Development of Commercial Wood Preservatives Efficacy

Chapter 20

Furfurylation of Wood: Chemistry, Properties, and Commercialization 1

2

3

Stig Lande , Morten Eikenes , Mats Westin , and Marc H. Schneider

Downloaded by CORNELL UNIV on August 4, 2012 | http://pubs.acs.org Publication Date: April 2, 2008 | doi: 10.1021/bk-2008-0982.ch020

4

1

Wood Polymer Technologies ASA, NO-1596 Moss, Norway Norwegian Forest Research Institute, NO-1432 Ås, Norway SP Swedish National Testing and Research Institute, Box 857, SE-50115 Borås, Sweden University of New Brunswick and Infinity Wood, Ltd., 989 Clements Drive, Fredericton, New Brunswick E3A 7J3, Canada 2

3

4

Furfurylated wood has proven to be a biological inert material and, thus, is an environmentally-benign alternative to wood treated with heavy metals or tropical hardwoods. The chapter will deal with the history, properties and uses of furfurylated wood to date.

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, C C A (salts of copper, chromium and arsenic), in several European countries and the U S A . Several alternative preservatives for wood are in use. Common for all these preservatives is their toxicity to fungi, bacteria, and insects along with non-target animals and humans. A n alternative to employing toxic compounds is chemical modification, which changes the wood structure and wood chemistry so that the wood is less susceptible to bio-deterioration. One promising chemical modification method is the furfurylation process, which uses furfuryl alcohol that polymerizes in the wood structure and reacts with the wood substrate.

© 2008 American Chemical Society In Development of Commercial Wood Preservatives; Schultz, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

337

338


History of Furfurylation Early research with furfuryl alcohol (FA) concerned FA-resins as a substitute to phenol formaldehyde resins (PF resins) for adhesives, molding compounds and inorganic composites. In the 1930's and 1940's several processes and patents concerning paper laminates, grinding wheels and metal casting moulds from FA-resin impregnated paper or inorganic compounds such as silicon carbide, were developed (7). Metal casting moulds and grinding wheels are still the main commercial applications for furfuryl alcohol. Research concerning modification of wood with furfuryl alcohol, hereafter referred to as "furfurylation" of wood, was initiated by the "pioneer of wood modification", Dr. Alfred Stamm, in the early 1950's. Additional early work on wood furfurylation was done by Irving Goldstein (2-4). Goldstein's process employed zinc chloride as a catalyst and mainly modified wood veneers. He reported that treatment with 90 % (weight/volume) solutions of F A resulted in products with high dimensional stability and increased resistance to fungal decay, alkali and acids (2, 4). This research lead to a small-scale production of furfurylated wood in the 1960's by Koppers Wood Inc. (USA). Among these products were laboratory bench tops, pulp mixer rotor-blades and knife handles (5). Anaya developed a process very similar to the Stamm/Goldstein process (6, 7). However, this process was never commercialized. The major problem with these processes was that zinc chloride used as a catalyst deploymerized the cellulose, which consequently reduced the strength properties of the modified wood. Professor Marc Schneider and Dr. Mats Westin developed, separately and more or less simultaneously during the early 1990's, alternative catalysts for the furfurylation process. They both based their polymerization process on similar paths of chemistry using cyclic carboxylic anhydrides, mainly maleic anhydride, as key catalysts (5-70). These novel systems lead to solutions that are chemically stable at room temperature and have good properties with regard to the impregnation process. The final furfurylated wood products, shown to be of high quality, are described in this chapter. The properties of this second generation of furfurylated wood are generally superior to those wood modified by the firstgeneration process developed Stamm and Goldstein (8).

Chemistry of Furfurylation The acid catalyzed polymerization of F A in wood has a very complex chemistry. The resulting polymer is a highly branched and cross-linked furan polymer that is chemically bonded to the wood structural components. However, the reaction parameters (i.e., type of and concentration of catalyst, the p H of the solution, processing temperature and time, and the presence of water) highly affects the final product by the degree and type of chemical bonds to wood, the

In Development of Commercial Wood Preservatives; Schultz, T., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2008.

339 major type of polymer, units, and the degradation of wood components (mainly depolymerization of cellulose and hemicelluloses by acid hydrolysis). The reaction types can be divided into several categories: a) homopolymerization of F A , b) co-polymerization of F A and additives or wood extractive substances, and c) grafting of F A or polymerized F A to wood cell wall polymers (77, 72).


Homo-polymerization of FA Early studies stated that the reaction rate is a simple function of the p H inaqueous furfuryl alcohol solutions (7). However, there are several competing reactions and the rate of the dominating reaction mechanism is affected by factors other than pH. These factors include temperature, FA/water-ratio, and the presence of oxygen and/or weak organic bases. In the presence of a powerful acid catalyst, such as para-toluene sulphonic acid (PTSA), the dominating reaction is a very rapid radical polymerization which results in a high degree of ring-opening reactions (75). However, these kinds of systems are not suitable to use for impregnation and modification of wood since they are not stable at room temperature. Solutions used for impregnation of wood must be stable during storage and during impregnation of wood, but should react readily when the impregnated wood is heated. In the initial phase of the polymerization of these stable systems, there are two competing condensation reactions as are illustrated in Figure 1 : Reactions 1 and 2a (14). The reaction products according to reaction 1 normally dominate. High reaction temperature and high concentration of furfuryl alcohol further suppress the formation of the ether bridges in reaction 2a (14).

Figure 1. Initial polymerization of FA (acid catalyzed condensation reaction).

Since the ether-bridge in reaction 2a is an unstable intermediate, it is believed that it undergoes reaction 2b to form formaldehyde as an intermediate



340 product.The terminal methylol group could later split off as formaldehyde. The latter reaction is highly temperature dependant. Initially, a linear polymer is formed. But as the polymerization progresses, cross-linking via reactions 3 and 4 occurs more frequently and the polymer becomes infusible. It is believed that formaldehyde, formed according to reaction 2b and by splitting off of terminal methylol groups, acts as a cross-linking agent according to reaction 4. However, the dominating cross-linking reaction is by condensation reaction 3, shown in Figure 2(1,14). Furthermore, it is commonly assumed that some cross-linking in the late stages of polymerization occurs through ring-opening reactions. A competing reaction to the F A ring-opening reaction is formation of levulic acid. This reaction, and oxidation reactions, can be suppressed by the presence of a weak organic base, e.g. triethanolamine (7,75). However, no significant difference in the polymer structure was found between polymerization carried out in the presence or absence of oxygen (14).

Figure 2. Branching (reaction 3) and cross-linking (reaction 4) of FA-polymer chains.

Reaction of FA to the Wood Cell W a l l Components The high degree of permanent bulking of the furfurylated wood cell wall is considered evidence of reactions between polymer and wood at an early reaction stage. Similarly to low-molecular weight PF-resins, F A impregnation leads to super-swelling of the cell wall (i.e. a slightly higher degree of swelling than for water). However, as PF-oligomers become more and more hydrophobic during polymerization, they begin to migrate out of the cell wall. This results in a



341 maximum anti-shrink efficiency (ASE) of approximately 50 % for PF resins. Since FA-oligomers are even more hydrophobic than PF-oligomers at the same degree of polymerization (DP), one would expect anti-shrink efficiencies below 50 % for furfurylated wood. Instead, the A S E for FA-modified wood can be as high as 80 %. This is a strong indication that grafting reactions occur between the FA-polymer and wood during the early stages of polymerization. Grafting of poly-FA onto cellulose was previously accomplished using ferric salts and peroxides as catalysts (16). However, these types of catalysts are too reactive to be used. Therefore, by using cyclic anhydrides, as described in this chapter, grafting reactions likely occur between the poly-FA and cellulose, hemicelluloses and, principally, the lignin in wood. Reaction 5 (Figure 3) is a possible reaction between F A and a guaiacyl unit of lignin (the predominant monomeric lignin unit in softwood lignin).

Lignin guaiacyl unit

Figure 3. Suggested reaction between FA and a lignin unit

Biological Resistance Furfurylation of wood can be done with different F A levels to obtain wood with varying degrees of modification, expressed as Weight Percent Gain (WPG). W P G is thereby an essential parameter when discussing the effects of furfurylation. A moderately low treatment level (WPG around 30%) will have only a minor effect on the physical properties of the modified wood. The major property improvement observed with a moderate W P G is increased resistance to biological degradation. Furfurylation has been shown to be a good method to increase the resistance against biological deterioration of wood using non-toxic chemicals. Various research groups have carried out laboratory and field tests on FA-modified wood. Irving Goldstein was the first to show that furfurylated wood has an increased resistance to biological decay. His formulation was based on high concentration of F A (90 %) that resulted in a high W P G . For commercial applications it is important to know the minimum treatment level required to


342 achieve a desired property, to minimize the cost of the process. The studies reported in this chapter are published in international journals (17, 18) and some tests conducted for Wood Polymer Technologies A S A (Porsgrunn, Norway).


Fungal Decay Resistance The results from laboratory testing (according to a modified version of A W P A Ε10, soil block test) of the decay resistance of furfurylated wood with basidiomycetes cultures of brown and white rot fungi are presented in Table 1 (mass loss caused by Postia placenta and Phanerochaete chrysosporium, respectively). The mass loss of furfurylated wood of medium and high W P G caused by brown rot decay are less than the mass loss of samples treated with C C A to a retention for Use Class 4, ground contact. Only for the furfurylated samples with a low treatment level have mass losses slightly higher than for CCA-treated samples. The conclusion is that the resistance to brown and white rot decay fungi is high for furfurylated wood treated to a W P G of 35 or more. Laboratory tests with terrestrial microcosms (TMC) are more realistic test set-ups than typical laboratory decay tests, as long as ordinary soils are used. Accelerated conditions are gained by controlling the temperature and humidity. The results from T M C testing in three types of soils are presented in Table 2 (mass losses caused by fungal decay). Prior to the tests, the samples were subjected to artificial weathering by leaching in water. The mass losses caused by leaching are shown in the third column. No differences were found between the untreated samples of pine sapwood, and pine sapwood samples free of extracts (extracted with acetone). Only minor differences were found between the mass losses caused by leaching of the control samples and leaching of the FA-modified samples. In-ground exterior field tests are the most realistic test setup that can be performed for testing decay properties. The climate and soil type will affect the rate of degradation. Therefore, outdoor exposure tests are normally run for at least several years using multiple test sites to obtain realistic results. A comparison of the performance of furfurylated mini-stakes in three different fields in Sweden, Simlângsdalen, Ultuna and Ingvallsbenning, are presented in Table 3. Although the type of decay is very different in these three fields, the performance of furfurylated wood is quite similar in all three fields. At the higher modification levels, furfurylation seems to provide a performance equal to or better than C C A in the retention level approved for Use Class 4.

Marine field tests ( E N 275) Wood used in marine applications is susceptible to marine borers. Laboratory tests exist for only some marine borers; the main option is to conduct


343


Table 1. Mass loss due to brown rot (Postia placenta) and white rot (Phanarochaete chrysosporium) decay in laboratory assays after 16 weeks. Mass loss caused by Postia placenta

Mass loss caused by Phanerochaete chrysosporium

Control I

Untreated

48.3

22.7

Furfurylated 1

W P G = 25 W P G = 125

9.2 1.7

4.5 2.4

Control II

Untreated

65.6

24.2

Furfurylated II

W P G = 23 W P G = 128

6.9 1.1

5.0 2.9

Control III

Untreated

60.0

20.0

Furfurylated III

W P G = 35 W P G = 75 W P G = 120

4.3 4.3 2.4

2.7

Untreated

44.1

15.2

4.0

1.1

Control IV

9 kg/m

CCA

3

-

Reprinted from Properties o f Furfurylated W o o d b y L a n d e et.al. from Scandinavien Journal o f Forest Research, www.tandf.no/forest, 2004, 19, 22-30, b y permission o f T a y l o r & Francis



Untreated Acetoneextracted

Modification level

1.5 1.6

Untreated

Birch Control

Furfurylated birch W P G = 40

5.6

85.8

42.8 25.2

94.5

7

73

6.6

30.0

5.4 2.1

41.9

3

51

5.2 2.8 1.8

58.5

62.9

7.1

28.2

2.9 1.3

17.1

4

57

8.5 5.0 1.9

14.8

20.1

R e p r i n t e d from Properties o f Furfurylated W o o d b y L a n d e et.al. from S c a n d i n a v i e n J o u r n a l o f Forest Research, www.tandf.no/forest, 2 0 0 4 , 19, 2 2 - 3 0 , b y p e r m i s s i o n o f T a y l o r & F r a n c i s

3

3

4 kg/m 9 kg/m

C C A (NWPCStandard no.l) Not leached Not leached

Not leached

Untreated

Pine Control III

Not leached Not leached

Untreated

5.7 2.1 1.0

76.7

2.1 2.4 1.7 0.6

78.0

2.4

Mass loss (%) Mass loss (%) Mass loss (%) Mass loss (%) due to in TMC 1 in TMC 2 in TMC 3 leaching,

Furfurylated pine W P G = 25

Pine Control II

Furfurylated pine W P G = 22 W P G = 41 W P G = 60

Pine Control I

Modification type (/chemical)

Table 2. Mass loss for treated and untreated pine sapwood, after 12 months in 3 different T M C s (soil microcosms).


345 Table 3. Condition of pine sapwood mini-stakes (8x20x200mm) after 8 years in three Swedish fields. Treatment


Furfurylation

CCA

Treatment level WPG kg/m 3

15 33* 50 >100 b

2.1 9.0

b

Index of Decay (0-100% decay) Site A Site Β Site C 92 19* 4 0

71

92*

17 0

25* 0*

100 21

96 62

92* 25*

100

100

100*

a

( N W P C Standard N o . l ) _

Untreated controls

a) Nordic Wood Preservation Council, b) CuO (19 wt-%); Cr03 (36 wt-%); As205 (45 wt-%) *Six years exposure. SiteA - Simlingsdalen, SiteB - Ultuna, SiteC Ingvaldsbenning

tests in an actual marine site. In Europe, marine testing is done according to European Norm EN-275. The furfurylated wood was found to perform well in marine tests. A l l untreated pine sapwood samples in the first set of controls were rejected when inspected after one year (Table 4, column to the right). Specifically, the controls were so severely attacked by Shipworms (Teredo navalis) that the teredo tunnels covered more than 90% of the X-ray pictures. The untreated samples were replaced with new controls and, at each annual assessment during the following years all controls failed due to teredo attack and were again replaced. Conversely, the furfurylated samples at medium to high modification levels were all rated sound after 5 years of marine exposure.

Termite resistance Much deterioration is caused by termites. Although termites only exist in certain areas, the economical value of the wood destroyed is greater than the value of wood destroyed by decay fungi. Furthermore, some termites are capable of destroying dry wood in buildings that are typically not susceptible to fungal degradation. Furfurylation of wood appears to reduce its susceptibility to termite attack; however, this is based on only a few tests. One laboratory termite test was done at the University of Hawaii, following A W P A E l Standard. For untreated pine the average mass loss was 66 %, which gives an A W P A rating 0 (total failure).


346 Table 4. Condition of pine sapwood (25 χ 75 χ 200 mm samples) after 5 years of exposure on test rigs in the bay outside Kristineberg (Sweden) Marine Research Station. Chemical retention


Chemical treatment Furfurylation

CCA (NWPC Stand. No. 1) a

Untreated controls

wpg 11 29 50 120

Replica 3

kg/m

b

4 18

_

b

η 5 5 5 5 6 6

8+5+5 +7

No. of samples classified as attacked sound Attack 5 5 5

-

-

Overall rating

reject (0-4) 5 4.0 0.0 0.0 0.0

-

_

6

5

1

-

_

_

8+5+ 5+7

Failed Sound Sound Sound

4.0 0.2

Failed Sound

4.0°

Failed

0

a) Nordic Wood Preservation Council, b) CuO (19 wt-%); Cr03 (36 wt-%); A s 0 (45 wt-%). c) Controls; 4 sets, each set lasted only one year 2

5

For untreated beech the average mass loss was 38 % with an A W P A rating of 4.0 (severe attack). For furfurylated pine sapwood the average mass loss was 8 % and the A W P A rating 7.4. However, for furfurylated beech the average mass loss was only 2 % ( A W P A rating 9.6). For approval of a treatment as effective against termite attack, a visual rating of at least 7 is required. However, the maximum single mass loss accepted is 5% with an average of 3%. Only the furfurylated beech fulfilled this criterion. Therefore, the general conclusion is that furfurylation provides good but not total termite resistance. In-ground mini-stake tests in fields with high subterranean termite activity were used to test termite resistance in outoor exposure. Generally, highly furfurylated wood performed very well. A l l stakes were rated sound after one year in the Indonesian Bogor field, whereas very little remained of the control stakes (93-98% mass loss). In a supplementary test in the Indonesian Bandung field, medium-furfurylated pine (WPG=43) also seems to be resistant to termite degradation (Table 5).

Mechanical and Physical Properties of Furfurylated Wood Chemical modification of wood can affect a wide range of properties. Cellulose, hemicelluloses, and lignin are intimately combined to form the


347 Table 5. Performance of mini-stakes (8 χ 20 χ 200 mm) in the Bogor and Bandung, two Indonesian termite fields

Material

Mod. level

Scots pine control


Furfurylated pine

W P G = 15 W P G = 43 W P G = 115

Agathis pine control Furfurylated Agathis

W P G =15 W P G = 115

Sengon control Furfurylated Sengon

W P G = 15 W P G = 115

Bogor field, weight loss (%), month

Bandung field, weight loss (%), month

3 33

6 47

12 93

3 10

6 66

12 85

3

16

62

-

-

-

1

1

2

4 0 0

62 0 0

65 0 0

36

7

95

2 0

2 1

61 2

16

85

98

2 0

9 1

85 1

Reprinted from Properties o f Furfurylated W o o d by Lande et.al. from Journal o f Forest Research, www.tandf.no/forest,

2004, 19, 22-30,

Scandinavien

b y permission o f

T a y l o r & Francis

anatomical structure of wood. Chemical modification introduces chemicals that react with one or more of the three wood constituents to varying amounts. Typically, the change in wood properties is related to the degree of the treatment. A slight chemical modification will have only a minor impact on wood properties, while high treatment levels will have a greater influence. The optimum treatment level is determined by the intended use and the manufacturing costs.. In furfurylation, the treatment level is conventionally expressed as W P G . By proper dilution of the impregnation solution, it is possible to obtain a W P G ranging from 10 to greater than 100% for low-density wood, with the W P G determined by the F A concentration and wood density. Low-density woods have a relatively large void volume, resulting in a high W P G compared to dense wood. Furfurylation of wood results in polymer formation in wood cell lumens and cell walls. Polymers formed in the cell wall are fixed and will to some extent replace adsorbed moisture, resulting in a permanently swollen or bulked cell wall. This "bulking" effect reduces dimensional changes with changes in moisture content. This dimensional stabilization can be expressed as the antishrink efficiency (ASE). The permanent bulking and grafting of F A polymer to



348 the cell structure will also affect the stiffness, strength and brittleness of the wood. Stiffness can be expressed as stiffness stabilization efficiency (SSE). For furfurylated wood, the volumetric A S E has been demonstrated to be high, even at low W P G . For example, at WPG=32 the A S E is close to 50 %, and at WPG=47 the A S E is approximately 70 % (see striped bars in Figure 4). The stiffness stabilization efficiency (SSE, Grey bars in Figure 4) is also high with moderate W P G (77). The increase in hardness is moderate at low WPGs but very high at high WPGs. The drawback is the decreased impact bending strength at medium to high WPGs (negative bars), although the drastic reduction shown by the bars in the diagram is to some extent an artifact caused by using much smaller samples than the standard specifies. In other impact bending tests of furfurylated wood with correct test specimen size, the results show a more modest reduction of impact strength (19). Furfurylated wood with W P G of about 30 % was also exposed to elevated temperature and humidity to investigate the long-term effect on impact bending strength. In general, this W P G level resulted in an impact bending strength loss of 25-35 % after exposure.

Environment and Eco-toxicity Successful commercialization requires low environmental impacts with the use and disposal of furfurylated wood. The high decay resistance of furfurylated wood is believed to be caused by inertness of the modified wood to fungal enzymes, rather than to toxicity. In an attempt to understand the protective mechanism, several studies were performed. Information obtained included toxicity of airborne emissions, toxicity of water leachates and potential release of toxic chemicals upon disposal. F A is a liquid that is classified as a hazardous chemical. F A is a strong solvent and a flammable liquid. Polymerized F A is non-toxic and safe to use and dispose. However, chemical reactions that take place during polymerization might leave unreacted byproducts, including monomeric F A , in the polymer. These could be phytotoxic and, therefore, contribute to the decay resistance. Several studies of the furfurylated wood have helped to understand these potential side effects. Leach water from furfurylated wood was tested for F A content, and only extremely small concentrations of F A were found. Radial fungal mycelium growth test of basidiomycetes on FA-contaminated malt agar growth medium showed no growth inhibition, even with 1000 to 4000 times the concentration found in the leaching water. This suggests that any unreacted F A in the furfurylated wood does not contribute to the resistance against decay fiingi. Leached water from furfurylated wood has also been tested for eco-toxicity at Toxicon Laboratory in Sweden. Results were compared to leach waters from



349

-75 % Figure 4. Impact strength, stiffness, dimensional stability and hardness of samples by treatment level (WPG) (Reprinted from Properties of Furfurylated Wood by Lande et. al from Scandinavien Journal of Forest Research, www.tandf.no/forest, 2004, 19, 22-30, by permission ofTaylor & Francis.)

other wood and building materials. The furfurylated pine wood had very low toxicity units, lower even than some untreated woods such as western red cedar, a tropical hardwood (azobé) and Scots pine (Table 6). Air emissions from furfurylated wood exhibited low concentrations of volatile organic compounds (VOCs). Results from an emission test performed by an accredited laboratory are in Table 7 (20). The results are emissions of V O C reported as Toluene equivalents (TVOC) after 1 and 4 weeks of outgassing. Furthermore, burning furfurylated wood was found to release no more toxic gases than burning untreated wood (Table 8).

Biocide Directive The European Commission has implemented a biocide directive to control and register all biocides in use. New biocides have to fulfill the requirements of this directive. However, since it has been demonstrated that furfurylated wood does not work by biocidal action, but by being biologically inert, the Commission ruled that furfurylated wood was not required to be registered under the biocide directive.


350 Table 6. Toxic Units (TU) of leaching water from structural materials.


Material Pine Pine - Heat Treated Pine - Furfurylated Western Red Cedar Azobé C C A - type C Concrete Recycled plastic

Algaetox 16-32 8-16 3.1 >32 8-16 >32

Development of Commercial Wood Preservatives Efficacy

Recommend Documents