives are already banned or were never used in many countries. Existing technology for broad spectra toxic preservation will give way to specifically targeted toxicity and technologies that do not depend on toxicity at all. Several leachable, corrosive, and degrading fire retardants have been restricted and research in this area ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

26 will be based on bound or bonded, non-hydroscopic, non-degrading fire retardants. Research is underway to develop new concepts in biomaterials protection.

QUALITY ASSURANCE AND TESTING There is a need to develop an assurance of quality in the use of biobased products in world markets especially where they are replacing traditional products made from other resources. This requires the need to develop codes and specification of each desired biobased composite product. In some local markets around the world, there may be no or zero code requirements at the present time. In order to assure the user of the composites that the product will perform in a certain way, codes should be developed to insure consumer confidence. Certainly for international markets, there will be a need to follow codes and specifications for the intended country. Without this, there is no hope of entering that country's markets with a new biobased composite. There is a need for research on properties and performance of each type of proposed composite. Research, however, can go on forever but, in most cases, there is not enough data to convince industry to start using a new resource over the one they presently use. Research can just push so far to get a new product into the market. At some point the market must pull the technology into use. There are many examples where there has been a strong research push but no market pull and so the technology remains in the research arena. We need to expand and standardize our testing program on both chemical and physical properties of biobased resources and composites made from them. We need to keep up with new experimental techniques that are being developed in other materials areas and utilize these new innovations to expand our knowledge base of our composites. We also need to establish a meaningful link between accelerated aging tests and the "real world". Many of the tests done in the laboratory are intended to speed up the time frame to failure of a given composite in a given environment but we have little understanding of exactly how much time the test represents in the world in which the composite will be expected to perform.

CONCLUDING THOUGHTS The future of biobased materials will be very exciting and dynamic. It will be driven by traditions, trends, costs, performance, availability of resources, and legislation. Of these, the most critical issue is costs. Logical, creative and futuristic ideas will have little chance of success if the economics are not positive. In the area of biobased resource utilization, there are several competing ideologies today that are driving public opinion. On the one hand we have a growing need to create jobs, expand recreational opportunities, and improve the standard of living. On the other hand, we have concerns about energy consumption, an expanding world population, maintaining wilderness areas, cleaning up the environment, and the consumption of our natural resources. There is no "right or simple" answer. There is no question that renewable, recyclable, and sustainable resources will play a major role in future world developments. Biobased resources and biobased materials ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

27 will be part of this dynamic future. We must not allow ourselves to be locked into a mental framework tied to past technologies or close our eyes to exciting new possibilities. By considering the entire biobased resource as a raw material for biobased composites, we are not limited to just one type of biobased resource. Wood will remain the major raw material for many biobased composites for those countries that have an abundant wood supply and where wood remains economic. Countries with limited or no wood supplies will rely more and more on non-wood biobased resources. The world is full of many varieties of renewable biobased resources and only a few of them are now being used for composites. Existing biobased composite markets will continue to grow but the real excitement will come in the new potential markets we have either lost to other resources or in totally new markets. We are no longer limited by shape, density, size, or texture. We are only limited by our imagination and our knowledge and understanding of how to achieve the highest level of performance from these biobased composites. We must interact in the materials community to develop our true place in the materials world. As in all competitions, there will be a natural selection process in materials usage where the survival of the fittest will be the resource that has the most to offer in a cost effective and environmental framework. We must lead market development not simply react to consumer needs. We must not only continue to make incremental improvements in present technology but develop whole new technologies that take one of the oldest resources into a leadership position in composites into the 21st century. The 21st century may well be known as the cellulosic era. There will be more products that are derived from biobased resources (Rowell 1998). This expanded role for biobased resources will not only take place in composites, but also in chemicals, paper and paper products, fuels, lubricants, bioenergy, and many other products. We have two choices: we can put our collective hands in our pockets and lament the passing of the "good old days" or we can put our collective minds and resources into a bright new future in biobased resources and composites. If we chose the new future, it will require vision, risk, capital and commitment. The future is in our hands.

REFERENCES HERZOG H (1997): Carbon dioxide capture, reuse, and storage technologies for mitigating global climate change, White Paper Final report, Massachusetts Institute of Technology, Cambridge, MA. Prepared under DOE Order No. DE-AF22-96PC01257. KRASSIG HA (1993): Polymer Monographs, Vol 2, Eslevier Press, New York, NY. MALONEY TM (1996): The family of wood composites. Forest Prod. J.46(2): 19-26


28 MARRA G (1979): Overview of wood as a material. J. Educational Modules for Materials Science and Engineering. 1(4):699-710. ROWELL RM (1997): Agro-based composites: Exploring the limits. In Proceedings: 18th Riso International Symposium on Materials Science, Polymeric Composites Expanding the Limits, S.I. Anderson, P. Brondsted, H. Liholt, A. Lystrup, J.T. Rheinlander, B.F. Sorensen, and H. Toftegaard, Eds., 127-133, Riso National Laboratory, Roskilde, Denmark. ROWELL RM (1998): The state of art and future development of bio-based composite science and technology towards the 21st century. Proceedings: The fourth Pacific Rim Bio-Based Composite Symposium, Y.S. Hadi, ed., Bogor, Indonesia, November 1-18.


29

Session 2 Broad-Scope Communications


30


31

SOLVING THE PROBLEM OF MISUNDERSTANDING AT THE INTERNATIONAL CONFERENCES, DEALT WITH ECOLOGY AND FORESTRY Julia V. Ivanova

Student of the 4th year, the interpreter of the delegation from Moscow State Forest University

Valentin A. Makuev Doctor of technical science, lecturer, dean of the part-time education faculty MFSU - Moscow State Forest University – 1, 1st Institutskaia St., Mitischi-5, Moscow Region, 141005 – Russia. ([email protected])

SUMMARY The paper deals with the wrong understanding of the meaning of the forest, technical and ecological terms and its inappropriate use at the international conferences, where the English language is the work language for participants, but not the native one. The paper describes some causes of misunderstanding and presents their examples, taken from conferences, congress and journals of forestry.

INTRODUCTION Specialists of various spheres of science and from different countries meet at international conferences to share their knowledge with each other and to receive new information. As the specialists are from different countries, so they speak different languages. So that, it was decided to speak one common language at the international conferences, that is English language, to make communication more productive. MATERIALS AND METHODS Causes of Misunderstanding Communication between representatives of different cultures at the international conferences is still not as productive as desired. There are some causes, which describe such misunderstanding at the international conferences, where the English language is the work language for participants, but not the native one. The first cause of misunderstanding at the conferences, dealt with ecological and forest matters, is that conferences include specialists of various majors, such as forest engineering, technology of wood, chemical technology of wood, forestry, garden & park and landscape design, machines and equipment for forestry and etc. Every major is characterized by huge amount of specific terms. It’s clear that a specialist of one major can’t know all terms of other majors and, consequently, doesn’t know them in foreign language. ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

32

The second cause consists in difference between languages of different countries. Every language has its own pronunciation, intonation, construction of sentences and etc. These characteristics have a great influence on people’s English, for whom English is not the native language. So, people from different countries speak “different” English. Sometimes the wrong pronunciation of one sound in a word leads to the wrong understanding of this word. And in some cases, it doesn’t allow to understand clearly the whole sentence or utterance. For example: [ju:s] use [ju:z] If the word “use” is pronounced as [ju:s], this word is noun and means “application, usage”. If this word is pronounced as [ju:z], this word is verb and means “to employ, to apply”. [‘prezent] present [pri’zent] If the word “present” is pronounced as [‘prezent], this word is noun and means “gift” or “real time, now”. If this word is pronounced as [pri’zent], this word is verb and means “to give, to make a gift”. The common mistake of speakers at the conferences, where English is a work language, is speaking with “raise” intonation at the end of the statement. It’s a rude mistake, because it makes the utterance hard to understand. The hearer may interpret this sentence as the interrogative or unconcluded sentence. But in fact, this sentence is concluded and should be pronounced with “fall” intonation. For example: “Skidded logs were recorded by diameter class.” (Journal of Forestry. August 2000.) If this sentence is pronounced with “raise” intonation, it is an unconcluded sentence. The hearer waits for continuing and thinks that logs, in this case, can be recorded not only by diameter class, but also by length class, sort class and something else. The constructions of sentences, especially complex sentences, are diverse in different languages. To translate complex sentences many people use the method, called “loan translation”. This method implies literal replacement words and expressions in narrative language by words and expressions in foreign language. This method is not suitable in this case, because such constructions may be difficult to understand for representatives from other countries. So, while translating complex sentences, it’s better to use stable English constructions or to divide this complex sentence into two or several more simple sentences. For example:


33 “Some recent initiatives that support improvements include the programme on forests for the Environmental Action Plan of the New Partnership for Africa’s Development (NEPAD), which, for instance, sets the key criteria for selecting projects to be funded under NEPAD through the Global Environment Facility/United Nations Environment Programme (GEF/UNEP) to include regional scope, capacity building, multifocus and programmatic approach.” (XII World Forestry Congress: Area A – Forests for people, 2003) This sentence is very long and full of organization and programme names. It’s hard to understand, especially during oral communication, and should be divided into two more simple sentences. “Some recent initiatives that support improvements include the programme on forests for the Environmental Action Plan of the New Partnership for Africa’s Development (NEPAD). This programme sets the key criteria for selecting projects to be funded under NEPAD through the Global Environment Facility/United Nations Environment Programme (GEF/UNEP) to include regional scope, capacity building, multifocus and programmatic approach.” The third cause is that the foreign dictionary of forestry and forest industries gives several possible versions of translation for one term. It’s difficult to remember all versions and seem to be unnecessary. That’s why the specialist chooses only one version and learns this word or word combination. However, if he hears another version of translation, he doesn’t recognize this term. Consequently, he doesn’t clear understand the meaning of the whole sentence. And if that term is often used in the paper, the whole report is hard to understand. For example: “forest regeneration, forest renewal, forest reproduction” “understory trees, underbrush, undergrowth, young growth” “bucking, cross cutting, shortening of tree length” “block, chuck, chump, piece of log, shock bolts” “secondary road, spar road, spur road, spur” These words in their groups have the same meaning. CONCLUSIONS To make the process of communication productive and effective, to help specialists to acquire new information and disseminate their useful knowledge, the following aspects should be taken into consideration: -

to know the basic terms of all ecological and forest majors in native language and in English. to learn not only terminology, but Grammatik and Phonetic of foreign language; to try thinking in English, it can help to build sentences in the right way. to have a notion about all possible versions of translation the basic forest, technical and ecological terms.


34

ACKNOWLEDGEMENTS Thanks are due to Valentin A. Makuev, doctor of technical science, lecturer of MSFU, dean of the part-time education faculty. Researchers and discussions of the main points of this paper will be made at the ICECFOP1 - 1st International Conference on Environmentally-Compatible Forest Products in Portugal on the base of the paper of Valentin A. Makuev, called “Choice and Formation of Harvester Park for Russian Logging Enterprises, Taking into Consideration Ecological Requirements”. REFERENCES MOZHAEV DV, NOVIKOV BN and RIBAKOV DM (1998): English-Russian and Russian- English Dictionary of Forestry and Forest Industries. It contains nearly 50,000 terms. It’s the 2nd issue, full. RUSSO, Moscow. FILIPCHUK AN (2002): Terminological Vocabulary of Forestry. VNIILM, Moscow. SKOROHODKO EF (1963): Questions of the Translation of English Technical Literature. KGU, Kiew. BORISOVA LI (1976): Questions of the Evaluation of Translations Quality in Modern Translation Science. Translation of the Science and Technical Literature. Moscow. LYNCH DL, ROMME WH and FLOYD ML (2000): Forest Restoration in Southwestern Ponderosa Pine: Journal of Forestry, August: 17-23.


35

THE CORKMARK PROCESS NATURAL IS GOOD BUT NOT ENOUGH Maria Carolina Varela1, Diogo Holstein Campilho2 and Manuel Holstein Campilho3 1 - Forest engineer, Internat. coord. of the “Quercus suber Network” of FAO/Silva Madetirranea Estação Florestal Nacional, Quinta do Marquês 2780-159 Oeiras. Portugal [email protected] 2 - Enologue and wine producer. Quinta da Lagoalva de Cima, 2090-222 Alpiarça, Portugal [email protected] 3 - Agronomic engineer and forest owner Quinta da Lagoalva de Cima, 2090-222 Alpiarça, Portugal [email protected]

SUMMARY The advent of closures made out of synthetic products was the stimulus of the process for creation of CORKMARK. Stoppers are the backbone of the cork oak chain. However the stopper is poorly noticeable for the majority of the wine consumers. In the absence of external sign at the wine bottles consumers are not able to exercise positive discrimination upon the type of the material the stopper is made. The CORKMARK logo was developed to respect the aesthetical harmony of wine bottles labels (figure in the text). Cork is the only natural product widely used for sealing wine and spirits. Cork is the outbark of the tree Quercus suber L. (cork oak), a species confined to the western part of the Mediterranean basin. Beyond being a natural product cork is one of the best examples of sustainable forest production. It selfregenerates after peeling-off and the process goes on during decades without injuring the trees. A considerable part of cork is a biologic product since it is produced without any contacts with biocides and other environment detrimental products. The process of the CORKMARK of was headed by FAO/Silva Mediterranea’s network, “Silviculture of Quercus suber”, under the responsibility of the international coordinator of the network. The major aspects of the CORKMARK process are presented: Objectives, technical and legal aspects related with the creation and use of a trade registered mark are synthetically presented, as well as the involvement of countries and institutions. CORKMARK is registered trade mark in European Union, United States of America, Switzerland, Australia and New Zealand. The need to promote cork as a wine sealing material on the basis of its natural profile is discussed having into consideration the need to improve the performances of the cork stopper. Enhancement of synergistic research among cork and wine sectors is discussed, as well as the role of research on cork promotion. The added value of CORKMARK for cork production and wine industry is discussed in the perspective that “Natural” is a crucial attraction for consumers but inefficient, even harmful for cork defence if mechanical defects or products able to induce organoletic disturbances on wine such as TCA (trichloroanisol, or “cork taint” as it is also known) are present at the cork stoppers. ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

36

Keywords: Cork, cork oak, Quercus suber, corkmark, stopper, natural product, forest sustainable management, TCA, “cork taint”, synthetic closure.

WHAT IS CORK Glamour is on cork, the outer-bark of the tree cork oak named by Lineu Quercus suber. Cork tissue is not exclusive from cork oak. It is a frequent component of the bark of various trees such as Ulmus campestris var. suberosa and Acer suberosum or it may even form a continuous renewable layer like in Phellodendron amurense (Figure 1) and Quercus variabilis (Natividade, 1950). Yet only cork oak is able to produce a pure cork’s out-bark which properties are gentle enough to assure safe and long-lasting sealing to wine and spirits’ bottles. So unique is the cork from cork oak that the word CORK is in botanical and forestry literature by rule used with implicit reference to cork oak. Henceforward so it will be used along this text.

Figure 1a and 1b- Phellodendron amurense, the velvet tree. It also produces a renewable bark of cork. Yet the qualities are not suited for stoppers. Photo- MC Varela, Židlochovice, Czech Republic, 2004

Cork Properties Any other natural material exhibits such a set of physical and chemical properties as cork.


37 Cork performances for stoppers are due to the convergence of complete liquid-proofing, very low specific gravity, high elasticity, high torsion angle, high stability under dry conditions. In it self cork is organoletic inert. Extremely low conductivity for heat, sound and electricity turn it into an outstanding natural material for insulation. Protection against chocks of fragile objects adds on cork use.

CORK, A SUSTAINABLE FOREST PRODUCTION Cork oak has its natural range at the western part of the Mediterranean basin (figure 1) . It is a frugal species that can grow in a wide variety of soils exception for hidromorphic and soils containing active calcareous. Versatile species it can grow in close forests or under low densities of less than hundred trees per hectare, allowing agro-silvo-pastoral systems under canopies. Cork oak forests provide diversified work that drives social benefits on human-depressed zones of the Mediterranean rural world that any of its potential species substitutes may ever do. The man-managed forests of cork oak shelter wildlife that finds no habitat on pines or eucalyptus forests.

Figure 2- Cork oak natural range (Natividade, 1950)

Alike any other forest product cork production is a carbon dioxide trapping process. The renewable status of forest products is “doubled” at cork by the fact that it is a bark that self regenerates on the living tree.


38

Figure 3- Cork debarking at Portugal (photo- MC Varela, 1999)

Cork production and harvesting is under legally regulation in Portugal and other cork oak countries after the intense research developed by several authors where Vieira Natividade is the outstanding name. First debarking is only allowed when the tree reaches 70cm on circumference at breast height (CBH) and 9 years is the minimum period for debarking. Total debarking’ length is also regulated by the “debarking coefficient” a factor defined on the basis of the tree’s CBH. The length of the first debarking shall be 1,5x CBH and it might increase to 3 as maximum. Pruning and tree felling are also regulated in Portugal and other cork oak countries. Forest sustainable management in cork oak forests is guaranteed whenever the stands exploitation respects the legislation, natural regeneration is assured and no soilaggressive practices are applied. Cork stripping is a tree-friendly operation and cork improves on quality after stripping. The first cork is designated by “virgin cork”. It has high porosity and low profile qualities for cork stoppers. The second stripping induces more continuous growth and decrease on porosity. By the third stripping on cork layer turns more continuous, more homogeneous, porosity tends to be tinier and elasticity improves (Natividade, 1950). Higher survival after fire hazards happens on stripped trees upon those covered by virgin cork due to the higher continuity of cork and the narrower porosity (Sousa Pimentel, 1882; Natividade, 1950).


39

THE CORK STOPPER, A BOUNTY FROM NATURE TO WINES The cork stopper is a benchmark for wine, the primary added value of wine’s history. It liberated wine from the irretrievable oxidation processes that desolate wine producers and lovers during centuries. Indeed cork was a bounty from nature to the efforts of D. Perignon, the French monk that by the end of the XVII century, engage himself on the search of an efficient sealing material for his magnificent champagnes. The cork stopper gave wine industry the confidence and impulsion to engage on investments towards high quality and incursion on distant markets.

Figure 4 a 4b– Cork stoppers, a piece of nature at the hands of wine lovers

Twenty two billions of stoppers are world wide used around the year. If synthetic materials could substitute cork that would mean direct environmental loads coming from the abandonment of cork stands and the vanishing of the associated ecological values. It would bring also an increase of non-biodegradable products that are CO2 releasers on their processing processes.

The Natural Qualities of the Cork Stopper The cork is stopper is “from cradle to grave” entirely environmental. And it is the only raw material for stoppers that is so. Apart from being a renewable natural product from sustainable forests, cork stoppers are recyclable (for other uses than beverage stoppers) and biodegradable. No synthetic material holds in simultaneity these qualities. The industrial processing of cork is deprived of environmental aggressions providing the water from the boiling process is decontaminated from tannins and phenols before flowing out. Cork stopper processing is mainly mechanical, the use of chemical products is confined to glues on agglomerated stoppers, which are by law restricted to glues authorized for products in contact with foodstuff. ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

40

Cork is a waist-less industry. The left-over from the entire stopper processing is fully used as raw-material for other products such as agglomerated stoppers, floor boards and insulation products for houses, water-pipe, etc. Agglomerated cork is also used for fishing canes butts, for car steering wheels, etc.

THE CORKMARK PROCESS By the nineties closures made out of synthetic materials improved the performances, pointing up that cork could have alternatives after three centuries of use. Cork prestige is present at the synthetic closures commercial labels that have corkcomposed designation. Imitation of visual aspects is also part of the synthetic stoppers approach. Since the stopper is by rule concealed at the yes of the wine consumer, to allow positive discrimination upon synthetic products the “Quercus suber Network” of FAO/Silva Madetirranea put into action a process for establishment of a hallmark for the cork defence (Figure 5). The main objective of CORKMARK is to be stamped outside the wine bottles in order to allow the consumers an easy way to distinguish the stopper rawmaterial.

Figure 5- The CORKMARK logo. The maim goal of CORKMARK is to promote the sustainable use of a raw-material that is natural, recyclable, biodegradable and renewable being patent at the outside of wine bottles to give wine consumers information about the stopper’s raw-material.

The CORKMARK process was developed during two years involving FAO/Silva Mediterranea adherent countries, more than sixty state, public, private and NGO institutions related to cork forest, cork industry and wine sector. Among others wine-related organisations the following gave support to the CORKMARK process: OIV - Office Internacional de La Vigne et du Vin, AREV - Assemblée des Régions Européennes Viticoles, AEVP- Associação das Empresas de Vinho do Porto IVP- Instituto do Vinho do Porto Sociedade Agrícola da Quinta da Lagoalva de Cima, S.A SOGRAPE - Vinhos de Portugal, S.A ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

41

CORKMARK is at the moment a registered trade mark at European Union, Australia, New Zealand, Switzerland and United States The goal of CORKMARK is to promote the sustainable use of cork, a raw-material that is natural, recyclable, biodegradable and renewable. Modern powerful publicity campaigns, able to manipulated consumers attitudes, inverting or inventing absurd qualities for products (Burrows & Sanness, 1999), are driving higher use of environmentally aggressive products and threatening forests economical stability while making opportunistic use of consumer’s anxiety for forest protection. Stoppers made-up about 90% of the added value generated at the cork chain. The other cork products exist as using the “waists” of stoppers. Their cost-effective profile is absolutely insufficient to keep the economical exploitation of near 2.000.000 of hectares of cork oak forests. If cork stoppers value would vanish great part of cork oak forests could disappear freeing the land for other forest species such as pines and eucalyptus, golf courses, agriculture, even speculative urbanism.

NATURAL IS GOOD BUT NOT ENOUGH The sustainable use of cork as the preferred raw-material for sealing bottled wine is grounded on two pillars – be a natural material and be satisfying. The allure coming from the natural status of a product is extremely attractive for environment concerned consumers providing the product performance is satisfying.

Cork Promotion and Research Promotion (in tight relation with rigorous quality standards) is a key element to keep the cork stopper use. Consumers’ eagerness upon natural and environmentally-friend is so strong that more frequently preference is given even if the product’s cost is higher. The outstanding natural qualities of cork are magnetism inaccessible to any synthetic material. Targeted to wine lovers cork promotion claims elegant marketing campaigns keeping in mind that for many of wine appreciators the stopper is secondary element, concealed and intangible. Moreover, some of the synthetic stoppers imitate cork’s visual aspect. Since cork production and processing is a distant issue for the majority of wine lovers, cork campaigns are reinforced if making use of unquestionable arguments based on scientific and research results. Criteria and indicators of the sustainable forest management process that are behind cork production shall be translated into marketing “language”.


42 The cork stopper is recyclable but not re-utilisable. At cork stoppers recycling is an environmental-friend process in contrary to the great majority of synthetic products. The cork production is a carbon dioxide trapping process and wild-life sheltering. However any of these arguments if lacking solid scientific back ground may look at exigent minds like a tale. Space for speculative contra-information is also open whenever campaigns are not solid based. Life cycle analyse on the various kind of closures is essential to ground cork campaigns.

Cork Stopper Performance and TCA or Cork-taint The ecological allure of cork is insufficient if the final product disappoints. Although industrial processes at cork industry are improving at a revolutionary pace there is still much to be done. Malfunction can still be seen at some mechanical aspects of the stoppers Extraction force, liquid absorption and sealing capability are characteristic that have been focused on research (e.g.-Chatonnet, et al.1998). Breakage, cracking and excessive dust are lowperformance aspects still happening on cork stoppers (Godden, et al. 2001). Apart from further research that might still be needed, the existing knowledge allows for good achievements providing rigorous selection is done throughout the various stages of the processing. However the main focus of the current controversy on the cork stopper is the occurrence of TCA or “cork taint”. TCA is the short-cut wording for a set of products that can contaminate and spoil wine’s aromas, e.g. TCA- 2,4,6 trichloroanisole, 2,4 DCA- 2,4 dichloroanisole, PCApentachloroanisole, PCP- pentachlorophenol, TCP- 2,4,6 trichlorophenol, etc (Pollnitz, et al., 1996). TCA is the outcome product of microbial activity, a detoxification mechanism whereby fungi remove chlorophenols (Pollnitz, et al., 1996). TCA is a complex phenomenon. According to the same authors the chlorophenolic precursors of TCA in the cork stopper can have various origins. To overcome the cork stopper’s imperfections cork industry and other state laboratories shall involve themselves in exhaustive research under tight synergy, contrary to the ongoing “individual” research lines. Since fungi and other TCA-producers micro-organisms activity is virtually eradicated under certain moisture thresholds, cork and cork stoppers shall be maintained at controlled levels of humidity according to the processing stage. At the beginning of the cork stopper processing, which happens some days after the raw- cork boiling process, till the bottling the stopper shall be keep at least under 8% of moisture (preferably 6%), especially during the storage process. ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

43

Rates of occurrence of TCA Newspapers, magazines and other opinion-maker media report numbers that vary from 1% to 32%. The sensorial ability for occurrence of TCA is not straightforward objective, even among experts as can be assessed by the results of the tests run by the Advanced Wine Assessment Short Course conducted by The Australian Wine Research Institute in April 1995. The rate of 4.8% was accessed on the basis of being “considered by at least 20% of the participants to be affected by recognisable cork taint” (Pollnitz, et al., 1996). Out of the 4.8% (18 bottles) only 3 were unanimously recognized by the entire set of 15 assessors of being tainted by TCA. The example illustrate that TCA detection is not easy. The obvious incongruity of the figures appearing in newspapers claims for anonymity inquiries to wine producers and final consumers together with laboratorial scientific assessment on bottles claimed to be tainted by TCA TCA off-flavour claims a profound research focused at the aim of establishing practical protocol for its eradication. Cork-taint is traceable by analysis such as gas chromatography coupled with mass spectrometry (GC/MS) or electron capture detection. Since the methodology allows also discriminating TCA origin on wine and on the stopper (Pollnitz, et al., 1996), the wine bottles confirmed to be spoiled through the cork stopper shall be cost-damage compensated. PERSPECTIVES According the results of the IPSOS’ inquiry made during 1998 (Chatonnet, et al.1998) the wine consumer was still against cork substitutes The increasing exhilaration of consumers towards natural products gives cork a bright future providing the cork stopper’s performance is rewarding.

REFERENCES Burrows J & Sanness B (1999): A summary of “The competitive climate for wood products and paper packaging: the factors causing on environmental promotions”Geneva Timber and Forest Discussion Papers- UN Economic Commission for Europe FAO- ECE/TIM/DP/16 Chatonnet P, Labadie D & Gubbiotti M-C (1998): Etudes comparative des characteristiques de bouchons en liège et en matériaux synthétiquyes. in Revue des Œnologues, 92 Godden P, Francis L, Field J, Gishen M, Coulter A. Valente P, Hoj P & Robinson ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

44 E.(2001): Wine bottle closures: sensory properties of a Semillon wine- performance up to 20 months post-bottling. In Wine Industry Journal. Vol 16 nº 5 Sep Oct 2001 Natividade, J. (1950): SUBERICULTURA. (In Portuguese). Reediting Dir. Geral Serviços Florestais, Lisboa 1990. 387 pp Pollnitz AP, Pardon KH, Liacopoulos D, Skouroumounis GK & Sefton MA (1996): The analyses of 2,4,6 trichloroanisole and other chloroanisoles in tainted wine and corks- The Australian Wine Research Institute, Australia


45

Session 3 LCA of Forest Products


46


47

METHODOLOGICAL CONSIDERATIONS ON LCA APPLIED TO WOODEN PRODUCTS Marco Fioravanti1, Paolo Frankl2, Linda Toracca1 1

DISTAF, University of Florence, FLORENCE - ITALY 2

Ecobilancio Italia s.r.l., ROME - ITALY

SUMMARY Wood is a versatile material with a dual nature it’s both an important raw material, renewable and easily worked, and a renewable source of energy capable of replacing fossil fuels. This makes the system of wood and wood products extremely complex. The properties of each half of wood’s dual nature are both profoundly interrelated and frequently in conflict. Wood residues generated from harvesting and sawing and the plywood production can be used for many different purposes, as raw material for production of particleboard, fibreboard, pulp, etc., but also as fuel to produce energy, (as electricity or direct heat), in the production of those same process. So the two natures are profoundly interrelated, which makes for difficulties in analysis. Therefore the whole complex but unique system should be looked as a whole. LCA studies undertaken by the Authors in the past few years on window frames, packaging, pallet etc have thrown up some methodological problems in the application of this methodology to wood products in the allocation of by-products and the assessment of biological CO2. This paper will focus on some methodological considerations with regard to these two problems, in order to improve the effectiveness of LCA as applied to wood products.

INTRODUCTION Importance of wood Wood is a versatile material with a dual nature: first as an important raw material, renewable and easy to work, then as a renewable source of energy able to replace fossil fuels. The use of wood as a raw material avoids emission of CO2 (C remains fixed in wood fibres in wood products) and can replace other kind of non-renewable raw materials. Wood used as a source of energy (electricity or direct heat) is replacing fossil fuels and consequently CO2 emission (from those fossil fuel) is avoided. [11] While wood burning does lead to an emission of CO2 but a forest in active growth from which that wood is taken can, if sustainably managed, absorb that CO2. This complexity presents us with a number of opportunities. In order to exploit this resource in the right way and obtain the greatest environmental efficiency from it, wood ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

48 must be managed, therefore we first have to thoroughly investigate and analyse the whole system, with all its flows and interactions.

LCA as a tool LCA (Life Cycle Assessment) is a well-known methodology for evaluating real impacts of a process on the environment. It is used to assess the environmental efficiency of a process in order to reach a reduction of impacts like raw material and energy consumption. To properly analyse the interaction any production process has with the environment, that process should be observed at every stage, from the extraction of raw material until the end of its lifecycle and the subsequent disposal of it. Thus we can provide objective, comparable information on existing interactions between a given product and the environment, compare different products or different options. For this reason it’s an indispensable tool for the study of wood “system” and to distinguish between the advantages and disadvantages of each of its uses and to evaluate different options. We must remember LCA was evolved for the study of single products, and so studying a complex system like wood’s, presents certain problems. Therefore every stage in a LCA must be defined clearly and unambiguously. The two critical aspects with reference to wood which emerged from recent LCA on windows, pallets and packaging, carried on by Authors [3] are: the definition of the role of by-products and the allocation of biological CO2. This paper sets out to investigate these problems and to suggest a starting point for discussion.

CRITICAL ASPECTS CO2 accounting The role of forests in Carbon cycle it’s well known. Trees fix CO2 in their fibres as they grow, acting as a Carbon sink. This C stock, trapped in wood fibres, is defined as biological CO2. This Carbon sink function stops when timber is felled. The destination of the carbon depends on the final use of wood. Consequently the use of wood has strong implications for the environment and plays an important role in environmental problems. Should we account wood’s biological CO2 as a credit, or not? LCA studies carried on by Authors on crates [3] show that we should. In the following example (fig.1), impacts of different end of life scenarios have been compared, using both approaches. Results show a great difference between the two.


49

Figure 1: Comparison between different end-of-life options with different approaches. (Ecobilancio Italia)

If we don’t look on biological CO2 as a credit of the material, we might be tempted to think that the end use of wood should be for combustion purposes, thereby avoiding the use of fossil fuels and the consequent emission of CO2. Incineration has the same impacts in both approaches because the biological CO2 credit vanishes with burning. Alternatively, if we consider biological CO2 credit for wood, its recycling produces the greatest environmental efficiency by maintaining and improving the carbon pool. Even controlled landfill has a more positive impact than incineration because, since CO2 is released very slowly during wood’s decomposition, it continues to act as a carbon pool. If we assume that a piece of wood contains a certain measurable quantity of CO2 fixed in its fibre and that this CO2 will remain fixed until that piece is burned or dumped (and then decomposes) we see how much more effective other uses of wood can be. For instance if we use wood to build something durable, we set up a carbon pool, a stock of carbon that is locked into what has been built. In the same way, by converting the old into the new and diverting it from incineration or disposal, recycling used timber products assumes an important role because it contributes to the preservation of this pool. By extending the lifespan of the original material as it does, recycling contributes to the expansion of wood’s carbon pool [4,5]. While the importance of the allocation of CO2 in the impact assessment of a product is fairly clear, what isn’t so clear is how that allocation should be made. We now know that not all harvested biomass C is released into the atmosphere in the year it’s harvested, as previous IPCC (Intergovernamental Panel on Climate Change) guidelines maintained [6], but we don’t yet know how to measure and allocate that CO2. This is an important point because the existence and the estimate of a carbon pool within wood products affects all evaluation of carbon balance. In may 1998 IPCC proposed three new approaches to the problem of harvested wood products (HWP) for the CO2 accounting [7].


50 Stock change approach This method evaluates net change in carbon stocks of forest and HWP. Variation in carbon stock is allocated to the country where the forest grows. Variations in the stock of wood products are allocated to the country where products are used and disposed of. Production approach Net variations in forest and HWP are both allocated to the country of production, regardless of the import and export of wood and wooden products. Atmospheric flow approach This method means assessing all atmospheric carbon exchanges for each country, where and when they occur.

Fig.2: Schemes representing production approach, stock change approach and atmospheric flow approach. (Hashimoto et al. 2002)

Each approach leads to different results, as a consequence of the different allocation suggested. According to this new approach, some countries already started evaluating the carbon pool of their wood products, looking for the most appropriate approach for the characteristics peculiar to that country and its wood system. The results are interesting because, first, they show how IPCC guidelines approach underestimates the environmental role of wood products as a pool, while the other approaches look at a wider picture, aiding understanding of the carbon budget, and determining the best management of timber products and establishing guidelines to increase efficiency in their use and disposal.


51 Data of net exchange in the Australian wood system [8, 9], shown in table 1, underline the differences existing between the different approaches. Table 1: HWP pool Australia 1998

Australia 1998

IPCC (Mt C)

Existing pool New products (input of the pool) Disposal of products (outputs of the pool) Budget Virgin wood harvested

Production (Mt C)

Stock change (Mt C)

58.9

65.7

5.7

4.1

- 4.9

- 3.6

59.7

69.2

- 4.9

- 3.6

- 5.7

Emission due to the changes in the pool

This is only a small example that should be extended to all different type of wood systems. It focuses on the role of wood products as a carbon pool. Emphasis should be made on the importance of management of this pool by means of environmental policy.

Allocation of by-products. Wood system can be defined as a multiple-output system. A raw material is the unique input of a system that has different outputs (the principal product and different byproducts). Wood residues generated from harvesting and sawn wood and plywood production can be used for a variety of purposes. As raw material for the production of, for example, particleboard, fibreboard, pulp, etc. and also as fuel to produce energy (electricity or direct heat) for the manufacture of those or other products. [10,11,12]. This means that it’s difficult to develop an analysis of the system. We have to study it as a whole, complex but unique system. So it’s clear that wood’s two natures are closely interrelated, which is why it’s difficult to develop an analysis of the system, and why the study of this complex but unique system is better approached as a whole. Due to this complexity, the allocations of impacts in a LCA study become critical. Let’s for example take the results from an LCA on window frame carried on by the Authors [3]. Analysing this study we see that the proportion of wood mass for the window frame is only 35%, while the remaining 65% goes beyond the process, into auxiliary products. That means we need 91 kg of virgin wood for a 32 kg window (fig.4). The leftover 58 kg can either be used for energy recovery or in the production of particleboard, fibreboard, pulp, paper, etc, depending on the characteristics of original material. An


52 LCA should consider these factors when defining system boundary, flow diagram and the functional unit, otherwise results could be misleading.

Fig.3: Illustration of wood system (elaborated from Jungmeier et al.)

Fig. 4: Life cycle of wood window frame (Ecobilancio).


53

We cannot attribute the full impact of the manufacturing process to the window frame, for it represents only 35% of material input. So we should include the materials surpluses to that process, even if they’re considered waste, so as not to overestimate the environmental impact of the window frame itself. We have to find a way to account for the impact of by-products. Two possible methods emerged from the conclusion of the Cost Action E9 – LCA of Wood and Wood Products [1,2,12]: system expansion and substitution. The first method, system expansion, suggests expanding system boundary by combining the material and energy aspects of wood, using a functional unit for products and energy. In this way we modify functional unit by adding auxiliary products and energy recovery to the principal product. Using the example of the windows frame, the functional unit will be not only the window frame but also energy recovered and particles for particleboard manufacturing (n window + n kW/h + n kg particles). Impact is thus shared among all the products involved in the process. The second method suggests substitution. In the LCA of a wood product, where waste wood can be used to produce energy, this energy can replace other types of energy, for instance that derived from fossil fuel. Use of by-product as raw material or energy recovery credits the primary product for the conservation of virgin wood or fossil fuels. In this way the impact is not divided among all elements in the process, but is reduced in proportion to the credits acquired. Functional unit remain only the primary product. Whatever choice is made the system should be investigated as a whole, studying quantity and fluxes of energy and materials in order to allocate impacts in the right way to achieve realistic and correct results and indications. Clear identification of the flows within the wood system is the starting point for establishing all the potential contained within wood, and the opportunities it offers. FINAL REMARKS In this analysis we wanted to focus on particular topics. LCA is the required methodology to obtain precise, accurate answers. But if we really want to understand wood’s potential, and to use this methodology to evaluate the opportunities wood offers, if we want to understand how we can treat and use wood throughout its full lifecycle certain questions have to be clarified. It is necessary to define a unique method to allocate biological CO2 in order to evaluate clearly wood products carbon pool. Currently the role of carbon pool of wood products is grossly underestimated by the method of calculation in the IPCC Guidelines. Then we must decide how to assess by-products, because they usually represent the greater part of the process’s input, and cannot therefore simply be treated as waste. A comprehensive analysis of the flow between different elements of the system is necessary to evaluate and establish which particular uses of wood and wood residues are the most environmentally efficient, while remaining economically viable. [4]. We must ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

54 remember that without economic justification for our choices, they could remain impracticable. It’s important to establish which kind of material we are able to produce with recycled wood, which kind of market it can have. We should continue along this route, thoroughly investigating the wood “system”, looking for an appropriate method for the allocations of impacts and credits for both the products and by-products of wood, aiming to arrive at a method tailored particularly to the requirements in wood’s complex nature.

REFERENCES [1] COST Action E9 (1998): WG1-4: Life Cycle Assessment of Forestry and Forest Products. [2] COST Action E9 (1998): WG3: End of life - Recycling, Disposal and Energy Production. November 8-9, Hamburg, Germany [3] ECOBILANCIO ITALIA s.r.l.(2001): LCA on wood products: window frames, pallets and crates for fruit and vegetable packaging - Final report of the project Ecobilancio: National Consortium Rilegno. [4] HASHIMOTO S, MORIGUCHI Y (2004): Six indicators of material cycles for describing society’s metabolism: application to wood resources in Japan: Resources, Conservation and Recycling 40: 201–223 [5] HASHIMOTO S, NOSE M, OBARA T, MORIGUCHI Y (2002): Wood products: potential carbon sequestration and impact on net carbon emissions of industrialized countries: Environmental Science & Policy 5: 183–193 [6] INTERGOVERNAL PANEL ON CLIMATE CHANGE (1994): Guidelines for National Greenhouse Gas Inventories [7] INTERGOVERNAL PANEL ON CLIMATE CHANGE (1998) Evaluating Approaches for Estimating Net Emissions of Carbon Dioxide from Forest Harvesting and Wood Products: Meeting Report Dakar, Senegal [8] JAAKKO POYRY CONSULTING PTY LTD (1999): Usage and life cycle of wood products: National Carbon Accounting System Technical Report n.8. [9] JAAKKO POYRY CONSULTING PTY LTD (2000): Technical report n.24: analysis of wood product accounting options for the national carbon accounting system: The Australian Greenhouse Office. [10] JUNGEMEIER G, McDARBY F, EVALD A, HOHENTHAL C, PETERSEN A, SCHWAIGER H, ZIMMER B (2003): Energy Aspects in LCA of Forest Products – Guidelines from Cost Action E9: The International Journal of Life Cycle Assessment 8(2): 99-105


55 [11] JUNGEMEIER G, WERNER F, JARNEHAMMAR A, HOHENTHAL C, RICHTER K (2002): Allocation in LCA of wood based products. Experiences of Cost Action E9 – Part I Methodology: The International Journal of Life Cycle Assessment 7(5): 290-294 [12] JUNGEMEIER G, WERNER F, JARNEHAMMAR A, HOHENTHAL C, RICHTER K (2002): Allocation in LCA of wood based products. Experiences of Cost Action E9 – Part II Examples: The International Journal of Life Cycle Assessment 7(6): 369-375


56


57

EVALUATION OF THE ENVIRONMENTAL PERFORMANCE OF PRINTING AND WRITING PAPER USING LIFE CYCLE ASSESSMENT

Ana Cláudia Dias, Margarida Louro, Luís Arroja, Isabel Capela Department of Environment and Planning, University of Aveiro, 3810-193 Aveiro, Portugal Tel: 351 234 370 200, Fax: 351 234 429 290, E-mail: [email protected]

SUMMARY The potential environmental burdens over the whole life cycle of printing and writing paper produced in Portugal from Eucalyptus globulus were quantified, using Life Cycle Assessment (LCA) methodology. The main stages considered in the life cycle were: forest, pulp production, paper production, final disposal, energy production, chemical production and transports. The results suggest that pulp production processes have an important contribution to water emissions, resulting in a major contribution to eutrophication. Besides, it plays a major role in renewable energy consumption. Energy production in the grid, printing and writing paper production and transports contribute significantly to air emissions and to non-renewable energy consumption, and, consequently to global warming, acidification and non-renewable resource depletion. Wastepaper landfilling assumes the predominant role in photochemical oxidant formation.

INTRODUCTION Life Cycle Assessment (LCA) is a technique that aims at studying the environmental aspects and potential impacts of a product’s or process’s life cycle, starting at the raw materials extraction through to the product manufacturing, until use and final disposal stages (ISO, 1997). The pulp and paper industry has been carrying out LCA studies (e.g. Eurosac and Eurokraft, 1996; Hedenberg et al., 1997; Axel Springer Verlag et al., 1998; Rafenberg and Mayer, 1998) in order to provide a comprehensive understanding of the environmental impacts of its activity and to assist the decision making process concerning new investments. In this study, LCA methodology was applied to printing and writing (P&W) paper produced in Portugal from Eucalyptus globulus. This paper grade represents about 60% of the production of paper and board in Portugal (CELPA, 2003). The goal of this study was to identify the stages (or processes within the stages) contributing most to the environmental impacts over the whole life cycle of P&W paper. The main stages considered in the life cycle were: forest, pulp production, paper production, final disposal, energy production, chemical production and transports.


58

METHODOLOGY The LCA methodology applied in this study is in accordance with ISO (International Organization for Standardization) standards: 14040, 14041, 14042 and 14043 (ISO, 1997; ISO, 1998; ISO, 2000a; ISO2000b), which distinguish four phases in a LCA study: goal and scope definition, inventory analysis, impact assessment and interpretation. The goal shall refer the intended application and the reasons for carrying out the study. In the scope, the system under study, the system boundaries, the functional unit, and the allocation procedures, among others, must be defined. In the inventory phase, all the relevant inputs and outputs of the system are identified and quantified, involving data collection and calculation procedures. These inputs and outputs may include the use of resources (raw materials and energy) and emissions to air, water and soil. In the impact assessment phase, the results of the inventory analysis are assigned to impact categories associated with resource use, human health and ecological consequences. This phase includes the following mandatory elements: selection of impact categories, category indicators and characterisation models, classification (assignment of inventory analysis results to the impact categories) and characterisation (calculation of category indicator results). Besides these mandatory elements, the impact assessment phase may also include the following optional elements: normalisation (calculation of the magnitude of category indicator results relative to reference value(s)), grouping (sorting and possibly ranking of the impact categories), and weighting (conversion and possibly aggregation of the indicator results across impact categories). In the interpretation phase the results from the inventory analysis and the impact assessment are combined together in accordance with the defined goal and scope in order to reach conclusions and recommendations.

Goal and Scope Definition The purpose of this study was to identify and assess the potential environmental burdens associated with the life cycle of the P&W paper produced in Portugal from E. globulus and consumed in Germany. The reason to carry out this study was to determine the relative contribution of the different stages (or processes) included in the life cycle. The German market was selected because about 75% of the Portuguese P&W paper is exported to European countries (CELPA, 2003), and data concerning the use and disposal of paper in Germany were available in Tiedemann et al. (2001). The system boundaries covered the following main stages: forest, pulp production, paper production, final disposal, energy production, chemical production and transports. The processes considered in each stage are presented in Table 1. ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

59

Since besides E. globulus pulp, the P&W paper also consumes pine pulp imported from Scandinavia, the production process of this kind of pulp and the production of the pine wood were also included in the system under study. The final disposal of wastepaper in Germany includes incineration, landfilling, and recycling (after sorting). Excluded from the system boundaries were: -

the production and maintenance of capital goods (e.g., buildings, machinery); the production of ink, toner, and any other material used during paper utilisation; the production of chemicals and fuels that represent less than 1% (in mass) of the functional unit. Exceptions to this cut-off criterion are energy-intensive processes, such as production of hydrogen peroxide and sodium chlorate, which are consumed in the pulp production processes (Strömberg et al., 1997).

The functional unit, i.e., the amount of paper for which the inventory data are quantified, was defined as 1 tonne of P&W paper, with a standard weight of 80 g/m2. In this study some allocation procedures were applied following ISO recommendations (ISO, 1998). Allocation is needed in multifunctional processes, such as multi-output and -input processes, and open-loop recycling, in order to allocate the input or output flows to each product. Wherever possible, allocation was avoided by unit process disaggregation or system boundaries expansion.

INVENTORY ANALYSIS

Forest Input data for the E. globulus forest, consisting of fertiliser and fuel consumptions, are typical data of the silvicultural and harvesting practices accomplished in Portugal (Emporsil and Soporcel, 1995), and refer to operations such as path opening, plantation, soil mobilisation and fertilisation, felling, debarking, and off-road hauling. Air emissions from fuel combustion were estimated using emission factors for diesel and petrol (Habersatter, 1998). The pine forest was characterised in this study using average data covering the silvicultural and harvesting practices carried out in Finland (KCL, 1999). Pulp Production For the E. globulus pulp production, average data provided by the Portuguese pulp and paper industry were used in this study. They refer to the woodyard, wood cooking, brownstock washing and screening, ECF (elemental chlorine free) pulp bleaching using chlorine dioxide produced on site, chlorine dioxide production, recovery of the chemicals used in wood cooking, energy generation from bark and black liquor (liquor from wood cooking), wastewater treatment in an activated sludge plant and solid waste landfilling. E. globulus pulp is produced with a consistency of 3.5% dry solids, being its production integrated with the production of P&W paper.


60

Table 1: Main stages in the P&W paper life cycle and corresponding processes

Stages Forest Pulp production Paper production

Processes

•

E. globulus forest Pine forest E. globulus pulp production Pine pulp production

•

P&W paper production

•

Landfilling Incineration Sorting Recycling: - Graphic paper production (from wastepaper) - Packaging paper production (corrugated and folding box, from wastepaper) - Tissue production (from wastepaper)

• • •

• •

Final disposal

•

•

•

Energy production

• •

•

• •

Chemical production

• •

• • • • •

Transports

• • • •

Heavy fuel oil production (consumed in the E. globulus pulp production process, in the P&W paper production process, and in the tissue production process) Natural gas production (consumed in the tissue production process) Lignite production (consumed in the tissue production process) Electricity produced in the grid (consumed in the P&W paper production process, in the wastepaper sorting and recycling, and in chemical production) Thermal energy produced in the grid (consumed in wastepaper recycling) Production of chemicals consumed in the E. globulus pulp production process (NaOH, CaCO3, H2SO4, NaClO3, S2, NaCl, H2O2) Production of chemicals consumed in the pine pulp production process (NaClO3, H2O2) Production of chemicals consumed in the P&W paper production process (PCC, CaO, CaCO3, CO2, optical brightener, starch) Production of chemicals consumed in the recycling production process (H2O2) Transport of E. globulus to the pulp mills Transport of pine to the pulp mills Transport of pine pulp to the P&W paper mill Transport of P&W paper to Germany Transport of wastepaper from the user to the several disposal alternatives Transport of the chemicals consumed in the E. globulus pulp production process Transport of the chemicals consumed in the P&W paper production process Transport of the heavy fuel oil consumed in the E. globulus pulp production process Transport of the heavy fuel oil consumed in the P&W paper production process


61

Data on the production of pine pulp are typical of the Finnish technology (KCL, 1999), and include the following steps of the production process: wood cooking, brownstock washing and screening, ECF pulp bleaching with oxygen delignification and reduced chlorine dioxide consumption, chlorine dioxide production, pulp drying, recovery of the chemicals used in wood cooking, energy generation from bark and black liquor, wastewater treatment in an activated sludge plant, sludge combustion and mill condensation power plant. In the pulp production processes, allocation procedures were applied because there is a co-production of surplus electricity that is exported to the national grids. In these cases, allocation was avoided by expanding the system boundaries. Thus, the production of a similar amount of electricity in the national grids was included in the boundaries, and its environmental burdens were subtracted from the environmental burdens of these processes. This means that the surplus electricity produced by the pulp production processes avoids the production of a similar amount of electricity in the national grids Paper Production The P&W paper production process includes E. globulus pulp transfer, pine pulp pulping, pulp refining, cleaning and screening, broke recovery, paper machine, finishing, wastewater treatment in an activated sludge plant and on site energy production from heavy fuel oil. Inventory data are average data provided by the Portuguese pulp and paper industry. Final Disposal The fluxes of P&W paper in Germany were taken from Tiedemann et al. (2001). About 6% of the consumed paper remains in use in the form of archives or books. About 76% of the wastepaper goes to recycling, 17% goes to landfilling and 7% goes to incineration. Of the recycled paper, about 85% is used to produce tissue paper, 15% is used for the production of packaging paper, and 5% is integrated in the production of graphic paper. The inventory data for the recycling processes were obtained in the literature and specialised databases (FEFCO et al., 1997; Habersatter, 1998; KCL, 1999), with the exception of the recycling into graphic paper, whose data were provided by IFEU (Institut für Energie und Umweltforschung Heidelberg GmbH). IFEU also provided data for the landfilling and incineration processes, representative of the German technology. The system boundaries expansion, to avoid allocation, was also adopted in this stage to deal with surplus electricity produced by landfilling, incineration and packaging paper production, and surplus thermal energy produced by incineration. The same approach was also used to avoid allocation in paper recycling, where new paper grades are produced, beginning a new stage of the life cycle. The allocation of the environmental burdens, between the P&W paper under study and these new paper grades, was avoided by assuming that the recycling processes displace the production of the same amount of paper, with similar functions, made from virgin fibre. Thus, the production of tissue ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

62 paper and packaging paper from virgin fibre was included in the system boundaries. On the other hand, the production of graphic paper from virgin fibre was excluded because it represents less than 1% in mass of the functional unit (cut-off criterion). Energy production Energy production includes the production of fuels and the production of electricity and thermal energy in the grid. The production of fuels comprises fuel extraction and processing and was characterised using data from the literature (Habersatter, 1998). The fuels considered in this study were heavy fuel oil, natural gas and lignite. Data on the production of electricity in the grid were collected for the following production models: Portugal, Spain, France, Finland, Belgium, UCPTE (Union for the Connection of Production and Transportation of Electricity) mix and Nordic mix (Habersatter, 1998). The production of thermal energy in the German grid was taken into account using data from IFEU. Chemical production The inventory data for the production of several chemicals consumed in pulp, P&W paper and recycling production processes were retrieved from the literature (Virtanen and Nilsson, 1993; Habersatter, 1998; KCL, 1999). The production of PCC (precipitated calcium carbonate) is an exception, because it takes place in the paper mills and the inventory data were directly provided by the producers. Transports The transport of wood, pulp, P&W paper, wastepaper, chemicals and fuels, by truck, ship and electric train, was considered using distances provided by the pulp and paper industry. The consumption of fuel and the corresponding emissions of each kind of transport were obtained from the literature (Habersatter, 1998). IMPACT ASSESSMENT

The impact categories considered in this study were: global warming for a time horizon of 100 years (GW), acidification (A), eutrophication (E), non-renewable resource depletion (NRRD) and photochemical oxidant formation (POF). The selection of these categories was conditioned by the parameters available from the inventory analysis. The inventory parameters assigned to each impact category and the sources of the characterisation factors that allow the aggregation of the parameters within each category, are shown in Table 2.


63 Table 2: Parameters assigned to the impact categories and sources of characterisation factors

Impact category Global warming (GW) Acidification (A) Eutrophication (E) Non-renewable resource depletion (NRRD) Photochemical oxidant formation (POF)

Parameters Non-renewable CO2, CH4, N2O SO2, NOx, HCl, NH3, HF, H2S NOx air, NH3 air, N water, NO3water, NH4+ water, P water, PO43-

Source of characterisation factors IPCC (2001) Hauschild and Wenzel (1998) Lindfors et al. (1995)

Crude oil, natural gas, coal

Lindfors et al. (1995)

CH4, halogenated hydrocarbons, aromatic hydrocarbons

Heijungs et al. (1992)

The optional elements of the impact assessment (normalisation, grouping and weighting) were excluded from this study because they are unnecessary for achieving the goal of this study. RESULTS Inventory Analysis Results Although the inventory results consist of an exhaustive list of parameters, only the most important ones were selected to be presented and assessed: renewable energy consumption, non-renewable energy consumption, non-renewable carbon dioxide (CO2), nitrogen oxides (NOx), sulphur dioxide (SO2), chemical oxygen demand (COD) and adsorbable organic halogens (AOX). Renewable CO2 emissions were excluded from this analysis, because they were assumed to be in balance with CO2 sequestration in the forest stage. Figure 1 shows the relative contribution of the several stages comprised in the life cycle to the inventory parameters. The contribution of the processes added to the system boundaries in order to avoid allocation was considered separately in the category called “credits”.


64

Forest

Pulp production

Paper production

Final disposal

Energy production

Chemical production

Transports

Credits

100 80

Contribution (%)

60 40 20 0 -20 -40 -60 -80

Ren. energy

Non-ren. energy

Non-ren. CO2

SO2

NOx

COD

AOX

Inventory parameters

Figure 1: Relative contribution of the stages considered in the life cycle of P&W paper to the inventory parameters

The pulp production stage consumes about 95% of the total renewable energy. This is an expected result because all the energy consumed in the pulp production processes are of renewable origin. The E. globulus pulp production process has a predominant role to this contribution because it represents about 80% of the pulp consumed in P&W paper. The credits given to recycling contributes to a reduction of almost 65% in the consumption of renewable energy in the whole system. This means that the recycling of P&W paper avoids the consumption of renewable energy, in the production of paper from virgin paper, equal to about 65% of the consumption of renewable energy in the whole system. About 35% of the consumption of non-renewable energy takes place in the production of energy, of which about 40% and almost 30% correspond to the non-renewable component of the electricity produced in the grid consumed in the P&W paper and in the tissue production processes, respectively. P&W paper production and transports are also important consumers of non-renewable energy, with a contribution to this parameter of about 20% each. In P&W paper production this is due to the use of fuel oil to produce on-site energy, while in transports about 65% of the contribution is associated with the transport of P&W paper and its distribution in Germany. A reduction of more than 30% in the total consumption of non-renewable energy is obtained considering the credits given to the system, mainly to recycling. The emissions of non-renewable CO2 have a profile similar to that of consumption of non-renewable energy. The main source of SO2 emissions is energy production, contributing to 40% of the total SO2 emissions. About 45% and 25% of these emissions are released in the production, in the grid, of the electricity consumed in the P&W paper and in the tissue production processes. The paper production stage has also a remarkable contribution to SO2 emissions (20% of the total SO2 emissions) due to on-site energy production. The ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

65 credits given to the system allow a reduction in the total SO2 emissions of about 25%, mainly due to the recycling processes. Approximately 50% of the NOx emissions are generated by transports, and among these, by the transport of P&W paper and its distribution in Germany. A reduction of more than 20% in NOx emissions is achieved due to the credits given, mainly to recycling. About 60% of the total COD is emitted in the pulp production stage, mainly in the E. globulus production process. The second most important contribution to this parameter, representing 30% of the total emissions, is that of paper production. Total COD emissions are reduced by almost 40% due to the credits given to recycling. Another important parameter in the pulp and paper industry is AOX, which is mostly generated in the pulp production stage due to the consumption of chlorine dioxide in pulp bleaching. The E. globulus pulp production process contributes to about 80% of these emissions. The credits given to the system, mainly those given to recycling, reduce AOX total emissions in approximately 55%.

IMPACT ASSESSMENT RESULTS The relative contribution of the different stages of the life cycle of P&W paper to the impact assessment categories considered in this study are shown in Figure 2. The category of “credits” accounts for the contribution of the processes added to the system boundaries in order to avoid allocation.

Forest

Pulp production

Paper production

Final disposal

Energy production

Chemical production

Transports

Credits

100 80

Contribution (%)

60 40 20 0 -20 -40

GW

A

E

NRRD

POF

Impact categories

Figure 2: Relative contribution of the stages considered in the life cycle of P&W paper to the impact categories


66 The largest contribution to global warming comes from the energy production (30% of the total global warming potential), mainly as a result of its emissions of non-renewable CO2. Final disposal, paper production and transports play also an important role to this impact category (approximately 20% of the total global warming potential, each one) due to their emissions of non-renewable CO2. In final disposal, methane emissions from paper landfilling are also of great importance. A reduction of about 30% in the total global warming potential is provided by the credits given to the system, mainly to recycling. Energy production is the most important stage regarding acidification, contributing to 30% of the total acidification potential, mainly due to SO2 emissions. Transports are important contributors as well, causing 20% of the total acidification potential, mainly as a result of its NOx emissions. The total acidification potential decreases about 25%, mostly as a result of the SO2 emissions “avoided” by paper recycling. The pulp production stage is responsible for 35% of the eutrophication potential, mainly generated by its COD and phosphorus emissions. Transports are also relevant contributors to this potential (almost 30% of the total) due to its NOx emissions. The credits given to the system, mainly those given to recycling, allow a reduction of the eutrophication potential in approximately 25%. NOx and COD emissions are of major importance to this reduction. Almost 30% of the non-renewable resource depletion potential is caused by the paper production stage. This contribution comes from the consumption of heavy fuel oil for on-site energy production. Energy production and transports contribute to 25% of the total non-renewable resource depletion potential. In the energy production stage, consumption of fuel oil and coal to produce electricity is predominant, while in the transport stage, consumption of fuel oil and diesel is the main cause of this contribution. The credits given to the system lead to a reduction of about 25% in the non-renewable resources depletion potential of the whole life cycle. Final disposal of wastepaper contributes to about 80% of the overall photochemical oxidant formation potential due to methane emissions from landfilling. In this case, the credits given to the system result in a small reduction (less than 10%) of the total photochemical oxidant formation potential.

CONCLUSIONS The following conclusions can be drawn from the analysis of the results obtained in this study: - the pulp production stage has an important contribution to water emissions, resulting in a major contribution to eutrophication. Besides, it is the major consumer of renewable energy; -

the paper production stage is the dominant contributor to the non-renewable resource depletion potential. It is also of great importance to air emissions (nonrenewable CO2 and SO2), to non-renewable energy consumption and to global warming;


67

-

although the final disposal stage has no relevant contribution to the analysed inventory parameters, it assumes the predominant role in photochemical oxidant formation and a significant role in global warming, due to its methane emissions;

-

the energy production stage contributes significantly to air emissions (nonrenewable CO2 and SO2), to non-renewable energy consumption, and, consequently, to global warming, acidification and non-renewable resource depletion. It is important to note that this contribution comes mainly from the production of electricity in the grid consumed in the P&W and tissue production processes;

-

transports are the major source of NOx, resulting in important contributions to acidification and eutrophication. In addition, they are of great importance to non-renewable CO2 and non-renewable energy consumption, and to the impact categories that consider these parameters: global warming and non-renewable resource depletion;

-

the forest and chemical production stages don’t contribute significantly to any of the inventory parameters and impact categories analysed;

-

the credits given to the system to avoid allocation lead to reductions in the values obtained for the inventory parameters and the impact categories that vary between almost 10% and 65%.

ACKNOWLEDGEMENTS The authors are grateful to the FCT (Science and Technology Foundation - Portugal) and to the program PRAXIS XXI for the financial support provided to the project “Life Cycle Assessment (LCA) - Environmental Impact Assessment of the Activity: From Eucalypt to Paper” (3/3.2/PAPEL/2323/95). The authors would also like to thank Portucel, Soporcel and Raiz for their collaboration in providing first hand information and inventory data. Finally, the authors would like to express their gratitude to IFEU, in the persons of Mr. Jürgen Giegrich and Mr. Andreas Detzel for helping with the data from Germany REFERENCES AXEL SPRINGER VERLAG, STORA and CANFOR: A Life Cycle Assessment of the Production of a Daily Newspaper and a Weekly Magazine. Axel Springer Verlag, Stora and Canfor, 1998 CELPA: Boletim Estatístico 2002. Associação da Indústria Papeleira, Lisboa, 2003 EMPORSIL and SOPORCEL: Manual Técnico: Instalação e Condução de Povoamentos Florestais com Eucalipto. Emporsil and Soporcel, 1995


68 EUROSAC and EUROKRAFT: LCA of Distribution in Paper Sacks. Eurosac and Eurokraft, 1996 FEFCO, GROUPEMENT ONDULÉ and KRAFT INSTITUTE: European Database for Corrugated Board Life Cycle Studies. FEFCO, Groupement Ondulé and Kraft Institute, 1997 HABERSATTER K: Inventaires Écologiques Relatifs aux Emballages. Swiss Federal Office of Environment, Forests and Landscape, Berne, 1998 HAUSCHILD M and WENZEL H: Environmental Assessment of Products, Volume 2: Scientific Background. Chapman & Hall, London, 1998 HEDENBERG Ö, JACOBSON BB, PAJULA T, PERSON L and WESSMAN H: Use of Agro Fiber for Paper Production from an Environmental Point of View. NORDPAP DP 2/54, SCAN FORSK-RAPPORT 682, 1997 HEIJUNGS R, GUINÉE JB, HUPPES G, LANKREIJER RM, UDO DE HAES HA, WEGENER SLEESWIJK A, ANSEMS AMM, EGGELS PG, VAN DUIN R and DE GOEDE HP: Environmental Life Cycle Assessment of Products, Guide and Backgrounds. CML, Leiden University, Leiden, 1992 IPCC: Climate Change 2001: The Scientific Basis. Cambridge Press, Cambridge, 2001 ISO: Environmental Management - Life Cycle Assessment - Principles and Framework, ISO 14040. International Organization for Standardization, Geneva, 1997 ISO: Environmental Management - Life Cycle Assessment - Goal and Scope Definition and Inventory Analysis, ISO 14041. International Organization for Standardization, Geneva, 1998 ISO: Environmental Management - Life Cycle Assessment - Life Cycle Impact Assessment, ISO 14042. International Organization for Standardization, Geneva, 2000a ISO: Environmental Management - Life Cycle Assessment - Life Cycle Interpretation, ISO 14043. International Organization for Standardization, Geneva, 2000b KCL: KCL-EcoData. Registered trademarks of the Finnish Pulp and Paper Research Institute, Espoo, 1999 LINDFORS L-G, CHRISTIANSEN K, HOFFMAN L, VIRTANEN Y, JUNTILLA V, HANSSEN O-J, RØNNING A, EKVALL T and FINNVEDEN G: Nordic Guidelines on Life-Cycle Assessment, Nord 1995:20. Nordic Council of Ministers, Copenhagen, 1995 RAFENBERG C and MAYER E (1998): Life Cycle Analysis of the Newspaper Le Monde: Int J LCA 3(3): 131-144


69 STRÖMBERG L, HAGLIND I, JACOBSON B, EKVALL T, ERIKSSON E, KÄRNÄ A and PAJULA T: Guidelines on Life Cycle Inventory Analysis of Pulp and Paper. NORDPAP DP 2/30, SCAN FORSK-RAPPORT 669, 1997 TIEDEMANN A, TIEDEMANN CB, BUSCHARDT A, GEORGI B, GOOSMANN G, GREGOR H-D, MEHLHORN B, MODI A, NEITZEL H, OELS H-J, SCHMITZ S and SUHR M: Life Cycle Assessment for Graphic Papers. Federal Environmental Agency, Berlin, 2001 VIRTANEN, Y and NILSSON, S: Environmental Impacts of Waste Paper Recycling. International Institute for Applied Systems Analysis, London, 1993


70


71

ECOLOGICAL ASPECTS OF FOREST ROADS NETWORK DEVELOPMENT Vladimir V. Nikitin1 1

Forest Transportation Department - Forest Faculty - Moscow State Forest University 141005 Mytistchy - Moscow – RUSSIA ([email protected])

SUMMARY Forests network are to have the certain density. Also forest network development plays an important ecological role. As a result of our investigations, mathematical model of forest roads development was suggested. The basic kinds of impact of roads to the forests were considered. They are taken into account in the model now as limitations

INTRODUCTION Nowadays in Russia public institutions are responsible for forest care and harvesting. These institutions are called forestries. A forest fund is assigned to a forestry which builds a road network in its forest necessary for fulfilling the functions of the forestry. Seeds, seedlings, saplings, fertilizers and other goods required in the forest, and also the harvested timber are transported by the roads. The total number of the types of goods transported by forest motor roads covers more than 100 types. MATERIALS AND METHODS The main principles of designing forest road networks were originated in the 19601970s. According to the principles forest road network should meet the following requirements: a) They should provide permanent access to every forest quarter; b) They should cover the whole area of the forest land, including areas with small forest density per hectare and non-forested areas subject to afforestation; c) Their average freight turnover per kilometre should be small; d) The road pavement should be strong enough while the cost of the roads should be minimal because of small freight turnover; e) The cutover area for the right-of-way of the road should be minimal; f) The road network and installations should be completely combined with the water conservation network of the forest land; g) The road network should be combined with fire breaks; h) Roads in forest parks and recreation forests should harmonize with natural landscape to emphasize and reveal the beauty of the forest; ICECFOP1 – 1st International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

72 i) It should be possible to improve the road pavement step by step if the traffic is more intensive. Obviously, the mentioned principles don’t take into account environmental aspects of designing a forest road network. This is confirmed by the mathematical model of developing a forest road network. It supposes such development of road network which will provide the minimum of the function

=

L

∑ ∑ (R D

i = 1

3

+

( k )

P

( k )

q

k = 1

i

)l

( k ) i

⎯⎯ →

min (3.1)

where: R(k) - the cost of building 1 kilometre of a road of the type k; P(k) - the cost price of a transportation unit by the road of the type k in rubles per a thousand of kilometres; qi , i=1,2,…D, - the mass of forest goods transported to the shipping station № i; l(ik) - the length of the section of the road of the type k, while transporting forest goods from the shipping station № i:

l

( k ) i

=

M

∑ m

l mrs =

− 1

l = 1

rs m

(xr − xs )2 + ( yr − ys )2 ,

r, s∈ M

(3.2)

where xr , yr , xs , ys are position data of a road junction № r and s belonging to the ensemble M, between which a section lm(rs) is situated. The optimum alternative of developing a road network on the basis of the model was chosen in the following way. The value of traffic intensity was calculated for each mth road section between points r and s:

N

m

⎛ = f ⎜ ⎝

D

∑

i =1

⎞ qi ⎟ ⎠

If the value Nm was more than the normative value for a road of the given type k, this was the ground for restructuring the road into type (k+1). Different alternatives of transportation routes were simulated at different stages of the iterative process. Therefore the value of the function I was changed. The minimum of I was chosen according to one of the gradient method.


73 The model we have supposed before does not meet the modern requirements of the environmental protection in the forest. It does not take into account environmental parameters and is therefore out of date. One of the types of the anthropogenic impact on the forest is building and maintaining roads. There are various forms and kinds of the impact (see table 1). To take into account environmental aspects of designing a forest road network we are going to employ a new environmental concept of applying a mathematical model mentioned above. The concept includes three stages. Stage 1. Especially valuable areas in the forestland are distinguished according to the criterion of the availability of the following elements within the limits of the area. -

Valuable landscape elements; Key biotopes; Flora and fauna diversity; Natural territories under state protection.

Stage 2. Areas with damaged and/or not regenerated plant and humus cover, which have lost their initial value, such as existing roads, dams, embankments and pipelines, are distinguished within each valuable landscape element distinguished at the first stage. Stage 3. Quantitative assessment of the level of landscape disturbance is carried out in the ratio of the disturbed area Fh to the total area of the plot Fw

K

h

=

F F

h w

Such assessment is carried out in the following ways: • • • •

For each landscape type; For each damage kind within the limits of the plot; Totally for each damage kind within the limits of the forestland; Totally for the whole area according to damage kinds.

The landscape disturbance degree can be assessed by means of coefficient Kh. Assessing the landscape disturbance degree is a launching pad for solving a more important problem, that is designing the landscape condition after the anthropogenic impact. In our research forest road construction is the impact. The more disturbed is the landscape, the more subject it is to the further damage during further recovery work. Changes in the flora and fauna diversity while developing a road network in the forestland are another object of designing. For instance traffic noise can impact on the behavior of wild animals by frightening them away from pastures and ponds and breaking their usual travel paths. Nowadays these elements of the common model of developing a road network in the forestland are being worked out.


74

Table 1. The character of environmental impact while developing a road network in the forest Forms and kinds of the road impact on the forest Stages of building and Damage factors functioning of a road Form Kind • Exploration work, building

Tracks of carriers and road-building machinery • Well-boring • Construction of temporary settlements •

Trenching, digging

•

Chemical soil, atmosphere and ground water pollution

Road maintenance

•

Mechanical Thermal

Long-lasting physical-chemical, chemical, thermal

• Explosions

• Fires

Soil cover

•

Watering

•

Traffic noise

Emergency situations

•

Short-term mechanical, physical-chemical, chemical

• Repair work

Hydrology change

•

Dust loading

•

Deflation

• • •

Underflooding Swamping Breaking travel paths of wild animals • Mechanical landscape disturbances • Soil, atmosphere and ground water pollution • Secondary landscape disturbances during recovery work

Table 2. Landscape disturbance classification. Category of landscape disturbance

Disturbance degree Kh

Almost not disturbed ecosystem Low-disturbed ecosystem Significantly disturbed ecosystem Badly disturbed ecosystem Totally destroyed ecosystem

KhKh>0,1 0,5>Kh>0,3 0,8>Kh>0,5 Kh>0,8


75

CONCLUSIONS We suppose we shall be able to work out draft designs of developing road networks for forest enterprises at the end of this year. The basic mathematical model in which the environmental aspects mentioned above will be taken into account as limitations will be used for this. The point is to exclude the alternatives of developing a road network causing further disturbance of the disturbed forest landscapes or reducing the flora and fauna diversity while searching for a minimum of the function in spite of the fact that such alternatives can be economically favorable.

REFERENCES ILJIN B.A.: Forest Roads. Leningrad Forest Technical Academy, Leningrad, 1980.


76


77

Session 4 Treatment of Effluents with Forest Residues


78


79

APPLICATION OF PINE BARK AS A SORBENT FOR ORGANIC POLLUTANTS IN EFFLUENTS Brás, I.1,2 *, Lemos, L.2, Alves, A.1 and Pereira, M.F.R.3 1

2

LEPAE - Laboratório de Engenharia de Processos, Ambiente e Energia, Departamento de Engenharia Química, Faculdade de Engenharia da Universidade do Porto, Rua Roberto Frias,s/n, 4200-465 Porto, Portugal

Departamento de Engenharia do Ambiente, Escola Superior de Tecnologia de Viseu, Instituto Politécnico de Viseu, Portugal

3

LCM– Laboratório de Catálise e Materiais, DEQ, Faculdade de Engenharia da Universidade do Porto, Rua Dr. Roberto Frias, 4200-465 Porto, Portugal *Corresponding author: Tel: +351 232 480 500 Fax: +351 232 424 651 email: [email protected]

SUMMARY Nowadays trace organic pollutants represent a major concern in water treatment systems. Activated carbon has been used for most applications aiming at the reduction of these kind of compounds in aqueous effluents, but regeneration needs and high operation costs led to a renewed interest in the search for alternative sorbents. Pine bark is an excedentary raw material from sawmills in Portugal, and therefore a profitable natural resource that has already been successfully tested in the adsorption of organochlorines from contaminated waters. This study aims at characterizing structurally and chemically the pine bark surface, in order to understand the nature of sorption occurring when a trace organic contaminant is present in aqueous effluents. Pentachlorophenol (PCP) was the trace contaminant used in the experiments. Scanning electron microscopy (SEM), mercury porosimetry, Fourier transform infrared spectroscopy (FTIR) and X-ray photoelectron spectroscopy (XPS) were the techniques used, in addition to classical chemical analysis and Solid Phase Micro Extraction (SPME) prior to Gas Chromatography with Mass Spectrometry (GC-MS) for PCP quantification. The pine bark proved to be a material with a very low porosity, low specific surface area, with a strong carbon aromatic content probably related with polyphenols and lignin composition. Sorption experiments showed a good correlation for the linear adsorption isotherm, as well as the desorption experiments. In the conditions tested, the average PCP removal after 24 hours was above 98%. This material proved to be an encouraging sorbent for cheap water remediation solutions.

INTRODUCTION The utilization of wood by men generates wastes such as the tree barks, which are an important, inexpensive and rather abundant resource. The bark represents 10-15% of the total weight of the tree (Kofujita et al., 1999) and its use as fuel is the most overspread utilization, requiring only the dryness to take advantage of the bark heating value. The tannins content with phenolic nature has been studied as a source of resins when ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

80 reacting with formaldehyde, enhancing its application in panel production with less requirement of this reactant. This tannin source has also been applied in the leather tannery industry (Jorge et al., 2001, Fradinho et al., 2002). Jorge et al. (2001) point out some other bark applications in the pharmaceutical industry, as biocide in agriculture due precisely to the bark tannin content, and as ionic exchange resins making use of pine bark surface properties. The wood and bark chemistry is different bearing the wood higher contents in carbohydrates and lower percentages of lignin and extractives. Bark also has in its composition polyphenols, known as tannins, and suberin that are not found in wood (Fengel and Wegener, 1984). Within barks of several species, the chemical composition changes in terms of ashes, lignin, polyphenols, extractives and carbohydrates contents for hardwood and softwood (Harun and Labosky, 1985; Kofujita et al., 1999). Fradinho et al. (2002) reported that the Pinus pinaster bark composition, in weight percent, is represented by 33% lignin, 11% polyphenolics, 39% polysaccharides, 17% extractives and less than 1% of ashes. The specificity of the pine bark chemistry encouraged several authors to evaluate its effectiveness in wastewater treatment applications namely for the sorption of heavy metals (Randall, 1977, Alves et al., 1993, Al-Asheh and Duvnjak, 1999, Vázquez et al., 2002, Santos et al., 2003) and organic pollutants (Brás et al., 1999, Haussard et al., 2003, Ratola et al., 2003) using a local, inexpensive and abundant resource, as referred above. The organic pollutants, specifically those having high toxicity and persistence in the environment, as most of the pesticides, are usually not efficiently removed from water at the range of small concentrations that they emerge and therefore become spread in the environment. To override this situation and control the fate and toxicity of these organic pollutants becomes necessary to apply adsorption processes with activated carbon or, if possible, with alternative materials to overcome the cost associated with the former conventional method. The sorption process can be attributed to the distribution of the organic compound between two phases, by accumulation in the solid phase active sites (adsorption) or by partitioning (Huang et al. 2003). Therefore, the study of sorption and desorption equilibrium is an important tool to quantify the extent of pollutant retention by the sorbent material, being the distribution coefficient the parameter that endorse such affinity. In systems where the distribution coefficient attained in the sorption ( K ds ) and in desorption ( K dd ) are different, the pollutant is considered to be, in some extent, immobilized in the solid matrix, being this immobilization stronger higher the divergence between K ds and K dd is. In this class of compounds we can find pentachlorophenol (PCP) used in the wood preservation industry and included in the large family of pesticides (Crosby, 1981). PCP is a hydrophobic ionisable organic compound (HIOC) once it can be found in the environment as protonated or anionic specie according to the surrounding ph, with a pka of 4.75. In its neutral form it is strongly hydrophobic (log kow 5.24) with low aqueous solubility (14 mg/l at 20ºc) and behave likewise the hydrophobic non-ionisable organic compounds (hoc) partitioning in the organic matter of the soil and organoclays by hydrophobic bonding (Boyd et al., 1988, Divicenzo and Sparks, 1997).

ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

81 The main objective of this study is to evaluate the physical and chemical properties of pine bark and its interactions with PCP, assessing the extent of PCP binding in the pine bark surface through sorption and desorption equilibrium studies.

MATERIALS AND METHODS

Pine bark preparation and characterization techniques The pine bark was collected in a sawmill in the north of Portugal. After grinding in a Reischt mill and sieved in Endecotts EFL 2000/1 automatic siever, the 150-450 µm size of bark was dried at 105ºC±2ºC for 48 hours in a Binder muffle. The textural characterization was performed by mercury porosimetry with a Poremaster-60 Quantachrome apparatus. The pine bark characterization was also done by proximate analysis with a Mettler TA 4000 thermal analyzer and by elemental analysis with a CARLO ERBA 1108 Elemental Analyser. The determination of surface charge was performed by pH drift tests described elsewhere (Rivera-Utrilla et al., 2001) and the basicity and acidity surface determination was previous reported elsewhere (Pereira et al., 2003). The pine bark DRIFT spectra were carried out in a Nicolet 510P FT-IR spectrometer. The spectra were recorded at 4 cm-1 resolution and 256 scans were accumulated prior to Fourier transformation and expressed in transmittance in the 4000450 cm-1 range. The elements at the surface were analysed by X-ray photoelectron spectroscopy (XPS) with a VG Scientific ESCALAB 200A spectrometer and the scanning electron microscopy (SEM) was performed by JEOL JSM-6301F microscope after sample preparation by pulverization and vaporization of carbon and gold in a JEOL JFC 100 apparatus at CEMUP (Centro de Materiais da Universidade do Porto).

Reagents Pentachlorophenol (PCP) was obtained from Supelco (cat n-4-8555). Sodium sulphate anhydrous p.a. and sulphuric acid 95-97% p.a. were purchased to Merck. The aqueous PCP solutions of various concentrations were prepared from an intermediate 5 mg/l PCP solution made from a 2440 mg/l stock solution in 0.1 m NaOH (Merck) and distilled and deionised water. The standards preparation for solid-phase microextraction calibration was done by dilution of the intermediate stock solution with distilled and deionised water at pH 2 saturated with Na2SO4. The ph adjustment was done with sulphuric acid 5 N.

Sorption studies The equilibrium sorption and desorption experiments of PCP on pine bark were conducted in closed glass 15 mL tubes in a shaker at 25 ºC. The required equilibrium time of 24 hours was previously determined in kinetic studies with solutions of 1 mg/L PCP in a system with solid:solution ratio of 1:100 (w:w). In each tube, 100 mg of pine bark and 5 mL of PCP solutions with concentrations between 0.03 and 5 mg/L at pH 2 ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

82 were added. After reaching equilibrium, the pine bark was allowed to settle at the bottom of the tubes and the maximum amount of solution was removed from each reaction vessel for PCP analysis and replaced by 5 mL of PCP free solution. The glass tubes were placed again in the shaker at 25ºC. After 24, hours the solutions were once more removed for analysis of PCP by solid-phase microextraction (SPME).

PCP analysis For PCP quantification, an 85-µm PA fiber (Supelco, Cat. Nº PN 57304) and the respective SPME sampling manual holder (Supelco, Cat. Nº 57330-U) were employed. The fiber was previously conditioned at 300ºC for 2 hours in the gas chromatograph injection port. For the extraction of PCP, 2 mL of the standard solution or sample were measured into a 4 mL amber vial and the fiber was immersed in the solution for 30 minutes, at room temperature (25±2ºC), with rapid and constant stirring. After this period, the fiber was removed and desorbed in the GC injection port for 3 minutes. The quantification of PCP was attained with an Agilent 6890 Series gas chromatograph (GC) equipped with a 5973 N Series mass spectrophotometer (MS) selective detector in SIM mode acquiring fragments with 264, 266 and 268 atomic mass unit (AMU). The interface temperature was 160ºC and the ion source (electron ionization) was set at 230ºC with electron energy of 69.9 eV, whilst the quadrupole mass filter was kept at 150ºC. The capillary column was a Hewlett-Packard 5MS (30m x 0.25 mm x 0.25 µm) (cat. HP 19091S-433). Helium (99,9990% purity) was the carrier gas, at a 1 mL/min constant flow through the column. The oven was initially set at 80ºC, and then raised to 260ºC at 15ºC/min. The injector was in splitless mode at 250ºC, closed for 3 minutes before purging with helium at 20 mL/min. The calibration curve of SPME for PCP quantification was performed with six standard solutions within a concentration range of 0.8 µg/L to 40 µg/L with good linearity (r2 = 0.9953). The detection limit calculated, by 3* S/N ratio, was 1.99 µg/L with a repeatability of 23.82% (as coefficient of variation, CV%) and an intermediate precision of 15,70% (as coefficient of variation, CV%).

RESULTS AND DISCUSSION 1.

Pine bark characterization

Pine bark is usually characterised in terms of its main constituents like cellulose, lignin or tannins, in order to take advantage of their potentials. Nevertheless, to evaluate its sorbent capacity it is important to know the surface properties and their ability to interact with the surrounding medium. Hence, the first step of the present work was to perform the characterization of the pine bark in order to realize the possible mechanistic interactions between the PCP and the surface. The physical properties obtained by mercury porosimetry are presented in Table I. The data are consistent with SEM images shown in Figure 1. In fact, the pine bark surface has low porosity and the few pores detected are classified as macropores (according to IUPAC with diameter higher than 50 nm) leading to a very low specific surface area. ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

83

Table I: Physical properties of pine bark. Values Real density (g/cm3) Pore volume (cm3/g) Surface area (m2/g) Average pore diameter (µm)

1.34 0.10 0.74 2.990-0.101

a)

b)

Figure 1. SEM images of sieved pine bark: (a) amplified 10000 times; (b) amplified 5000 times

The proximate and elemental analyses allow us to know the overall composition of the bark while the XPS and FTIR information take into account the surface chemistry. The volatiles, fixed carbon and ash content obtained by proximate analysis were 73.16%, 26.47% and 0.37 % (w:w in dry basis), respectively. The organic matter content of this material can be found by the sum of volatiles and fixed carbon, attaining a value of 99.63% (w:w in dry basis). By elemental analysis it was possible to verify that the organic matter was made up of 55.80% of C, 5.97% of H, 37.59% of O and 0.64% of N (w:w). With the XPS, 82.18% of C, 17.21% of O and 0.63% of N were identified at the surface of pine bark (atomic percentage, with spectrum peak areas divided by the corresponding sensitivity factors – C: 1.00, O: 2.93 and N: 1.80) taking into account that the low mass of hydrogen is not detected by the technique. The chemical bond analysis was made by curve fitting using the XPSPEAK 4.1 software, deconvoluting the C1s peak in four subpeaks (Figure 2) with binding energy corresponding to different functional classes: C1 concerns to the aliphatic and aromatic C-C and C-H bonding (285 eV); C2 to non carbonyl oxygen bond, C-OH and C-O-C (286.4 eV); C3 to carbonyl oxygen bond, C=O ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

84 (287.8 eV); and C4 to carboxyl oxygen, COOH and COO- (289.2 eV) (Sinn et al., 2001; Shchukarev et al., 2002). The highest percentage was attained for carbon-carbon bonding, with 73%, giving an idea about the strong magnitude of aromatic bonds assigned to the pine bark lignin and extractable fraction. The hydroxyl bonds represent 18% of the carbon associated to the holocellulose and phenolic content of the bark. The less representative bonding classes were the C3 with 6.4% and the C4 with 2.3%, probably related with the extractable content of the bark.

Figure 2. C1s deconvoluting peak (affected by a surface charge of 4.37 eV).

Another technique applied to complete the pine bark characterization was the FTIR, although the surface complexity and heterogeneity unable to readily characterize the absorbed infrared light at specific frequencies. In fact, the information acquired is only qualitative in the study of specific structures at the surface. The spectrum can be divided in four regions, namely the region between 4000-2500 cm-1 for stretching vibrations of X-H bonds, 2500-2000 cm-1 for the triple bonds, 2000-1500 cm-1 for double bonds and the known finger print region between 1500-650 cm-1. As can be seen in Figure 3, around 3400 cm-1 the spectrum represents a broad band due to the bond between the oxygen and the hydrogen stretching vibration, including hydrogen bonding. The band at 2900 cm-1 is assigned to the stretching C-H bond in the aromatic and aliphatic structures and the absorption identified around 1450 cm-1 is also related with C-H deformation. It can also be distinctly identified the band at 1610 cm-1 and at 1510 m-1 due to the aromatic C=C skeletal vibrations and a weak band around 1735 cm-1 caused by the stretch vibrations in C=O carbonyl structure. One important broad band is identified around 1160 cm-1 as result of the asymmetric stretching of C-O-C in the cellulose and hemicellulose (Vazquez et al, 2000).


85

Transmitance

Comentário: Alterar o nome dos eixos para inglês

4000

3500

3000

2500

2000

1500

1000

500

-1

Wave number (cm )

Figure 3. DRIFT spectra of pine bark.

With the FTIR analysis it was possible to certify the XPS results identifying the weight of the C=C aromatic bonding, the phenolic and polysaccharides hydroxyl group along with the C-O-C bonds and the double bonds between the carbon and oxygen atoms in the carbonyl and carboxyl groups. These kinds of structures are important in the perception of possible interactions with the surrounding molecules and the association they establish. For instance, functional groups like carboxylic and hydroxylic can be classified as acids once they can loose the associated hydrogen proton; on the other hand, the carbonyl group with the strong electronegative oxygen atom can accept hydrogen atoms behaving as a base. The pine bark acidity and basicity determination showed once again the importance of the surface acidic groups with values of 5.58 meq HCl/g for the acidity and 0.08 meq NaOH/g for the basicity. The attempt to evaluate the pH of the point of zero charge - pH at which the surface has an equal number of negative and positive charges, showed that the pine bark has always an acidic behavior with an equilibrium pH around 4.4 (Figure 4).

14 12 Final pH

10 8 6 4 2 0 0

2

4

6

8

10

12

14

Initial pH

Figure 4. Determination of the pine bark surface pHpzc using the pH drift method.


86 The presence of oxygen and hydrogen atoms on the pine bark surface along with its aromatic structure may induce that the solute molecules can establish hydrogen bonds and hydrophobic interactions with the sorbent.

2. Sorption Although other models are mentioned in literature, the most applied to describe sorption in wastewater systems are the linear (equation 1), the Freundlich (equation 2) and the Langmuir (equation 3).

qe = K d * C e (1) qe = K F * C e

qe =

1/ n

(2)

Q 0 * b * Ce (3) 1 + b * Ce

In these equations, qe and Ce are the solute concentration in the solid and in the aqueous phase, respectively. For each equation, Kd is defined as the distribution coefficient in the linear isotherm, KF and 1/n are related to the sorption capacity and to the energy distribution of the sorption sites, respectively, for the Freundlich isotherm. For the Langmuir isotherm, b is associated to the sorption energy and Qo is considered to be the maximum sorption capacity related to the total cover of the surface. Figure 5 represents the sorption and the desorption equilibrium experimental data. For the sorption evaluation, the parameters obtained for each model are shown in Table II. In the range of concentration studied, limited by the water solubility of PCP, the sorption is depicted by a linear isotherm although small deviations were observed, which can be assigned to experimental errors. Although the Freundlich model presents a better correlation factor, the linear model was preferred by its simplicity. A Kd of 1.86 L/g was obtained, an expected value when compared with similar materials, like peat that was tested as a sorbent for PCP by Tanjore and Viraraghavan (1997). These authors achieved a best correlation with the Freundlich isotherm with KF of 1.58 L/g and 1/n of 1.02, this last parameter very close to the unity, meaning that the linear isotherm was probably an acceptable adjustment. In spite of the good correlation factor obtained, the Langmuir isotherm can not be considered a suitable model to fit the experimental points since they are far away from the saturation region. This is reflected in the large errors associated to the Langmuir parameters. Table II. Coefficients obtained for the sorption isotherms by nonlinear regression. Linear

Freundlich

Kd (L/g)

r

2

KF (L/g)

1.86±0.06

0.978

0.84±0.14

Langmuir

1/n

r

2

o

Q (mg/g)

b*10-3 (L/g)

r2

1.18±0.04

0.993

9±45

0.2±1.0

0.976


87

In this study, if kinetic data is taken into account, the PCP average removal attained was above 98%. Previous work showed that the pine bark had affinity for hydrophobic organic compounds. Brás et al. (1999) performed mini-column dynamic studies and reached to the evidence of high sorption efficiencies for organochlorine pesticides in water, with yield of removal above 93%, except for Lindane with a significant lower value of 38%. The best yield of removal attained was for DDT with a value higher than 99%. Ratola et al. (2003) reported the sorption study of Lindane and Heptachlor by pine bark conducting equilibrium experiments, presented the Freundlich isotherm as the best model to describe the equilibrium data with KF of 0.928 L/g and 1/n of 1.01 for Lindane and 70.778 L/g and 1/n of 1.21 for Heptachlor. Once verified the good sorption capacity of the pine bark, it was intended to study the reversibility of the phenomena. A preliminary evaluation of desorption from the pine bark surface was performed using a batch washing (Figure 5).

0.250

qe (mg/g)

0.200 0.150 0.100 0.050 0.000 0

20

40

60

80

100

120

140

Ce (ug/L)

Figure 5: Pentachlorophenol sorption and desorption data for pine bark: adsorption(); desorption ();

The Kd attained in the desorption, 6.14 L/g (r2=0.937), was higher than the sorption Kd. The deviation of the desorption equilibrium data relatively to the sorption data is representative of sorption irreversibility, providing evidence that the PCP establish a strong bonding with the solid phase. Karichkof et al. (1979) stated that the linear isotherm is often indicative of partitioning. Indeed, if a PCP partitioning exists between the organic solid phase and the aqueous phase, induced by the strong affinity of the organic pollutant to the organic phase, is expected that the reverse situation would not occur. Chen et al. (2004) studied the sorption and desorption of PCP in soils and also accomplished that the sorption had a linear behaviour and found strong evidences of hysteresis. They suggest that the chemical composition and the condensed rigid physical character of lipids enable them to interact more strongly with the hydrophobic organic solutes via weak van der Waals bonds. General reasons accepted to cause sorption hysteresis are related to experimental artefacts, irreversible binding to specific sorption sites, slow rates of desorption and ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

88 sorbing molecule entrapment (Chen et al., 2004). To override the effects of experimental artefacts a blank experiment was conducted without pine bark but following the same experimental procedures and the sorption and desorption experiment was conducted under the same conditions. Owing to the pine bark low porosity it is unlikely that the entrapment of the PCP molecules was the cause of the observed hysteresis. Accordingly, it is reliable to conclude that the sorption irreversibility is assigned to the irreversible binding to specific sites of the pine bark or slow rates of desorption. In effect, hydrophobic compounds with such high log Kow as PCP will have strong affinities to organic matter matrices with representative aromatic content as pine bark. Whichever the case, it can be stated that the PCP in the pine bark can be considered to be stabilized, reducing effectively the mobility of this hazard pollutant in the environment. CONCLUSION

From the results of the present work, it can be conclude that: 1. Pine bark is a material with small porosity, with the few pores identified in the range of macropores, and therefore with small specific surface area; 2. The bark has very low content in ashes with the organic matter representing more than 99% of the weight. XPS and FTIR characterization of the surface chemistry show the expected presence of C-C aliphatic and aromatic bonds and C-O bonds from the hydroxyl, phenolic carbonyl and carboxyl groups, the last two assigned to the extractable fraction of the bark; 3. The sorption of PCP can be described by a linear isotherm, suggesting that the sorption is due to the PCP partitioning between the aqueous and the solid phase. 4. The conducted desorption experiment show that the PCP is strongly bonded to the pine bark surface. The main conclusion of the present work is that the pine bark can act as an effective sorbent for PCP, being a good alternative to conventional sorbents, which almost totally immobilizes this organic pollutant with all the benefits associated with the use of local and abundant residues.

REFERENCES

Al-Ashe, S., Duvnjak, Z. (1999): Sorption of heavy metals from synthetic solutions and industrial wastewater using plant materials: Water Quality Research Journal of Canada 34 (3): 481-503. Alves, M.M., Beça, C.G.G., Guedes de Carvalho, R., Castanheira, J.M., Pereira, M.C.S., Vasconcelos, L.A.T. (1993): Chromium removal in tannery wastewaters "polishing" by Pinus sylvestris bark: Water Research 27(8): 1333-1338. Boyd, S. A., Shaobai, S., Lee J. F., Mortland M. M. (1988): Pentachlorophenol sorption by organo-clays: Clays and Clay Minerals 36 (2): 125-130. ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

89 Brás, I.P., Santos, L., Alves, A. (1999): Organochlorine Pesticides Removal by Pinus Bark Sorption: Environmental Science Technology 33 (4): 631–634. Chen, Y., Chen, H., Xu, Y., Shen, M. (2004): Irreversible sorption of pentachlorophenol to sediments: experimental observations, Environment International 30 (1): 31-37. Crosby, D.G (1981): Environmental chemistry of pentachlorophenol: Pure and applied chemistry 53: 1051-1080. DiVincenzo, J.P. and Sparks, D.L. (1997): Slow sorption of Pentachlorophenol on soil: Concentration effects: Environmental Science & Technology 31 (4): 977-983. Fengel D, Wegener G: Wood – Chemistry, Ultrastructure, Reactions. De Gruyter, Berlin, 1984. Fradinho, D.M., Neto, C.P., Evtuguin, D., Jorge, F.C., Irle, M.ª, Gil, M.H., Jesus, J.P. (2002): Chemical characterisation of bark and alkaline bark extracts from maritime pine grown in Portugal: Industrial Crops and Products 16 (1): 23-32. Harun, J. and Labosky, Jr, P. (1985): Chemical constituents of five northeastern barks: Wood and Fiber science 17 (2): 274-280. Haussard, M., Gaballah, I., Kanari, N., de Donato, P., Barrès, D., Villieras, F. (2003): Separation of hydrocarbons and lipid from water using treated bark: Water research 37 (2): 362-374. Huang, W., Peng, P., Yu., Z., Fu, J. (2003): Effects of organic matter heterogeneity on sorption and desorption of organic contaminants by soils and sediments: Applied Geochemistry 18 (7): 955-972. Jorge, F.C., Brito, P., Pepino, L., Portugal, A., Gil, H., Costa R. P. (2001): Aplicações para as Cascas de Árvores e para os Extractos Taninosos: Uma Revisão, Silva Lusitana, 9 (2): 225-236. Karickhoff, S.W., Brown, D.S., Scott, T.A. (1979): Sorption of hydrophobic pollutants on natural sediments: Water Research 13 (3): 241-248 Kofujita, H., Ettyu, K., Ota, M. (1999): Characterization of major components in bark from five Japanese tree species for chemical utilization: Wood Science and Technology 33 (3): 223-228. Pereira, M.F.R., Soares, S.F., Órfão, J.J.M., Figueiredo, J.L. (2003): Adsorption of dyes on activated carbons: influence of surface chemical groups: Carbon 41 (4): 811-821. Randall J.M. (1977); Variations in effectiveness of barks as scavengers for heavy metal ions: Forest products Journal 27 (11): 51-56. Ratola, N., Botelho, C., Alves, A. (2002): The use of pine bark as a natural adsorbent for persistent organic pollutants – study of lindane and heptachlor adsorption: Journal of Chemical Technology and Biotechnology 78 (2-3): 347-351.


90 Rivera-Utrilla, Bautista-Toledo, I., Ferro-Garcia, M.A., Moreno-Castilla, C. (2001): Activated carbon modifications by adsorption of bacteria and their effect on aqueous lead adsorption: Journal Chemical Technology and Biotechnology 76 (12): 1209-1215. Santos, C., Machado, R. Correia, M. J.N., Carvalho, J. R. (2003): Biosorption of Copper by Grape- Stalks and Pine- Bark Biomasses: The European Journal of Mineral Processing and Environmental Protection 3 (1): 108-118. Shchukarev, A., Sundberg, B., Mellerowicz, E., Persson, P. (2002): XPS study of living tree: Surface and interface Analysis 34 (1): 284-288. Sinn, G., Reiterer, A., Stanzl-Tschegg, E. (2001): Surface analysis of different wood species using X-ray photoelectron spectroscopy (XPS): Journal of Materials Science 36 (19): 4673-4680. Tanjore, S., Viraraghavan, T. (1997): Effect of oxygen on the adsorption of pentachlorophenol by peat from water: Water, Air and Soil Pollution 100 (1-2): 151162. Vásquez, G., Freire, S., González, J., Antorrena, G. (2000): Characterization of Pinus Pinaster bark and its alkaline extracts by diffuse reflectance Fourier transform infrared (DRIFT) spectroscopy: Holz als Roh- und Werkstoff 58 (1): 57-61. Vázquez, G., González-Álvarez, J., Freire, S., López-Lorenzo, M., Antorrena, G. (2002): Removal of cadmium and mercury ions from aqueous solution by sorption on treated Pinus pinaster bark: kinetics and isotherms: Bioresource Technology 82 (3): 247-251.


91

Session 5 Environmental Impact of Wood Preservation


92


93

CREOSOTE SPILLS AS A CAUSE OF ODOURS IN RIVER WATERS. IDENTIFICATION OF THE SOURCE AND STRATEGIES ADOPTED FOR MINIMIZING ITS IMPACT Emese Borsiczky, Fabián-Ángel Sánchez-Alonso, Jordi Martin-Alonso FORESTE S.L. Pg. Cantí 8, 2-2, 08005 Barcelona (Spain). E-mail: [email protected]. Phone/Fax: +34 93 309 1838

Keywords: creosote, drinking water, environmental impact, Black Locust ABSTRACT Studies on the environmental impact of creosote are common, but they mainly deal with contamination in groundwater and sediments, as well as studies on bioremediation, partitioning of Polycyclic Aromatic Hydrocarbons (PAH) into water, microbial degradation, etc. This study presents the strategy used for the identification of creosote constituents causing odour in the surface water of a river used for supplying drinking water to the Barcelona’s Metropolitan Area, as well as the attempt to trace the companies responsible for the contamination. It also shows an alternative concept to avoid further pollution of the environment: the complete elimination of wood preserving chemicals by using naturally resistant wood.

INTRODUCTION

Tastes and odours in drinking water are frequently major complaints of consumers. Volatile organic compounds are usually associated with these effects. Several odour episodes in the surface water of the Llobregat river (N.E. Spain), which is used for drinking water production for Barcelona and its metropolitan area, showed a high abundance of Polycyclic Aromatic Hydrocarbons (PAH’s). These unexpected results were related to creosote, the raw material used by wood-preserving factories located at the banks of the river. Creosote is a complex mixture obtained during the distillation of coal tar produced by high temperature carbonisation of bituminous coal (AWPA, 1971). The composition of creosote depends on the coal tar origin (Johnson, 1993; Mueller, 1991; Novotny, 1981). Yet, it contains roughly 90% of polycyclic aromatic hydrocarbons (PAH), numerous phenolic compounds, nitrogen, sulphur and oxygen heterocycles and also a small fraction of benzene, toluene and xylenes. The PAH’s found in creosote are well-known to be mutagenic and carcinogenic (Arvin, 1992; USEPA, 1987; Marvin, 1995). Additionally, they have a teratogenic effect (Arvin, 1992). Creosote is considered to be a carcinogen by the International Agency for Research on Cancer (IARC, 1985). The United States Environmental Protection Agency has published a proposed intent to cancel registration of creosote for all uses except as a wood-preservative (US EPA, 1984) as well as to assess the health effects of creosote (US EPA, 1987, 2003). In addition, the European Community has regulated the presence of PAH in drinking water to a maximum of 0.1 µg/L for the sum of four ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

94 compounds in drinking water plus a stricter limit for benzo(a)pyrene of 0.010 µg/L (EC Directive 98/83/EC). Finally, the European Community has also strongly restricted the marketing and use of creosote (EC Directive 2001/90/EC). The nature and significance in terms of environmental and human health of PAH’s in creosote lead to extensive studies of their occurrence and distribution. Thus, creosote contamination in groundwater (Arvin, 1992; Middaugh, 1991; Mueller, 1993; Pastorok, 1994), sediments (Marvin, 1995; Pastorok, 1994) and water (Ventura, 1995; 1998) has been reported in the literature. Other studies related to bioremediation (Mueller, 1991; 1993), partitioning of PAH’s into water (Lee, 1992), and microbial degradation (Flyvberg, 1993) have also been carried out. This study presents comparisons of different analytical methods for the identification of compounds causing odour in a river supplying water to Barcelona, and the attempt to trace the responsible enterprises for the contamination. MATERIALS AND METHODS

Sampling

Samples were collected in 1 L amber glass bottles, with screw lined Teflon caps, filled until overflow in order to prevent loss of volatile compounds by the presence of headspace. Samples were stored at 4ºC and analysed within 24 h. Preconcentration

Closed loop stripping analyses (CLSA) were carried out using a commercial CLSA apparatus (Brechbüler, Switzerland) according to the method developed by Grob (1973). For liquid-liquid extraction (LLE), the USEPA method 550 was used. Gas chromatography (GC-FID)

For CLSA analyses a Fisons (UK) model 5300 gas chromatograph equipped with a FID detector was used. For LLE extracts, a Konik 3000 C (Spain) gas chromatograph equipped with a FID detector was used. Gas Chromatography - Mass Spectrometry (GC-MS)

A VG TRIO 1000 (VG instruments, UK) spectrometer was used for GC-MS analyses. High Performance Liquid Chromatography (HPLC)

A Kontron (USA) liquid chromatograph equipped with two Kontron model 420 pumps, an autosampler model 465 and a Kratos Analytical Spectraflow 980 fluorescence detector was used for all HPLC analyses (λ exc = 280 nm, λ em = 389 nm).


95 GEOGRAPHICAL SITUATION

Figure 1 shows a general scheme of the different sampling points located along the river Llobregat. The two wood preserving factories (A and B), the affected waterworks plants (WW1 and WW2), and a wastewater treatment plant (WWTP) that allows the treatment of waste waters from the city of Manresa (65,000 inhabitants) are also displayed. The effluent of factory B is included in the Manresa’s WWTP, the sewers of this wastewater facility are of the combined type that can overflow during rainfalls. As both factories are located in the Cardoner river, a tributary of the Llobregat river, sampling point nº 1 can be considered as the background concentration of PAH’s in the Llobregat river, whereas nº 2 and 3 are the background level in the Cardoner river. Sampling point nº 4 is a sewer including the effluent of factory A, therefore, nº 5 is exclusively affected by this factory. Sampling points nº 6, 7 and 8 are related to the B factory; nº 6 refers to untreated effluent of a combined sewer that includes factory B, and the other two correspond to raw and treated wastewaters from the Manresa’s WWTP. Finally, sample nº 9 to WW 2 can be affected by both A and B factories.

Figure 1: Geographical situation

RESULTS AND DISCUSSION

Several odour events in the Llobregat river water were detected in the waterworks plants which caused downtime. Closed-loop stripping analysis, a routine technique for the identification of volatile organic compounds at trace levels was primarily used. During these episodes of creosote contamination, the GC-FID chromatograms of CLSA extracts from river water with odour problems were different to those usually obtained. The extracts were analysed by GC-MS, and revealed high abundance of 2-3 ring PAH.


96 These unexpected results required further analysis using the LLE and GC-MS method (Fig. 2). The results showed the presence of heavier PAH’s not previously detected by the CLSA method. The high levels of PAH’s could be correlated to the existence of two creosote wood-preserving factories (named A and B) located in the upper course of the river. Table 1 lists the peak identification for the different compounds related to creosote found both in river water spills and in the raw material supplied by the two wood preserving factories.

Figure 2: GC-MS chromatogram of a river water extract (LLE method). Peak numbering as reported in Table 1

Nº

Table 1: Compounds in river water and creosote identified by GC-MS Compound Nº Compound Nº Compound

1 2 3 4 5

Toluene Ethylbenzene m+p-Xylene o-Xylene Phenol

16 17 18 19 20

Biphenyl 2,6-Dimethylnaphthalene 1,6-Dimethylnaphthalene C2-Naphthalene 2,3-Dimethylnaphthalene

31 32 33 34 35

6 7 8 9 10 11 12 13 14 15

2,3-Dihydroindene (1H)-indene o-Cresol C2-Phenols 2,4-Dimethylphenol Naphthalene Benzo(b)tiophene Isoquinoline 2-Methylnaphthalene 1-Methylnaphthalene

21 22 23 24 25 26 27 28 29 30

Acenaphtene Dibenzofuran C3-Naphthalene (9H)-Fluorene C1-Biphenyls C1-Dibenzofuran C1-(9H)-Fluorene C2-Biphenyls Dibenzothiophene Phenanthrene

36 37 38 39 40 41 42 43 44 45

Anthracene Acridine (9H)-Carbazole C1-Phenanthrenes 4H-cyclopenta (d,e,f)phenanthrene Fluoranthene Pyrene Benzo[a]anthracene Crysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[a]pyrene Dibenzo(a, h)anthracene Benzo(g, h, i)perylene Indeno(1, 2, 3, c, d)pyrene


97

Figure 3: GC-MS profiles of creosotes used by A and B factories during the first event. Peak numbering as reported in Table 1.

As PAH’s represent the major constituents of creosote, that contains compounds which are of significance in terms of environmental and human health and are regulated by EEC and USA legislations, particular emphasis was devoted to quantify PAH’s by LLE with HPLC. The HPLC profiles of the two creosotes used as a raw material in the two wood preserving facilities and river samples were very similar, but the ratio naphthalene / isoquinoline (N/I) was considerably different in the GC-MS profiles (Fig. ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

98 3, see peaks 11 and 13) This ratio was 15.7 for creosote A and 0.55 for creosote B. Therefore it was useful in this case for the univocal identification of the factory responsible for the episode of contamination. Table 2 displays the levels of PAH’s EEC priority pollutants measured by HPLC in river water during two significant creosote episodes. Samples were collected from the control points when odour was detected. The first episode is referred to a dumping of the A factory. High levels of PAH were found in sampling point nº 4, and no overflow was observed at point nº 6. Sample nº 8, the outlet of the WWTP, presented relatively low levels of PAH attributable exclusively to factory B, therefore, its influence was not important because these wastewaters were subsequently diluted into the Cardoner river by a factor of 50. These results confirmed that creosote from points nº 9 to WW2 was mainly spilled by factory A, and this conclusion is reinforced because their chromatograms compared to those obtained for sampling point nº 8 were different in their N / I ratio. TABLE 2: Average levels (ng/l) of PAH EC priority pollutants measured by HPLC in river water during two creosote spills. Sampling point First episode Second episode 1 n.a. 13 2 n.a. 31 3 7 n.a. 4 30450 85 5 n.a. 15 6 n.a. 2000 µg/l 7 n.a. 15420 8 256 481 9 878 1456 10 218 n.a. WW 1 in 459 687 out 82 n.w. WW 2 in 368 489 out 18 n.w. 1

Sum of Fluoranthene, Benzo[b]fluoranthene, Benzo[k]fluoranthene, Benzo[ghi]perylene, Indeno[1, 2, 3 - cd]pyrene (EC Directive 80/778/EC)

Benzo[a]pyrene,

WW1 provides drinking water to cities around Barcelona; WW2 provides drinking water to Barcelona city; n.a. means not analysed; n.w. means the treatment plants were not working.

Several months later, another creosote episode was detected. All the controlled sampling points presented the same chromatographic profiles. Both A and B factories supplied us again their creosotes used as a raw material. Their chromatograms were identical and even the N/I ratio was the same. The explanation given by the companies is that there is only one creosote supplier in Spain, but depending on market prices it may also be imported from France. It is known that the composition of creosote depends on its origin. Therefore, we concluded that both factories used different creosotes when the first episode occurred (Spanish in factory A and French in factory B) whereas they both employed Spanish creosote during the last incident. The average levels of PAH during the second event are also displayed in Table 2. Trace levels of PAH were detected at points only affected by factory A. Apparently abnormal results were obtained for points related to factory B. The heavy rain falling during this ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

99 event explains the results. An important amount of creosote from factory B was leached during the beginning of the rain to the combined sewer and then to the wastewater treatment plant, thus explaining the high levels of PAH measured in points nº 6 and 7 (2000 µg/l and 15420 ng/l, respectively). A low concentration (481 ng PAH/L) was measured at the outlet of the WWTP (point nº 8), so its efficiency to remove the PAH’s was higher than 99%. Because of dilution, negligible levels from this point were spilled to the Cardoner river. Nevertheless, after the first run-off creosote from factory B (point nº 6) entered directly into the river (point nº 9) because of a wastewater sewer overflow due to the intensive rainfall. Therefore, the levels of PAH measured in samples nº 9 to WW 2 came directly from factory B. After the episodes of contamination, factory A isolated its creosote circuit in order to avoid accidental spillages, although leachates coming from its storage area still may affect the river. Factory B changed the product used as wood preservative from creosote to an inorganic mixture which includes boric acid and chromium (III) as main components. After these changes, no creosote episode has been detected in the raw water of the Llobregat river. Furthermore, this case can be considered as an example on how dispersion on the environment along with the analytical limit of detection can jeopardise the total impact of a given compound. The only way to eradicate negative impacts from wood preservatives is by eliminating them completely, as for example by using untreated wood. RECOMMENDED ALTERNATIVES

This contamination of surface waters intended for drinking water production acted as a trigger for considering the introduction of naturally resistant wood in the Spanish marketplace that may replace hazardous containing woodstuff. The most naturally resistant wood in Europe is Black Locust (Robinia pseudoacacia), so we consider it may play an important role in the European marketplace, as it is common throughout the temperate regions of the continent, exhibits excellent mechanical characteristics and is relatively affordable (Hanover 1989). Due to the high flavonoids content in the heartwood (Smith et al. 1989), it shows high biological resistance against wood destroying fungi and can endure for several decades in the soil. However this useful tree is underutilized, therefore more research on its properties would provide a better knowledge to promote a more efficient use by replacing treated wood in some specific cases. Hungary is the main European producer of Black Locust (Keresztesi, 1988). Unfortunately, the adaptation of a series of thermal power stations to substitute fuel for firewood as a way to reduce CO2 emissions, along with its non negligible improvement, has lead to some unexpected negative effects on the Hungarian wood market: -

The price of the firewood used in households has increased by 20-25 % within one year.

-

Less raw material is available for furniture panels (fibreboard, chipboard, etc.)


100

-

In many cases carpentry timber is also sold as fuelwood.

-

Due to the lack of enough quantity of firewood in the vicinity of the power thermal stations, thousands of tracks per year are transporting firewood to the thermal stations from long distances, deteriorating regional roads, disturbing local traffic and polluting the environment.

Black Locust is an ideal species for biomass production. But for a sustainable use in Hungary, specific fuelwood plantations would be necessary in the vicinity of the thermal power stations.

CONCLUSIONS

Employing several analytical techniques, odour events in river water entering two waterworks of the Barcelona’s Metropolitan Area were associated to creosote, allowing to deduce the responsible factory in two different pollution episodes. Black Locust may cost-effectively replace treated wood in many applications, but its availability in Hungary, one of the biggest EU suppliers, is under risk because of the substitution of fuel for firewood in some thermal power stations.

ACKNOWLEDGMENTS

We are thankful to Dr Francesc Ventura for its helpful contribution to this work.

REFERENCES

Arvin E., and Flyvbjerg J. (1992). J. IWEM, 6, 646-652. American Wood-Preservers Association “The AWPA Book of Standards”, AWPA. Washington DC, (1971). EC Directive 80/778/EC. J. Off. European Community (1980). EC Directive 98/83/EC. J. Off. European Community (1998). EC Directive 2001/90/EC. J. Off. European Community (2001) Flyvbjerg J., Arvin E., Jensen B.K. and Olsen S.K. (1993). J. Cont. Hydrol., 12, 133180. Grob K. (1973). J. Chromatogr, 84, 255-273. Hanover, J.W. (1989). Proc. Conf. on Fast Growing Nitrogen Fixing Trees, 1989, Marburg, W. Germany.


101 IARC. (1985). IARC Monographs. Polynuclear Aromatic Compounds. Part 4. WHO, IARC, Lyon (France), 35, 137-140. Johnson N., Sadler R., Shaw G.R. and Connell D.W. (1993). Chemosphere, 27, 11511158. Keresztesi, B. (ed). (1988). The Black Locust. Akademiai Kiado, Budapest, Hungary. Lee L.S., Rao P.S.C. and Okuda I. (1992). Environ. Sci. Technol., 26, 2110-2115 . Marvin C.H., Lundrigan J.A., McCarry B.E. and Bryant D.W. (1995). Environ. Toxicol. Chem., 14, 2059-2066. Middaugh D.P, Mueller J.G., Thomas R.L., Lantz S.E., Hemmer M.H., Brooks G.T. and Chapman P.J. (1991). Arch. Environ. Contam. Toxicol., 21, 233-244 Mueller J.G., Lantz S.E., Blattmann B.O. and Chapman P.J. (1991). Environ. Sci. Technol., 25, 1055-1061. Mueller J.G., Lantz S.E., Ross D., Colvin R.J., Middaugh D.P. and Pritchard P.H. (1993). Environ. Sci. Technol., 27, 691-698. Novotony M., Strand J.W., Bimith S.L., Wiester D. and Schwende F.J. (1981). Fuel., 60, 213-220. Pastorok R.A., Peek D.C., Sampson, J.R. and Jacobson M.A. (1994). Environ. Toxicol. Chem., 13, 1929-1941. Smith A.L, C.L. Campbell, M.P. Diwakar, J.W. Hanover, and R.O. Miller (1989). Holzforschung 43, 293-296. US EPA. (1984). Fed. Register, 49, 33328. US EPA / 600 / 8-88 / 025. Health Effects Assessment for Creosote. July (1987), PB88 179395. US EPA. (2003). Fed. Register, 68, 68042-68044. Ventura F., Matia Ll., Romero J., Boleda Mª R., Martí I. and Martín-Alonso J. (1995). Water. Sci. Technol., 31, 63-68. Ventura F., Boleda Mª R., Lloret, R. and Martín-Alonso J. (1998). Wat. Res., 32, 503509. Wise S.A., Benner B.A., Byrd G.D., Chesler S.N., Rebbert R.E. and Schantz M.M. (1988). Anal. Chem., 60, 887-894.


102


103

Session 6 Heat-Treatment of Wood


104


105

HYGROSCOPICITY IN HEAT-TREATED WOOD: EFFECT OF EXTRACTIVES Marcos M. González-Peña1; Martin C. Breese2; Callum A. S. Hill3 1, 2, 3

School of Agricultural and Forest Sciences, University of Wales, Bangor. Thoday Building, Deiniol Road, Bangor, Gwynedd, LL57 2UW, United Kingdom ([email protected], [email protected], [email protected])

SUMMARY Following thermal treatments, the hygroscopic properties of wood were examined in two hardwood species and two softwood species, in the extracted (E) and nonextracted (NE) conditions. Samples were exposed to three temperatures (190°, 210° and 230° C), for four periods (30, 60, 120 and 240 minutes), at atmospheric pressure, under a nitrogen flow. Heat-treated samples exposed to controlled humidities in the adsorption cycle were characterised by lower equilibrium moisture content (e.m.c) than controls in all cases. The most severe treatment reduced e.m.c. by about 40%. E samples showed a faster rate of sorption than the NE specimens. This effect faded as the severity of the treatment increased. In all treatments, the e.m.c. was always lower for the NE set. Major differences in the adsorption isotherm were found between E and NE samples following heat treatment using the milder schedules; the percentage of e.m.c. reduction remained almost constant for the E samples, but decreased for the NE samples. Losses in dry weight (WL) ranged from 2.0 to 14.7 % for softwoods, and from 2.3 to 22.9% for hardwoods. No ultimate effect of the extraction on WL could be determined. The trend was that at the lowest temperatures, WL in E samples was higher, whilst the opposite was generally found at the highest temperature. Volume loss ranged from 0.0% to 5.8% for softwoods and from 0.1% to 16.0% for hardwoods. Antiswelling efficiency (ASE) ranged from 5% to 70% for softwoods and from 6% to 66% for hardwoods.

INTRODUCTION

Investigations on the effect thermal treatment of wood in reducing its hygroscopicity have been conducted for more than 80 years (Tiemann, 1920). Reduced hygroscopicity of thermally treated wood leads to enhanced decay resistance without using additional chemicals (Farahani et al., 2001, Welzbacher and Rapp, 2002), and also to improved dimensional stability (Burmester, 1973; Feist and Sell, 1987; Bekhta and Niemz, 2003). Increased concerns on the treatment, use, and final disposal of wood treated with biocides, have stimulated technological studies into processes for the heat-treatment of wood and wood panels in the last two decades (Evans, 2003). Such research has led to the establishment of production facilities in various European countries, with increased commercial activity (Gohar and Guyonnet, 1998; Rajasuo, 2002). Heat-treatment of wood is proposed as an option producing lower environmental impact than conventional preservative treatments (Militz, 2002). Currently, the commercial presence of heattreated wood is modest though, and it is recognised that quality controls are required (Syrjänen and Kangas, 2000; Welzbacher and Rapp, 2002).


106

A better understanding of the changes that occur in the wood as a result of thermal treatment, as well as studies on the optimal processing conditions and the properties of the resulting products are all essential steps for the commercial establishment at a larger scale of these tentatively environmentally friendly technologies. Information regarding the effect of wood extractives on the properties of heat-treated wood is to some extent limited. Earlier research indicates that the presence of extractives might accelerate breakdown of wood polymers during heating (Bourgois et al., 1989). No information exists regarding the effect of the initial extractive content on the hygroscopicity or on the dimensional stability of thermally treated wood. The study of the influence of extractives on the properties of heat-treated wood in particular, and its bearing on the overall environmental impact of these technologies in general, are of major importance. It may be that a variable behaviour of the material, depending on the extractive content on harvesting the timber, could influence (positively or negatively) the quality consistency of the heat-treated wood. At a more general appraisal, the initial extractive content may influence negatively the Life Cycle Assessment of the heattreated wood, due to a large emission of volatile organic compounds (VOCs) during processing. This may overshadow the much improved rating of heat-treated wood in service compared to untreated air-dried wood, as far as VOCs indoor emission is concerned (Manninen et al., 2002). This paper presents results from a larger study that involves the characterisation of seven locally grown heat-treated hardwoods and five softwoods. Small, clear specimens of these species were heat-treated in the before- and after-extraction states and the hygroscopicity of the resultant product compared. Other changes in physical properties such as density, anisotropy and dimensional stability were also reviewed. The present work is part of an agenda to develop schedules for the heat-treatment of British-grown woods, as an alternative method of producing wood products with enhanced decay resistance and improved dimensional stability for above ground and indoor uses.

METHODS Species and heating schedules

Data reported in this paper are from Beech, Hornbeam, Ash, Douglas fir, and Western Red cedar. One set of wood blocks (20 x 10 x 5 mm, T x R x L) were cold water extracted for 48 hours, and then hot-water extracted for 8 hours. To avoid degradation, the specimens were vacuum oven dried overnight at 60° C and at 100 mbar. Samples were subsequently Sohxlet extracted in a toluene: methanol: acetone solution (4:1:1 v/v) for 6 hours, and finally oven dried as above. Heat-treatment of the material was conducted in a temperature chamber (Binder VD115), at three different temperatures (190°, 210° and 230° C ± 1° C), for four different periods (30, 60, 120 and 240 minutes) under nitrogen atmosphere. A small, constant flow of nitrogen (+150 mbar) was kept at all times. Six replicates, previously dried to 0% moisture content, were used for each treatment. Samples were loaded into the oven which had been previously heated to the ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

107 treatment temperature. Treatment time was counted from the moment when the centre of the block had reached the required temperature, previously determined as 1.5 hr on average. These samples were identified as E. Another set of closely matched samples was heat-treated without previous extraction under the same conditions. Samples from this set were identified as NE. In order to determine the remaining extractive content in the NE samples after the heat treatment, Sohxlet extraction was performed to selected species from both sample sets for 6 hours. The remaining extractive content in the NE samples was determined by subtracting the weight of low-temperature rotor-evaporated extractives (non-volatised breakdown products) from the E samples. It has been determined that lignin in the solid residue resulting from thermal treatments, particularly when the treatment is in the presence of water, reaches a maximum breakdown level, and subsequently repolymerises to give a new lignin of lower molecular weight. This so-called pseudolignin is, to different degrees, soluble in organic solvents, including acetone and ethanol (Garrote et al., 1999). Hence, the second Sohxlet extraction (after the heat treatment) was performed using hot water only. The dried by-products were dissolved in ethyl acetate for one week, to complement the gravimetric analysis using GC (Varian Gas Chromatograph CP3380, column DB5, FID temperature 275° C). Helium was used as the gas carrier for the column. The temperature program began at 80° C (1 min), thereafter raising to 200° C at 10° C min−1 and holding for 30 min.

Hygroscopicity tests

For the adsorption cycle, from the oven dry condition, control and treated samples from both E and NE sets were placed in small airtight containers containing saturated salt solutions to produce five relative humidities (RHs), namely 11%, 33%, 43%, 75% and 92%, in a hermetic chamber with controlled temperature at 25° C ± 0.5° C. Additionally, water vapour sorption of control and treated samples from selected treatments was determined using an automated Dynamic Vapour Sorption (DVS) instrument (Surface Measurement Systems), also in the adsorption cycle, from the oven dry condition, at 25° C ± 0.5° C. A wood sample of ca. 10 mg with dimensions of 0.5 x 10 x 10 mm (T x R x L) was used in the instrument’s microbalance for each run. This sample was taken at 1/3 depth from the edge of each of the heat-treated samples studied. This was regarded as representative of the whole sample. The schedule for the DVS was set to 10 different RHs (0, 5, 10, 20, 40, 60, 72, 82, 90 and 95 percent). These RHs are generated by mixing dry nitrogen and saturated water vapour, using mass flow controllers. Data on mass change was acquired every 20 seconds. At each RH an inbuilt algorithm was set to consider that equilibrium had been reached when the slope of an adjusted tangent line to the curve of mass change with respect to time for the last three minutes of data was smaller than 0.002% min-1. In order to elucidate the possible differences in both sets of samples in the sorption mechanism a two-hydrate model based on the Hailwood-Horrobin theory of sorption was used to fit the adsorption isotherm curves. Adsorption times required to reach e.m.c. for each relative humidity were also compared using the experimental data from the DVS. ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

108

Dimensional stability and density changes

Anisotropy changes were calculated by the differences in the anisotropy ratio due to the treatment, evaluated from the oven dry to the fully saturated condition. The dimensional stability brought about by the heat treatment was determined by means of the antiswelling efficiency (ASE) according to Hill and Jones (1996). Dry weight losses (WL) were determined for all the treatments, as were the effect of time and temperature in these changes. RESULTS AND DISCUSSION

After heat treatment remnant extractive content and breakdown products were determined from the NE Ash samples. Water extracts after heat-treatment gave a maximum yield of 1.89% of the oven dry (OD) weight. It was found that the difference to the content of breakdown products in the E samples was very small (maximum 0.53 %). This amount did not affect the statistical analysis of WL due to the heat treatment. As a further extraction step might have obscured the response of the material during the hygroscopicity and dimensional stability tests, it was decided not to further extract all the remaining sample sets from the other species before performing the subsequent studies. Hence, WL due to the treatment is, in all cases, the apparent WL without correction for possible extractive or breakdown products remaining in the treated samples. As shown in the chromatograph in Figure 1, little evidence of remaining polar extractives was found in the NE samples (Figure 1). mV 10.0

7.5

5.0

Untreated Ash

190°C pre-extracted 190°C no pre-extracted

230°C pre-extracted 230°C no pre-extracted

2.5

0.1

10 Internal control: acetic acid

20

30

40

Retention time

Figure 1. Gas chromatograph of water extracts from Fraxinus sp. after heat treatments at 190° C and 230° C for two hours. The trace at the rear is from the untreated sample. At the front is from acetic acid.

The presence of non-polar compounds in the NE samples after the heat treatment was not evaluated. Nevertheless, according to Nuopponen et al. (2003), non-polar fats and waxes disappear from heat-treated softwood at temperatures from 120° to 180° C. ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

109 Above 200° C, all the resin acids disappear from the heat-treated material. It can be presumed that most of the extractives (polar and non-polar) had disappeared after the thermal treatment schedules used for this work. For both sample sets (E and NE), and at the two temperature extremes tested in this experiment (190° C and 230° C), the major polar product found in the water extract was acetic acid. This was expected, since during the heat-treatment acetyl groups of hemicellulose hydrolyse and acetic acid is formed (Zaman et al., 2002). This result is in line with the findings of Manninen et al. (2002), who determined that of the individual compounds, the most abundant VOCs from the heat-treated wood were 2furancarboxaldehyde, acetic acid and 2-propanone. No attempt was made to identify the smaller peaks. A relatively higher amount of acetic acid was found in the NE samples treated at 190° C for 2 hours, compared to the other treatments, as seen by a higher peak for such treatment. However, the content of acetic acid was much lower than the content of acetic acid detected in the water extracts from the untreated Ash (Figure 1, at the rear). In the untreated hardwood acetic acid was probably formed by an incipient hydrolysis of the acetyl groups of the hemicelluloses during the water extraction procedure. However, even at room temperature, acetic acid has been detected as the largest emission among all the VOCs from untreated hardwoods (Risholm-Sundman, et al., 1998). This indicates the temperature instability of the acetyl groups in hardwood hemicelluloses. Our results indicate that if heat-treated wood is to be used for indoor purposes, some harmful compounds might still be released into the air, particularly acetic acid. In addition to causing irritation to the respiratory system, this compound is annoying to the sense of smell in humans, eliciting negative emotions (anger and disgust) (VernetMaury et al., 1999). A surface finishing (e.g. a lacquer) may be prescribed in order to reduce such emissions. The behaviour of the adsorption isotherms was similar for all the heat-treated species. In both treated and control samples, the isotherm curve was sigmoid in nature, and of the Type II form (Sing, 1998). For clarity the analysis of representative isotherm curves for E and NE samples is referred to Western Red cedar (WRC), from the data obtained in the DVS. Gross deviations from the behaviour of this species are indicated in the text. Figure 2 shows the isotherms for control and treated samples at 190° C and 230° C in both E and NE states.


110

20 control

Equilibrium moisture content (%)

190° C

15

230° C

10

5

0 0

10

Cell Wall Porosity Reseach Group

20

30

40

50

60

Relative Humidity (%)

70

80

90

100

© University of Wales Bangor

Figure 2. Modelled sorption isotherms using a two-hydrate model from the Hailwood-Horrobin theory for Western Red cedar in the E (dashed line) and NE (solid line) conditions, for control samples and heattreated specimens at 190° and 230 °C for two hours. Symbols are experimental data for the E (open circle) and NE (filled circle) conditions, obtained in the DVS.

The e.m.c data obtained by storage over the saturated salt solutions at low RHs lacked the resolution required for the size of the sample used in this work. Not even small differences could be appreciated at the lowest RH used (11%), equivalent to an e.m.c. in untreated wood smaller than 2%. This fact is reflected in some overlapping of e.m.c.s at the lowest RH studied (Figure 3, Table 1). Moreover, control E samples had, in some instances, slightly lower e.m.c.s than treated samples at the lowest RH studied. In contrast, the resolution of the DVS (which detects mass changes up to 0.1 ppm), gave a very good resolution even at the lowest RH tested, 5% (equivalent to an e.m.c. in untreated wood smaller than 1 %). Save for the lowest relative humidity in the E status, the e.m.c. of heat-treated samples was, in most cases, smaller than the corresponding controls. Large differences were detected at RH higher than 40%. In spite of multiple references that timber with high extractive content generally shows lower hygroscopicity (Ajuong and Breese, 1997 and references given therein), we did not find such behaviour in all the four untreated species reported in this study. For example, control extracted WRC (7.01% average extractive content), showed a slightly lower hygroscopicity compared to unextracted control samples at RH higher than 40% (Figure 2). We attribute this to the extraction procedure, which involved multiple swelling and drying steps to the samples. This process might have induced collapse of the pore structure of the cell wall, which tended to prevent further sorption in the unheated material, at least for the first adsorption cycle (the only one analysed in this work). A noticeable effect on the E samples was the


111 higher reactivity to water and the lower time required to reach equilibrium with the surrounding environment, as described below.

18 16

Moisture adsorbed (%)

14 12 RH 92%

10 8

RH 75%

6 4

RH 43% RH 33%

2

RH 11%

0 0

1

2

3

4

5

6

7

8

9

10

11

Weight loss due to the treatment (%)

Figure 3. Effect of weight loss (WL) on hygroscopicity in Western Red cedar in the extracted (dashed line, open circle) and non-extracted (solid line, filled circle) conditions. Average values of moisture adsorbed are shown.

The major constituents of wood have been listed in the following order of decreasing moisture adsorptivity (Christensen and Kelsey, 1958): hemicelluloses, cellulose and lignin. The differences reflect the lower concentration of hydrophilic groups in lignin compared to carbohydrates and the presence of non-reactive crystalline cellulose, compared to the amorphous nature of the hemicelluloses. Since moisture is adsorbed largely in the amorphous polymers that compose the cell wall, the moisture adsorbed is inversely related to the presence of these substances. Thus, the effect of the heat treatment on the hygroscopicity of wood is closely related to the heat stability of these polymers, and to the chemical composition of the resultant product. As seen in Figure 2, the reduction of hygroscopicity was noticeable from the milder treatments, and from the lowest RH. This indicates that the most hydrophilic substances, namely hemicelluloses, were rapidly modified or decomposed, even at the lowest temperature of treatment (190° C). Hygroscopicity of treated samples varied depending on the temperature of treatment and the relative humidity of exposure (Figure 2). In the case of the heat-treated NE samples, the effect of temperature of treatment was small for the lower range of relative humidities (up to a RH of 40%). At higher RH, the effect changed; whereas the e.m.c reduction of the sample treated at 190° C (compared to the e.m.c. for control) remained fairly constant (to about 30%), the reduction the sample treated at 230° C continued diminishing as the RH increased, reaching a maximum reduction of 41%. This difference is probably due to the continuous breakdown of the hemicelluloses at higher ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

112 temperatures, as denoted by the increasing WL (Figure 4), and also to an incipient modification of the amorphous regions of cellulose. A proportionally larger content of less hydrophilic lignin as wood decomposed is probably also involved in the lower hygroscopicity shown in wood at higher temperatures of treatment. The rate of reduction on hygroscopicity decreased with time of treatment for all the three temperatures tested.

16%

14%

Weight loss (%)

12%

10%

190°C [E] 210°C [E] 230°C [E]

8%

190°C [NE] 210°C [NE] 230°C [NE]

6%

4%

2%

0% 0

50

100

150

200

250

Time (min)

Figure 4. Weight loss (%) against time and temperature of treatment in Western Red cedar in the extracted [E] and non-extracted [NE] conditions. Bars represent ± one standard deviation.

In the case of the E samples treated at 190° C, the reduction in hygroscopicity reduced as the RH increased, in spite of an overall larger WL of the E samples in the treatment at this temperature (Figure 4). When the treatment was at 230° C the hygroscopicity was similar to the NE sample, the reduction in hygroscopicity continuing as the RH increased, but the rate of reduction was smaller than for the NE sample. As explained above, it is believed that the E samples underwent a process of collapse of the cell wall micropores due to the multiple drying and swelling cycles involved in the extraction process. This phenomenon probably led to an increase in the density of the cell wall substance, and thus to an increased weight loss compared to the unextracted material. A higher WL in the E sample did not equate with a lower hygroscopicity probably due to a higher affinity towards water as compared to the NE samples (Figures 5 and 6), and most probably due to an increased internal sorption surface due to the extraction process revealed after the heat-treatment. According to Stamm (1964), the sorption surface area of wood is proportional to its moisture content at a RH of 30%. From data acquired using the DVS, it can be determined that the surface in the E sample treated at 190° C is reduced only by 20% (Figure 2), whereas that for the NE sample was reduced by 30%, despite the disappearance of bulky extractives. This may indicate that the sorption surface for the E sample was not further reduced due to the treatment because the porous matrix had already collapsed due to the extraction, or that a larger adsorption surface due to the extraction process was revelled after the heat treatment. ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

113

Date: 28 March 2004 Time: 1:04 pm File: WRCT11ne.XLS Sample: WRCedarNE-T11

Temp: 25.7 °C Meth: porto.SAO M(0): 10.9

DVS Change In Mass Plot dm

Target RH

12

100 90

10 80 70 60 6

50 40

Target RH (%)

Change In Mass (%)

8

4 30 20

2

10 0

0 0

100

200

300

400

500

600

700

800

900

Time/mins

Cell Wall Porosity Reseach Group

1000

1100

1200

1300

© University of Wales Bangor

Figure 5. Change in mass (%) as a function of time to reach equilibrium moisture content as determined using the DVS. The figure shows the moisture uptake of wood measured as mass change (solid line) at 10 increasing relative humidities (dashed line). NE sample of Western Red cedar after heat-treatment at 230°C for 2 Hr.

100

Relative humidity at equilibrium (%)

90 80 70 60 50 40 Control [NE] 190° C [NE]

30

230° C [NE]

20

Control [E] 190° C [E]

10

230° C [E]

0 0

300

600

900

1200

1500

1800

Time to reach equilibrium (min)

Figure 6. Time to reach the equilibrium moisture content for Western Red cedar in the extracted [E] and nonextracted [NE] conditions. Heat treatments at 190° and 230° C for two hours.


114 For the treatment at 230° C the effect of the extraction procedure is probably masked by the ultrastructural changes that had taken place during the heating treatment, and the real effects of the extractives may be resolved. From the result from this work, it is proposed that the cell wall polymers in the NE samples are decomposed and volatised to a larger extent compared to the E samples (Figure 4) at higher temperatures. A larger decomposition renders the NE samples less hygroscopic than the E samples. This result should be taken with caution, because the study was performed only in the first adsorption cycle. Keith and Chang (1978) found that the first adsorption cycle from heat-treated wood always gave lower e.m.c.s than untreated material, compared to subsequent sorption cycles. The treated material always showed reduced hygroscopicity as compared to control, though. In order to elucidate the possible differences in both sets of samples in the sorption mechanism, a two-hydrate model from the Hailwood and Horrobin (1946) theory (HH) was used to fit the adsorption isotherms. The HH theory assumes the water-polymer system to be a solution. Water adsorbed onto the polymer, M, at equilibrium with the relative pressure h, is assumed to exist in two states: water in solution with the polymer (dissolved water, Ms), and water combined with units of the polymer to form hydrates, Mh. The theory is based on the equilibria between polymer, hydrated polymer and the dissolved water, forming an ideal solid solution. There are two types of equilibria: one between the dissolved water and the water vapour of the surroundings, where K is the equilibrium constant, and n number of equilibria between the dissolved water and hydrates, designated with n equilibrium constants Ki (i= 1...n). A third fundamental constant, Mp, is defined as the molecular weight of the polymer unit that forms a hydrate. According to Simpson (1973), the two-hydrate model gives a larger moisture content required for the complete combination of the polymers to form hydrates in untreated wood, as compared to the single-hydrate model. The model is given by the equation: M =

K Kh + 2 K 1 K 2 K 2 h 2 1800 ⎛ Kh ⎜⎜ + 1 M p ⎝ 1 − Kh 1 + K 1 Kh + K 1 K 2 K 2 h 2

⎞ ⎟⎟ ⎠

Where M, Mp, K, and h are as defined above, and K1 and K2 are the equilibrium between the dissolved water and the first and second hydrates, respectively. This equation was adjusted by nonlinear regression; all the parameters were constrained at a value ≥ 0 (Table 1). Calculations using the parameters in Table 1 for the two-hydrate model, give the maximum Mh for untreated NE Hornbeam of 5.31%, an increase from the 3.56% produced using the single-hydrate water model. In both cases, R2 was > 99.9%. Papadopoulos and Hill (2003) have estimated that at least 8.6 mmol g-1 of OH groups per gram of dry wood exist in unmodified softwood. A moisture content of ca. 5% yields a value of 2.8 mmol g-1, assuming that each molecule of hydrated water is associated with a primary sorption site in the polymer. It appears that the two-hydrate equation from the HH model still underestimates the moisture content required for the complete combination of the polymer with water to form hydrates for both untreated and heat-treated wood. ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

115 However, the following deductions can be drawn from the results from the model: the constant K expresses the activity of dissolved water per unit relative vapour pressure, and its value should be unity if it has the same activity as liquid water (Okoh and Skaar, 1980). The values for K vary between 0.68 and 0.88, indicating that the dissolved water has lower activity than the liquid water in both untreated and heat-treated wood. For several treatments, the constant K2 was assigned the value of zero by the algorithm, indicating that the best fit with non-negative parameters was for the single-hydrate model. For the heat-treated wood, the constant Mp (the molecular weight of the polymer substance necessary to associate with one molecular weight of water), exhibits a tendency to increase as the temperature or the time of treatment increased. According to Spalt (1958), this effect implies that a lower proportion of sites are made available for water sorption. The trend observed for the physical constants K, K1, and Mp obtained in this study were found to be in line and in the same order of magnitude with those previously reported by Spalt (1958) and by Papadopoulos and Hill (2003) for esterified wood. Heat-treated wood always gave a lower Mh than untreated wood, when it was calculated using the parameters of Table 1. An analysis of the hydrated water, Mh, calculated from the HH two-hydrate model for the isotherms obtained in the DVS (Figure 2), indicated that the surface-bound water experienced a similar reduction (37%) irrespective of the temperature of treatment in either set of samples (E or NE). This suggests that the accessible OH groups were almost equally reduced at any temperature, slightly more in the E samples treated at 230° C. In contrast, the dissolved water, Ms, suffered no change at the lower temperature of treatment for the E samples, and diminished with the treatment at 230° C by 27%. For the NE samples the Ms was reduced sharply from the lower temperature (25% with respect to control). At 230° C, the Ms had been reduced by more than 40%. Thus, the difference in the way water was sorbed in the E and NE samples was dictated by the amount of dissolved water held. It is proposed that the reduction in the Ms water occurs due to a stiffening and collapse of the cell wall micropores, rather that by the further inaccessibility of the OH groups. The mechanism of this collapse in the E condition was different to that in the NE condition. Probably, in the NE samples, the collapse was larger due to the loss of extractives during the thermal treatment. As far as the dimensional stability is concerned, the ASE ranged from 6 to 66% for hardwoods, and from 5 to 70% for softwoods. Higher ASE was reached in the NE condition in all cases. Maximum losses in volume were 5.8% for softwoods and 16.0% for hardwoods. It is proposed that the reduction in hygroscopicity is in part an artefact of the reduction of the density of the material.


116 Table 1. Mean values for experimental e.m.c. at five levels of relative humidity produced over saturated salt solutions for untreated and heat-treated wood samples in the E and NE conditions (Std. Dev. in parentheses), and physical constants from the Hailwood-Horrobin model (K, K1, K2 and Mp). See text for abbreviations.

Treatment

11%

33%

43%

75%

92%

K

K1

K2

Mp

Beech-E

Control E 190 °C 2 Hr 210 °C 1 Hr 210 °C 4 Hr 230 °C 2 Hr

1.4 (0.18) 1.8 (0.23) 1.7 (0.27) 1.5 (0.20) 1.6 (0.95)

3.4 (0.16) 3.9 (0.24) 3.0 (0.55) 2.9 (0.59) 2.3 (0.15)

4.9 (0.44) 4.9 (0.09) 4.1 (0.41) 3.5 (0.95) 3.2 (0.23)

10.7 (0.17) 10.5 (0.52) 8.5 (0.19) 7.4 (0.34) 6.9 (0.38)

18.0 (0.68) 17.3 (0.38) 14.7 (0.71) 11.8 (0.05) 10.7 (0.40)

0.811 0.824 0.858 0.819 0.812

1.928 3.909 5.730 5.438 4.792

0.000 0.000 0.000 0.000 0.000

352.5 404.7 558.2 586.4 622.4

Control NE 190 °C 2 Hr 210 °C 1 Hr 210 °C 4 Hr 230 °C 2 Hr

1.7 (0.33) 2.0 (0.75) 1.1 (0.43) 1.2 (0.33) 1.2 (0.41)

4.2 (0.53) 3.6 (0.18) 3.0 (0.05) 2.5 (0.03) 2.4 (0.26)

5.4 (0.15) 4.7 (0.22) 3.9 (0.11) 3.2 (0.15) 3.4 (0.36)

11.7 (0.43) 10.2 (0.03) 7.9 (0.29) 7.0 (0.31) 6.3 (0.26)

20.3 (0.35) 17.0 (0.10) 13.6 (0.37) 11.8 (0.37) 10.5 (0.02)

0.844 0.838 0.854 0.834 0.853

3.434 4.736 4.189 3.539 6.729

0.000 0.000 0.532 0.000 0.721

372.8 439.4 627.5 613.1 834.1

H Beam-E


1.3 (0.32) 1.8 (0.25) 1.8 (0.40) 1.2 (0.30) 1.3 (0.19)

3.7 (0.15) 3.8 (0.23) 2.4 (0.46) 2.7 (0.08) 2.7 (0.28)

5.1 (0.12) 5.2 (0.46) 3.9 (0.28) 3.8 (0.03) 3.4 (0.20)

11.2 (0.08) 10.4 (0.22) 8.1 (0.16) 7.8 (0.33) 7.2 (0.28)

19.5 (0.08) 17.7 (1.01) 14.5 (0.16) 12.3 (0.41) 11.4 (0.40)

0.844 0.837 0.875 0.782 0.800

2.104 4.489 5.603 2.567 3.791

0.549 0.052 0.000 0.000 0.000

405.7 424.0 614.6 468.1 557.1

H Beam-NE


1.3 (0.13) 1.5 (0.26) 1.3 (0.18) 1.1 (0.03) 0.8 (0.12)

4.3 (0.14) 4.0 (0.83) 3.0 (0.20) 2.7 (0.12) 3.0 (0.18)

5.5 (0.15) 4.6 (0.04) 4.1 (0.21) 3.5 (0.24) 3.8 (0.05)

12.0 (0.23) 10.1 (0.15) 8.0 (0.10) 7.5 (0.18) 7.2 (0.25)

21.1 (0.29) 17.5 (0.15) 14.0 (0.08) 12.1 (0.22) 11.1 (0.15)

0.878 0.845 0.882 0.797 0.839

2.021 4.119 4.758 2.623 1.460

2.095 0.000 1.464 0.000 8.055

472.3 438.5 730.5 505.3 815.5

WR Cedar-E


1.3 (0.05) 2.5 (0.51) 1.8 (0.42) 1.5 (0.46) 1.2 (0.33)

4.6 (0.48) 4.2 (0.67) 3.4 (1.29) 2.9 (0.33) 2.0 (0.21)

4.9 (0.07) 4.9 (0.41) 4.3 (0.08) 4.1 (2.00) 2.6 (0.30)

9.7 (0.63) 9.1 (0.16) 8.8 (0.62) 7.6 (0.30) 7.0 (0.54)

15.2 (0.30) 15.2 (0.92) 14.2 (0.24) 12.0 (1.33) 11.3 (1.08)

0.855 0.861 0.819 0.807 0.773

2.581 21.42 5.271 5.828 0.898

6.648 0.509 0.000 0.151 0.000

638.9 598.0 488.3 567.7 452.2

WR Cedar-NE


1.0 (0.12) 1.5 (0.60) 0.9 (0.20) 0.8 (0.54) 1.0 (0.79)

3.5 (0.44) 3.1 (0.39) 2.8 (0.30) 2.8 (0.46) 2.4 (0.59)

4.1 (0.38) 4.4 (0.18) 4.1 (0.13) 2.8 (0.18) 3.1 (0.23)

9.2 (0.46) 8.4 (0.24) 7.4 (0.16) 7.1 (0.49) 5.9 (0.80)

15.2 (0.53) 13.5 (0.37) 11.9 (0.40) 10.5 (0.16) 9.8 (0.37)

0.820 0.811 0.852 0.684 0.855

2.542 4.738 1.618 0.972 5.421

0.435 0.161 6.464 0.000 1.388

469.8 512.0 793.5 355.9 931.9

D. Fir-E


1.5 (0.20) 2.5 (0.30) 2.3 (0.46) 1.8 (0.25) 1.7 (0.33)

4.0 (0.43) 4.2 (0.26) 3.2 (0.09) 3.8 (0.29) 2.8 (0.46)

4.6 (0.20) 4.9 (0.30) 4.7 (0.21) 3.4 (0.22) 3.4 (0.05)

9.2 (0.74) 9.9 (0.52) 9.6 (0.41) 8.0 (0.47) 7.2 (0.31)

13.9 (0.36) 14.5 (0.19) 14.2 (0.42) 12.8 (0.72) 11.7 (0.27)

0.768 0.791 0.787 0.833 0.844

4.255 9.367 5.100 9.11 8.244

0.000 0.065 0.000 0.000 0.000

407.0 443.8 430.0 581.5 666.0


1.6 (0.37) 1.8 (0.23) 1.4 (0.63) 1.4 (0.34) 1.4 (0.16)

3.9 (0.48) 4.6 (0.84) 3.3 (0.41) 3.1 (0.73) 2.8 (0.26)

5.3 (0.52) 5.0 (0.09) 4.5 (0.38) 3.8 (0.39) 4.0 (0.65)

10.2 (0.49) 9.8 (0.41) 8.8 (0.28) 7.9 (0.25) 7.0 (0.32)

15.8 (0.11) 15.2 (0.43) 13.5 (0.32) 12.3 (0.05) 11.1 (0.45)

0.786 0.837 0.759 0.798 0.833

3.739 6.648 3.101 4.287 6.996

0.434 1.891 0.000 0.000 1.141

407.5 571.2 400.7 512.9 747.9

Beech-NE

Sp.

D. Fir-NE

Equilibrium moisture content (%) at 25 °C at RH:


117

CONCLUSIONS

In addition to possible reductions on the amount of VOCs typically emitted by untreated wood in service, heat treatments described in this work offer significant improvements in moisture-related properties of British woods, for its potential use in indoors applications. The treatments reduced the hygroscopicity of wood and the amount of swelling that it undergoes, and changed the rate at which it changes moisture content. Reduction of wood hygroscopicity was rapid using the lower treatment temperature. It was found that the actual reduction should be determined at a relative humidity higher than 40%, where the real differences between treatments are apparent. The effect of the extractives on the hygroscopicity of heat-treated wood was somehow overridden by the extraction process; the non-extracted samples showed a slightly larger reduction in hygroscopicity than the extracted set did.

ACKNOWLEDGMENTS

The first author wishes to thank the Mexican Council for Science and Technology (CONACYT) for a PhD bursary. Thanks are due to Coed Cymru (Welsh Woods), for kindly providing part of the wood used for this study.

REFERENCES

Ajuong, E.-M. A. and M. C. Breese (1997). "The role of extractives on short-term creep in compression parallel to the grain of Pai wood (Afzelia africana Smith)." Wood and Fiber Science 29 (2): 161-170. Bekhta, P. and P. Niemz (2003). "Effect of high temperature on the change in color, dimensional stability and mechanical properties of spruce wood." Holzforschung 57 (5): 539-546. Bourgois, J., M. C. Bartholin, et al. (1989). "Thermal-Treatment of Wood - Analysis of the Obtained Product." Wood Science and Technology 23 (4): 303-310. Christensen, G. N. and K. E. Kelsey (1958). "The sorption of water vapour by the constituents of wood: determination of sorption Isotherms." Australian Journal of Applied Science 9: 265-282. Evans, P. (2003). "Emerging technologies in wood protection." Forest Products Journal 53 (1): 14-22. Farahani, M. R. M., C. A. S. Hill, et al. (2001). The effect of heat treatment on the decay resistance of Corsican pine sapwood. 5th European Panel Products Symposium, Llandudno, North Wales, the Biocomposites Centre. pp 303-308.


118 Garrote, G., H. Dominguez, et al. (1999). "Hydrothermal processing of lignocellulosic materials." Holz Als Roh-Und Werkstoff 57 (3): 191-202. Gohar, P. and R. Guyonnet (1998). Development of wood rectification process at the industrial stage. 4th Symposium International on Wood Preservation, Cannes, France, The International Group in Wood Preservation. 6 p. Hailwood, A. J. and S. Horrobin (1946). "Absorption of water by polymers: analysis in terms of a simple model." Transactions of the Faraday Society 42B: 84-102. Hill, C. A. S. and D. Jones (1996) "The dimensional stabilisation of Corsican pine sapwood by reaction with carboxylic acid anhydrides - The effect of chain length" Holzforschung 50 (5): 457-462. Keith, C. T. and C. I. Chang (1978). Properties of heat-darkened wood. I Hygroscopic properties. Montreal, Eastern Forest Products Laboratory: 20 p. Manninen, A. M., P. Pasanen, et al. (2002). "Comparing the VOC emissions between air-dried and heat-treated Scots pine wood." Atmospheric Environment 36 (11): 17631768. Militz, H. (2002). Thermal treatment of Wood: European Processes and their background. 33th Annual Meeting of The International Group in Wood Preservation, Cardiff, Wales, The International Group in Wood Preservation. 19 p. Nuopponen, M., T. Vuorinen, et al. (2003). "The effects of a heat treatment on the behaviour of extractives in softwood studied by FTIR spectroscopic methods." Wood Science and Technology 37 (2): 109-115. Okoh, K. I. A. and C. Skaar (1980). "Moisture sorption isotherms of the wood and inner bark of ten southern U. S. hardwoods." Wood and Fiber 12 (2): 98-111. Papadopoulos, A. N. and C. A. S. Hill (2003). "The sorption of water vapour by anhydride modified softwood." Wood Science and Technology 37 (3-4): 221-231. Rajasuo, P. (2002). Heat-treated wood - Product classification and quality control in Finland. COST Action E22 Environmental Optimisation of Wood Protection. Working Group 3 Report. Workshop and Meetings in Tuusula, Finland, 3-4 June 2002. 14 p. Risholm-Sundman, M., M. Lundgren, et al. (1998). "Emissions of acetic acid and other volatile organic compounds from different species of solid wood." Holz Als Roh-Und Werkstoff 56 (2): 125-129. Simpson, W. (1973). "Predicting equilibrium moisture content of wood by mathematical models." Wood and Fiber 5 (1): 41-49. Sing, K. S. W. (1998). "Adsorption methods for the characterization of porous materials." Advances in Colloid and Interface Science 76-77: 3-11. Spalt, H. A. (1958). "The fundamentals of water sorption by wood." Forest Products Journal 11: 288-295.


119 Syrjänen, T. and E. Kangas (2000). Heat treated timber in Finland. 31th Annual Meeting of The International Group in Wood Preservation, Kona, Hawaii, USA, The International Group in Wood Preservation. 9 p. Tiemann, H. D. (1920). The kiln drying of lumber. Philadelphia, J.B. Lippincott Company. pp 188-190. Vernet-Maury, E., O. Alaoui-Ismaïli, et al. (1999). "Basic emotions induced by odorants: a new approach based on autonomic pattern results." Journal of the Autonomic Nervous System 75 (2-3): 176-183. Welzbacher, C. R. and A. O. Rapp (2002). Comparison of thermally modified wood originating from four industrial scale processes - durability. 33th Annual Meeting of The International Group in Wood Preservation, Cardiff, Wales, The International Group in Wood Preservation. 13 p. Zaman, A., R. Alen, et al. (2000). "Thermal behaviour of Scots pine (Pinus sylvestris) and silver birch (Betula pendula) at 200-230 degrees C." Wood and Fiber Science 32 (2): 138-143.


120


121

Session 7 Nonwood-Fibre Products


122


123

MINIMIZING ENVIRONMENTAL BURDEN OF OIL PALM TRUNK RESIDUES THROUGH DEVELOPMENT OF LAMINATED VENEER LUMBER PRODUCT Kamarulzaman Nordin1, Mohd Ariff Jamaludin1, Mansur Ahmad1, Hashim W. Samsi2, Abdul Hamid Salleh2 and Zaihan Jalaludin2 1

Department of Furniture Technology, Faculty of Applied Sciences, Universiti Teknologi MARA 40450 Shah Alam, Selangor, MALAYSIA ([email protected]) 2

Product Development Division, Forest Research Institute Malaysia (FRIM) Kepong, 52109 Kuala Lumpur, MALAYSIA ([email protected])

SUMMARY Being the largest producer of palm oil with a global market share of about 50%, the Malaysian palm oil industry is set to grow even more in the coming decades with everincreasing oil palm plantation areas. As such, substantial amount of residue is expected to originate each year from oil palm re-plantation. Similar to the forest products industry, environmental concerns in the palm oil industry are also becoming serious issues that need stern consideration by the stakeholders. In ensuring the future growth of Malaysian palm oil industry, efficient use of field residues is therefore in need to minimize the environmental burdens associated with the disposal of the oil palm residues. In this paper, initiative that have been undertaken to utilize residues from oil palm re-plantation, particularly the oil palm trunk (OPT) for the production of laminated veneer lumber (LVL) was described. The aim was to access the bending and compression strength of the OPT LVL and to compare them with Malaysian oak (rubberwood), timber species that is commonly used in the manufacture of furniture in Malaysia. Development effort to further improve the strength properties of the OPT LVL was also discussed. OPT LVL was found to have comparable bending and compression strength to solid Malaysian oak. With such promising findings, the palm oil industry would benefit through development of by-products with higher valueadded while continuing efforts to reduce the overall environmental burden and placed the industry on a new environmentally sustainable platform.

INTRODUCTION

The oil palm tree (Elaeis guineensis) which was introduced in 1917 is one of the most important commercial crops in Malaysia and has brought in enormous sums of foreign exchange in export earnings which accounted for 4.0% of the nation’s total export of merchandise in the year 2000 (Ministry of Primary Industry, 2001). Also known as the “golden crop”, it has positioned Malaysia as the leading nation in oil palm production, being the largest producer and exporter of palm oil in the world with a market share of about 50% and 58% respectively (Mohd Nasir, 2003). With the increasing trend of world’s demand for oils and fats (Yusof, 2002), the oil palm industry in Malaysia has an important role to play in fulfilling the growing global need for oils and fats. This has led to the rapid growth of oil palm plantation area in the country which correspondingly saw growth in oil palm biomass residues being generated throughout harvesting and ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

124 processing activities as well as during replanting. It is reported that in 1997, Malaysia produced about 13.2 million tonnes of oil palm biomass, including trunks, fronds, and empty fruit bunches (Kamaruddin et al., 1997). This figure is expected to increase substantially when the total planted hectarage of oil palm in Malaysia reached 5.10 million hectares in 2020, as indicated by Jalani et al., (2002). A recent figure indicated that oil palm plantation areas in Malaysia has expanded from 3.5 million hectares in 2001 to 3.8 million hectares in 2003 (Malaysia Oil Palm Board, 2004). Environmental concerns in the palm oil industry are also becoming serious issues that need stern consideration by the stakeholders. As an industry that has demonstrated a high level of attention to its own environmental impact, the Malaysian oil palm industry is poised well to meet the challenges facing it in the new globalize era and paradigm shift in trade. Over the years, the oil palm industry has cooperated well with the World Wide Fund for Nature (WWF) global initiative towards more sustainable palm oil production (Andrew et al., 2003) and has developed many sound practices to minimize the negative environmental burdens of the oil palm industry such as the adoption of Better Management Practices (BMPs) which are regarded as standard operational practices. The implementation of zero burning technique for replanting on a commercial scale (Mohd Hashim et al., 1993) and in new plantings (Jamaludin et al., 1999; Ramli, 1999) has been a major factor in minimizing air pollution by plantations. In line with the need to preserve a clean environment as well as to achieve a vision of zero-waste strategy in the Malaysian palm oil industry, research and development activities are also focused on the utilization of oil palm biomass such as empty fruit bunches (EFB), oil palm fronds (OPF) and oil palm trunks (OPT). This oil palm biomass has great potential to be converted into high value-added and useful incomegenerating products. Numerous research and development efforts undertaken on the utilization of EFB contemplated mainly on the production of pulp for paper making (Tanaka et al., 2002; Astimar et al., 2002) while a handful can also be found on the production of medium-density fiberboard (Ridzuan et al., 2002), oil palm fiber mattress and agricultural mats, high quality organic fertilizer, charcoal briquette and roof tiles (Mohamad Husin et al., 2002). Intensive research work in generating technologies to convert OPF and OPT for the manufacture of commercially viable composite panel products have also proven successful in most cases. The production of medium-density fiberboard (Laemsak and Okuma, 2000), particleboard (Chew, 1987) and cementbonded particleboard (Kochummen et al., 1990), fiber reinforced cement board (Abraham et al., 1998), fiber plastic composite (Liew et al., 2002) and plywood (Ho et al., 1985) from OFB and OPT have shown to be technically feasible. The production of blockboard (Mohamad et al., 2001) as well as furniture (Mohamad et al., 1989) from OPT lumber have also been investigated with promising potentials. Some of those above-mentioned processes are being commercialized through pilot plant studies and small-scale production evaluations with the help of companies in the private sector. Despite all these extensive research work on oil palm biomass, study on the utilization of these residues particularly the OPT for the production of laminated veneer lumber (LVL) could not be traced in the literature. As a result, Universiti Teknologi MARA (UiTM) and Forest Research Institute Malaysia (FRIM) have taken the initiative to explore the potentials of OPT for the manufacture of LVL with the hope to develop a product with high value-added and simultaneously help reduced the overall environmental burden and placed the industry on a new environmentally sustainable ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

125 platform. In this paper, only the properties of OPT LVL in terms of bending and compression strength were discussed. Other issues related to the manufacture of OPT LVL such as surface roughness, wettability, surface tension, as well the evaluation of product made from OPT LVL will be addressed in a companion article.


Samples of mature OPT logs of approximately 30 years old were obtained from a replantation site in Perak and were transported to a furniture factory in Sungai Petani, Kedah. The factory had wide experience producing veneer and LVL from Malaysian oak (formerly known as rubberwood) for its furniture production. The OPT logs were peeled to 4.0 mm thick veneers and were later dried to approximately 7% moisture content. The 500 mm wide × 780 mm long × 25mm thick LVL, which consisted of 9 layers of veneer, was bonded together using urea formaldehyde (UF) glue. Three different configurations of OPT LVL were produced as illustrated in Figure 1. They were LVL made entirely of OPT veneers (OPT LVL), hybrid OPT LVL with 2 layers of Malaysian oak veneers (OPT-2MO LVL) as well as the one with 3 layers of Malaysian oak veneers (OPT-3MO LVL). In order to evaluate the strength properties of the manufactured LVL, bending and compression strength tests were conducted according to the standard as stipulated in Japanese Agricultural Standard for LVL JAS: SE-10. A total of thirty LVL bending specimens of 90 mm wide × 550 mm long and compression specimens of 25 mm wide × 50.8 mm long were cut from the LVL panel manufactured and were conditioned to a moisture content of 10 to 12% in an environment of about 200C and 65% relative humidity (RH) before they were tested for strength in bending and compression. Three-point loading flatwise static bending tests were conducted on each of the specimens prepared until failure occurred and the bending modulus of rupture (MOR) and modulus of elasticity (MOE) were calculated. Compression perpendicular to the grain tests were also conducted to failure prior to calculating the compression strength.

(a)

(b)

(c) OPT veneers

Malaysian oak veneers

Figure 1 : Three different configurations of OPT LVL produced. (a) OPT LVL (b) OPT-2MO LVL and (c) OPT-3MO LVL.


126 RESULTS AND DISCUSSION

The bending and compression strength results for the LVL are presented in Table 1. The mean bending modulus of rupture (MOR) and modulus of elasticity (MOE) for the OPT LVL board were 41.0 MPa and 4,219 MPa, respectively. In comparison with LVL made with the inclusion of Malaysian oak veneers, the results of this study showed that the bending properties of the LVL were greatly improved. For OPT-2MO LVL, the mean MOR and MOE values escalated to 49.1 MPa and 6,828 MPa, which shows an increase of about 19.8% and 61.8%, respectively. The inclusion of another layer of Malaysian oak veneers (OPT-3MO LVL) markedly enhanced the bending properties of the LVL with the MOR and MOE values improved to 59.7 MPa and 7,613 MPa, an increase of approximately 45.6% and 80.4%, respectively. Analysis of variance (ANOVA) test conducted on the mean bending MOR and MOE values between LVL boards of the three different configurations revealed that significant different exists between them. Comparison-of-means results in Figure 2 show that the OPT-3MO LVL boards undoubtedly produced the highest strength in terms of MOR and MOE and were significantly higher than the OPT-2MO LVL and OPT LVL boards. Boards made with the inclusion of 2 layers of Malaysian oak veneers (OPT-2MO LVL) apparently produced significantly greater bending strength than those made entirely of OPT veneers (OPT LVL). Thus, the inclusion of Malaysian oak veneers significantly improved the OPT LVL bending properties. Table 1: Mean bending and compression strengths of OPT LVL with three different configurations. Bending strength Material

No. of samples

OPT LVL

Comp. strength (MPa)

+ (%)

Ovendry density (kg/m3)

MOR (MPa)

+ (%)

MOE (MPa)

+ (%)

30

41.0 (7.8)

-

4,219 (774)

-

21.1 (2.5)

-

545 (32.2)

OPT-2MO LVL

30

49.1 (4.2)

19.8

6,828 (319)

61.8

26.6 (2.9)

26.1

572 (29.1)

OPT-3MO LVL

30

59.7 (5.4)

45.6

7,613 (533)

80.4

31.6 (3.3)

49.8

589 (20.0)

Note: Values in parentheses denote standard deviations.

In terms of compression strength, a similar trend to bending was also observed which saw a substantial improvement of the mean compression strength, as additional layers of Malaysian oak veneers were included in the OPT LVL. The OPT LVL with the inclusion of Malaysian oak veneers were found to be 49.8% (OPT-3MO LVL) and 26.1% (OPT-2MO LVL) stronger than LVL made entirely of OPT veneers (Table 1). The compression strength between the three different configurations of LVL boards was significantly different based on ANOVA test. The improvement in bending and compression strength could be attributed to the densification of the LVL board as a result of the inclusion of more uniform and higher density Malaysian oak veneers during manufacturing. This is evident from Table 1, which saw the increase in density ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

127 of the LVL board when additional layers of Malaysian oak veneers were included. Other factors such as the introduction of UF glue, and dispersion or removal of strength reducing veneer characteristics could perhaps be the contributing factor as well.

80.00

a b

Bending and Compression Strength (MPa)

70.00

60.00

50.00

a

OPT-3MO LVL OPT-2MO LVL

b

OPT LVL c

c 40.00

a 30.00

b c

20.00

10.00

0.00 Bending MOR

Bending MOE (X 100)

Compression

Means followed by the same letter are not significantly different Figure 2: Comparison-of-means tests results for bending MOR and MOE and compression strength between three different configurations of OPT LVL.

Another interesting finding that needs to be highlighted is the reduction in strength properties variation. Being a monocotyledonous species, OPT properties were stipulated of having great variations between outer parts of the stems and towards the centre of the stem (Lim and Khoo, 1986). In contrasts to solid OPT lumber (Table 1 and 2), this study indicated that, by utilizing OPT in the form of LVL, the variations between boards was significantly reduced, implicating that more uniform board properties could be produced. The ANOVA test conducted confirmed that there was no significant difference between boards in terms of the mean bending MOR and MOE values. Furthermore, with the inclusion of a few layers of Malaysian oak veneers, more promising results were achieved. This could potentially make the OPT wastes technically feasible for the production of LVL. In comparison to solid Malaysian oak (Table 1 and 2), with the exception of bending MOE, the bending MOR and the compression strength of the OPT LVL with 3 layers of Malaysian oak veneers produced slightly higher values than solid Malaysian oak. Despite having lower bending strength, the compression strength of the OPT LVL with 2 layers of Malaysian oak veneers was very much comparable to solid Malaysian oak. ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

128 Such promising results presented in this study showed that with further research and development efforts, that is still underway, OPT veneers have the potential to be utilized in combination with Malaysian oak veneers for the production of LVL. Table 2: Bending and compression strength of solid OPT and Malaysian oak. Bending strength MOR MOE (MPa) (MPa)

Material

*

Compression strength (MPa)

Oven-dry density (kg/m3)

Solid OPT lumber*

8 – 45

800 - 8,000

5 – 25

220 – 550

Solid Malaysian oak lumber*

58

8,800

26

530

Killman and Lim (1985)

CONCLUSIONS

This study showed that the properties of OPT LVL were improved by combining OPT veneers with several layers of Malaysian oak veneers during the process of LVL manufacturing. Such combinations have resulted in the improvement of bending and compression strength of LVL produced entirely from OPT. In addition, such practice also produced LVL board with far less variation in strength properties as compared to solid OPT properties. The OPT LVL produced from this study was found to possess bending and compression strength comparable to solid Malaysian oak. With further research and development, overall performance of the OPT LVL could be improved for commercial utilization of OPT wastes for LVL manufacturing in the near future.

ACKNOWLEDGEMENTS

The authors are grateful to Universiti Teknologi MARA (UiTM) and Forest Research Institute Malaysia (FRIM) for their continuous support throughout this study.

REFERENCES

ABRAHAM JM, ZAKARIA MA, MOHD NOR MY and SIMATUPANG MH (1998): Suitability of Kraft Pulp from Oil Palm trunk for cellulose fiber reinforced cement boards: Journal of Tropical Forest Products 4 (2): 159-165. ANDREW N, BELLA R and THOMAS V (2003): Responding to Global Demands for Sustainable Palm Oil: Industry-WWF Collaboration: International Planters Conference, 16-17 June 2003. 18 pp. ASTIMAR AA, MOHAMAD H and ANIS M (2002): Preparation of Cellulose from Oil Palm Empty Fruit Bunches via Ethanol Digestion: Effect Of Acid and Alkali Catalysts: Journal of Oil Palm Research 14 (1): 9-14.


129 CHEW LT (1987): Particleboard Manufactured from Oil Palm Stems: A Pilot Scale Study: FRIM Occasional Paper No. 4, Forest Research Institute Malaysia (FRIM), Kepong, 8 pp. HO KS, CHOO KT and HONG LT (1985): Processing, Seasoning, and Protection of Oil Palm Lumber: Proc. of the National Symposium on Oil Palm by-products for AgroBased Industries, Kuala Lumpur, Malaysia, November 1985. JAMALUDIN N, MOHD PADZIL A and RAHMAN S (1999): Zero Burning in Jungle to Oil Planting: Proceedings of the PORIM International Palm Oil Congress on Emerging Technologies and Opportunities in the Next Millennium, Palm Oil Research Institute Malaysia (PORIM), Kuala Lumpur, Malaysia, pp. 243 -251. JALANI S, YUSOF B, ARIFFIN D, CHAN KW and RAJANAIDU N (2002): Prospects of Elevating National Oil Palm Productivity: A Malaysian Perspective: Oil Palm Industry Economic Journal 2 (2): 1-9. KAMARUDDIN H, MOHAMAD H, ARIFFIN D and JALANI S (1997): An estimated availability of oil palm biomass in Malaysia: PORIM Occasional Paper No. 37, Palm Oil Research Institute Malaysia (PORIM), Bangi, 100 pp. KILLMAN W and LIM SC (1985): Anatomy and Properties of Oil Palm Stem: Proc. of the National Symposium on Oil Palm by-products for Agro-Based Industries, Kuala Lumpur, Malaysia. KOCHUMMEN AM, WONG WC and KILLMAN W (1990): Manufacture of Cement Board using Oil Palm Stems: Unpublished IDRC Final Report. LAEMSAK N and OKUMA M (2000): Development of boards made from oil palm frond II: Properties of binderless boards from steam-exploded fibers of oil palm frond: Journal of Wood Science 46 (4): 322-326. LIEW KC, JALALUDDIN H, PARIDAH MT, KHAIRUL ZAMAN MD and MOHD NOR MY (2000): Properties of oil palm frond-polypropylene composite: Proc. of the Utilization of Oil Palm Tree-Oil Palm Biomass: Opportunities and Challenges in Commercial Exploitation pp. 116-118. LIM SC and KHOO KC (1986): Characteristics of Oil Palm Trunk and Its Potential Utilization: The Malaysian Forester 49 (1): 3-22. MALAYSIA OIL PALM BOARD (2004): Oil Palm planted area according to state in hectares: Malaysian Oil Palm Statistics 2003: Retrieved from: http://161.142.157.2/home2/home/rev03_area2.html MINISTRY OF PRIMARY INDUSTRIES (2001): Statistics on Commodities 2001: Ministry of Primary Industries Malaysia. MOHAMAD H, RIDZUAN R, ANIS M, WAN HASAMUDIN WH, KAMARUDDIN H, ROPANDI M and ASTIMAR AA (2002): Research and Development of Oil Palm Biomass Utilization in Wood-based Industries: Palm Oil Developments No. 36: 1-5. ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

130

MOHAMAD H, ABDUL HALIM H and REDZUAN R (2001): Blockboard from Oil Palm trunk: MPOB Information Series, MPOB TT No. 110, 2 pp. MOHAMAD H, ABDUL HALIM H and RIDZUAN R (1989): Manufacture of furniture from oil palm trunk: Proceedings of the PORIM International Palm Oil Development Conference, 5-9 September, Palm Oil Research Institute of Malaysia, Kuala Lumpur, Malaysia. MOHD HASHIM T, TEOH CH, KAMARULZAMAN A and MOHD ALI A (1993): Zero Burning - An Environmentally Friendly Replanting Technique: In PORIM International Palm Oil Congress, Update and Vision (Jalani Sukaimi et al., eds.), Palm Oil Research Institute of Malaysia, Kuala Lumpur, pp. 185–194. MOHD NASIR A (2003): Palm Oil Products Exports, Prices and Export Duties: Malaysia and Indonesia Compared: Oil Palm Industry Economic Journal 3 (2): 21-31. RAMLI AM (1999): Adopting the Zero Burning Technique in New Clearings: Mentiga’s Experience: The Planters 75 (881): 391–401. RIDZUAN R, STEPHEN S and MOHD ARIFF J (2002): Properties of Medium Density Fiberboard from Oil Palm Empty Fruit Bunch Fiber: Journal of Oil Palm Research 14 (2): 34-40. TANAKA R, PENG LC and WAN ROSLI WD (2002): Preparation of cellulose pulp from oil palm empty fruit bunches (EFB) by processes including pre-hydrolysis and ozone bleaching: Proc. of the USM– JIRCAS Joint International Symposium ~ Lignocellulose - Material of the Millennium: Technology and Application, Penang, Malaysia, pp. 33-38. YUSOF B (2002): Palm Oil and Its Global Supply and Demand Prospects: Oil Palm Industry Economic Journal 2 (1): 1-10.


131

MANUFACTURING OF FIBER COMPOSITE MEDIUM DENSITY FIBERBOARDS (MDF) BASED ON ANNUAL PLANT FIBER AND UREA FORMALDEHYDE RESIN Sören Halvarsson 1, 2, Magnus Norgren 1, Håkan Edlund 1 1

Department of Natural and Environmental Sciences, FSCN, Mid Sweden University, SE-851 70 Sundsvall, Sweden ([email protected]) 2

Metso Panelboard AB Department of Research, Technology and Development (RTD) SE-851 50 Sundsvall, Sweden ([email protected])

SUMMARY Production of fiber composite materials such as Medium Density Fiberboard (MDF) and particleboard (PB) is in general based on wood as a raw material. However, cereal straws and other annual agriculture waste materials have regained an interest as a potential raw material for production of MDF. The cereal straws are among the most common lignocellulosic materials that are easily accessible, nonexpensive and renewable. The aim of this investigation was to produce high performance MDF based on wheat straw and urea formaldehyde (UF) resin. The usage of UF-resin for wheat straw MDF-panels has so far resulted in acceptable strength properties but poor moisture resistance and thickness swelling (TS). Application of melamine modified UF-resin for wood based MDF has improved the moisture resistance of produced MDF panels. In this investigation two commercial melamine modified UF-resins were used as binders (adhesives) in the production of wheat straw MDF. Hammer milled wheat straw was treated with water and sulfuric acid (0.6 %) before refining. The reason was to improve the curing conditions of the UF-resins by a reduction of the pH and the pH-buffering capacity of refined wheat straw fiber. Refining of wheat straw was performed at slightly lower pressure and retention time compared with refining of wood material. However, a lot of fines and dust (wheat straw fibers < 0.5 mm) were generated during refining. A high resin content of the melamine modified UF-resin was necessary (15 %) to compensate for the high ratio of wheat straw fines and dust. Final panel properties of wheat straw MDF could meet the requirements of the MDF standard (EN 622-5:1997), including the TS. Strength properties as internal bond (IB) and modulus of rupture (MOR) were increased as a function of density. Thickness swelling was reduced as a function of density. The usage of wheat straw as a raw material in combination with a melamine modified UF-resin, as an adhesive, is a possible route for manufacturing of high performance Medium Density Fiberboard.

INTRODUCTION Annual plant fiber as an alternative raw material for production of fiber composite materials such as Medium Density Fiberboard (MDF) and particleboard (PB) has got a regained interest (Metso Panelboard, 2002). The increased cost for wood based raw materials and environmental considerations have lead to an intensive speculation of new and suitable lignocellulosic materials for production of MDF. The most frequent referred alternative non-wood materials are flax, bagasse, hemp, reed and cereal straws ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

132 such as rice straw and wheat straw (Younquist et al. 1994), (Russell CW, 1996), (Boyd AL, 2001). A common renewable fiber material for production of composite materials is wheat straw. Wheat straw is growing in the temperate climate zone and is one of the major cereals in several regions globally. In these areas wheat straw have a central roll in the agriculture life. However, some attentions have occurred concerning the environmental issues due to the troublesome handling of agriculture waste material. In several places the agriculture residuals is normally burned and locally air pollution has become a problem. Waste disposal of agriculture waste materials is also troublesome due to limited disposal area and a relatively long time for natural enzymatic degradation of agriculture waste materials. In the past several attempts have been made to produce MDF based on wheat straw and urea-formaldehyde UF-resin. The manufacturing of wheat straw MDF and PB in combination with UF-resin has resulted in panels with mediocre panel properties (Sauter SL, 1996), (Markessini and Roffael, 1997). Strength- and bending properties, internal bond (IB), modulus of rupture (MOR) and modulus of elasticity (MOE) have resulted in mediocre panel properties and close to the MDF standard (EN 622-5:1997). The major disadvantage of straw-based Fiberboard products was the unacceptable water uptake or thickness swelling (TS) (McLauchlin and Hauge, 1998), (Eroglu and Istek, 2000), (Mo et al, 2003) The troublesome TS of wheat straw panels is in general explained by an improper surface wetting of the applied resin on straw materials due to a waxy layer on the surface (Sauter SL, 1996), (Markessini and Roffael, 1997). The thin waxy surface layer on wheat straw results in reduced resin/water absorption and was thought to restrict the usage of water based resins such as urea-formaldehyde and phenol formaldehyde resins. However, pressurised refining of wheat straw will at appropriate processing conditions disintegrates the thin waxy layer and removes extractives from refined fiber. The improved surface wetting and resination seemed to result in enhanced board properties (Lawther et al, 1996), (Han and Kawai, 2001). Another general accepted method to reduce thickness swelling of MDF and PB is to add a hydrophobic component such as wax or to increase the hydrophobic component in UF-resin by modification with melamine. In some investigations the poor panel properties have been associated with an unfavourable pH and pH buffering capacity of the straw material or refined fibers. A higher pH and pH buffering capacity can be observed for wheat straw compared with wood. In general, a reduced curing rate of an UF-resin is expected for resinated fiber at pH between 7 and 9 or in combination with a high pH-buffering capacity (Johns and Niazi, 1980), (Wasylciw W, 2001), (Johns et al, 1985). The ultrastructure of wheat straw is to some extent more complicated than wood. A relatively large number of different cell types and cell elements can be found, including reinforcing fibers, parenchyma cells, vessel elements and epidermal cells. The straw is also growing during one season and develops shorter fibers and thinner cell walls compared with perennial wood organisms (Zhai and Lee, 1989), (Donaldson et al. 2001). Therefore, the potential of generate dust during refining is higher for annual plant materials than for wood materials. The chemical composition of wheat straw will also be different compared with wood materials. Wheat straw and wood materials


133 contains almost equivalent amount of cellulose. However, the hemicellulose content is higher and the lignin content is lower when comparing with wood samples. The objective with this investigation was to manufacture high performance wheat straw MDF. The main idea was to improve the curing conditions by adding a small amount of acid in a pretreatment step (0.6 % sulfuric acid) and to add an extra amount of UF-resin to a level of around 15-19 %. Furthermore, to compare two commercial UF-resin at different melamine contents (different hydrophobic conditions). Panels were manufactured in a complete pilot MDF process line including hammer milling of wheat straw, acid pre-treatment, pressurized refining, blowline resination, tube flash drying, forming of fiber mat, prepressing and pressing.

MATERIALS AND METHODS Raw material

The wheat straw material (Triticum aestivium) was harvested in the Uppsala district in Sweden. The wheat straw was cut to a length of around 20 to 30 cm and bailed in 250 kg bales. Together with the wheat straw a small amount of leaf and wheat seed was observed in the bales. The bales were stored in dry and cold conditions before hammer milling. Resin addition and resin formulation

Two commercial Urea Formaldehyde (UF) resins were manufactured by Akzo Nobel (Casco Products AB, Sundsvall, Sweden) and selected as adhesives for the trials with wheat straw. Cascorit UF 1134 is a typical UF-resin for MDF products on the European market of E1 grade (UF-resin). The second commercial resin selected was Cascorit UF 1106 of a water resistance grade consisting of 7-10 % melamine (UMF-Resin). The major difference between selected resins was the melamine content. Properties of used resins are presented in Table 1. Table 1: Properties of two commercial urea formaldehyde resins supplied by Akzo Nobel, Casco Products AB. Resin Properties

Unit

Appearance Solid Content Viscosity Density pH Gelation Time1 Free Formaldehyde Solubility Storage Stability Melamine Content

% mPas kg/m³ [-] s % parts week %

1

Cascorit 1106 UMF-resin

Cascorit 1134 UF-resin

White 64 100 - 150 1.27 7.5 - 9 100 < 0.1 >10 8 7-10

White 64.5 100 - 300 1.27 7.5 - 9 100 - 200 < 0.1 4 1-5

Casco analysis method IAR 040


134

The target resin content (RC) was set to 15 %. The melamine modified UF resin was also added at a higher RC, 19 %. Both resin grades were mixed with a hardener (ammonium chloride) to a level of 1.0 % and a retarder (hexamethylenetetraamine) to 0.2 % based on resin, dry basis (db). The mixed resins were diluted to a 50 % resin solid content before blowline addition. Wax emulsion

The anionic wax emulsion Boardwax B100 was supplied from Emutech AB, Sweden. The solid content of the wax emulsion was 56 %. Target addition of wax emulsion was set to a level of 1.0 % based on wheat straw (db) and added into the infeed screw at a wax solid content of 30 %. Hammer milling

The wheat straw was size reduced by using two connected hammer mills. The first (infeed) hammer mill was of a Jeffery Swinghammer Shredder, size 15x8 and installed with 25 mm screens. The second hammer mill from Kamas Industri AB, kvarn H-30.C, (Karlstad, Sweden) was equipped with oval screening holes with an aperture of 10 mm. The rotation speed of the first hammer mill was set to 1000 rpm and the second hammer mill was adjusted to 1500 rpm. A production rate of 60 kg/h was chosen and the length of the size reduced wheat straw was in the range of 10 – 15 mm. Two fans and a cyclone were connected to the hammer milling system. Acid pretreatment

Sulfuric acid was added to a target level of 0.6 % (db) of the wheat straw in the pretreatment equipment before defibration/refining. Steam and water were used to increase the moisture content and acid to lower the pH and pH buffering capacity values. The acid pretreatment was performed in a continuous process set to 70 °C and to a retention time of 11 minutes, see Figure 1. The water pretreatment and steam heating resulted in an MC level of around 80-100 %. The pretreated wheat straw material was collected in plastic bags and was stored for 2 hours before refining. Analysis of fiber moisture content, pH and pH-buffering capacity was performed. Refining in a laboratory horizontal defibrator (OHP 20)

The acid water/steam pretreated wheat straw material was refined in a pressurized single disk refiner. Type OHP 20 laboratory defibrator with a plate diameter of 508 mm (20") and a horizontal digester (preheater). Refining was carried out at a rotation speed of 1500 rpm and at a preheater pressure of 0.6 MPa. Refined fiber was vented from the refining house into the blowline. The fiber was resinated in the blowline and dried in a connected continuous flash dryer. The retention time in the defibrator system was 2 minutes. The average moisture content of dried fiber was below 10 %, see fiber production process data in Table 2.


135

Raw Material Input

Chemical Input

Mixer Screw

Water Input

Steam for heating

Conveyer Screw

Analysis Temperature pH MC

Figure 1: Schematic drawing of the water/chemical pretreatment. Addition of water or chemicals in the inlet of the mixer screw followed by steam heating in the conveyer screw.

Three different trials were performed, denoted 15 % UF-resin, 15 % UMF-resin and 19 % UMF-resin. In the first trial (15 % UF-resin) the resin type was (Cascorit UF 1134) and the RC was set to 15 % (db). In trials (15 % UMF-resin) and (19 % UMF-resin) the melamine modified UMF-resin (Cascorit UF 1106) was set to 15 % (db) and 19 % (db). Table 2: Experimental average values of defibrator processing parameters. Trial Defibrator Production Preheating pressure Preheating temperature Preheating time Housing pressure Rotational speed Segment Disc clearance Resin Content Flash tube dryer Burner temperature Air inlet temperature Air outlet temperature 1

Unit RC 1

UF-resin 15 %

UMF-resin 15 %

UMF-resin 19 %

kg/h kPa ºC s kPa rpm mm %

69 600 165 120 600 1500 5821 0.68 15

55 600 165 120 600 1500 5821 0.59 15

64 600 165 120 600 1500 5821 0.63 19

ºC ºC ºC

394 145 86

334 121 86

372 142 78

Resin Content


136 Forming, prepressing and pressing

Fiber mats were formed manually in a forming box (300 x 300) mm followed by a cold prepressing at 1.5 MPa pressing during 60 seconds. The prepressing operation increased the homogeneity and density of the formed fiber mats resulting in an easier handling and following pressing of the fiber material. Pressing was performed in a laboratory press. The press sequence was optimized for each of the three trials. An initial pressure of 4.0 MPa and a press-time (press factor) of approximately 14 s/mm was used. A total of 15 fiber mats were hot platen pressed at 190 ºC. The final wheat straw fiber panel dimension was (300 x 300 x 16) mm. Target density was set to 750 kg/m³. Experimental analysis and evaluation of fiber and panel properties

The analysis of fiber distribution was performed by measurement of a fiber sample in water suspension by an optical method, PQM 1000. The pH measurements and pHbuffering capacity determination was performed by a modified method of the analysis performed by (Johns and Niazi, 1980). Titration was only performed by acid titration (0.025 N H2SO4), due to the high pH of the wheat straw material. Finished panels were conditioned for 1 week under 20 ºC and 65 % relative humidity. The un-sanded panels were evaluated according to the European Standard for medium density Fiberboard EN 622-5:1997. All analyses are summarized in Table 3. Table 3: Performed analyses of raw material, fiber and MDF according to EN 622-5:1997 pH pH-buffering capacity PQM 1000 pH pH-buffering capacity Bending strength, MOR Static modulus of elasticity, MOE Internal bond, IB Thickness swell, 24h, TS Water absorption, 24h, WABS Density profile Determination of resin content (Kjeldahl Analysis)

Raw material Fiber

Fiberboard (Panel)

RESULTS AND DISCUSSION Pretreatment of wheat straw

Pressure refined wheat straw had a lower pH and lower buffering capacity compared with the hammer milled wheat straw. Pretreatment of hammer milled wheat straw was performed by addition of 0.6 % sulfuric acid, water and steam. The acid pretreatment of hammer milled wheat straw particles reduced the pH level from 7.9 to around 6.4, see Figure 2. The slightly alkaline UF-resin increased the pH of resinated wheat straw fiber to 6.8. However, a reduction of pH is normally obtained of wood furnishes and wheat straw during pressurized refining at temperatures levels above 150 - 160 ºC. The ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

137 retention time in the defibrator system has also an influence of the pH. Several degradation processes of lignocelluloses during refining initiate formation of acidic components, presumable formation of carboxylic groups (Lawther et al, 1996), (Han and Kawai, 2001).The pH-buffering capacity of wheat straw particles was analyzed for both acid-pretreated and untreated hammer milled straw materials. Untreated hammer milled wheat straw was processed by the same method in the pretreatment step but without addition of acid. The pH-buffering capacity of the untreated wheat straw particles was close to a level of 27 ml but was reduced to around 22 ml after acid pretreatment, see Figure 2. Decreased pH-buffering capacity of fiber materials used for MDF manufacturing will improve the gel time and curing conditions of applied UF-resin (Johns and Niazi, 1980). The pH of refined fibers will be influenced of the type of raw materials used, added chemicals, resin types and contents, and finally the specific processing conditions during pretreatment and refining.

8.0 7.5 7.0

Untreated Wheat Straw

6.5

Acid Treated Wheat Straw

pH

6.0 5.5 5.0 4.5 4.0 3.5 3.0 0

5

10

15

20

25

30

Added Acid [ml] 0.025 N H2SO4

Figure 2: pH-buffering capacity of untreated and acid pretreated hammer milled wheat straw particles.

Unresinated wheat straw fiber combined with acid pretreatment resulted in the lowest pH-buffering capacity. Titration resulted in 19.0 ml volume of a 0.025 N H2SO4 solution to reach a pH of 3. The difference between unresinated refined wheat straw fiber and resinated fiber (15 % UF-resin) was minor and a pH-buffering capacity of 20.5 ml volume was observed. Resinated wheat straw fiber of the melamine modified UFresin (15 % UMF-resin) consumed 22.5 ml of the acid solution. Increased resin content of UMF-resin from a level of 15 % to 19 % increased the pH-buffering capacity to 25 ml (19 % UMF-resin), see Figure 3.


138

7.0 6.5

Acid Treated Wheat Straw Fiber

6.0

Resinated fiber (15% UF-Resin)

pH

5.5

Resinated fiber (15% UMF-Resin)

5.0

Resinated fiber (19% UMF-Resin) 4.5 4.0 3.5 3.0 0

5

10

15

20

25

30

Added Acid [ml] 0.025 N H2SO4

Figure 3: pH-buffering capacity between refined, unresinated and resinated wheat straw fiber.

A decrease of pH and a reduction of the pH-buffering capacity values of resinated fiber will improve the curing rate and approach the curing conditions of wood fiber (Johns and Niazi, 1980), (Johns et al. 1985), (Sauter SL, 1996). The curing rate of applied UFresin during pressing should consequently increase for acid treated wheat straw fiber. However, in this case a minor addition of acid (0.6 %) to wheat straw resulted in a small reduction of pH and of the pH-buffering capacity. The most pronounced effect of the pH-buffering capacity was achieved by addition of different types of UF-resins (UFresin/UMF-resin). Moreover, the amount of UF-resin added increased the buffering capacity. The effect of other parameters such as refining conditions, moisture content of fiber, hot plate press temperature, pressing time, initial surface pressure etc. will probably have a more significant effect of the curing conditions of applied UF-resin than the effect of a light acid pretreatment. However, the effect of acid pretreatment of wheat straw can be necessary for full-scale manufacturing of MDF to keep a constant level of pH and pH-buffering capacity due to a wide range of raw material parameters.

Fiber length distribution of wheat straw fiber

The fiber length distribution of wheat straw fiber samples was analyzed in the PQM 1000 system. Length distribution curves of refined wheat straw fiber samples followed together as a function of the fiber length, see Figure 4. The plotted curves are independent of resin type, resin content or other fiber treatments. Moreover, a typical MDF fiber sample of refined Scandinavian pine has been plotted as a reference in Figure 4.


139

10

Wheat straw fiber

9

15% UF-Resin

Relative distribution [%]

8

15% UF-Resin

7

15% UMF-Resin 6

15% UMF-Resin 5

19% UMF-Resin 4

Pine Fiber Reference 1 3

Pine Fiber Reference 2

2 1 0 0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

8.0

PQM fibre length—linear scale [mm]

Figure 4: PQM 1000 fiber length distribution. Fiber weight distribution vs. fiber length.

The average fiber length of wheat straw was relatively short and calculated values were around 0.9 to 1.0 mm compared with the fiber length of wood fiber, approximately 2.5 mm in average fiber length. Refined wheat straw generated a much higher amount of short fiber compared with the refined pine furnish. The amount of fiber straw particles less than 0.5 mm was more than twice of the amount of pine fiber. Short fibers and dust particles will absorb a higher amount of resin compared with wood fiber due to the larger surface area. A higher amount of resin must probably be added to the wheat straw fiber material to compensate for the higher amount of dust compared with wood fiber. Refined Scandinavian pine has consequently better opportunities to generate effective fiber/fiber bondings in finished panels, due to longer fiber and less fines and dust.

Mechanical properties of wheat straw fiberboard

Testing of mechanical properties of produced panels resulted in values close or better than MDF standard (EN 622-5:1997) The highest IB, MOR and MOE values obtained of wheat straw panels were achieved for samples at a resin content of 15 % and 19 % of the melamine modified UF-resin type (UMF-resin). Presented data are based on measured values of pressed panels, representing 3 to 6 panels for each trial, see Table 4. Internal bond of MDF produced by addition of 15 % UF-resin showed an IB average of 0.46 MPa, MOR was estimated to 20 MPa and MOE was in the level of 3.3 GPa. The use of melamine modified UF-resin (15 % UMF-resin) contributed to a major increase of IB to 0.67 MPa, MOR and MOE resulted in values of 22.9 MPa respectively 3.3 GPa. Furthermore, an additional improvement of mechanical properties was achieved for the highest resin content. Resulting in an IB of 0.79 MPa, MOR and MOE correspond to 27 MPa and 3.7 GPa. However, the average densities of produced panels were also increased and a higher level of density will contribute to improved panel properties, which will be discussed, in following sections. A common method to improve TS of wood based panels is to increase the melamine content of used UF-resin (Pizzi A, 1994), (Maylor R, 1995). The thickness swelling of wheat straw panels based on the 15 % and 19 % melamine modified UF-resin was ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

140 dramatically reduced. A reduction of TS from 18 % to 12 % and 8.8 % was observed. The same behavior was also observed for the water absorption, see Table 4. Table 4: Mechanical properties of wheat straw fiber MDF based on a typical UF-resin for MDF (UFresin) and melamine modified UF-resin (UMF-Resin) at different Resin Content (RC). Property

Unit

UF-Resin RC =15 %

UMF-Resin RC =15 %

UMF-Resin RC =19 %

EN Standard

Internal Bond (IB)

MPa

0.43

0.67

0.83

> 0.6

Modulus of Rupture (MOR)

MPa

21

23

27

> 20

Modulus of Elasticity (MOE)

GPa

3.27

3.32

3.71

>3

Thickness Swelling (TS) 24h

%

18

12

8.8

< 12

Water Absorption (WABS) 24h

%

57

44

33

Average Density 1

kg/m3

726

770

790

1

500 - 800

Average density measured at the IB analysis

Density and Internal Bond (IB)

Panel properties are strongly dependent of the panel average-, surface- and core density. Increased density of PB and MDF will generally result in enhanced IB, MOR, and MOE (Schulte and Fruhwald, 1996). Commercial wood based MDF has been optimized to get favorable bending properties and hard panel surfaces. A typical MDF has a density profile with a high surface density above 1000 kg/m³ and a core density around 650 kg/m³. The average density will be in the level of 750 kg/m³. IB of a typical MDF panel will in most cases be correlated to the average density. However, a better way to describe MDF properties is perhaps to relate panel properties to a relevant type of density. The vertical density profile (density as a function of thickness) and corresponding IB of wheat straw panels were measured. The densities of the actually breaking points during IB testing were registered, see Figure 5. IB of wheat straw MDF as a function of average density respectively minimum density is plotted in Figure 6 and Figure 7. In both figures, IB increases as a function of density. Panels were produced in three different trials as described above containing different types and levels of resin, 15 % of the UF-resin, 15 % and 19% of the UMF-resin. The industry MDF standard is also presented in Figure 5 and 6. A comparison of data in Figure 5 and Figure 6 shows a shift of measured IB-values to lower densities and to a more homogenous curve with less scattering of IB values. A better way to evaluate IB is probably to relate the measured IB to the minimum density obtained from corresponding density profile, see Figure 5. In this investigation the relation between IB and minimum density was more pronounced compared with the curve of IB as a function of average density. The effects of resin type and increased resin content were complex to evaluate due to the variation of densities. IB of MDF close to a fixed average density e.g. 750 kg/m3 shows the highest IB values for the UMF-resin of 19% ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

141 RC corresponding to 0.75 MPa. Lower IB values were observed for MDF based on the 15 % UMF-resin and IB at 750 kg/m3 density corresponds to 0.6 MPa. The lowest IBvalue was estimated to around 0.5 MPa for MDF panels based on the UF-resin, see Figure 6. Moreover, all panels above 760 kg/m3 average density were approved according to MDF standard.

1200

Density [kg/m³]

1000

800

600

400

200

0 0

2

4

6

8

10

12

14

16

Thickness [mm]

Figure 5: The vertical density profile of a 16 mm wheat straw MDF-panel with a surface density of 1000 kg/m³ and a core density of 640 kg/m³. The vertical line in the middle is the IB-breaking point. The horizontal dashed line represents core density.

1.2

15% UF-resin

1.1 15% UMF-resin

Internal Bond [MPa]

1.0 0.9

19% UMF-resin

0.8 MDF Standard

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 500

550

600

650 700 750 800 3 Average Density [kg/m ]

850

900

950

1000

Figure 6: IB vs. average density of wheat straw MDF panels based on a typical resin for MDF (UF-resin) and a melamine modified resin (UMF-resin). ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

142

However, only minor improvements of IB can be observed of MDF presented as a function of minimum density for different types of resins or resin content in the whole density interval, see Figure 7. The increased resin content of the UMF-resin shifted the IB curve to slightly higher levels. However, in Table 4 the overall average values of IB increased with the resin level. The IB increase is presumably an effect of the increased MDF density.

1.2 1.1

15% UF-resin

Internal Bond [MPa]

1.0

15% UMF-resin

0.9 19% UMF-resin

0.8

MDF Standard

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 500

550

600

650

700

750

800

850

900

950

1000

3

Minimum Density [kg/m ]

Figure 7: IB vs. Minimum density of wheat Straw MDF panels based on typical resins for MDF (UF-resin) and a melamine modified resin (UMF-resin).

Bending property MOR

The bending property modulus of rupture (MOR) of wheat straw MDF panels was plotted as a function of average density, see Figure 8. The bending strength increased from 16 MPa to 30 MPa. However, a large variation of MOR values was observed for the straw MDF. The bending strength of samples based on the 15 % UMFresin showed a random variation between 20–25 MPa over an extended average density interval of 720 to 780 kg/m3. A possible explanation of the variations of MOR as a function of average density is probably related to the density profile. Bending properties are strongly dependent of the proportion between surface density and core density. A variation of the vertical density profile between different samples can result in a constant MOR as a function of the average density. Surface density and density profile will probably have the most determining effects of the bending properties.


143

MOR [MPa]

40 35

15% UF-resin

30

15% UMF-resin

25

19% UMF-resin

20

MDF Standard

15 10 5 0 600

650

700

750

800

850

900

950

1000

3

Average Density [kg/m ]

Figure 8: MOR vs average density of wheat straw MDF-panels based on a typical resin for MDF (UF-resin) and a melamine modified UF-resin (UMF-resin).

Thickness swelling and water absorption

The major problem of MDF based on agriculture waste materials such as wheat straw is the water uptake and sensitivity to hygroscopic conditions. The mechanism of thickness swelling can be described as an uptake of water and swelling of single fibers in a fiberresin matrix. Hydrophilic components in the wheat straw materials will adsorb water in the porous lignocellulosic material and develop swelling of fibers, which initiate a thickness increase of the MDF. Thickness swelling (TS) of wood based MDF panels will normally decrease as a function of density. However, the swelling behavior of fiberboards is also depending on the density interval investigated. TS as function of density in the density interval of 200 kg/m³ to 500 kg/m³ (soft boards) has been reported to increase as function of density (Rowell et al. 1995). This reversed density-thickness swelling behavior was explained by an internal swelling of fibers and minor fiber compression during pressing. Internal swelling allows swelling without causing the panel to enlarge due to the free space available inside the panel. In the density range above 650 kg/m³ a more complex relationship between thickness swelling and density exists due to the vertical density profile. A published investigation of thin discrete layers inside a wood based MDF showed a high contribution of thickness swelling of the high-density surface layers compared with the core layer (Wang and Winistorfer, 2001). In this case the TS was plotted as a function of average density, see Figure 9. TS varied between 8.0 and 24 % and was reduced as a function of average density. Most of the samples above 760 kg/m³ were approved according to MDF-standard except for panels based on the traditional UF-resin. However, MDF based on wood furnish and the traditional UF-resin will in most cases result in lower TS. For all densities tested, this study showed that wheat straw MDF based on the UF-resin was not sufficient for approval according to MDF-standard.


144

TS [%]

The more hydrophobic character of the UMF-resin undoubtedly reduced thickness swelling of the wheat straw MDF and showed a stronger dependence of density compared with the UF-resin. Thickness swelling of MDF based on the UMF-resin samples at the lower resin content varied between 9 % and 14 %. Additional reduction of TS was achieved for MDF samples at the higher resin loading of the UMF-resin and densities above 770 kg/m³. The lowest TS observed was in the range of 8 % to 10 %, see Figure 9.

30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0

15% UF-resin 15% UMF-resin 19% UMF-resin MDF Standard

600

650

700

750

800

850

900

950

1000

3

Average Density [kgIm ] Figure 9: Thickness swelling vs average density of wheat straw MDF-panels based on a typical MDF resin (UF-resin) and a melamine modified UF-resin (UMF-resin).

Several investigations have shown that modification of hydrophobic components in MDF reduce the thickness swelling of wood based MDF (Younquist JA, 1986), (Rowell and Keany, 1991), (Karr and Sun, 2000). In this investigation the hydrophobic component was represented by the amount of melamine in applied UF-resin. Reductions of TS and water absorption of wheat straw MDF were obtained by usage of a melamine modified UF-resin. CONCLUSIONS

Manufacturing of wheat straw MDF based on a traditionally UF-resin will in most cases result in improper panel properties. Adding a melamine modified UF-resin to wheat straw fiber improved the MDF properties and especially the thickness swelling. The pretreatment of hammer milled wheat straw by an addition of 0.6 % sulfuric acid reduced the pH of the wheat straw from 7.9 to 6.4. Moreover, pH buffering capacity values were slightly reduced. The curing conditions of UF-resin were by this approach a small step closer to curing conditions similar to UF-resin and wood based raw materials. Refining of wheat straw resulted in a shorter average fiber length and a higher amount of short fibers, below 0.5 mm (fines and dust) compared with typical wood based furnish (pine). The resin bonding capacity of refined wheat straw fiber will ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

145 consequently be lower. A resin loading of 15 % was necessary to get acceptable wheat straw MDF properties. Additional increase of the resin content to a level of 19 % resulted in improvements of the overall average values of IB, MOR and MOE. However, the obtained improvements of panel properties were also linked to the increased panel density. Several of investigated MDF properties could be related to the panel density. The strength property IB was closely related to minimum density. MOR increased as a function of average density in a more diffuse way. Thickness swelling was reduced as a function of average density. Melamine modified UF-resin improved the panel properties of wheat straw MDF. The most promising panel property was thickness swelling. The usage of traditional UFresin in MDF panels revealed an unacceptable thickness swelling of 14 % to 24 %. A melamine modified UF-resin reduced the thickness swelling of the wheat straw MDF and resulted in values below 12 %, which is approved according to the MDF-standard.

ACKNOWNLEDGEMENTS

The authors are grateful for the financial support of this research from the Swedish foundation "Stiftelsen för Kunskap- och Kompetensutveckling, KK-stiftelsen" and participant companies Metso Panelboard AB and Akzo Nobel, Casco Products AB.

REFERENCES

BOYD AL (2001): AgriFiber composite panels: Environmentally responsible product performance: Proc. 35th International Particleboard/Composite Material Symposium, Washington state university, Pullman, Wash., USA: 117-124 DONALDSON J, HAUGE J and SNELL R (2001): Lignin distribution in Coppice Poplar, Linseed and Wheat Straw: Holzforschung 55: 379 - 385 EN 622-5:(1997): Fiberboards – Specifications – Part 5: Requirements for dry process boards (MDF) EROGLU H and ISTEK A (2000): MDF manufacturing from wheat straw: Inpaper international: 11-14 HAN G and KAWAI S (2001): Development of high-performance UF-bonded reed and wheat straw MDF: J Wood Sci 47: 350-355 JOHNS WE and NIAZI KA (1980): Effect of pH and buffering capacity of wood on the gelation time of urea-formaldehyde resin: Wood and fiber 12(3): 255-263 JOHNS WE, RAMON RM and YOUNGQUIST JA (1985): Chemical effects of mixed hardwood furnish on panel properties: Proc. 19th International Particleboard/Composite Material Symposium, Washington state university, Pullman, Wash., USA: 363-377


146 KARR GS and SUN XS (2000): Strawboard from vapor phase acetylation of wheat straw: Ind Corp and Prod Sci: 11, 31-41 LAWTHER JM, SUN R and BANKS WB (1996): Effect of steam Treatment on the chemical Composition of Wheat Straw: Holzforschung 50: 365-371 MAYLOR R (1995): New melamine modified binders for moisture resistant MDF: Proc. symp: Wood adhesives 1995, USDA Forest Service, Forest Prod. lab. MARKESSINI E and ROFFAEL E (1997): Panels from annual plant Fibers bonded with Urea-Formaldehyde Resins: Proceedings of the 31th International Particleboard/Composite Material Symposium, Washington state university, Pullman, Wash., USA: 147-160 MCLAUCHLIN AR and HAUGE JRB (1998): Panel products from UF-Bonded nonwood Fibers – a viable proposition? : Proc. 2 nd European Panel Products Symposium, Llandudno, Wales, UK 142-152, October 1998 METSO PANELBOARD AB, (2002): Straw and other annual fibers searching for a future: Metso Panelboard Customer Magazine 2: 12-17 MO X, CHENG E, WANG D and SUN X S (2003): Physical properties of mediumdensity wheat straw particleboard using different adhesives: Ind Corp and Prod 18: 4753, PIZZI A: Advanced wood adhesives technology: Marcel Dekker Inc: NewYork, ISBN 0-8247-9266-1, 1994 ROWELL RM, KAWAI S and INOUE M (1995): Dimensionally stabilized, very low density Fiberboard: Wood and Fiber Sci 27(4): 428-436 ROWELL RM and KEANY FM (1991): Fiberboards made from acetylated bagasse fiber: Wood and Fiber Sci 23(1): 15-22 RUSSELL CW (1996): The straw resource: A new Fiber basket?: Proc. 30th international particleboard/composite material symposium, Washington state university Pullman, Wash., USA: 183-190 SAUTER SL (1996): Developing composites from wheat straw: Proc. 30th International Particleboard/Composite Material Symposium, Washington state university, Pullman, Wash., USA: 197-214 SCHULTE M and FRUHWALD A (1996): Some investigations concerning density profiles, internal bond and relating failure position of particle board: Holz als Roh-und Werkstoff 54: 289-294 WANG S and WINISTORFER PM (2001): Determination of layer thickness swell of MDF and OSB by optical technique: Forest Products Journal (in press).


147 WASYLCIW W (2001): Straw-based composite panels - attributes, issues and UF bonding technology: Proc. 35th International Particleboard/Composite Material Symposium, Washington state university, Pullman, Wash., USA: 145-153 YOUNQUIST JA, ENGLISH BE, SCHARMER RC, CHOW P and SHOOK SR (1994): Literature review use on non-wood plant fiber for buildings material and panels, General technology report FPL-GTR-80: USDA Forest Service, Forest Prod. Lab., Madison, WIS YOUNQUIST JA and KRZYSIK A and ROWELL RM (1986): Dimensional stability of acetylated aspen flakeboard: Wood and Fiber Sci 18(1): 90-98 ZHAI HM and LEE ZZ (1989): Ultrastructure and topochemistry of delignification in alkaline pulping of wheat straw: J. Wood Chem. and Tech 9(3): 387- 406


148


149

MANUFACTURING PARAMETERS ON THE PROPERTIES OF CEMENT-BONDED BOARDS USING SUGARCANE BAGGASE Dwight A. Eusebio

Supervising Science Research Specialist - Chief, Composite Products Section Program Coordinator, Housing Materials and Construction Technologies Forest Products Research and Development Institute Department of Science and Technology College, Laguna 4031 – PHILIPPINES Tel.: +63-49-536-2377 – Fax: +63-49-536-3630 - [email protected]

ABSTRACT Cement-bonded boards (CBBs) were produced using different sizes of sugar cane bagasse as raw material, general-purpose Portland cement as binder and aluminum sulfate [Al2(SO4)3] as accelerator. The effects of material ratio (1:1, 1:2, 1:3 by weight), soaking time (24, 48, 72 hrs for bagasse and 8, 16, 24 hrs for rattan shavings) and particle size (retained 8, 4, 2 mm and passing 2-mm screen) at different board densities (0.60, 0.80, 1.00 g/cm3) were determined. The highest values for modulus of rupture (MOR) and nail-head-pull-through (NHPT) were obtained from boards with a density of 1.00 g/cm3 containing bagasse particles retained at 2-mm screen and bagasse:cement ratios of 1:2 and 1:1, respectively. Similarly, the lowest values for thickness swelling (TS) and water absorption (WA) were obtained from the same particle size but with board densities of 0.60 g/cm3 and 1.00 g/cm3, respectively.

Keywords: Cement-bonded boards, sugarcane bagasse, bagasse:cement ratio

INTRODUCTION

Cement-bonded boards (CBBs) are rousing interest because of their potential application in the construction industry. Production of the boards is likely to increase in the near future because agro-forest materials and cement are available in the country. CBB can help ease the housing backlog besetting the Philippines. From 1993 to 1998, new dwellings installed due to population growth and upgrading needs reached 2,853,000 units, while the housing backlog totalled 843,000 units. In the next six years (1999-2004), the total housing requirement is 3,940,984 units (Housing Need Target 1999-2004), which include new housing units, household growth, doubled-up household, replacement, homeless and upgrading needs. CBBs have specific advantages over conventional resin-bonded composite panels, such as their resistance to termites, fire, rot and fungi. They have gained favor throughout the industry due to their extended applications compared to plywood, resin-bonded particleboard and other allied products. ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

150

Many of the laboratory and pilot scale studies on CBB mostly dealt with the effect of wood/cement ratio, water/cement ratio, accelerators, density, board thickness and soaking time on their properties. The rest focused on the use of ITPS and agricultural residues such as abaca fiber wastes, rattan shavings, talahib, cotton stalks and bagasse as raw materials. No studies on the effect of particle size and its consequent effect on the amount of cement as binder, amount of cement setting accelerators, board density and soaking time have been reported thus far. As of March 1997, there were 37 sugar mills operating throughout the country with an average weekly production of 63,566 metric tons (MT). According to the AgroIndustrial Research Laboratory Department of the Sugar Regulatory Administration (1996-1997), in calendar year 1994-1995 about 18,505,264 MT of sugar cane was processed, resulting in 1,647,022 MT of raw sugar and 5,331,196 MT of bagasse. Other by-products such as filter cake and molasses yielded 529,965 MT and 684,864 MT, respectively. Advanced studies have been done in producing CBB particularly by shortening the pressing time using steam injection pressing, hot pressing and carbon dioxide injection technologies (Eusebio, et. al. 1994 and Simatupang, et. al. 1991). In one study, rattan shavings were used as raw material for the rapid curing of CBB by hot pressing with the incorporation of sodium silicate (Na2SiO3) and sodium bicarbonate (NaHCO3). Results revealed that a pressing time of 15 minutes is enough to produce boards with properties comparable to those produce by 24 hour conventional pressing (Eusebio 2000). This present study aimed to produce CBBs using different sizes (retained on 8, 4, 2 mm mesh sizes and passed mesh 2) of sugar cane bagasse. Also, the effects of board density (0.60, 0.80 and 1.00 g/cm3), and soaking time of the materials (prior to mixing with cement) on board properties were determined

METHODOLOGY Material Preparation

Sugar cane bagasse was used as the reinforcing material. General-purpose Portland cement (Type I) and aluminum sulfate [Al2(SO4)3] were used as binder and cement setting accelerator, respectively. Sugarcane bagasse was screened using three different mesh sizes of 8, 4 and 2 mm. Particles retained at these mesh sizes were used in the study and designated as R8, R4 and R2. Those that passed mesh 2 size but retained in mesh 1 was also used and designated as P2. Particles that passed mesh 1 were discarded because they were too fine and contained very short fibers. Soaking

The materials were soaked in water to leach out extractives that could inhibit cement setting. Soaking times were 24, 48, 72 hrs. After each soaking period, the particles were washed with clean water. ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

151

Mixing

The required amounts of cement, bagasse and chemical were weighed as per experimental design. Water for hydration was fixed at 1:1 water:cement ratio. The preweighed materials were then mixed manually in a plastic basin. Mixing was approximately 5 min per mixture until the raw materials were completely coated with cement paste. Mat Forming, Pressing and Clamping

Each mixture was spread evenly into a forming box placed over a stainless caul plate to form a mat. To densify the mats, these were compressed to 10 mm thickness using a laboratory hydraulic press. These were then clamped for 24 hrs for final setting and hardening. The boards were tested after 28 days curing and conditioning at ambient temperature. Property Testing

The modulus of rupture (MOR) and nail-head-pull-through (NHPT) were tested using a Universal Testing Machine. Thickness swelling (TS) and water absorption (WA) were determined by immersing the test specimens in water for 24 hrs. Statistical Analysis

Analysis of covariance (ANACOVA) in complete randomized design (CRD) was used to analyze the data on MOR, NHPT, TS and WA. Treatment means were further evaluated using Duncan’s multiple range test (DMRT)

RESULTS AND DISCUSSION

The MORs of CBBs at 0.60, 0.80 and 1.00 g/cm3 densities containing four particle sizes (PS) of bagasse with four levels of Bag:Cem ratio (BC) and three levels of soaking time (ST) are presented in Figs. 1 to 3. Results showed that as the density of the board was increased, the MOR also increased. This could be attributed to the high compaction ratio that led to better interparticle contact. The effect of soaking on MOR varied depending on board density. At low density (0.60 g/cm3), MOR did not significantly increase with longer soaking time (Table 1a). At higher densities (0.80 & 1.00 g/cm3), however, MOR was highly affected by soaking time (Tables 2a & 3a). This indicates that a longer soaking time effectively removed more sugars believed to be detrimental to the strength development of cement. The effect of PS, ST and BC ratio on MOR was significant for all board properties (Tables 1a, 2a & 3a). The interaction among PS, ST and BC was further analyzed by DMRT and the results were presented in Tables 1b, 2b and 3b for 0.60, 0.80 and 1.00 g/cm3 densities, respectively. The lowest MOR was obtained from boards containing P2 particles and 1:3 BC. Although the amount of cement was highest at 1:3 BC, it could be speculated that sufficient bonding did not take place because of the large ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

152 surface area of fine particles. Also, short fibers did not contribute much in resisting the applied load. On the other hand, the low MOR values for boards containing R8 particles with 1:3 BC could be due to the low compaction ratio. Mixtures with 1:3 BC had low mat height, thus compressibility was low, resulting in poor interparticle bonding. The maximum MOR values for boards with densities of 0.60 g/cm3 and 0.80 g/cm3 were obtained from boards containing R4 bagasse soaked for 48 hrs and 1:1 BC. The MOR values for the first six and first three combinations of PS x ST X BC of boards at 0.60 g/cm3 and 0.80 g/cm3 were not significantly different. On the other hand, the highest MOR value at 1.00 g/cm3 density was exhibited by boards with R2 x 72 x 1:2 combination. The first six combinations revealed insignificant differences in MOR including combination R4 x 48 x 1:1. It was the same for 0.60 g/cm3 and 0.80 g/cm3 densities and the first four combinations of boards with 0.80 g/cm3 density. It can therefore be considered that the most appropriate combination for the three density levels is R4 x 48 x 1:1.

2

MOR, kgf/cm

1 :1

B a g :C em ra tio

100

1 :2

1 :3

80 60

2 4 h rs

4 8 h rs

7 2 h rs

40 20 0 R8

R4

R2

P2

R8

R4

R2

P2

R8

R4

R2

P2

P artic le siz e F ig . 1 . M O R o f C B B w ith a d e n sity o f 0 .6 0 g /c m 3 a t d iffe re n t b ag a sse p a rtic le size a n d a s a ffe cte d b y b a g :c e m ra tio an d so a k in g tim e .

B ag:C em ratio

100 80

48 hrs

24 hrs

2

MOR, kgf/cm

1:1

1:2

1:3

72 hrs

60 40 20 0 R8

R4

R2

P2

R8

R4

R2

P2

R8

R4

R2

P article size Fig. 2. M O R of C B B w ith a density of 0.60 g/cm 3 at different bagasse particle size and as affected by bag:cem ratio and soaking tim e.


P2

153 B ag:C em ratio

MOR, kgf/cm

2

100 2 4 h rs

80

1 :1

1 :2

1 :3

7 2 h rs

4 8 h rs

60 40 20 0 R8

R4

R2

P2

R8

R4

R2

P2

R8

R4

R2

P2

P article size 3

F ig. 3 . M O R o f C B B w ith a d en sity o f 1 .0 0 g/cm at d ifferen t b agasse p article size an d as affected b y b ag:cem ratio an d so akin g tim e.

The NHPT strength of the boards at 0.60, 0.80 and 1.00 g/cm3 densities with different PS is presented in Figs. 4, 5 and 6, respectively. NHPT increased as the density was increased, indicating the highly significant effect of density (Tables 1a, 2a & 3a). The effect of PS, ST and BC on NHPT was significant only for 0.80 and 1.00 g/cm3 densities (Tables 1a, 2a & 3a). DMRT results on the interaction among PS x ST x BC are shown in Tables 2b and 3b for 0.80 and 1.00 g/cm3 densities, respectively. The interactions significantly affected NHPT except for boards with 0.60 g/cm3 density. The highest NHPT values were obtained from boards with PS of R2 and BC of 1:1 for both 0.80 and 1.00 g/cm3 densities. ST affected the NHPT strength of the boards as shown in Figs. 4 to 6. But as presented in Tables 1a, 2a and 3a, this effect was significant only for 0.60 g/cm3 density, indicating a decrease in NHPT as ST was extended to 48 and 72 hrs. It was not very clear as to why NHPT values decreased with longer soaking time. At densities of 0.80 g/cm3 and 1.00 g/cm3, some boards containing bagasse soaked for 24 hrs showed NHPT values equal to or higher than boards with bagasse soaked for 48 or 72 hrs. This trend could be attributed more to the density variation within the boards that overshadowed the effect of ST.

B a g :C em ra tio

100

1 :1

1 :2

1 :3

NHPT, kgf

80

2 4 h rs

60

4 8 h rs

7 2 h rs

40 20 0 R8

R4

R2

P2

R8

R4

R2

P a rtic le siz e

P2

R8

R4

R2

F ig . 4 . N H P T o f C B B w ith a d e n sity o f 0 .6 0 g /c m 3 a t d iffe re n t b a g a sse p a rtic le siz e a n d a s a ffe c te d b y b a g :c e m ra tio a n d so a k in g tim e .


P2

154 B a g :C e m ra tio

100 2 4 h rs

NHPT, kgf

80

1 :1

4 8 h rs

1 :2

1 :3

7 2 h rs

60 40 20 0 R8

R4

R2

P2

R8

R4

R2

P2

R8

R4

R2

P2

P a rtic le siz e F ig . 5 . N H P T o f C B B w ith a d e n sity o f 0 .8 0 g /c m 3 a t d iffe re n t b a g a ss e p a rtic le siz e a n d a s a ffe c te d b y b a g :c e m ra tio a n d so a k in g tim e .

NHPT, kgf

100

B a g :C em ra tio

2 4 h rs

1 :1

4 8 h rs

1 :2

1 :3

7 2 h rs

80 60 40 20 0 R8

R4

R2

P2

R8

R4

R2

P2

R8

R4

R2

P2

P a rtic le siz e F ig . 6 . N H P T o f C B B w ith a d e n sity o f 1 .0 0 g /c m 3 a t d iffe re n t b ag a sse p a rtic le siz e a n d a s a ffec te d b y b a g :c e m ra tio a n d so a k in g tim e .

A general trend showed that increasing the amount of cement, i.e., 1:3 BC, impaired the NHPT strength of the boards. The nail could easily be pulled through when force was applied. This could be attributed to poor particle-to-particle interface bonding. On the other hand, boards with 1:1 BC exhibited high NHPT strength in all density levels with very few exceptions as in the case of 1.00 g/cm3 density boards with BCs of 1:2 and 1:3 using bagasse soaked for 24 and 48 hrs, respectively. TS results after 24 hrs of water immersion of boards with 0.60, 0.80 and 1.00 g/cm3 densities are shown in Figs. 7 to 9, respectively. There was no definite trend that ST would improve the TS of the boards at all density levels. The effects of density variation within boards and ST on TS were insignificant as revealed by ANACOVA (Tables 1a, 2a & 3a). TS values of boards containing P2 particles were generally high compared to boards with R8 and R4 particles, and more pronounced at 1.00 g/cm3 density (Fig. 9). The effect of PS, ST and BC on TS was significant only for 0.80 g/cm3 density. The most acceptable PS, ST and BC combination was R2, 24 and 1:3, respectively, which gave the lowest TS value of 0.42% (Table 2b).


155 10

B ag:C em ratio

8

24 hrs

1:1

1:2

1:3

72 hrs

48 hrs

TS, %

6 4 2 0 R8

R4

R2

P2

R8

R4

R2

P2

R8

R4

R2

P2

P article size F ig. 7. TS of C B B w ith a density of 0.60 g/cm 3 at different bagasse particle size and as affected by bag:cem ratio and soaking tim e.

10

24 hrs

TS, %

8

Bag:Cem ratio

1:1

48 hrs

72 hrs

1:2

1:3

6 4 2 0 R8

R4

R2

P2

R8

R4

R2

P2

R8

R4

R2

P2

Particle size 3

Fig. 8. TS of CBB with a density of 0.80 g/cm at different bagasse particle size and as affected by bag:cem ratio and soaking time.

B ag:C em ratio

10

2 4 h rs

4 8 h rs

1 :1

1 :2

1 :3

7 2 h rs

TS, %

8 6 4 2 0 R8

R4

R2

P2

R8

R4

R2

P2

R8

R4

R2

P2

P article size 3

F ig. 9 . T S o f C B B w ith a d en sity o f 1 .0 0 g/cm at d ifferen t b agasse p article size an d as affected b y b ag:cem ratio an d so akin g tim e.

WA of CBBs at 0.60, 0.80 and 1.00 g/cm3 densities is shown in Figs. 10, 11 and 12, respectively. It can be noted that WA decreased as the density was increased considering all the parameters. This implies that a high compaction ratio at 1.00 g/cm3 density contributed to better particle-to-particle contact, eliminating too many voids that could have accommodated water intake. The effect of density was highly significant (Tables 1a, 2a & 3a).


156 WA tended to decrease as ST was extended from 24 hrs to 72 hrs with very few exceptions as in the case of boards of 1.00 g/cm3 density. ANACOVA results revealed (Tables 1a, 2a & 3a) the highly significant effect of ST on WA.

B ag:C em ratio

100

24 hrs

WA, %

80

1:1

1:2

48 hrs

1:3

72 hrs

60 40 20 0 R8

R4

R2

P2

R8

R4

R2

P2

R8

R4

R2

P2

P article size Fig. 10. W A of C B B w ith a density of 0.60 g/cm 3 at different bagasse particle size and as affected by bag:cem ratio and soaking tim e.

B a g :C em ra tio

100

WA, %

80

2 4 h rs

1 :1

1 :2

4 8 h rs

1 :3

7 2 h rs

60 40 20 0 R8

R4

R2

P2

R8

R4

R2

P2

R8

R4

R2

P2

P a rtic le siz e 3 F ig . 1 1 . W A o f C B B w ith a d e n sity o f 0 .8 0 g /c m a t d iffe re n t b a g a sse p a rtic le siz e a n d a s a ffe c te d b y b a g :c e m ra tio a n d so a k in g tim e .

B ag:C em ratio

WA, %

100

1 :1

1 :2

1 :3

80

2 4 h rs

60

4 8 h rs

7 2 h rs

40 20 0 R8

R4

R2

P2

R8

R4

R2

P2

R8

R4

R2

P2

P article size 3

F ig. 1 2 . W A o f C B B w ith a d en sity o f 1 .0 0 g/cm at d ifferen t b agasse p article size an d as affected b y b ag:cem ratio an d so akin g tim e.

The effect of PS was more pronounced at board densities of 0.60 and 0.80 g/cm3, indicating an increase in WA from coarse particles (R8) to finer ones (P2). Fine particles tend to absorb more water considering their large surface areas and more volume per unit weight.


157

Table 1a. ANACOVA on the MOR, NHPT, TS and WA of bagasse-cement board at 0.60 g/cm3 density. MOR NHPT TS WA SOURCE OF VARIATION d MS F-value MS F-value MS FMS F-value f value Particle size (PS) 3 3343.59 151.09 ** 4493.26 132.67 ** 12.84 10.05 **4591.69 228.69 ** Soaking time (ST) 2 11.59 0.52 ns 1099.97 32.48 ** 3.27 2.56 ns 648.83 32.31 ** PS x ST 6 270.27 12.21 ns 799.35 23.60 ** 1.76 1.38 ns 207.05 10.31 ** Bag:Cem ratio (BC) 2 3102.23 140.19 ** 4964.40 146.58 ** 102.64 80.32 **4152.18 206.80 ** PS x BC 6 108.48 4.90 ** 74.00 2.19 ** 10.60 8.30 ** 346.37 17.25 ** ST x BC 4 91.35 4.13 ** 8.86 0.26 ns 2.19 1.72 ns 10.73 0.53 ns PS x ST x BC 12 92.69 4.19 ** 21.11 0.62 ns 2.04 1.60 ns 51.38 2.56 ** Density 1 1439.45 65.05 ** 1097.45 32.40 ** 0.10 0.08 ns 712.41 35.48 ** CV = 19.73 (MOR), 26.69 (NHPT), 54.23 (TS), 8.11 (WA) R2 (%) = 0.78 (MOR), 0.78 (NHPT), 0.68 (TS), 0.91 (WA)

Table 1b.DMRT of the different properties of bagassecement board at 0.60 g/cm3 density. TREATMENT R2 24 1:1 R2 24 1:2 R2 24 1:3 R2 48 1:1 R2 48 1:2 R2 48 1:3 R2 72 1:1 R2 72 1:2 R2 72 1:3 R4 24 1:1 R4 24 1:2 R4 24 1:3 R4 48 1:1 R4 48 1:2 R4 48 1:3 R4 72 1:1 R4 72 1:2 R4 72 1:3 R8 24 1:1 R8 24 1:2 R8 24 1:3 R8 48 1:1 R8 48 1:2 R8 48 1:3 R8 72 1:1 R8 72 1:2 R8 72 1:3 P2 24 1:1 P2 24 1:2 P2 24 1:3 P2 48 1:1 P2 48 1:2 P2 48 1:3 P2 72 1:1 P2 72 1:2 P2 72 1:3

MOR 59.62 bcd 63.77 ab 44.81 i-l 46.13 h-k 44.29 i-m 36.99 nop 22.05 ef 24.90 de 16.12 hi 33.02 abc 28.96 cd 19.39 fgh 37.14 a 34.04 ab 25.89 de 36.45 a 36.30 a 31.05 bc 23.18 ef 23.63 e 15.11 hij 24.95 de 22.94 ef 18.98f gh 25.35 de 24.89 de 21.96 ef 17.30 gh 14.95 h-j 11.80 ijk 22.14 ef 16.71 h 10.81 jk 24.46 de 10.85 jk 8.83 k

WA 83.91 a 53.12 ij 49.66 j-l 73.12 cd 48.28 kl 41.50 mn 75.13 bc 52.12 ijk 40.84 n 68.82 ef 47.82 kl 40.22 n 67.00 f 45.30 lm 38.73 no 69.00 def 49.53 j-l 39.06 n 67.30 f 45.63 lm 39.42 n 62.13 gh 42.13 mn 34.90 o 60.84 h 40.29 n 34.68 o 74.76 bc 66.45 f 60.32 h 77.41 b 65.96 fg 53.13 ij 71.82 cde 55.80 I 51.00 jk


158

Table 2a. ANACOVA on the MOR, NHPT, TS and WA of bagasse-cement board at 0.80 g/cm3 density MOR NHPT TS WA SOURCE OF VARIATION df MS F-value MS F-value MS F-value MS F-value Particle size (PS) 3 6533.84 145.07** 5418.43 65.35** 23.36 20.08** 2153.46 101.67** Soaking time (ST) 2 309.50 6.87** 16.18 0.20ns 0.60 0.52ns 71.49 3.38** PS x ST 6 1047.09 23.25** 2030.49 24.49** 8.99 7.73ns 101.00 4.77** Bag:Cem ratio (BC) 2 6220.82 138.12** 6571.08 79.25** 110.47 94.94** 558.49 26.37** PS x BC 6 143.59 3.19** 322.28 3.89** 3.40 2.93** 111.44 5.26** ST x BC 4 244.15 5.42** 154.13 1.86ns 0.14 0.12ns 37.41 1.77 ns PS x ST x BC 12 108.06 2.40** 268.81 3.24** 3.37 2.90** 36.55 1.73ns Density 1 3232.32 71.77**1763.21 21.27** 0.06 0.06ns 807.27 38.11** CV = 15.08 (MOR), 21.67 (NHPT), 37.23 (TS), 11.25 (WA) R2 (%) = 0.79 (MOR), 0.72 (NHPT), 0.81 (TS), 0.83 (WA)

Table 2b.DMRT of the different properties of bagasse-cement board at 0.80g/cm3 density. TREATMENT R2 24 1:1 R2 24 1:2 R2 24 1:3 R2 48 1:1 R2 48 1:2 R2 48 1:3 R2 72 1:1 R2 72 1:2 R2 72 1:3 R4 24 1:1 R4 24 1:2 R4 24 1:3 R4 48 1:1 R4 48 1:2 R4 48 1:3 R4 72 1:1 R4 72 1:2 R4 72 1:3 R8 24 1:1 R8 24 1:2 R8 24 1:3 R8 48 1:1 R8 48 1:2 R8 48 1:3 R8 72 1:1 R8 72 1:2 R8 72 1:3 P2 24 1:1 P2 24 1:2 P2 24 1:3 P2 48 1:1 P2 48 1:2 P2 48 1:3 P2 72 1:1 P2 72 1:2 P2 72 1:3

MOR 59.62 bcd 63.77 ab 44.81 i-l 46.13 h-k 44.29 i-m 36.99 nop 48.90 ghi 55.70 def 37.56 m-p 55.68 def 55.58 def 44.94 i-l 67.87 a 58.65 b-e 48.73 g-j 55.30 d-g 62.74 abc 49.90 f-i 43.01 i-n 45.24 h-k 24.55 r 43.87 i-n 41.68 j-o 40.57 k-o 56.50 c-f 52.63 e-h 41.75 j-o 35.32 op 31.83 pq 17.22 s 46.42 h-k 31.93 pq 22.28 rs 38.27 l-p 28.00 qr 23.18 rs

NHPT 61.30 abc 59.40 abc 45.45 e-h 65.05 ab 54.85 cd 36.90 h-k 65.90 a 48.50 def 38.45 g-j 41.70 f-i 27.10 lmn 19.20 n 48.10 def 41.50 f-i 27.00 lmn 49.37 def 56.30 bcd 45.15 e-h 55.45 cd 47.00 d-g 40.45 f-i 61.00 abc 41.40 f-i 25.70 lmn 48.55 def 27.05 lmn 21.20 mn 55.25 cd 33.20 i-l 22.15 mn 52.50 cde 30.15 j-m 24.20 lmn 45.40 e-h 28.50 k-n 21.50 mn

TS 3.90 ef 0.52 lm 0.42 m 6.05 abc 1.42 i-m 0.64 klm 6.64 a 1.67 i-l 0.54 lm 6.62 a 1.67 i-l 0.96 j-m 5.89 abc 2.21 hi 1.24 i-m 4.41 de 0.96 j-m 0.83 j-m 5.64 abc 1.72 ijk 1.08 i-m 5.42 bcd 1.33 i-m 0.87 j-m 5.21 cd 1.95 hij 1.57 i-l 6.37 ab 3.24 fg 1.71 ijk 5.41 bcd 2.86 gh 2.19 hi 6.12 abc 3.29 fg 1.60 i-l


159

Table 3a ANACOVA on the MOR, NHPT, TS and WA of bagasse-cement board at 1.0 g/cm3 density. MOR NHPT TS WA SOURCE OF VARIATION df MS F-value MS F-value MS F-value MS F-value Particle size (PS) 3 10933.54 194.16** 10309.07 53.48** 66.94 37.40** 712.67 66.99** Soaking time (ST) 2 289.15 5.13** 116.15 0.60ns 2.26 1.27ns 152.02 14.29** PS x ST 6 968.72 17.20** 1537.61 7.98** 1.66 0.93ns 40.96 3.85** Bag:Cem ratio (BC) 2 8995.31 159.74** 14206.08 73.69** 159.34 89.03** 39.60 41.32** PS x BC 6 385.11 6.84** 983.95 5.10** 8.75 4.89** 146.21 13.74** ST x BC 4 391.07 6.94** 854.89 4.43** 3.06 1.71ns 15.65 1.47ns PS x ST x BC 12 459.19 8.15** 470.44 2.44** 3.16 1.77ns 36.09 3.39** Density 1 7351.03 130.54** 7430.22 38.54** 2.04 1.14ns 878.26 82.56** CV = 13.74 (MOR), 21.82 (NHPT), 35.97 (TS), 9.53 (WA) R2 (%) = 0.82 (MOR), 0.64 (NHPT), 0.79 (TS), 0.87 (WA)

Table 3b. DMRT of the different properties of bagasse-cement board at 1.0 g/cm3 density. TREATMENT R2 24 1:1 R2 24 1:2 R2 24 1:3 R2 48 1:1 R2 48 1:2 R2 48 1:3 R2 72 1:1 R2 72 1:2 R2 72 1:3 R4 24 1:1 R4 24 1:2 R4 24 1:3 R4 48 1:1 R4 48 1:2 R4 48 1:3 R4 72 1:1 R4 72 1:2 R4 72 1:3 R8 24 1:1 R8 24 1:2 R8 24 1:3 R8 48 1:1 R8 48 1:2 R8 48 1:3 R8 72 1:1 R8 72 1:2 R8 72 1:3 P2 24 1:1 P2 24 1:2 P2 24 1:3 P2 48 1:1 P2 48 1:2 P2 48 1:3 P2 72 1:1 P2 72 1:2 P2 72 1:3

MOR 77.11 a 74.48 ab 62.94 c-f 63.04 c-f 64.82 cde 54.48 g-i 68.73 bcd 78.07 a 63.34 c-f 70.61 abc 69.17 bcd 52.21 hij 74.86 a 77.62 a 62.23 def 28.37 p 34.20 nop 42.16 klm 56.59 fgh 58.18 e-h 44.34 k-m 58.67 e-h 59.16 e-h 37.48 m-o 60.12 e-g 64.29 c-f 44.94 klm 46.34 jk 39.40 lmn 31.09 op 49.17 I-k 41.37 lmn 29.80 p 54.20 ghi 40.86 lmn 31.46 op

NHPT 100.25 a 83.65 cd 74.30 c-g 86.85 bc 82.40 cde 74.30 c-g 97.35 ab 80.30 c-f 64.10 g-k 68.00 f-i 58.10 h-m 38.00 op 71.85 d-h 58.65 h-l 42.20 nop 71.85 d-h 53.20 j-n 50.70 k-o 80.30 c-f 75.65 c-g 49.80 k-o 61.90 g-k 70.25 d-h 66.40 f-j 62.30 g-k 50.05 k-o 44.15 m-p 76.70 c-g 50.70 k-o 36.40 op 70.50 d-h 54.40 i-n 34.35 p 71.15 d-h 47.40 l-p 33.35 p

WA 42.74 c-f 27.34 klm 24.30 mno 43.60 b-e 28.75 i-l 22.07 o 46.06 b 27.94 jkl 23.07 no 44.74 bcd 28.95 i-l 23.40 no 40.89 ef 27.32 klm 25.68 lmn 51.13 a 44.98 bcd 32.62 h 40.09 fg 27.75 jkl 24.44 mno 40.74 ef 29.63 h-k 29.44 h-k 39.73 fg 28.79 i-l 28.66 i-l 45.59 bc 37.41 g 28.26 i-l 42.10 def 40.25 fg 30.80 hij 43.86 b-e 37.26 g 31.53 hi


160

CONCLUSIONS AND RECOMMENDATION

The mechanical properties in terms of MOR and NHPT are directly proportional to the increase in board density. Maximum MOR and NHPT values are obtained from boards at 1.00 g/cm3 density level. There is no remarkable indication that TS decreases as the density is increased, but WA improves as the boards become denser. The effect of soaking time is insignificant, indicating that prolonging the immersion of materials from 24 hrs to 72 hrs prior to mixing with cement does not enhance board properties. Stated in another way, soaking time is not directly proportional to increasing board properties.

The highest MOR value for bagasse-cement boards is obtained from boards containing particles retained at 2-mm screen with 1:2 BC. Bagasse boards containing the same particles that are retained at 2-mm screen exhibit the highest NHPT but with 1:1 BC. For dimensional stability, boards with 1:1 BC show the highest TS and WA values at all density levels but with no definite trend as to the effect of particle size.

It is recommended that materials retained at 2 mm mesh be used. A ratio of 1:2 BC should be applied for CBBs with bagasse at 1.00 g/cm3 density level.

LITERATURE CITED

AGRO-INDUSTRIAL RESEARCH LABORATORY DEPARTMENT. Industrial Research and Development, Sugar Regulatory Administration - CY 1996 - 97. CADIZ, E.V. 1998. Raw materials availability for the forest-based industry. Paper presented during the Industry Dialogue and R&D Planning Workshop held at the Maquiling Breeze Resort, Springdale Subdivision, Los Baños, Laguna on 15 April 1998. EUSEBIO, D.A. 2000. Rapid Curing of Cement-Bonded Boards From Agricultural Residues. Proc. of the 7th Inorganic-Bonded Wood and Fiber Composite Materials Conference. Sun Valley Resort, Sun Valley, Idaho, USA. Sep 25 – 27, 2000. EUSEBIO, D.A.; S. KAWAI; Y. IMAMURA and S. SASAKI 1994. IsocyanateInorganic Bonded Composite III. Rapid Production of Cement Bonded Particleboard by Steam Injection Pressing. Journal of the Japan Wood Research Society. Vol. 40, No. 9. HOUSING NEED TARGET 1999-2004. National Shelter Program.


161

Session 8 Alternative Methods for Wood Preservation


162


163

ENVIRONMENTAL FRIENDLY WOOD PROTECTION – FURFURYLATED WOOD AS ALTERNATIVE TO TRADITIONAL WOOD PRESERVATION Stig Lande1, Mats Westin2 and Marc H. Schneider3 1

Wood Polymer Technologies ASA – Haakon VII’s g 1 – P.O. Box 1431 Vika - NO-0115 OSLO - NORWAY ([email protected])

2

SP – Swedish National Testing and Research Institute – P.O. Box 857 – SE-50115 BORÅS – SWEDEN ([email protected])

3

University of New Brunswick and Woodtech Incorporated – 989 Clements Drive – Fredericton, New Brunswick – E3A 7J3– CANADA ([email protected]) Correspondence to: Stig Lande, E-mail: [email protected]

ABSTRACT Wood modified with furfuryl alcohol, ‘furfurylated wood’, is currently being marketed as a non-toxic alternative to traditional preservative treated wood (wood impregnated with biocides). Over the last decade modernised processes for furfurylation of wood have been developed by the authors. These new systems do not add metals or halogens to the product, which is important for an environmentally acceptable product. This paper deals with the environmental aspects and durability of furfurylated wood. Results from several decay tests, emission analysis studies and ecotox tests are presented. The results show that furfurylated wood is highly decay resistant. Furthermore, no significant increase in eco-toxicity of leaching water was found and degradation through combustion does not release any volatile organic compounds or poly-aromatic hydrocarbons above normal levels for wood combustion. Hence, durability enhancement by furfurylation of wood is not believed to be harmful to the environment.

Keywords: furfuryl alcohol; leaching; modified wood; toxicology; wood preservation

INTRODUCTION

Due to the increasing awareness of the hazards of using toxic compounds for wood preservation, limitations have been introduced to the production, trade and use of the most utilized wood preservative, CCA (salts of copper, chromium and arsenic), in several European countries and the USA. Several alternative preservatives for wood are in use. Common for all preservatives is that their mode of action is their toxicity to fungi, bacteria and insects. An alternative to preservation of wood by toxic compounds is chemical modification, which changes the wood structure and wood chemistry so that the wood becomes less susceptible for bio-degradation.


164 The use of furfuryl alcohol (FA) as a wood modification agent has been known for some decades. (Goldstein 1955, Goldstein 1960, Goldstein & Dreher 1960, Anaya et al. 1984, Amaury et al. 1988). In the beginning of the 1990s, Schneider and Westin (Schneider 1995, Westin 1996, Westin et al. 1998) simultaneously developed new catalytic systems for furfurylation of wood. These new systems resulted in relatively cheap manufacturing processes for furfurylated wood with good properties with respect to mechanical strength and stability and biological deterioration. In this study we present results from several tests related to decay properties and environmental aspects of using furfurylated wood. It is of great importance for successful commercialisation that environmental impacts caused by use and disposal of such materials are low. An important question arises: Is the earlier reported high decay resistance (Westin et al. 1998, Westin et al. 2002) based on toxicity of the product or are other protective mechanisms involved? To show the potential in using furfurylated wood as substitute for preserved wood we present results from several decay tests in laboratory and field. In an attempt to answer the question of protective mechanism, results are presented from chemical analysis of leaching water from furfurylated wood, mycological studies on these leaching waters and mycological studies on furfuryl alcohol contaminated water. In addition emissions to air in constant climate, emissions from combustion and eco-toxicological tests on leaching water from furfurylated wood.


Preparation of Wood Materials

The wood material used for all tests was Scots pine (Pinus sylvestris L.) sapwood. For the basidiomycete and terrestrial microcosm tests (extended EN standard 807), wood blocks, 150 × 50 × 120 mm, were sawn from a single beam of Scots pine sapwood. For the field tests in ground and close to ground, 8 × 20 × 200 mm mini-stakes were used, and for the marine field test 25 × 70 × 200 mm specimens were used. The first step of the treatment was a full cell (Bethel process) vacuum-pressure impregnation in a laboratory stainless steel reactor (30-litre capacity). Four different furfuryl alcohol (FA) concentrations were used: 92%, 48%, 30% and 15% FA, respectively, in the waterbased impregnating solutions. The catalysts used for all levels were citric acid and cyclic carboxylic anhydrides and buffering salts were also added. The vacuum step was applied for 45 minutes using a membrane pump, the treating solution was introduced and a pressure of 12 bar was applied for 2h. The specimen surfaces were wiped clean from excess liquid and weighed. The specimens were wrapped in aluminum foil and cured/reacted in an oven at 103ºC for 16 hours. The aluminum foil was then removed and the specimens were post-cured/dried for another eight hours at 103ºC in the oven. All surfaces of the specimens were sanded slightly, in order to avoid the effect of polyFA coating. The resulting modification levels were measured as weight percent gain (WPG), and the four resulting levels were 99-120%, 40-45%, 25-35% and 1017%WPG, respectively. The commercial product called VisorWood has approximately 30% weight gain (WPG) relative to dry weight before modification. The VisorWood material was produced in the industrial pilot plant of Wood Polymer Technologies ASA (WPT) in Moss, Norway and used for the emission tests, leaching tests and eco-tox tests. ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

165

Laboratory testing in three types of unsterile soil (extended EN standard 807)

Modified wood specimens (5 × 15 × 30 mm) were buried to ¾ of their length in the soil in three types of terrestrial microcosms, TMCs. These were: compost soil (TMC 1), soil from the Simlångsdalen test field (TMC 2) and soil from a conifer forest (TMC 3). The characteristics of the conifer forest soil resembled the characteristics of the soil in the Ingvallsbenning test field. Specimens were removed after 6 and 12 months and the weight loss for each specimen was calculated. The weight loss values for 32 weeks were interpolated from the 6- and 12-month data.

Field tests in ground contact (modified version of EN 252)

Mini-stakes (8 × 20 × 200 mm) were put out in ground contact in three test fields, buried to ¾ of their length. The fields were located in: Simlångsdalen (N56°43’ E13°07’), Ultuna (N59°49’ E17°38’) and Ingvallsbenning, all in Sweden. Simlångsdalen is a field with dominating brown rot decay. Ultuna is a fertile field with dominating soft rot and bacterial decay and Ingvallsbenning is an acidic conifer forest field with a dominant white rot decay. fields the rating was done in accordance with the European standard EN 252, in which no decay (sound) is rated 0, slight decay 1, moderate decay 2, severe decay 3 and failure is rated 4. An Index of Decay was calculated from the average rating figures of each group of stakes, when rated according to the standard. In this system an index of 0 means that all stakes are sound and 100 means that all stakes have failed. Field test in sea water with marine borer activity

A marine field test of furfurylated wood at four modification levels according to EN 275 was performed. Reference treated wood were two levels of NWPC Standard No.1 (2.6% and 0.6% water-free chromated copper arsenate salt in water solution). In January 1999, all specimens were placed on rigs close to the sea bottom near Kristineberg Marine station, Northwest of Gothenburg, Sweden.

Chemical Analysis of Leaching Water from VisorWood

Ten samples of furfurylated Pinus sylvestris L. (15 × 25 × 50 mm) were subjected to leaching as described in EN 84 (1997). The water for all ten samples was changed ten times during two weeks and each leachate was saved separately. The leachate samples for days 1, 3, 5, 7 and 9 were analysed by HPLC for the content of furfuryl alcohol. Leachates from days 1, 5 and 9 were also saved for use as growth media for fungi, as described below.


166 Radial Growth of Basidiomycetes on VisorWood Leaching Water Media

In addition to leaching water from furfurylated wood, ten samples of untreated Pinus sylvestris L. (15 × 25 × 50 mm) were subjected to leaching as described above. Malt (16 g) and agar (8 g) were added to 400 ml pooled leaching water samples from both treated and untreated wood for days 1, 5 and 9 in separate bottles and then sterilized by autoclaving before being poured into 9 cm Petri dishes (20 ml). Six replicates for both treated and untreated leachate water were inoculated with a 3 mm cylinder of actively growing mycelium placed upside down for each of the three wood decaying fungi; Coniophora puteana (Schumacher ex Fries) Karsten (BAM Ebw. 15), Gloeophyllum trabeum (Persson ex Fries) Murrill (BAM Ebw. 109) and Coriolus versicolor (Linnaeus) Quélet (CTB 863 A). The radial growth was recorded in 4 directions daily for 7 days with an accuracy of 1 mm.

Radial Growth of Basidiomycetes on furfuryl alcohol containing medium

Furfuryl alcohol was mixed directly into malt-agar growth medium in concentrations of 0.2, 0.5 and 1.0 mg/ml. Six replicates of each concentration and control sample without FA were inoculated with three wood decaying fungi as described above. The radial growth was recorded in 4 directions daily for 7 days with an accuracy of 1 mm.

Emissions of volatile organic compounds (VOC) to air

Emissions of VOC from furfurylated wood (VisorWood) were sampled and analysed according to European standards ENV 717-1 (1998) and ENV 13419-1 (1999). Furfurylated wood was loaded at 1.0 m2 per m3 into a test chamber (225 dm3, polished stainless steel) conditioned to 23°C and 45% (±3%) relative humidity (RH). The test chamber was ventilated with an air velocity of 0.1-0.3 m/s so that the chamber was theoretically ventilated once per hour (±0.05). Air samples were taken after 3, 10 and 28 days by pumping air through Tenax and Sep-pack (C-18 polymers coated with 2,4dinitrophenylhydrazine) adsorption tubes for VOC and aldehyde sampling. The VOC were quantified by thermal desorption of the Tenax adsorption tubes followed by GC/MS analysis. The aldehydes were eluted with acetonitrile and analysed by HPLC with UV detection. Appropriate internal standards were used in each case.

Fire test and analyses of smoke gases

Fire tests were conducted on VisorWood and Scots pine specimens according to ISO 5659-2 (1994) (test in non-ventilated box). The smoke gases were analysed by GC-MS, GC-FID and FTIR spectroscopy, (SP, Fire Technology 2002).

Ecotoxicological test on leaching water from furfurylated wood

Wood specimens (50 × 25 × 15mm) were taken from larger furfurylated boards (400 × 100 × 25mm), impregnated and cured at the WPT laboratory to a furfurylation WPG of 30%, and from untreated pine control boards. The specimens were leached in water ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

167 according to EN 84 (1997). The specimens were extracted in 100 ml aliquots/specimen for 24 hours at room temperature. Water volumes were exchanged daily except during weekends (ten times in all, according to the standard). In order to find NOEC and EC50 values for leaching water from furfurylated wood the following tests were carried out on a broad range of dilutions on the pooled leachates: Growth Inhibition in Green Algae (OECD TG 201, 1984), Daphnia, Immobilisation Test (OECD TG 202, 1984), Acute Toxicity in Zebra Fish (OECD TG 203, 1992) and “Microtox” (ISO/DIS 11348-3, 1996). Eetvelde et al. (1998) used a method to express toxic units (TU) as a value for toxic response. TU is defined as the dilution ratio of test water giving the EC50 value, i.e. the more diluted the higher the TU value.

RESULTS AND DISCUSSION Laboratory tests in terrestrial microcosm (TMC)

The results from TMC testing in three types of soils can be seen in the right-hand columns (weight loss caused by decay) of Table 1. Prior to the test, the samples were leached and the resulting weight losses can be seen in the third column. There was no obvious difference between the untreated and the acetone-extracted pine sapwood. There were only minor differences found between the weight losses from leaching of the control samples and leaching of the low-furfurylated samples, whereas for the medium- and high-furfurylated samples the leachable substance amount is reduced when the furfurylation level is increased (Table 1). The efficacy of furfurylation is clearly shown by the low weight loss due to decay in all soil types, compared to the high weight loss of the control samples. Furthermore, the CCA-treated specimens were heavily decayed in the compost soil unlike the furfurylated. Table 1: Results for furfurylated wood after 32 weeks or 12 months in 3 different TMCs (n=10). Modification type (/chemical)

Pine Control I

Modification level

Weight loss during leaching %

untreated Acetone-extracted

2.4 2.1

65.5 (6.9) 61.4 (6.3)

61.6 (1.7) 60.0 (3.1)

20.1 (2.0) 14.8 (2.6)

2.4 1.7 0.6

2.6 (0.6) 1.2 (0.3) 0.8 (0.4)

3.4 (1.2) 2.1 (0.2) 1.6 (0.3)

8.5 (0.5) 5.0 (0.7) 1.9 (0.4)

Furfurylated pine WPG = 22 WPG = 41 WPG = 60

Weight loss (%) in TMC 1 (Compost soil) 32 week expo

Weight loss (%) in TMC 2 (Simlångsdalen soil) 32 weeks exposure

Weight loss (%) in TMC 3 (Conifer forest soil) 12 months exposure

Pine Control II

untreated

not leached

59.5 (15.6)

32.1 (16.8)

17.1 (7.8)

CCA (NWPCStandard no.1)

4 kg/m³ (class AB) 9 kg/m³ (class A)

not leached not leached

25.4 (8.6) 15.6 (3.5)

4.8 (4.3) 2.5 (2.4)

2.9 (0.6) 1.3 (0.4)


168

In-ground field test with mini-stakes

A comparison of the performance of furfurylated wood mini-stakes in Simlångsdalen with the performance in Ultuna and Ingvallsbenning is shown in Table 2. Table 2: Condition of pine sapwood mini-stakes (8x20x200mm) after 8 years in three Swedish fields. No. of Treatment level Index of Decay (0-100% decay) Treatment W (kg/m³) replica Simlå U Ingvall

PG Furfurylation

ngsdalen

15 33* 50 >100

CCA (NWPCa Standard No.1)

6, 6, 6 6, - , 6, 6, 6 6, 6, 6

92 19* 4 0

2.1b 9.0b

6, 6, 16 6, 6, 16

100 21

-

29, 29, 8

100

Untreated controls

-

a Nordic Wood Preservation Council

b CuO (19 wt-%); CrO (36 wt-%); As O (45 wt-%) 3 2 5

ltuna

sbenning

71

92*

17 0

25* 0*

96 62

92* 25*

100

100*

* Six years exposure

Although the decay type is very different in these three fields, the performance is quite similar in all three fields. At the higher modification levels furfurylation seem to provide a performance equal to or better than CCA in the retention level approved for use in HC4, 9 kg/m3 (based on crystal water free salt, which corresponds to 12 kg/m3 based on preservative product).

Marine field test

All untreated pine sapwood samples in the first set of controls were rejected at the assessment after one year (see Table 3, column to the right). They were so heavily attacked by Teredo navalis (shipworms) that the teredo tunnels covered more than 90% of the X-ray pictures. They were replaced with new controls and at each assessment during the forthcoming years all controls had failed due to teredo attack and were replaced. The reference CCA-treated samples at low retention level (4 kg/m3) have failed with an average service life of 3.2 years. Even one CCA reference sample at the high retention level is slightly attacked by teredo. The furfurylated samples at the lowest level (WPG=11) have been given an overall rating of “failed” although two severely attacked samples still remain in test. However, all furfurylated samples at rather low (WPG=29), medium (WPG=50) and high modification level (WPG=120) were all rated sound after 4 years of testing.


169

Table 3: Condition of pine sapwood samples (25 x 75 x 200 mm) after 4 years of exposure on test rigs in the bay outside Kristineberg Marine Research Station. Rating, No. of samples Overall Aver. Chemical No. Terenid rating Service retention classified as of attack Chemical treatment

Furfurylation

(kg/m3)

sound

attacked

rejected

(0-4)

5 5 5 5

5 5 5

2 -

3 -

3.6 0.0 0.0 0.0

Failed Sound Sound Sound

-

4b 18b

6 6

5

1

6 -

4.0 0.2

Failed Sound

3.2 -

-

8+5+5+7

-

-

8+5+5+ 7

4.0c

Failedc

1.0c

11 29 50 120

CCA a

(NWPC Stand. No.1)

Untreated controls

life

replica

WPG

-

a Nordic Wood Preservation Council 1year

(years)

b CuO (19 wt-%); CrO (36 wt-%); As O (45 wt-%) c 4 sets, each lasted 3 2 5

Chemical analysis of leaching water from VisorWood

Average FA concentration in the leachate was 0.23 mg/l. The initial concentration (first exchange) was slightly higher and the final slightly lower, Figure 1. This represents a total of 12.3 g of leached FA per m3 of wood. Detection limit of the HPLC analysis was 0.05 mg/l.

Figure 1. Concentration of furfuryl alcohol in combined leaching water (n=10) (±1 SD for the HPLC method).

Radial growth of Basidiomycetes on VisorWood leach water media

The results from the radial growth for the three Basidiomycetes on day 7 are presented as relative growth on media contaminated by leaching water from furfurylated wood and untreated pine in Figures 2 and 3 respectively. There were small but significant differences in total growth for VisorWood leachate media; Pinus control leachate and ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

170 pure agar solutions on day 7. Even though total growth on VisorWood leachate media and Pinus control leachate on day 7 were less than on the controls, the growth rates for the last 4 days were the same. Growth of the fungi for some solutions was slightly slower during the first 3 days, but thereafter growth proceeded at the same speed as for fungi growing on malt agar alone.

Figure 2. Radial growth relative to control of Coniophora puteana, Gloeophyllum trabeum and Coriolus versicolor (±1 SD, n=6) on malt-agar medium made of EN84 leaching water from furfurylated wood on day 1 (black), day 5 (white) and day 9 (grey).

Figure 3. Radial growth relative to control of Coniophora puteana, Gloeophyllum trabeum and Coriolus versicolor (±1 SD, n=6) on malt-agar medium made of EN84 leaching water from P. sylvestris L. on day 1 (black), day 5 (white) and day 9 (grey).

Radial growth of Basidiomycetes on furfuryl alcohol containing medium

The results from the radial growth for the three Basidiomycetes on day 7 are presented in Figure 4 as radial growth on FA-contaminated medium relative to the control. No significant growth rate inhibition of any tested fungi could be detected in control ICECFOP1 – International Conference on Environmentally-Compatible Forest Products Oporto, Portugal, 22-24 September 2004

171 samples with no furfuryl alcohol present. A concentration of 1.0 g/l FA in the growth medium did not inhibit or reduce growth rate of the fungi. In these experiments the FA concentration in the growth media was 1000-4000 times higher than the highest concentration found in the EN 84 leachates.

Figure 4. Radial growth relative to control of Coniophora puteana, Gloeophyllum trabeum and Coriolus versicolor on agar medium containing furfuryl alcohol at 0.2 (black), 0.5 (white) and 1.0 mg ml-1 (grey).

Emissions of volatile organic compounds to air

VOC and aldehyde emission results are presented in Table 4. Furfural is the major emission after 3 and 10 days at 214 and 107 µg/m3 respectively. Furfural is a natural hydrolysis product from wood hemicellulose, and occurs also in untreated wood. Sampling after 28 days showed that, at 60 µg/m3, furfuryl alcohol is the major component. All the other compounds in Table 1 had values below those of furfuryl alcohol, and the amount of each compound decreased rapidly with time. The total VOC from VisorWood was low. The major component was furfural, and the emissions of this compound were reduced to less than ¼ of the starting value after one month. The same pattern was seen for most other detectable compounds except furfuryl alcohol, which was emitted at a constant, low level during the testing period.


172 Table 4: Volatile organic compounds (VOC) analysed by GC-MS after sampling air from a climate chamber containing FA-treated wood after 3, 10 and 28 days. Compound

Concentration in climate chamber (µg/m3) 3 days

10 days

28 days

214 50 44 30 18 15 8 7 6 3 3 3 1

107 ? 25 16 5 4 3 3 2 1 1 1 1

47 60 13 9 2 1 2 1 1 1

Furfural Furfuryl alcohol Tetrahydrofuran-methanol Toluene Furandiene Benzene/alcohol Benzaldehyde Hexanal Xylene/ethylbensene α-pinene Methylcyclohexane Butyl acetate MIBK/hexanone

Fire test and analyses of smoke gas

Smoke gas analyses for optical density, CO, VOC and PAH are presented for three different combustion scenarios in Table 5. All reported values were noticeably lower for VisorWood compared to the reference material, except for the CO concentration in the 25 kW/m2 scenario, which was slightly higher.

Table 5: Results from smoke gas analysis of VisorWood and untreated pine. Mode

Optical density, Ds max CO concentration, ppm VOC produced, mg PAH produced, mg

25 kW/m2

25 kW/m2 pilot flame

50 kW/m2

Reference

VisorWood

Reference

VisorWood

Reference

VisorWood

664 5900 590 2.3

352 6000 370 1.8

402 3200 260 1.1

70 2200 120 1.0

131 160 51 1.7

70 77 28 32

8-16

Concrete