Adhesive Properties and General Characteristics
An adhesive or formulation is generally a mixture of several materials. The extent of mixture and the ratio usually depend upon the properties desired in the final bonded joint. The basic materials may be defined as those substances, which provide the necessary adhesive and binding properties.
Solvents are employed in many systems to provide vehicle and viscosity control. In some cases, low molecular weight resins of high fluidity are added to the basic resin to help control viscosity.
Fillers such as metallic oxides, mineral powders and various fibers are sometimes used to control reinforcement, decrease shrinkage, lower the coefficient of thermal expansion, control temperature operating ranges, and in some instances provide a more satisfactory system for a special environmental condition. Fillers are also used to control viscosity, especially if a thixotropic paste is desired. The most common of these are the ultrafine mesh silicas such as Cab O Sil. Most fillers also lower the cost of the system. They also prevent waste by virtue of improving the handling properties. They are often referred to as extenders.
Catalysts and hardeners are employed to activate the resin systems, especially where thermosetting resins are concerned, in order to speed up hardening and make the adhesive system practical. Acids, bases, salts, alcohols, sulfur compounds, and peroxides are a few of the basic catalyst materials. The selection must be based upon knowledge of the mechanics of polymerization reactions, which account for the curing or hardening of adhesives. Catalysts are very important in forming the final joint. The amount of catalysis is critical. Overcatalying may result in a poor joint, and the same holds true for under catalyzing.
There are several classes, types, and groups of adhesives. These have been classified as to use, chemical composition, mode of application, setting factors, vehicle, etc. The first general classifications to be considered are structural and nonstructural adhesives. These classifications are sometimes difficult to clarify. A structural adhesive would normally be defined as one, which can be employed where; joints or load carrying members associated with primary design are required. This type of adhesive will be subjected to large stress loads. The term structural bonded joints equate structural with the importance of its mission. In this concept, a further definition may be required where primary structural means loss of the aircraft or vehicle through joint failure and secondary structural means severe damage and impairment of the mission. The criteria in many cases have been defined on the basis of bond strength using the arbitrary top value strength of 1000 psi.
This is considered by many as a very poor definition, and since many people have disagreed on these terms, this discussion is included only to raise the question and allow the individual concerned to draw his own conclusion. It must be considered because a large portion of companies has specifications placed in these general categories. The major problem is that they all differ in context.
Nonstructural adhesives are not capable of supporting appreciable loads and are generally required to locate parts in an assembly. They will be employed many times where only a temporary bond is required. Their failure would not usually result in the loss of a vehicle. Adhesives, sealants, and coatings usually fall into this category, but could be responsible for the full accomplishment of the mission.
The type of adhesive material is easier to define and usually falls into three categories
1. Thermosetting resins are synthetic organic substances, which can be converted by chemical reaction into a permanently hard, practically infusible, and insoluble solid. These resins are high molecular weight polymers, which react by polymerization to form hard substances, usually rigid and possessing high strength properties. Thermosetting resins usually have a high modulus of elasticity, do not support combustion, and resist the action of most chemicals. When reacted, the thermosetting system will not be liquefied by heat but will deteriorate or decompose under heat ranges beyond its limitations. We might compare this to the baking of bread. Once it is catalyzed (baking powder), further baking will only burn it.
2. The thermoplastic resins are often employed in metal and plastic bonding and usually adhere well to both. They do not lend themselves to use as good load bearing adhesives, especially if they would be subjected to elevated temperatures. They will soften when heated and harden when cooled. An example here would be butter, placed in a molten liquid state by heating and becoming solid upon cooling.
The more common thermoplastic resins include the polyvinyls, acrylics, polystyrenes, celluloses, and polyamides. They are sometimes used effectively with thermosetting resins for specific formulations.
3. Elastomeric resins are used widely for modification of the thermosetting systems. They generally fall into a distinct class e.g., natural and synthetic rubber. A true elastomer is usually defined as a material that will stretch twice its original length without inherent loss of elastic properties. When used as a modifying agent for other resins, they usually induce flexibility and increase peel strength of the systems. They are often used alone or in slightly modified form for sealants, but lack the strength to be used alone for structural applications. Examples of this class are the butyls, nitriles, polysulfides, and neoprenes.
A thermosetting system, 100 percent reactive when in a pure state, the epoxies are very desirable and more widely used than any other chemical type. Epoxy is one of the newer types and has penetrated more fields of manufacturing operations in a shorter space of time than any of its predecessors. The epoxies have been formulated from more materials than any other class. They are very versatile and can be formulated to do any job, limited only by heat. Some formulations will withstand 800°F for short periods. The heat ranges of the epoxies are usually determined by the catalysts utilized to harden the system. The many catalysts used with epoxies produce systems of variable properties. The most common are the aromatic amines and cyclic anhydrides. The amines produce the low temperature cure cycles and limited heat range, while the anhydrides usually require higher cure cycles and withstand higher operational temperatures. Table 1 shows the general properties of a basic epoxy resin hardened by various catalysts with 301 stainless steel adherends.
Epoxies are available in liquid, paste, and film forms (supported and unsupported). The two component systems are more widely used because of the extended shelf life. They may be stored for long periods and, naturally will not activate until mixed. A few of the early epoxies were one part, in a stick form that was heated prior to application, which proved to be impractical from an application standpoint. Epoxy adhesives are not widely used in the film form unless they are modified, that is, alloyed with another adhesive system.
The epoxies are not affected by bond line thickness as compared to many other structural adhesives. This is important for application and processing, because the epoxies require very little pressure, and become very fluid when heated prior to the gel or B stage. The thin bond line is preferred, but is sometimes difficult to control with a paste or liquid. If bond line control is essential, it may be accomplished by utilizing glass beads of the desired size in the resin, which do not adversely affect the mechanical strength unless used excessively. The bond line control is one of the prime advantages of the film type adhesives, especially if a carrier is utilized. When utilizing the epoxy with a carrier, care must be exercised in pressure application, because bond line starvation will occur, due to the very fluid state of the resin under heat and pressure.
The epoxies have low peel and impact strength as compared to many other structural adhesives because of their brittle nature after cure. To improve the undesirable properties, they are alloyed with various other adhesive systems to produce a system to meet the demand of design requirements.
Adhesion. The expoxies have high specific adhesion to metals, glass, plastics, ceramics, paper, concrete, wood, and various other substrates. Because of their brittle nature, epoxies are not recommended for bonding the rubbers and elastomeric adherends, although they will adhere to these types of materials. The epoxies can be formulated to create mixtures of low viscosity and improved wetting, spreading, and penetrating action. If the substrate to be joined is cleaned and processed properly, adhesion presents very few problems.
Cohesion. When properly cured, the cohesive properties are considered very good, but are usually the limiting strength factor. The adhesive properties are superior to the cohesive properties in most formulations, thus cohesive failures will be experienced during testing from room temperature to the maximum operating limits of the system.
100 percent solids. The epoxies in the unmodified state cure without releasing water or other condensation byproducts. This makes them desirable where contact pressures are necessary for manufacturing. They are also convenient for bonding such materials as glass or thermoplastics, where high heat and pressures would be unsatisfactory. This characteristic also makes them desirable as potting compounds, since the possibility of air bubbles or inclusions is reduced. The addition of silver, carbon, or other conductors has proven very successful in varying the electrical properties of epoxies without the problems of discontinuities in the bond line and also without adversely affecting the mechanical properties of the system.
Low shrinkage. The epoxies cure with only a fraction of the shrinkage of vinyl type adhesives such as polyesters and acrylics consequently less strain is built into the glue line, and the bond is stronger. The shrinkage can be reduced to a fraction of 1 percent by incorporation of silica, aluminum oxide, or other organic fillers. A shrinkage factor of 3 percent would be considered extremely high for epoxies.
Low creep. They maintain their shape under prolonged stress better than thermoplastics and many thermosetting systems. This is an important asset in favor of the use of epoxies, because creep is considered a major problem in structural adhesive bonding, and an area of prime concern by designers. Creep, in all probability, has hampered the use of adhesives and plastics in the building industry more than any other single factor.
Resistance to moisture and solvents. The epoxies are resistant to moisture. Moisture does not effect an epoxy in the least but will migrate through the joint and deteriorate the substrate. When epoxy bonded joints are subjected to moisture or water immersion, the failures usually occur at the interface. This indicates the importance of proper surface preparation of the adherends. Their resistance to solvents is considered outstanding and accounts for their rapid advancement in the coating field. Because fluids do migrate through an epoxy with little or no effect to the system the substrate problem does exist, Which makes other systems more desirable for use in long term exposure to such fluids as fuels, although when modified with an elastomeric system, for example, they may possess very desirable properties in these areas.
Versatility applicable to modification. The properties of the epoxy may be changed by
Varying of the base resin and curing agents.
Varying cure cycles, both temperature and cure time.
Alloying the compound with another resin.
Compounding the various fillers. This may affect the cost factor, but the economics of epoxies are governed more by the type of catalyst utilized.
They are effective barriers to heat and electric current, yet at the same time may be modified easily for conduction of electricity. They are versatile in applying due to their wide range of modification, and may be applied manually, semiautomatically, or automatically.
The phenolics or phenol formaldehyde resins are formed by the condensation reaction of phenol and formaldehyde. This material was discovered in 1872. The phenolics are very rigid, strong, and have excellent resistance to fungi. They have moderate to good resistance to moisture, and very good high temperature properties. The phenolic resins have been used extensively in the lamination of plywood and in filament wound structures. They enjoy a wider range in the structural adhesive category when alloyed with other materials.
There are two basic classes of phenolic resins resoles and novalacs, and both begin as phenol alcohols. They are catalyzed with either an acid or an alkali. Regardless of the formulation of phenolic resins, they are considered to have high resistance to deteriorating influences encountered in service. They would not be considered excellent in resistance to stresses caused by thermal expansion, and extenders should not be used in attempts to correct this weakness. When combined or alloyed with other adhesive systems, they become excellent structural adhesives and are widely used in this manner throughout the aerospace industry.
The nitrile rubbers are elastomers and copolymers of unsaturted nitriles and dienes. The nitriles are not used as structural adhesives in this form, but yield many one part adhesives that are used for bonding small nonstructural parts, especially in the electronics and plastics The nitrile rubbers, when prepared for use as a cement, are milled on tight cold mill rolls, broken down, and rendered soluble in some type of solvent. The most widely used nitrile rubber adhesives are cured by the solvent escape drying method, but they may be catalyzed by the utilization of sulfur compounds and cured at room or elevated temperatures. The nitriles are available from the manufacturers in a variety of formulations, but the important role of this rubber system for structural adhesives comes as a result of being alloyed or mixed with another resin. The nitriles, like the phenolic resins, do not have the desired properties for structural bonding when used alone, but, for example, if the nitriles and phenolics are combined their mechanical properties change to a system with excellent properties for structural use. The nitriles give the rigid resins flexibility that produces high peel strengths and better than average shear strengths.
The vinyl polymers do not stand alone as a structural adhesive, but hundreds of adhesives are formulated by the use of this class of polymer. Vinyl is the univalent radical CH2CH, derived from ethylene, a compound which undergoes polymerization to form high molecular weight resins. More generally, the term vinyl polymer has been used to include a variety of resins, plastic films, and elastomers obtained by polymerizing monomers having one or more unsaturated double or triple bonds, including diolefins, such as butadiene, vinyldienes such as vinyldiene chloride or methyl methacaylate, and unsaturated compounds such as maleic anhydride.
The vinyls are important to adhesive bonding not only from the adhesive standpoint, but because the films derived from these substances are widely used as vacuum bags, slip sheets, etc. The more widely used ones are polyvinyl chloride, polyvinyl alcohol, and polyvinyl fluoride.
Neoprene was the first synthetic elastomer developed that possessed properties comparable to natural rubber. It is defined as an oil resistant synthetic rubber obtained by polymerizing chloroprene. The neoprenes were limited in use due to the cost factor until the shortage of natural rubber in World War II. At that time, neoprene was the only synthetic rubber available for use in adhesives and as a result, formulators began to experiment with it. They found that neoprene adhesives were just as good, if not better in many cases, than those based on natural rubber. The neoprenes are used in three general capacities in the adhesive industry. They are used structurally when alloyed with another resin, as a rubber cement and as a noncuring tacking paste. The neoprene cements are usually dispersed in solvents such as toluene, which is one of the more widely used. It may be dissolved in mixtures of aromatic and aliphatic hydrocarbons.
The neoprene cements may be cured at room temperature or by the use of heat, depending upon the accelerator used. Magnesium and zinc oxides are two of the more common accelerators, which effect a slow, room temperature cure.
The maximum operating temperature does not usually exceed 170°F and would show signs of degradation if used at that temperature for long periods of time. As neoprene ages, traces of hydrochloric acid are formed by decomposition of the chlorine containing molecules this acid tends to deteriorate most fabrics such as cotton, rayon, linen, etc.
A wide variety of polyurethanes can be formed by cross linking highly reactive isocyanates with various polyols. This elastomeric material provides a bond which resists not only the shear and tensile stresses satisfactorily but has very high impact resistance and excellent cryogenic properties. This has brought them into widespread use in space applications, especially for insulation problems. They are also widely used as sprayable coatings for aircraft wing assemblies, and for bonding solid propellants. They are utilized for bonding metal to metal, elastom eters, foam, plastics, nylon, glass, ceramic, and the fluorocarbons. Due to the flow characteristics they are not considered a good material for honeycomb construction.
The cohesive strength is usually better than the adhesive strength, but good cohesive failures are obtained by careful processing with a majority of formulations. They yield from 3000 to 5000 psi in shear at room temperature, but shear strength varies with cure conditions and pressure. There is a correlation between the bond line thickness and shear strength, the ideal bond line thickness being in the range of 2 to 6 mils. They will yield up to 8000 psi in shear at 423°F, but are limited to approximately 250°F at elevated temperatures.
The polyurethanes are not considered ideal. They pose processing problems due to their reaction with water and their gaseous nature. The systems that are MOCA catalyzed require hot mixing to diffuse the catalyst into the resin. The ratio of MOCA to resin has varying effects on the final joint and should be carefully controlled. They may be degassed before application, but the amount of degassing affects the pot life. When applied before gelling starts, they are very fluid and sometimes tend to cause starved bond lines. This has been controlled in special applications by the addition of 6 to 12 percent of nylon fibers.
Another problem associated with the polyurethane system applies to storage. Storage must be maintained that will inhibit fractional crystallization in fact, it should inhibit any crystallization and water accrual in the raw material.
In summary, the polyurethanes are excellent cryogenic materials, exhibit excellent shock properties, are more difficult to process than many systems, suffer from excessive creep at room temperature, and show changes in properties on aging, some of which are undesirable. They still hold the answer to cryogenic application, but have poor elevated temperature strength.
Silicones are semi inorganic polymers made up of a skeleton structure of alternate silicone and oxygen atoms with various organic groups attached, and are thermosetting type resins. A large variety of the RTV (room temperature vulcanizing) compounds are formulated utilizing the silicone resins. They do not possess the mechanical properties to be used as structural adhesives, but are widely used as sealants and potting compounds. The unit was designed to pass heat from the electronic equipment to the outside radiation system.
The silicones vary in curing temperatures from room temperature to 250°F, depending on the formulation and vulcanizing agent. The majority of these systems require only contact pressures during cure.
The silicones have many very desirable characteristics as listed
1. Good high temperature properties. They have good thermal and oxidative stability at temperatures up to 600°F and will withstand short exposures up to 800°F.
2. Silicones are good thermal insulators, which accounts for their utilization as thermal insulation and heat sinks.
3. They have good low temperature properties when compared to many other systems. The methyl silicones have brittlepoints at 100°F. but the methyl phenyl silicones may be used to 175°F.
4. They maintain good electrical properties over a wide temperature range.
5. They have adequate resistance to aging and weathering and remain, stable when exposed to ozone, corona, and sunlight.
6. They have fair resistance to water and moisture.
7. Generally, all silicones will withstand radiation however; the most effective group is the silicone resins, followed closely by the silicons rubbers. In all probability, the most outstanding characteristic is their ability to resist combined heat and radiation.
8. The ease of handling and low temperature cures brands the silicones for future growth.
The silicones are handicapped by low shear strength and many do not possess the adhesion or tack quality level desired. Adhesion may promoted by the use of primers. They deteriorate under constant contact with fuels, which limits their usage in fuel areas.
The reaction of organic acids and alcohols produces a class of materials called esters. When the acids are polybasic and the alcohols are polyhydric, they can react to form very complex esters. They are usually called alkyds and have long been useful as surface coatings and glass reinforced plastics. This same principle, utilized with various modifications, brings the polyesters into the adhesive field. The polyester adhesive systems cure rigid, and have a temperature operating range up to 500°F. They reveal poor adhesion to metals, especially aluminum.
The polyesters are attacked by most solvents and have a high shrinkage rate when compared to other adhesives. Shrinkage rates may run as high as 4 percent. Attempts are made to combat the high shrink rate by the utilization of fillers such as calcium carbonate and aluminum silicate.
Recently a new polyester has been developed that is flexible. The evaluation of this system is incomplete but indications are that it adheres better to metals than many of the earlier polyester systems.
The acrylics are a group of thermoplastic resins formed by polymerizing the esters or amides of acrylic acid. They are usually transparent, low viscosity, polymerizable liquids and were developed primarily for use as liquid locknuts. They are now used as adhesives, but are more important pertinent to structures as transparent sheets (plexiglass and Lucite).
The acrylic adhesives have indefinite shelf life when stored at ambient temperature with access to oxygen. When oxygen is excluded by applying the material in a thin film between two mating surfaces, gelation occurs at room temperature in a matter of minutes. To prevent gelation before application, the liquid is packaged in a low density polyethylene container permeable to oxygen. Curing may be accelerated by elevating the temperature using an oven, heat lamp, or press. The acrylics have been cured successfully in a vapor degreaser when small details are being joined. The heat causes gelation to occur before the solvent extracts the adhesive from the joint. Perchloroethylene, with a boiling point of 250°F results in a more rapid cure with less leaching than trichloroethylene with a 190°F boiling point. Also, the vapor degreaser removes the thin film of liquid that is kept from curing by contact with the air.
If the liquids are applied to sensitive electromechanical devices, be sure the uncured surface liquid is removed. Outgassing and condensation of volatiles in sealed systems may cause problems in service or storage.
Certain metals, such as zinc, cadmium, and gold, do not promote cure of these materials. For these metals an organometallic activator is supplied in solution in a chlorinated solvent, which is applied directly to the metal and allowed to dry. Zinc and cadmium plated surfaces can also be activated by a chromic rinse prior to sealing the surface.
Strength is far too low for the acrylics to be used as structural adhesives in lap joints, but resistance to torque shear is outstanding for joining cams, sleeves, pulleys, and gears to shafts in lieu of conventional fasteners.
Rosin (sometimes called colophony)
Spirit soluble thermoplastic materials are available in two forms gum rosin, the more popular form of which is obtained by distillation of the exudation fom pine trees, and wood rosin, which is prepared from pine trees.
Rosin adhesives are used for metal container labeling either as hot melts or in solvent solution, often with added plasticizers. It is also used in the modification of other resins. Rosin is used in the powder form for bonding wood components in aircraft, but is not usually termed a structural adhesive. It has very good strength, and good water and moisture resistance, but poor resistance to fuels and solvents.
Polysulfide rubber adhesives
The polysulfide adhesives are synthetic polymers obtained by the reaction of sodium polysulfide with organic dichlorides such as dichloro diethyl formal, alone or mixed with ethylene dichloride.
The polysulfides are used as adhesives where high strengths are not required, but are used more often as sealants. They are sometimes used as binders for solid propellants. This system offers good resistance to light, oxygen, oils, and solvents, and impermeability to various gases. They adhere well to almost any adherend, but have poor tensile properties. They exhibit poor properties when subjected to high humidity conditions and have an operating temperature range of -67 to +250°F.
The polysulfides are usually procured in two component paste or liquid systems, have good shelf life, and require no special storage facilities. They may be catalyzed, mixed, and frozen for several days to eliminate production handling problems. The polysulfides are considered a wise choice (if the service requirements do not exceed their capabilities) because of the economy involved.
A typical formulation of ceramic adhesives may contain silica, sodium nitrate, boric acid, and ferric oxide. These materials are heated above 2000°F, blended, and then crystallized. A hard frit is formed, dried, milled, and then passed through a screen of the desired size to ensure uniform grain size. Oxides and water are then added and the results are ceramic adhesives. The viscosity may be controlled by the amount of water added to the mixture.
In recent years, investigations have been carried out to adapt the ceramic adhesives to structural bonding. They possess high shear strengths up to 1500°F (1800 psi in shear) and may reach 5000 psi in shear at room temperature.
The prime disadvantages are low peel and flexural strengths coupled with very high temperature cure cycles. Much attention has been diverted from the ceramic systems since the newer polyaromatics became a reality. At present, the ceramics are not practical but research is currently in progress to improve the unfavorable properties they possess. The ceramics must be improved to become a sound structural adhesive but show great promise as an encapsulation material for high temperature rocket nozzles and nose cones.
A special purpose proprietary cyanoacrylatc adhesive (Eastman 910) is a one part, clear, watery liquid. It is free of solvents and cures at room temperature in contact with many surfaces without the addition of a catalyst or hardener. The system sets by an anionic polymerization mechanism, which is catalyzed by weak bases such as traces of moisture on most surfaces in contact with the atmosphere. Surfaces such as phenolic, polyester, polyethylene, and polystyrene plastics tend to inhibit the curing rate, and may be pretreated with a diluted solution of an activator, phenylethylethanolamine (910 surface activator). However, most common metals, glass, wood, and rubber surfaces bond very rapidly and cure in a matter of minutes.
Shear strengths up to 4000 psi can be obtained, but peel and impact strengths are poor. When exposed to elevatrd temperatures the mechanical properties are poor and aging as low as 160°F reflects degradation. When exposed to temperatures above this, the adhesives turn yellow and decompose. The prime advantage is the fast cure, and relatively little effort its required for well mated joints. It is widely used for bonding small electrical components and as a tacking adhesive however, it is expensive.
Problems have been encountered on production lines, as the adhesive presents an operator hazard because of its strong and rapid adhesion to the skin. Curing may be slow when the humidity is low, but this problem may easily be solved by placing an open container of water near the parts being bonded, but this material should never be placed in an oven for cure.
Adhesive Materials and Properties
THE COMPONENTS OF AN ADHESIVE
The components of the adhesive mixture are usually determined by the need to satisfy certain fabrication properties of the adhesive, or properties required in the final joint. The basic component is the binding substance which provides the adhesive and cohesive strength in the bond it is usually an organic resin but can be a rubber, an inorganic compound or a natural product. Other constituents of the adhesive fulfill other functions.
This is employed as a solvent vehicle for other adhesive components and also to provide the viscosity control which makes a uniformly thin adhesive coaling possible. Occasionally, liquid resins are added to control viscosity.
Catalysts and Hardeners
These are curing agents for adhesive systems. Hardeners effect curing by chemically combining with the binder material and are based on a variety of materials (monomeric, polymeric, or mixed compounds). The ratio of hardener to binder determines the physical properties of the adhesive and can usually be varied within a small range. Thus, polyamides combine with epoxy resins to produce a cured adhesive. Catalysts, which themselves remain unchanged, are also employed as curing agents for thermosetting resins to reduce cure time and increase the cross linking of the synthetic polymer. Acids, bases, salts, sulphur compounds and peroxides are commonly used and, unlike hardeners, only small quantities are required to effect curing. The amount of catalyst is critical and poor bond strengths result where resins are over or under catalyzed.
Accelerators, Inhibitors and Retarders
These substances control the curing rate. An accelerator is a substance that speeds up curing caused by a catalyst by combining with the binder (a catalyst may have the same effect but will not lose its chemical identity during the process). An inhibitor arrests the curing reaction entirely whereas a retarder slows it down and prolongs the storage and/or the working life of the adhesive.
There are many chemically inert ingredients which are added to adhesive compositions to alter their end use or fabrication properties. Modifiers include fillers, extenders, thinners, plasticizers, stabilizers, or wetting agents, and each material is used for a special purpose.
Fillers are non adhesive materials which improve the working properties, permanence, strength, or other qualities of the adhesive bond and those commonly used are wood flour, silica, alumina, titanium oxide, metal powders, china clay and earths, slate dust, asbestos and glass fibres. Some fillers may act as extenders.
Extenders are substances which usually have some adhesive properties and are added as diluents to reduce the concentration of other adhesive components and thereby the cost of the adhesive. Extenders often have positive value in modifying the physical properties of the glue line by providing reinforcement to resins which would otherwise craze. Common extenders are flours, soluble lignin and pulverized partly cured synthetic resins. Thinners are generally volatile liquids which are added to an adhesive to modify the consistency of other properties. Plasticisers are incorporated in a formulation to provide the adhesive bond with flexibility or distensibility, Plasticisers may reduce the melt viscosity of hot melt adhesives or lower the elastic modulus of a solidified adhesive. Stabilizers are added to an adhesive to increase its resistance to adverse service conditions such as, light, heat, radiation, etc. Wetting agents promote interfacial contact between adhesive and adherends by improving the wetting and spreading qualities of the adhesive.
One objective of this handbook is to indicate the basic properties of the different adhesives types. To a large extent the mechanical properties depend on the thermosetting or thermoplastic nature of the bond and the following general discussion of these differences provides background information to the detailed listing of the adhesives types (based on the major chemical ingredient) which follows.
The thermoplastic adhesives are classified under the general categories of thermoplastic resin and thermoplastic rubber adhesives. As a class, thermoplastic adhesives are fusible, soluble, soften when heated and are subject to creep under stress. Unlike the thermosetting resins, they do not change chemically in establishing a bond. The thermoplastic nature of these materials confines their application as adhesives to low load assemblies formed from metals, ceramics, glass, plastics and porous materials based on paper, wood, leather and fabrics and which are not subject to severe service conditions. Hot melt adhesives, which fall into this class, are being increasingly employed for fast assembly of packaging materials and plastic film laminates.
Thermoplastic resin adhesives are based on various synthetic materials (typified by the polyamide, vinyl and acrylic polymers and cellulose derivatives) or on natural products such as rosin, shellac, oleoresins and the mineral waxes. The important hot melt adhesives are invariably compounded from polyethylene, vinyl polymers and co polymers, polystyrene, polycarbonates, polyamides and other polymers. Additives, including plasticisers, fillers, and reinforcing materials, are frequently compounded with the resins to confer particular properties on the adhesive. With the exception of pastes, these adhesives are available in the same forms as the thermosetting adhesives, i.e. liquid forms can be solutions, dispersions, or emulsions of the polymer and other modifying components in a volatile medium. Solid forms are also available as films (supported and unsupported), pellets, sticks, or extruded cord lengths, suitable for machine application. Other solvent free liquid forms (100% solids systems) contain the thermoplastic material as a monomer or as a pre polymer which requires a catalyst to bring about polymerization to a high molecular weight solid.
Thermoplastic rubber adhesives are some of the most versatile industrial adhesives currently used. The rubber based adhesives discussed in the following pages include natural and reclaim rubbers and synthetic elastomers such as polychloroprene (neo prene), butyl, styrene butadiene and acrylonitrile butadiene (nitrile). Most of the elastomers are available in solvent and latex forms or as water dispersions and other types are supplied with vulcanising agents. The thermoplastic rubber adhesives are generally modified with fillers, plasticisers and compounding ingredients. The types of rubber and solvent vehicle used partly determines the physical and chemical properties of the adhesives and the many compounding techniques employed result in widespread variations in strengths, tack ranges, drying rates, environmental resistance and other properties.
Heat or solvent activation is used to convert film adhesives into the fluid state prior to bonding. Solvent activation is applicable only to situations where an adherend is porous enough to permit solvent release by absorption and diffusion and heat activation is employed where adherends are impermeable and able to withstand the temperatures involved. Heating also has the effect of curing any thermosetting component which may be present in the adhesive. Both techniques are also used prior to bonding to activate substrates which have previously been coated with a solvent base adhesive and dried to a tack free state. Bonding is usually carried out under heat and pressure after joint assembly. Solid thermoplastic adhesives of the hot melt type rely on heat to render them fluid and on cooling to bring about the setting action.
An unusual cure mechanism is displayed by the cyanoacrylates which are an example of chemically blocked materials. When confined between close fitting parts, these one component liquid monomers undergo polymerization in a very short period (often 15 s). The thin moisture film which is usually present on exposed surfaces is sufficient to harden these materials if the glue line is thin enough.
As a group, the thermosetting adhesives form bonds which are essentially infusible and insoluble through the action of heat, catalysts or combinations of these. In contrast to thermoplastics, the thermosetting resins display good creep resistance and provide the basis for many structural adhesives intended for high load applications and exposure to severe environmental conditions such as heat, cold, radiation, humidity and chemical atmospheres. Thermosetting adhesives include materials of natural origin such as animal glues, soybean and vegetable proteins, casein and miscellaneous water based adhesives as well as synthetic products based on epoxy, phenolic, polyester, polyaromatic and other thermosetting polymers.
Water based adhesives prepared from low strength materials of animal or vegetable origin were the earliest adhesives used and are still, important for furniture and plywood manufacture, paper and packaging materials, and similar applications where low strength and a limited durability to outdoor conditions are acceptable. In addition there are thermosetting rubber resin adhesives and other blends referred to as thermosetting thermoplastic resin adhesives. These adhesives have increased toughness and strength while their improved resilience enhances stress distribution properties. Examples of the latter class are the phenolic resins modified with nylon or various vinyl resins. The characteristics of these materials are, in general similar to those of thermosetting adhesives and hence many of these products are employed as structural bonding agents for metal to metal. Epoxy resins are modified with poly sulphides to improve their flexibility and are thermosetting materials. However, polysulphide adhesives often function as sealing materials and may be thermoplastic or thermosetting according to formulation and cure.
Thermosetting adhesives are supplied as liquids, pastes and solids. Liquid types are generally one or two component systems which are already non solvent, containing 100% solids materials, or react to become so by catalytic action. Some liquid adhesives contain a volatile solvent which is non reactive and which acts as a dispersant or improves the handling and processing properties of the system. The curing agent for a liquid system may be a powder which requires to be melted before mixing the components. As a result of added modifying agents, pastes are usually thixotropic and may be applied to vertical joints as non sag adhesives which will not flow out during assembly and cure of a bonded structure. Film forms may be supported or unsupported and of various thicknesses. They have the advantages of easy, clean handling and can be cut to conform to the shape of the joint. The shelf life of film and one component types is increased by refrigeration but in the case of some film adhesives cold storage is essential to prevent room temperature curing.
Natural product thermosets, like animal glue, set by loss of solvent. Many two component liquid types, such as epoxy resins, cure by catalytic action with or without the aid of heat. Other two parts thermosetting rubber resin adhesives can be vulcanized at room temperature but otherwise curing with heat and pressure is necessary. Some rubber resin adhesives which are used to bond unvulcanised rubber to metal cure during subsequent vulcanization of the rubber while other adhesives, employed to bond already vulcanized rubber to metal, are cured separately. Structural film adhesives invariably require heat and pressure to realise the maximum mechanical properties and curing temperatures ranging from 150 250°C with bonding pressures up to 100 N/cm2 are not uncommon. Post cures are often an additional processing requirement for structural adhesives where optimum strength is sought.
Rubber Resin Blends
There are innumerable adhesives in which rubbers and resins, both natural and synthetic, are blended to obtain combinations of desired properties of both types of material. Blended adhesives may be employed for structural or general purpose bonding according to the type of resin and rubber used and their ratio in a formulation. Those consisting mainly of thermosetting resins modified with synthetic rubber are used for the structural bonding of metal and other rigid materials. Phenolic nitrile and phenolic neoprene adhesives are examples of this type, in which the rubber component serves to improve the flexibility of the cured bond and promote its resistance to impact or shock loading. Thermosetting resins alone lend to be brittle. Adhesives based on rubber, with a certain amount of natural or synthetic resin as a modifying component, represent the other end of the scale. In practice, the various types of rubber are rarely used alone as adhesives but are invariably modified with resins to improve such properties, as tack, cohesive strength, specific adhesion to surfaces and heat resistance. Within these extremes are numerous formulations in which various ratios of resin to rubber are used. These adhesives have a wide range of applications which include bonding of textiles bonding of synthetic fabrics to wood and metal affixing wallboards and tiles lamination of paper, metal foil and plastic films laying of flooring materials and various other industrial or domestic applications.
The structural rubber resin adhesives are available as films or tapes (supported or unsupported on fabric carrier cloths) and occasionally as solvent solutions. The films are cured at elevated temperatures up to 200°C and under bonding pressures ranging from 30 100 N/cm2. Post curing is often included to ensure optimum mechanical properties for the cured adhesive. Liquid types are dried to remove solvent and then processed as film ad hesives. The non structural rubber resin adhesives are generally supplied as solutions in organic solvent mixtures and can be applied by brush, spray, dio or roller coater, spatula, or flow techniques. Because these adhesives rely on a loss of solvent before adhesive action can take place the shelf life and working life are usually indefinitely long provided the solvent content is maintained. With porous adherends the assembly can be made with wet adhesive and time allowed for solvents to escape by diffusion through the material. Where solvents have a high volatility assembly times may be as short as 15 min by which time the substrates have lost the tackiness required for contact bonding. Impermeable materials are coated with adhesive and bonded together only after the bulk of the solvent has been dried off to leave the adhesive in a tacky state. Light assemblies can frequently be handled after a few hours but heavier assemblies require a setting period of at least 24 h. Maximum joint strength is not realised until after a few days following the removal of residual solvent traces. Wet bonding generally produces joints having good strength and durability but poor solvent resistance. Optimum performance is given by heat curing (according to manufacturers instructions) which has the effect of removing the trace solvents otherwise retained by these adhesives and which act as plasticiscrs and increase the thermoplasticity of the system. Heat also promotes cross linking of the adhesive constituents and thereby increases the creep resistance of the joint. Processing conditions depend on the adhesive with bonding pressures ranging from 10 300 N/cm2 according to joint factors such as rigidity, dimensions, closeness of fit and glue line thickness. Low bonding pressures are more satisfactory for glue lines exceeding 0.2 mm. Curing schedules range from 1 h at 80°C to 20 30 min at 140°C, with optimum properties resulting from the longer curing periods.
PROPERTIES OF BASIC ADHESIVES TYPES
This section has been prepared almost entirely from published material appearing in technical books and journals. The length of an entry is not indicative of the importance of the adhesive type under consideration since the amount of information available was found to vary considerably. Due regard has been paid to technical information in the trade literature received from the various adhesives manufacturers. Manufacturers literature was found particularly useful for confirmation of such adhesive properties as colour, available form, processing factors, and applications. These data necessary to supplement material from published sources and effort has been made to keep the section free from any trade bias.
It has already been noted earlier that adhesives based on the same material may show considerable variation in their properties where modifying materials have been added to the formulation. Properties are dependent, not only on the adhesive composition, but also on the conditions under which it is prepared and used. Because of these possible variations any values given in this section should be regarded as representative of the probable behaviour of a basic type of material used under certain conditions. This is further complicated by the fact that a large number of commercial adhesives are blends of two or more basic adhesive types in particular both natural and synthetic rubbers and resins are often used together and form the basis of the numerous resin rubber or rubber resin adhesives which are available. Consultation with the manufacturers concerned is strongly recommended where detailed information is required on the behaviour of specific commercial adhesives under various service conditions.
Thermoplastic resins based on acrylates (properties of polymethyl methacrylate types are discussed below) or derivatives (amides and esters).
Available as emulsions, solvent solutions, and monomer polymer mixtures (one or two components) with catalysts (liquid or powder). One component liquids which polymerize under ultra violet radiation are available.
Emulsion solvent types set by evaporation and absorption of solvent. Polymer mixtures set through polymerization by heat, ultra violet radiation and/ or chemical catalysts.
Solvent types set over a period of 20 days at 20°C or 6 h at 80°C. Polymer mixture setting times depend on polymerizing method used chemical action, 14 d at 20°C or 4h at 80°C ultra violet action, 5 h exposure heal action, 2 h at 5.5° C followed by 8 h at 80°C. Bonding pressures range from contact to 17 N/cm2.
Resistance to weathering and moisture varies from poor (solvent types) to excellent (polymer mixtures). Change from transparent to yellow colour may occur with time (over 1 yr period). Not affected by alkalies, non oxidizing acids, salt spray, petroleum fuels but attacked by alcohols, strong solvents and hydrocarbons (aromatic and chlorinated). Highly resistant to ultra violet exposure. Service temperature range of acrylic resin adhesives is 60°C to 52°C.
Light structural assemblies based on acrylic plastics to themselves, wood, glass, metals, rubber, leather and fabrics colourless jointing of decorative plastic laminates production line assembly of components (ultra violet effective here) outdoor applications such as plastic name plates aluminium foil work windshields, instrument panels, lenses and optical components in aircraft, marine, and automotive industries. One type, n butyl methacrylate (alone or modified with Canada Balsam) is used as an optical cement. Ross 24 (modified type) is a heat setting cement which is transparent (= 1.485) and has good thermal shock resistance.
Physical Testing of Adhesives
A detailed description of the various test methods that have been developed for adhesive bonds is beyond the scope of this handbook. A short outline, in chart form, of commonly used lest specimens follow the remarks on the evaluation of adhesive strength. Non destructive test methods and the effects of adverse service conditions on bond strength are dealt with in subsequent paragraphs and the section concludes with a list, of titles of widely accepted standard test methods.
Specialised testing methods are required for the evaluation of the strength properties of adhesives. In addition to joint strength determination these methods provide a means for checking the efficacy of the processes used to make the bonds. The joint strength is invariably dependent on bonding technique factors such as adhesive application, adherend pretreatment and the adhesive curing conditions. Bonding conditions also determine the repro ducibility of test results and complete information on a number of variables is, therefore, necessary before undertaking an adhesive evaluation. The following particulars are essential for the fabrication of reliable test specimens.
Instructions for the preparation of the adhesive
Adherend surface pretreatment procedures recommended for the adhesive under consideration. (Special treatments may be involved where certain environmental tests are envisaged)
Adhesive application and processing before bonding. (Attention to coating thicknesses and their control, or drying conditions is often important).
Manufacturers specified conditions for joint assembly (temperatures, humidities or times)
Adhesive curing conditions relating to bonding temperatures, pressures and times.
Test specimens required to give reproducible failing strengths need to be carefully designed and prepared unsatisfactory bonds will result from the faulty execution of any stage in the assembly process. Most of the standard test methods employ test specimens of definite shape and size, which have to be machined to specified tolerances. Most methods specify the number of specimens to be tested in order to obtain a reliable result because testing factors and slight differences between adhesive batches prepared under identical conditions lead to joint strength variations. Ten or more specimens may be required to give meaningful data. The equipment used for testing is important and will influence the reliability of strength values obtained. Some variations in the performance between machines of the same type are to be expected. Often, test machine accuracy is greatest over a limited working range of the loading capacity (usually 10 90%), and test specimens should fail at loads within this span. The rate at which test specimens are stressed is another factor influencing the strength values obtained for adhesive bonds. Standard test methods generally specify the testing rate although it may be better to adopt a rate, which more closely resembles the stressing rate that an actual assembly is likely to experience.
The data obtained from test methods is useful for comparing the performance of several adhesives prior to the selection of one for a particular assembly job. It must be emphasised that the test specimen rarely simulates the actual configuration of an assembly and that the test data cannot therefore, be relied upon to predict the performance of the assembly in service. The same limitation applies to test specimens removed from an assembly these are unlikely to represent the behaviour of the whole structure. Short of testing an assembly under service conditions it is necessary to adopt a test specimen and method, which simulate the assembly and its working environment as closely as is practicable. The testing procedure finally employed must produce results that are likely to show a good correlation with the results that would he obtain in tests on assemblies. In this respect, selected standard test procedures are frequently employable without modification.
ASSESSMENT OF DURABILITY AND STRENGTH PARAMETERS
Fatigue testing refers to the repeated application of a specified load or deformation on a bonded specimen. Tests may be conducted under static or dynamic conditions (or both separately if necessary) according to the data required to evaluate an adhesive under service conditions.
Static fatigue properties are determined by measuring the maximum loading sustained by an adhesive over a given time. Various weight loadings applied to shear or tensile specimens provide a measure of the time required for bond failure.
Dynamic fatigue properties are measured by cycling test specimens with specified minimum to maximum stress loading for a given period or number of cycles or until failure. Cycle frequencies usually vary within the range 5000 to 107 Hertz. In addition to frequency, fatigue life is determined by amplitude, temperature and mode of stressing these variables must be specified along with the extent of loading. These tests do not determine the damping properties or elastic moduli of adhesives.
There is no standard test for measuring the distortion or dimensional change in a bonded specimen under sustained loading (Creep). Deformation of the adhesive is generally measured by noting the dimensional change occurring when a bonded specimen is subjected to a constant load for a specified time and temperature. Room temperature creep is known as cold flow. Higher temperatures usually increase the rate of creep significantly.
Creep tests are often carried out to determine joint deformation when stressed below the failing load required to break the bond. Joints may be loaded by springs (ASTM D2294 64T) or dead weights (MIL A 5090E) to maintain constant loading in a specified environment. Optical measurement of the shift in scribed reference lines on a lap joint edge is a useful method of creep assessment. Alternatively, the relaxation, or the ability of an adhesive to restore to its former state, may be optically determined on removal of stress. Rigid thermosetting adhesives display little or no creep under stress in contrast to thermoplastic or plasticised adhesives. Prolonged stressing of thermoplastic adhesives always reduces bond strength.
The shear strength of beams composed of adhesive laminated strips may be determined by flexural loading. The load is applied to the mid span to develop maximum shear stress and delamination in the centre layer of adhesive. The method gives higher shear strengths than are obtained with tensile or compressive shear specimens because the resistance of the adhesive to shear failure is increased by the compressive loading normal to the glue line.
Peel tests involve complex stress distributions. Peel strengths vary with the speed of testing (particularly with low modulus adhesives) and the forces needed to start and sustain peeling action are determined by the physical properties of the adherends, test specimen geometry (adherend thickness and width), and the adhesive strength characteristics. Peel strength increases with adherend thickness and adhesive thickness but decreases with adhesive modulus of elasticity. Steel adherends give higher peel strengths than aluminium adherends of similar thickness. Low peel strengths are usually a feature of brittle adhesives with high tensile strengths. Peeling rates of 15.2 cm/min for adherend widths of 2.54 cm are commonly specified.
The test specimens described previously may be used to determine the effects of adverse environments on an adhesive bond although no single test (or series) exists which will enable the user to predict its service life. A suitable test will provide information on the permanence of the bond when it is exposed to deteriorating circumstances such as temperature changes leading to oxidation, thermal degradation or softening of the adhesive. Other destructive hazards include low temperatures, sunlight and radiation, water, chemical reagents, oils and biodeterioration. Test specimens and procedures should be selected to simulate the type of service conditions envisaged for the bonded assembly. Consideration must be given to the adherend material for certain tests, e.g. for the evaluation of the acid resistance of an adhesive, certain metals would be unsuitable as adherends. Some of the environments which often provide the basis for unfavourable long term conditions of exposure for adhesive bonds are discussed here.
Adhesives and adherends are affected by high, low and varying temperatures. Elevated temperatures may decompose adhesive materials by oxidation or thermal degradation. Long exposure to moderate temperatures often leads to polymerization changes in adhesives. Displacement of bonded surfaces occurs where high or low temperatures accentuate differences between the thermal coefficients of expansion for adherends and adhesives stresses are set up at the interface which influence bond strength. Low temperatures embrittle many adhesives causing a reduction in their peel and cleavage strengths.
Test chambers with heating or cooling units can be employed for environment simulation. Adhesive durability is best determined at the service temperature higher test temperatures should be regarded for their comparative value only. Destructive temperature effects may become apparent in hours or years.
The long term ageing or weathering properties of bonded structures are difficult to predict since there are no standard short term permanence tests. Actual long time weathering is usually a reliable guide to adhesive durability although variations in exposure conditions over long test periods can make data difficult to interpret. Rainfall, humidity and temperature vary widely with locality. Accelerated weathering tests designed to reduce the long exposure periods are useful if the results can be correlated with actual weathering. Several tests have been adopted in the U.K. for providing a uniform testing procedure for military equipment. Referred to as I.S.A.T. (Intensified Standard Automating Trials), these tests are claimed to be equivalent to prolonged storage under existent weather conditions.
The permanence of a bond may be affected by exposure to external chemical agents or by the latent chemical reactivity of the adhesive for an adherend.
Several tests have been specified to evaluate adhesive bond strength on exposure to reagents such as acids, alkalies, water, sea water, petrol, organic solvents and lubricating oils. The deterioration in adhesion is sometimes dependent on reagent concentration. Temperature and exposure period should also be considered as test factors. Other tests are concerned with the effects of atmospheric constituents, which are known to cause adhesive deterioration, e.g., salt spray or ozone.
Chemical constituents in the adherends, such as plasticisers, can migrate into the adhesive and destroy adhesion. Additionally, the byproducts of an adhesive curing reaction may attack the adherend at the interface and cause loss of adhesion.
Certain adhesive types based on natural products such as casein, cellulose, dextrin or protein, etc., are subject to attack by bacteria, fungi, insects and rodents. Tests are available to check the effectiveness of preservative agents for adhesive formulations otherwise subject to biodeterioration.
The effects of light, either artificial or natural, on bonded glass or optical assemblies, involving transparent or translucent materials, may be important. Adverse effects include the loss of adhesive strength and the discoloration of the glue lines following photochemical changes in the adhesive. Light is unlikely to present a hazard for impervious adherend structures but is often an important factor with glass and transparent or translucent plastics.
Nuclear radiation is known to effect structural changes in high molecular weight polymers but has been scarcely studied for adhesive systems. The advent of nuclear technology and space research can be expected to produce test methods for this type of environment soon. An extensive literature on radiation induced changes in polymers provides a basis for studying adhesive performance. Some selected references appear at the end of the section.
Polyvinyl Acetate Wood Adhesives
Polyvinyl acetate is a thermoplastic polymer that has gained wide acceptance over the years as a raw material for the adhesives industry. Modified or unmodified, in solution or emulsion form, and as homopolymer or copolymer, it exhibits a versatility that makes it suitable for bonding a wide variety of substrates. In particular, it is capable of producing strong and durable bonds on wood and wood derived products, and this has been a major contributing factor to the tremendous growth of polyvinyl acetate based adhesives in recent years, from almost nothing in the early 1930s to an estimated worldwide production of all types of 2 million tonnes in 1977. It is familiar to people all over the world as the binder for interior and exterior emulsion paints, as the so called cold glue that replaced the heated pot of animal glue for carpentry, and as the white glue used in millions of households as a general purpose adhesive.
Vinyl acetate is a colorless flammable liquid with a viscosity of 0.4 cP at 20°C, a solubility in water of about 2% at 25°C, a boiling point of 72.7°C, and a characteristic odor. This is the starting point for the production of polyvinyl acetate (PVA), and, in conjunction with other vinyl monomers, acrylic esters, dialkyl maleates and fumarates, ethylene, and certain other monomers, of a range of specialty copolymers and terpolymers, while graft polymers can be produced with monomers such as styrene that will not copolymerize with vinyl acetate.
It is not certain when the first polymerization of vinyl acetate to polyvinyl acetate was performed. During the period between 1915 and 1925 the free radical initiation of polymerization of various vinyl monomers was widely studied and by 1930 polyvinyl acetate was commercially available. It was, however, only in the years following World War II that polymers of vinyl acetate began to be used in significant quantities, particularly in the paint and adhesive industries, and since then, PVA has shown the same rapid expansion as most other well known thermoplastic materials. Today some 50 major manufacturers and countless small suppliers around the world make material available in solid form, dissolved in solvent, or dispersed in water, as homopolymer, copolymer, or terpolymer, for a variety of applications that include paint manufacture, production of general purpose and specialized adhesives, textile coating, sizing and sealing of paper and related products, concrete additives, and production of sealants. Formulated products span the entire range of viscosities, from thin liquid to heavy paste, and may be fast or slow drying. Dried films may be clear or opaque, pigmented or unpigmented, flexible or brittle, hard or soft, as required. The product may be internally or externally plasticized or unplasticized, and may dry hard and tack free or pressure sensitive. With this versatility it is hardly surprising that polyvinyl acetate is now available almost anywhere in the world, with factories in most developed countries. At this point, the United States and Germany are the major producers, with companies such as Air Products, Borden, Monsanto, and Union Carbide in the United States, and Hoechst in Germany.
Chemistry of Polyvinyl Acetate
The steps involved in the manufacture of polyvinyl acetate are shown schematically in Figure 1.
A. Production of Vinyl Acetate Monomer
Originally, vinyl acetate monomer was produced by reacting acetylene and acetic acid together with suitable catalysts.
The earlier technique was to carry out this reaction in liquid phase but this method was replaced in the early 1930s by more efficient gaseous phase processes performed at high temperature.
This process remained viable for as long as calcium carbide was readily and cheaply available. With the increasing cost of energy, however calcium carbide has become steadily more expensive and less readily available, while an ever increasing range of chemicals has been made available by the rapid expansion of the petrochemical industry. These two factors have combined to make the modern method of production increasingly attractive. Here ethylene is used as the starting point for the production of both acetylene and acetic acid. Acetylene is made by removing hydrogen from ethylene, while acetic acid is obtained by oxidizing the ethylene to acetaldehyde, which is oxidized further to acetic acid.
B. Polymerization of Vinyl Acetate
Vinyl acetate monomer may be polymerized by many of the conventional polymerization techniques, including mass polymerization, solvent polymerization, and emulsion polymerization. The reaction is usually initiated and controlled by the use of free radical or ionic catalysts, although experimental methods of catalysis, including redox catalysis or activation by light, may be used for specialized products. The polymerization is characterized by three successive stages of reaction initiation, the growth of the polymer, and termination.
Initiation occurs when a free radical or ion attaches itself to a vinyl acetate molecule. This leads to a rearrangement of the electrons in the double bond, transferring the reactive site to the vinyl acetate monomer.
The initiator is usually a free radical derived from a peroxide such as benzoyl, lauroyl, or even hydrogen peroxide, although other initiators, such as persulfates, may also be used.
This highly reactive initiated molecule reacts with further monomer molecules by the same transfer mechanism, retaining the terminal reactive site for further growth.
Growth of the macromolecule is terminated when the reactive site is removed, either by combination with the reactive site of some other molecule, or by transfer of the reactive site to some other molecule.
Selection of the initiating catalyst, the ratio of catalyst to monomer, and the reaction conditions allows control over the average molecular weight of the polymer formed, and also of the degree of branching, if any, in the macromolecule.
In the case of polyvinyl acetate the polymerization may be carried out by a wide range of techniques, including mass polymerization, solution polymerization, and emulsion polymerization. Most of the poly vinyl acetate produced is made using emulsion polymerization techniques, and this is particularly true of those grades used in the production of wood adhesives. Vinyl acetate is particularly suited to emulsion polymerization, owing to the relatively high solubility of the monomer in water, and the complete solubility of the polymer in the monomer. In this process the vinyl acetate monomer is dispersed by means of relatively high speed stirring in water that contains suitable emulsifiers or protective colloids. A more or less stable suspension of monomer particles will be formed in the water. To this suspension the initiator is added, and typically the mixture will be heated to a temperature, which will substantially speed up the rate of reaction, while allowing it to remain controllable. As polyvinyl acetate is soluble in the monomer, the reaction will take place within the individual droplets of the suspension, producing a stable emulsion of polyvinyl acetate. Typical commercial polymer emulsions will contain between 40 and 60 parts by weight of the polymer, a very low residual level of monomer, and a viscosity between 0.1 and 20 Pa sec. In addition, emulsions intended for use as the basis of formulated wood adhesives will also have a relatively large particle size, usually within the range 0.3 5 × 10 6 m.
Other characteristics of the polymer emulsion that will goven its suitability for use in a specific application will include the molecular weight degree of copolymerization or plasticizing, if any film strength and film forming properties at low temperature and ability of the emulsion to withstand both the mechanical effects of mixing and changes in temperature, especially where freezing may be involved. Manufacturers of PVA emulsions will supply most or all of this information for their products to assist in the selection of the most suitable grade.
Formulating a Pva Based Adhesive
A. General Considerations
Formulating a PVA based wood adhesive, a number of factors, sometimes conflicting, must be borne in mind. It follows that the final product will often be a compromise in which the conflicting factors have been carefully considered in order to give most weight to those, which seem to be the most important in the specific application. Factors to consider will include the following
Where the adhesive is to be used for bonding wood to wood, consideration must be given to whether the wood will be hardwood or softwood. For the hardwoods an adhesive of high solid content is usually advantageous, while with softwoods, adhesives of lower solid content may be used. In addition, if the species to be glued is known to be oily, incorporation of a wetting agent or solvent will assist in adhesive penetration.
In many cases, however, wood will be only one of the substrates. The other substrate can vary from concrete in the case of adhesives for parquet or mosaic wood blocks to decorative laminates which may be cellulosic or plastic. Although, in general, PVA adhesives are used principally on cellulosic materials because of their exceptionally good adhesion to such surfaces, special applications may call for wood to be bonded to rubber, to foams, both flexible and rigid, to synthetic or natural fibers, or even to metal or other nonporous surfaces. Each of these will impose restrictions, sometimes severe, on the freedom of the formulator, and corresponding limitations on the applications for which the formulated adhesive may be suitable.
2. Surface Preparation
The preparation of the surfaces to be glued will also influence the formulation to a certain extent. In the case of adhesives for gluing wood to wood, inaccurately machined surfaces may necessitate the formulation of gap filling adhesives in order to produce a satisfactory bond over the entire surface. Adhesives may need to be formulated to give good adhesion to greasy, loose, or dusty surfaces or to seal a very porous substrate. In certain applications the adhesive must be capable of bonding surfaces that have been coated or lacquered,
The viscosity of the formulated adhesive will largely be governed by the method of application of the adhesive. Application methods will include manual application by brush, roller, smooth or notched trowel or spray, machine application from smooth or embossed rollers, with or without a doctor blade, or by cascade coaters, by nozzle or jet, or even mechanical extruders or sprays. Consideration should always be given to making the adhesive as easy to use as possible, especially where it will be handled by unskilled people unfamiliar with the proper handling of adhesives. This applies particularly to adhesives intended for household use.
4. Assembly Conditions
Intricate or multicomponent assemblies will demand an adhesive with a long open assembly period in order to enable all the components to be brought together and placed under pressure before the adhesive has started to dry. At the other extreme, applications such as core composing, the pressure station is usually very short, require an adhesive that can develop high bond strength very quickly. High production rates will similarly demand a quick setting adhesive. Application conditions involving high or low temperatures or humidities will also influence the formulation. The formulator must also consider whether or not the glued article requires machining. If this is the case, fillers must be chosen carefully or eliminated in order to minimize damage to cutters. Adhesives containing solvent should only be used in well ventilated locations.
5. Service Conditions
Although PVA adhesives are not, in general, used for joints that are under continuous load or subjected to high temperatures or high humidity, these adhesives can be formulated to give better performance under such conditions. The conditions under which the completed assembly will be expected to operate should always be taken into account when designing the adhesive.
Again, the use to which the completed article is to be put may influence the formulation. The appearance may be marred by unsightly glue lines, in which case it may be necessary to design the adhesive to dry completely transparent, or even to tint the adhesive so that the glue line will be less obtrusive. Tinting or pigmenting the adhesive may be essential the finished article is to be stained, as squeezed out adhesive may seal the surface in the vicinity of the glue line and prevent subsequent penetration of stain in that area.
7. Storage Conditions
If the adhesive is likely to be stored under adverse conditions, attention must be given to ensuring that it has adequate freeze thaw stability. The maximum storage life of the adhesive.
Very often the most serious limitation will be that of formulating to a particular price. In this regard, ease of application, reliability, and spread rate are factors to take into account, as it may be the case that a relatively expensive adhesive will prove more economical in the specific application than an inferior but cheaper product. When comparing prices it is important to take the density of the products into account, especially if they are sold in units of mass.
B. Formulating and Compounding
A number of different components will normally be incorporated into a PVA wood adhesive. Each of these has a specific function in the finished product. To formulate successfully, it is necessary to understand not only the performance criteria, but also the function of these components.
1. The Base Polymer
Since the PVA emulsion provides the major proportion of the adhesive strength, and will in many cases be the only binder in the formulation, it is worth considering its function in some detail. While the mechanism of adhesion is not fully understood, adhesion probably occurs as a result of secondary forces, principally Debye forces and London dispersion forces operating at close range, with some hydrogen bonding with the cellulosic fibers especially where polyvinyl alcohol has been used as the protective colloid. In addition, mechanical bonding occurs as a result of adhesive penetration into the open cell structure of the wood. All these combine to produce an excellent bond, which in a properly formulated adhesive will be stronger than the wood itself.
Since the PVA is in emulsion form, coalescing of the particles must occur in order to produce a continuous film during the drying process; this will happen only if the drying takes place at a temperature above the solidification temperature of the polymer particles. If this condition is not met, the particles will not coalesce properly, leading to a loss of mechanical properties in the dried film. If the temperature at which evaporation of the water takes place is significantly below this minimum film forming temperature or white point, no coalescing will take place, and a white, chalky film will result with no mechanical strength whatsoever. This white point is thus an important aspect of the base PVA to consider when formulating a wood adhesive.
Viscosity of the emulsion will be determined by the solid content of the emulsion, the particle size distribution, and the emulsifier or protective colloid system used, and may vary between very wide limits. For production of wood adhesives it is most common practice to use grades with a coarse to medium particle size (in the range 0.3 10 × 10 6 m) and a solid content between 40 and 60% by weight of the total emulsion. Even with these restrictions viscosity may still be anywhere between 0.05 and 50 Pa sec, however, and it will usually be necessary to modify this viscosity in the finished product.
The resistance of the dry film from a PVA emulsion to water is mainly dependant on the type and quantity of protective colloid used. Where this is polyvinyl aocohol, the water resistance of the dried film is generally poor. It is possible, however, either by using emulsions protected by cellulose based colloids, or by incorporating certain additives, to produce PVA adhesives that have a fair resistance to water.
Because polyvinyl acetate is a thermoplastic polymer, it loses cohesive strength as the temperature increases. In general, higher molecular weight polymers lose less strength at elevated temperatures than those of lower molecular weight, but the differences are not great. In addition, thermoplastic polymers are subject to cold creep, which is the tendency for a fully dried film to flow slowly under a sustained load. Plasticized grades are much more subject to cold creep than are unplasticized grades. To a certain limited extent, the tendency to cold creep can be reduced, but in general PVA wood adhesives are not suitable for applications in which the glue line is highly stressed or subject to temperatures above about 50°C and a combination of these two factors will completely rule out the use of PVA wood adhesives.
Different grades of PVA emulsions will also have different drying times and will therefore offer the possibility of formulating adhesives with long or short open assembly times. All the larger manufacturers issue comprehensive data sheets for their various grades from which it is possible to select, with a good deal of precision, the best grade to use as a starting point.
2. Other Binders
In addition to the PVA emulsion it is common practice to incorporate other binders into the formulation of various purposes. Probably the most widely used cobinder is polyvinyl alcohol. Incorporation of polyvinyl alcohol into a formulation increases the cold creep resistance of the dried film, but reduces its water resistance, especially where the polyvinyl alcohol has a low degree of hydrolysis. In addition, polyvinyl alcohol will increase the open assembly time of the adhesive substantially. Use of high molecular weight polyvinyl alcohols is common in low cost formulations, as these produce relatively high viscosity solutions allowing incorporation of extra water into the formulation. Adhesives containing polyvinyl alcohol will exhibit good machine stability and running properties and faster initial bond strength development.
The use of starch as an additive is also common practice. Here the major advantage is the cost reduction, which is again achieved at the expense of the water resistance.
Because of their affinity for water, starches will also extend significantly the open assembly time of the adhesive. As starches are particularly susceptible to microbial attack, care must be taken to ensure that the formulated adhesive is adequately protected. A wide variety of starch types may be incorporated; including pre gelatinized, water soluble, and oxidized starches. Where borax is used to solubilize the starch, the compatibility of the starch solution with the base emulsion must be checked. Borax has the effect of insolubilizing poly vinyl alcohol, and may therefore destroy the protective colloid of the PVA emulsion, thereby destabilizing the emulsion.
While cellulose derivatives such as carboxymethyl cellulose are often added, their function is usually that of viscosity modifier rather than additional binder, and the proportion is invariably small. Thermosetting resins such as phenol, resorcinol, or urea formaldehyde resins are occasionally added to PVA wood adhesives to improve their water resistance. Their use will be more fully discussed subsequently. For specialized application, a range of other binders may be incorporated, especially where wood is to be bonded to some other substrate. Thus all or part of the PVA homopolymer may be replaced by an ethylene vinyl acetate copolymer for the lamination of polyvinyl chloride film to wood, while vinyl acrylate polymers or copolymers may be added to improve adhesion to nonporous substrates. While dextrins may be added to PVA emulsions used in packaging applications, they are not normally added to PVA wood adhesives.
Plasticizers may be regarded as high boiling solvents with very low vapor pressures at the operating temperature of the adhesive. They will thus remain permanently in the dried film. Plasticizers form a film around the particles of the dispersion, increasing the distance between them and thus lowering the forces between them. In addition to increasing the flexibility of the dried film, they also lower the minimum film forming temperature of the adhesive. However, they also increase the tendency of the film to creep under load and should thus be used with caution in wood to wood adhesives. They should be avoided completely in adhesives for critical applications, especially highly stressed structures, where the use of a solvent to promote film forming characteristics is preferred.
One of the substrates is flexible, they may be used to improve adhesion and match the characteristics of the adhesive film more closely to those of the substrate, especially where this is flexible. Only plasticizers that are compatible with PVA should be used. Those in common use include esters, particularly alkyl phthalates such as dibutyl phthalate, and aromatic phosphates such as tricresyl phosphate, which are chiefly used where flame retardancy is a consideration. Speciality Plasticizers are not normally used in wood adhesives. Some physical properties of plasticizers in common use are shown in Table 1. The commonly used plasticizers are more or less immiscible with water; the addition of plasticizer to the base emulsion does not present undue difficulty. It is good practice to add the plasticizer slowly while stirring the emulsion vigorously. Once added, the stirring rate may be decreased, but stirring should be continued for at least 30 min to ensure thorough dispersion and to allow the plasticeser to solvate the emulsion particles. Plasticiser will seldom be added at a level above 10% based on polymer solids in formulations for wood adhesives.
Aminoresin Wood Adhesives
Aminoresins are polymeric products of aldehyde reaction with compounds carrying - NH2 or -NH groups. Such groups are mainly amide groups, such those in urea and melamine. They constitute the most important members of this class of compounds, more so than the amine groups as in the case of aniline. Formaldehyde is the main aldehyde used. Other aldehydes, such furfural, are generally not used for wood adhesives. The advantage of aminoresin adhesives (or amino plastic adhesives as they are often called) are their (1) initial water solubility (this renders them eminently suitable for bulk and relatively inexpensive production), (2) hardness, (3) nonflammability, (4) good thermal properties, (5) absence of color in cured polymers, and (6) easy adaptability to a variety of curing conditions.
Although many amidic and aminic compounds have been investigated for use in production of aminoresins, only urea and melamine and, in rare cases aniline, are extensively used. Thermosetting aminoplastic resins produced from urea and melamine are built up by condensation polymerization. Urea and melamine are reacted with formaldehyde, which results in the formation of additional products, such methylol compounds. Further reaction and the concurrent elimination of water, leads to the formation of low molecular weight condensates which are still soluble. Higher molecular weight products, which are insoluble and infusible, are obtained by further condensing the low molecular weight condensates.
Urea and melamine formaldehyde (UF and MF) resins have a great deal in common as regards the chemical and physical characteristics of both the cured and uncured resins. MF is superior to UF because of its superior water and heat resistance, hardness, and shorter curing time under less drastic conditions. The greatest disadvantage of these aminoplastic resins is their bond deterioration, caused by water and moisture. This is due to the hydrolysis of the aminoplastic or amino methylenic bond, which is the same for both UF and MF resins.
The higher resistance of MF resins to water attack is due to the considerably lower solubility of melamine in water. (Melamine dissolves in hot water only, whereas urea dissolves in cold water as well.) Therefore, UF adhesives are used for interior application only MF or melamineurea formaldehyde (MUF) resins can be employed successfully even for rather severe outdoor conditions. If full exterior grade quality is needed, it is safer to use phenolic type resins rather than aminoplastic resins.
Chemistry of Aminoresins
A. Urea Formaldehyde Condensation
The reaction between urea and formaldehyde is very complex. The combination of these two chemical compounds results in both linear and branched polymers, as well as tridimensional networks, in the cured resin. This is due to a functionality of 4 in urea (due to the presence of four replaceable hydrogen atoms), and a functionality of 2 in formaldehyde. The most important factors determining the properties of the reaction products are (1) the relative molar proportion of urea and formaldehyde, (2) the reaction temperature, and (3) the various pH values at which the condensation takes place. These factors influence the rate of increase of the molecular weight of the resin. Therefore, the characteristics of the reaction products differ considerably when lower and higher condensation stages are compared especially solubility, viscosity, water retention, and rate of curing of the adhesive. These all depend to a large extent on molecular weights.
The reaction between urea and formaldehyde is divided into two stages. The first is the alkaline condensation to form mono, di, and trimethylolureas. (Tetramethylolurea has never been isolated.) The second stage is the acid condensation of the methylolureas, first to soluble and then to insoluble cross linked resins. On the alkaline side, the reaction of urea and formaldehyde at room temperature leads to the formation of methylolureas. When condensed, they form methylene ether links between the urea molecules. The products from urea and formaldehyde, and from mono and dimethylolureas, are as follows
The reaction also produces cyclic derivatives uron, monomethyloluron, and dimethyloluron.
In weak alkaline solutions, the first product of the reaction is a complex (II), which is capable of rearranging itself exothermically into monomethylolurea. On acidification, the complex (II) eliminates water, resulting in unstable trimethylene urea hydrate (IV). The hydrated azomethin (III), which was identified by Fahrenhorst in urea formaldehyde resins, is regarded as characteristic of the intermediate stage of the reaction.
The electron theory provides a possible bonding mechanism between azomethin groups and either the solvent or other resin molecules. In this bonding, theelectrons of the C=N bonds and the free electron pair on the nitrogen atoms are involved. The higher pH stabilizes the degree of polymerization or association by permitting the formation of ionic complexes with water or the solvent. The lower pH causes the loosening of the water from the hydrated azomethin groups, allowing association. The resin eventually proceeds to the liophobic stage.
Indirect evidence strongly points to the existence of this mechanism as well as of the mechanism proposed by the classic theory of UF resin formation. However, no conclusive evidence of the participation of structure IV in the UF resinification process has yet been obtained. The association through azomethine type intermediates has been mentioned to explain resin formation and to interpret the mechanism of etherification of methylol groups under acid, neutral, and alkaline conditions. This theory opposes the classic theory.
In the first reaction, monomeric methyleneurea is formed as a result of the intramolecular loss of water. An unsaturated azomethine group is formed, followed by rapid polymerization. This gives the insoluble end product. The other reactions are condensation polymerizations in which the methylolureas are merely the building blocks of the polymers and of the insoluble end product. The polymers formed in both cases are mainly linear polymers obtained by the intermolecular splitting off of water. Under certain conditions water may also be split off intra molecularly. Cyclic compounds called urones are then formed. In both cases, further splitting off of water and formaldehyde leads to the formation of hardened or cured resins.
B. Melamine Formaldehyde Condensation
The condensation reaction of melamine (V) with formaldehyde is similar to the reaction of formaldehyde with urea. Formaldehyde first attacks the amino groups of melamine, forming methylol compounds.
Formaldehyde addition to melamine occurs more easily and completely than to urea. The amino group in melamine accepts easily up to two molecules of formaldehyde. Thus up to six molecules of formaldehyde are attached to a molecule of melamine. The methylolation step leads to a series of methylol compounds with two to six methylol groups.
Because melamine is less soluble in water than urea, the hydrophilic stage proceeds more rapidly in MF resin formation than in UF condensations. Therefore, hydrophobic intermediates of the MF condensation appear early in the reaction. Another important difference between MF and UF is that the MF condensation and curing occurs not only under acid conditions, but also under neutral or even slightly alkaline conditions.
The mechanism of the further reaction of methylolmelamines to form hydrophobic intermediates is the same as for UF resins, with splitting off of water and formaldehyde. Methylene and ether bridges are formed and the molecular size of the resin rapidly increases. These intermediate condensation products constitute the large bulk of the commercial MF resins. The final curing process transforms the intermediates to the desired MF insoluble and infusible resins through the reaction of amino and methylol groups which are still available for reaction.
A simplified schematic formula of cured MF resin has been given by Koehler and Fry . They emphasize the presence of many ether bridges besides unreacted methylol groups, and the methylene bridges. This is because in curing MF resins at temperatures of up to 100°C, no substantial amounts of formaldehyde are liberated. Only small quantities are liberated during curing up to 150°C. However, UF resins curing under the same conditions liberate a great deal of formaldehyde.
Wohnsieldler, Updegraff, and Hunt have tried to correlate the best physical properties of melamine formaldehyde resins with the degree of curing or condensation. They have found that the various properties attain their peak at different degrees of reaction. However, the best physical properties of MF resins were always associated with significant cross linking.
C. Aniline Formaldehyde Condensation
When aniline and formaldehyde are reacted in equal amounts, under neutral or alkaline conditions, N methylolanilines are formed. These form rapidly N methyleneanilines by eliminating water. If heated further, they form soluble and fusible aniline formaldehyde resins (VI).
In the presence of acids, especially hydrochloric acid, the amino group is protected by the formation of aniline hydrochloride. In this case the formaldehyde attacks the free paraposition on the aromatic ring. With aniline formaldehyde molar ratio of 11 the hydrochloride of the p aminobenzylalcohol is formed. The free base is produced by neutralizing the solution of p aminobenzylalcohol hydrochloride. Water is split off and long linear chains of brittle aniline formaldehyde resins (VII) of low mechanical strength are formed by condensation polymerization.
By using an excess of formaldehyde over the equimolar amounts, the linear chains of this product are cross linked, with the formation of methylene bridges. The resins produced have a three dimensional network of good mechanical strength. In these resins both aminomethylene linkages (CH2 NH ) and methylene linkages ( CH2) between the aromatic nuclei are present. This indicates that the formaldehyde is able to attack both the para and ortho positions of the aromatic rings (exactly as in phenolic resins), as well as the aromatic amino group (as in aminoplastic resins).
Amino groups are able to induce nucleophilicity in aromatic nuclei, higher than the hydroxy groups of phenols. Therefore, aniline and some of its derivatives (m hydroxyaniline and phenylendiamine) are sometimes used as terminal grafted groups of linear phenolic resins to accelerate curing. In these resins, the aromatic rings and the amino groups form bonds generally in a 75 8025 20 molar proportion, respectively. This gives rise to both aminomethylene ( CH2 NH ) and methylene ( CH2 ) linkages.
D. Reaction Kinetics Urea Formaldehyde
The kinetics of the formation and condensation of mono and dimethylolureas and of simple urea formaldehyde condensation products has been studied extensively. The formation of monomethylolurea in weak acid or alkaline aqueous solutions is characterized by an initial fast phase followed by a slow bimolecular reaction. The reaction is reversible. The formation of methylolurea is bimolecular and its dissociation monomolecular. The rate of reaction varies according to the pH with a minimum rate of reaction in the pH range 5 8 for a molar ratio of 11 for urea/formaldehyde and a pH of ± 6.5 for a 1 2 molar ratio.
The 12 urea/formaldehyde reaction has been proved to be three times slower than the 11 molar ratio reaction. The dehydration of formaldehyde (present largely as methylene glycol) and the formation of the urea anion are considered to be the controlling factors.
The rapid initial addition reaction of urea and formaldehyde is followed by a slower condensation, which results in the formation of polymers. The rate of the condensation of urea with monomethylolurea to form methylenebisurea (or UF dimers) is also pH dependent. It decreases esponentially from a pH of 2 3 to neutral pH value. No condensation occurs at alkaline pH values. The marked influence of the pH range on the reactions rates indicates that such reactions are of the hydrogen ion catalyzed type.
The initial addition of formaldehyde to urea in dilute solutions (0.1 M) is reversible, and is subject to general acid and base catalysis. The forward bimolecular reaction has an activation energy of 13 kcal/mol. The reverse unimolecular reaction has an activation energy of 19 kcal/ mol. The proposed mechanism of the acid catalysis is that of a protonated formaldehyde carbocation with urea. The alkaline catalyzed reaction proceeds instead through the reaction of the urea anion with formaldehyde. The subsequent reaction of monomethylolurea with formaldehyde in dilute solution, to give dimethylolurea, corresponds closely to the 11 monomethylolurea formation reaction in type, reaction mechanism, and activation energies
It is also reversible in concentrated solutions (2 4 M) at pH 7.0, and at 35°C, the addition reactions have the same rate constants as in dilute solutions and the reactions are very similar. No trimethylolurea is detectable in the reactions of urea and formaldehyde in dilute solutions containing a 6 8 M excess of formaldehyde . The rates of introduction into the urea molecule of one, two, and three methylol groups have been estimated to have the ratio 931, respectively. The formations of NN dimethylolurea and of trimethylolurea are also bi molecular, and their decomposition monomolecular. The formation of N,N dimethylolurea from monomethylolurea is about 1. 5 times that of monomethylolurea from urea. The decomposition of N, N dimethylolurea to monomethylolurea is three times that of monomethylolurea to urea.
No reaction was found between two molecules of dimethylolurea under the conditions chosen. It appears that (1) the amide group in urea is more reactive than that of monomethylolurea and (2) the methylol group in the latter is more reactive than in dimethylolurea. The results cannot be interpreted in terms of either dimethylene ether formation between urea molecules, or dehydration of methylolureas to methyleneureas followed by polymerization. However, an analysis done by the same author s of the insoluble fractions of urea formalde hyde condensates indicates that the properties of the condensates are the same as those produced by stepwise condensation to form methylene bridges between urea residues.
Methylenebisurea undergoes further condensation with formaldehyde and monomethylolurea , behaving like urea. The capability of methylenebisurea to hydrolize to urea and methylolurea in weak acid solutions (pH 3 5) indicates the reversibility of the methylene link and its lability in weak acid moisture. It explains the slow release of formaldehyde over a long time in particleboard and other wood products manufactured with UF resins.
Equilibrium constants for urea formaldehyde methylolureas have been theoretically derived, and found to agree with experimental results. They can be used to derive formulas, thereby quantifying these components in solution of different initial formaldehyde concentrations. Study of the addition and condensation reactions of urea and formaldehyde have led some authors to conclude that the addition process is a continuous one, with the hydrogen ion acting as a negative catalyst by addition to the nitrogen atom of the urea molecule. The condensation, on the other hand, reaches a state of equilibrium. In both reactions, the hydroxyl ion can be regarded as an indirect catalyst.
A few studies have been done on the pH changes in urea formaldehyde solutions during reactions. The main cause of the observed pH drop was the Cannizzaro reaction, disproportioning formaldehyde to methanol and formic acid. Oxidation by atmospheric oxygen also contributed. The latter could be eliminated by working in an inert atmosphere. This, or the use as a surface layer of a solvent which will not mix with water, inhibits the formation of methylene ureas. Using ammonia as neutralizing agent, an initial rise in the pH of the reaction mixture has been observed. This has been attributed to the decomposition of unstable methylolamines. The effect of methanol on the reactions has also been investigated. It slightly retards the addition reactions. Methanol hinders the formation of methylene links in the condensation reactions.
E. Reaction Kinetics Melamine Formaldehyde
Melamine and formaldehyde react in similar way to urea and formaldehyde, although basic differences are evident in the reaction rates and mechanism. The primary products of reaction are methylolmelamines, and evidence indicates that such compounds are formed only at ambient or higher temperature, except in acid pH ranges. The reaction is reversible throughout the pH range. Its forward rate is proportional to either [melamine] [HCHO], or [melamine] [H+CHOH] or [melamine] [HCHO], according to the pH of the media.
Phenolic Resin Wood Adhesive
Two main classes of adhesives aminoplastic and polyphenolic adhesives. The polyphenolic adhesives are the more significant for the production of weather and boil proof wood products.
As early as the nineteenth century, it was known that resinous materials could readily be formed by coreacting phenols and aldehydes. From the beginning of this century, much effort has been directed toward their commercial exploitation. Baekeland showed in 1907 that the simplest starting material, phenol and formaldehyde, form products of commercial importance under the correct reaction conditions.
Phenolic resins thus became the first true synthetic polymers ever to be developed. It may have been expected that by now, their chemical and physical structure would have been completely elucidated, but even now their structure is far from being completely clear. The polymers derived from the reaction of phenols and formaldehyde differ in one important aspect from other polycondensation products polyfunctional phenols can form a variety of isomerides with different chain lengths. Other products derived from polycondensation reactions, such as polyamides and polyesters, are mixtures of molecular chains of various lengths. However, in this case only one structure is possible for a molecule of a certain chain length.
For this reason, kinetic and related studies are feasible if one assumes that the growth of a chain proceeds from one molecule to the next in a smooth and regular manner. Polyfunctional phenols may react with formaldehyde in both the ortho and para positions to the hydroxyl group. This means that the condensation products exist as numerous positional isomerides for any chain length. This particularly not only makes kinetic studies extremely difficult, but also makes the organic chemistry of the reaction very complex and tedious to unravel.
The result has been that although phenolic resins developed commercially as early as 1908, and the first completely synthetic resin ever to be developed, their chemistry is still only partially understood. It may be argued with some justification that such a state of affairs is immaterial, because satisfactory resins for many uses have been developed on purely empirical grounds during the last 70 years. However, it cannot be denied that gradual understanding of the chemical structure and mechanism of reaction of these resins has helped considerably in introducing commercial phenolic resins which have been designed for certain applications, and which are capable of performances undreamed of in formulations developed by the empirical rather than the scientific approach. Knowledge of phenolic resin chemistry, structure, characteristic reactions, and kinetic behavior remains an invaluable asset to the adhesive formulator in designing resins with specific physical properties.
Chemistry of Phenol Formaldehyde Condensations
A. Reaction Mechanisms
Phenols condense initially with formaldehyde in the presence of either acid or alkali to form a mehtylolphenol or phenolic alcohol, and then dimethylolphenol. The initial attack may be at the 2, 4, or 6 position. The second stage of the reaction involves the reaction of the methylol groups with other available phenol or methylolphenol, leading first to the formation of linear polymers and then to the formation of hard cured, highly branched structures.
Novolak resins are obtained with acid catalysis, with a deficiency of formaldehyde. A novolak resin has no reactive methylol groups in its molecules, and therefore is incapable of condensing with other novolak molecules on heating without hardening agents. To complete resinification, further formaldehyde is added to cross link the novolak resin. Phenolic rings are considerably less active as nucleophilic centres at an acid pH, due to hydroxyl and ring protonation.
However, the aldehyde is activated by protonation, which compensates for this reduction in potential reactivity. The protonated aldehyde is a more effective electrophile.
The substitution reaction proceeds slowly and condensation follows as a result of further protonation and the creation of a benzylcarbonium ion which acts as a nucleophile.
Resols are obtained as a result of alkaline catalysis and an excess of formaldehyde. A resol molecule contains reactive methylol groups. Heating causes the reactive resol molecules to condense to form large molecules, without the addition of a hardener. The function of phenols as nucleophiles is strengthened by the ionization of the phenol, without affecting the activity of the aldehyde.
Megson states that reaction 2 (in which resols are formed by the reaction of quinone methides with methylolphenols or other quinone methides) is favored during alkaline catalysis. A carbonium ion mechanism is, however, more likely to occur. Megson also states that phenolic nuclei can be linked not only by simple methylene bridges, but also by methylene ether bridges. The latter generally revert to methylene bridges if heated during curing, with elimination of formaldehyde.
The difference between acid catalyzed and base catalyzed processes is (1) in the rate of aldehyde attack on the phenol, (2) in the subsequent condensation of the phenolic alcohols, and to some extent, (3) in the nature of the condensation reaction. With acid catalysis, phenolic alcohol formation is relatively slow. Therefore, this is the step that determines the rate of the total reaction. The condensation of phenolic alcohols and phenols forming compounds of the dihydroxydiphenylmethane type is instead rapid. The latter are therefore predominant intermediates in novolak resins.
Novolaks are mixtures of isomeric polynuclear phenols of various chain lengths with an average of five to six phenolic nuclei per molecule. They contain no reactive methylol groups and consequently cross link and harden to form infusible and insoluble resins only when mixed with compounds that can release formaldehyde and form methylene bridges (such as paraformaldehyde or hexamethylenetetramine).
In the condensation of phenols and formaldehyde using basic catalysts, the initial substitution reaction (i.e., the formaldehyde attack on the phenol) is faster than the subsequent condensation reaction. Consequently, phenolic alcohols are initially the predominant intermediate compounds. These phenolic alcohols, which contain reactive methylol groups, condense with either other methylol groups to form ether links, or more commonly, with reactive positions in the phenolic ring (ortho or para to the hydroxyl group) to form methylene bridges. In both cases water is eliminated.
Mildly condensed liquid resols, important of the two types of phenolic resins in the formulation of wood adhesives, have an average of less than two phenolic nuclei in the molecule. The solid resols average three to four phenolic nuclei but with a wider distribution of molecular size. Small amounts of simple phenol, phenolic alcohols, formaldehyde, and water are also present in resols. Heating or acidification of these resins causes cross linking through uncondensed phenolic alcohol groups, and possibly also through reaction of formaldehyde liberated by the breakdown of the ether links.
As with novolaks, the methylolphenols formed condense with more phenols to form methylene bridged polyphenols. The latter, however, quickly react in an alkaline system with more formaldehyde to produce methylol derivatives of the polyphenols. In addition to this method of growth in molecular size, methylol groups may interreact with one another, liberating water and forming dimethylene ether links (CH2 O CH2 ). This is particularly evident if the ratio of formaldehyde to phenol is high. The average molecular weight of the resins obtained by acid condensation of phenol and formaldehyde decrease hyperbolically from over 1000 to 200, with increases of the molar ratio of phenol to formaldehyde from 1.251 to 101.
As was expected, the molecule is nonplanar. The benzene rings are inclined equally and in opposite directions at angles of 52° to the plane of the angle, while the angle is approximately 10° greater than the tetrahedral angle. The heat evolved during the phenol formaldehyde reaction is 4.1 kcal/mol of formaldehyde added to phenol, and 16.9 kcal/mol of water is eliminated during the subsequent condensation. This indicates a total heat liberation of some 100 cal/g of anhydrous reactants.
B. Nature of Mechanism Methylene and Methylene Ether Bridges
Lilley and Osmond separate the mechanism of condensation of phenol with formaldehyde into two phases. In the early stages, when water is present, condensation is almost definitely ionic. At a later stage, when the reaction proceeds in unreacted phenol as the solvent, after dewatering (or in the final stage when reaction is mainly in molten polymer), it is almost definitely nonionic.
The reactions are shown as involving an added proton, but similar arguments apply by solvolysis with a suitable ionizing phenolic medium.
A methylene bridge is more likely to be formed than an ether link under acid conditions and when free positions in uncompletely reacted phenol are available. This can be shown by considering the behavior of the concentration of the carbonium ion when varying the hydrogen ion concentration. In alkaline conditions, the concentration of carbonium ions is low, but by considering the formation of the ions a similar conclusion can be reached.
It appears, therefore, that ethers can be formed only within limited and critical pH ranges, and then only at slow rates. Thus the formation of many ether links is unlikely. These critical ranges do not coincide with technological practice. Lilley points out the need for research on the distribution of methylol groups in phenolic resin syrups, and on the occurrence of ether links in such resins. All available evidence is against their occurrence in large numbers when free nuclear positions are available, as opposed to the monoreactive phenol alcohols used as models, where ethers form undoubtedly in larger numbers.
Referring to the nonionic mechanism, which is likely to occur in the later stages of phenolic resin formation and in the heat hardening of phenolic alcohols, Lilley shows that since the reaction cannot involve a carbonium ion, it will probably use an intermediate of the quinone methide type. This compound is probably generated by a hydrogen bonding mechanism. This has been shown to occur in phenol alcohols and dibenzyl ethers. In the case of a monoalcohol, the reaction will be
A radical of the quinone methide type is likely to have a considerably long life. It yields a dibenzyl ether easily and a methylene bridge with difficulty. Work done by Van Euler and Hultzsch supports the idea that the intermediate appears through hydrogen bonding.
This nonionic mechanism not only explains the known reactions of phenol monoalcohols, especially as far as ether formation is concerned, but also covers fully substituted dialcohols.
The former reaction is much more likely to occur and the formation of ethers is improbable. Experimental evidence supports this view. Although fully blocked monoalcohols undoubtedly lead to dibenzyl ethers in good yield, this is not the case of partially or nonsubstituted phenol alcohols. Thus ether forms with great difficulty from saligenin and homosaligenin and in very poor yield.
Lilley believes that the formation of dibenzyl ethers depends on the distribution of activation in the phenol alcohol. He suggests that for an alcohol containing a free position, subject to activation by the phenolic hydroxyl group, such activation will oppose ether link formation. Substituents in the benzene ring also affect ether stability. The basis for these ideas was obtained by comparing the behavior of two alcohols, 4 chlorosaligenin and 4 nitrosaligenin.
Each of these has a free ortho position, and they are comparable as regards the number of orientation of available reactive positions. However, the activations of their ortho positions are opposite in polarity. It has been shown that the formation of methylene groups is bound up with the d charge effect of the free ortho position. In the nitro compound, the ortho position is more positively charged. The lone pair of the methylol group provides the most nucleophilic point for the attack of the carbonium ion, and therefore ether formation should predominate.
In the chloro derivative, the negative charge effect is increased and methylene linkages should predominate. Other factors must be taken into consideration. The nitro group directly deactivates the 6 position, so that reaction there is unlikely. However, the chloro group does not affect that position. The nitro group opens positions 3 and 5 for attack, so that the chances of methylene bridge formation are increased. Similar effects were observed on p cresol. Therefore, the ether is more likely to form in the case of nitrosaligenin and less likely to form with chlorosaligenin than at first expected. However, results obtained clearly demonstrate that the proportion of ether links formed depend a great deal on the activity of the free ortho position.
One can conclude that in the hardening of phenol alcohols which have free positions for substitution by methylol or its equivalents, a methylene bridge rather than an ether link will preferably be formed . This happens even more for fast reacting phenols such as meta cresol, resorcinol, and phloroglucinol, where the charge in ortho position is considerably increased. It gives rise to a nonionic reaction condition. Under ionic conditions, there is a limited range of pH values in which ether can be formed. However, this does not correspond with the working conditions of generally used pH. The rate of such a formation is slow whatever the position of equilibrium. The case of a fully sub stituted phenol monoalcohol, where ether link formation is more probable, is therefore not parallel to that of one with free positions. This can be seen by comparing the low yield of ether from 4 chlorosaligenin with 4,6 dichlorosaligenin (Zinke and Ziegler).
This yields 70% of the corresponding ether if heated to 130°C. The use of phenol alcohols as models is therefore of doubtful validity. Phenolic resins occurring in typical practice are likely to contain far fewer ether links than suggested by conclusions drawn on the basis that monoreactive monoalcohols are reliable models for study.
Unfortunately, there is little experimental proof of the effect of pH on yields of ethers and methylene compounds. Lilley states that the reactions of phenolic resins were deduced from those of urea formaldehyde resins. The two systems contain identical electronic configurations, and it therefore appears justifiable to suggest that a carboniumion mechanism covers both cases.
In a later study, Rossouw, Pizzi, and McGillivray found that ether links form at pH 4.3 5.0, and that no ether links form at pH 9 when phenol, resorcinol, and phloroglucinol are used as the model compounds. The rate of the formation and decomposition of the ethers and the amount formed was observed during kinetic experiments and was found to decrease for phloroglucinol, resorcinol, and phenol.
It is important to note that extremely fast acting phenols, such as phloroglucinol, do form ether links. (It was thought that they do not form ether links because of the instability of their hydrpxybenzyl alcohols.) The reason why it is impossible to isolate, for instance, the phloroglucinol ether links is because such ethers form and decompose in the first half hour of reactions at ambient temperature. However, the ether links formed by phenol are much slower to form and to decompose. They are therefore stable enough to be isolated and detected.
The range of ether formations given by Lilley on deductions derived from aminoplastic resin chemistry are probably valid. A pH of 5 is still in the range of ether formations advocated by Lilley (explain pH 6), and pH 9 9.5 is too far from the pH 8 advocated for their formation. Whatever the case, it is certain that ether links form in very narrow pH ranges.
C. Acid Catalysis
Consideration must be given to the possibility of direct intervention by the catalyst in the reaction. Hydrochloric acid is the most interesting case of acid catalyst, and ammonia of an alkaline catalyst. When the phenol formaldehyde reaction is catalyzed by hydrochloric acid, two mechanisms possibly come into operation. Vorozhtov has proposed a reaction route which passes through the formation of bischloromethyl ether (Cl CH2 O CH2 Cl). Ziegler has suggested a route through the formation of a chloromethyl alcohol (Cl CH2 OH) as intermediate. The second route appears to be the more probable one. Both hypotheses agree that chloromethylphenols are the principal intermediates. The chloromethylphenols have been prepared and isolated by various means. They are highly reactive compounds which, with phenols, form dihydrox ydiphenylmethanes and complex methylene linked multiring polyphenols. Reaction is highly selective and takes place in the para position.
D. Alkaline Catalysis
Different mechanisms of alkaline catalysis have been suggested according to the alkali used. In the case where caustic soda is used as the catalyst, the type of mechanism, which seems the most likely is that which involves the formation of a chelate ring similar to that suggested by Price and Sachanandor. The chelating mechanism may initially cause the formation of a sodium formaldehyde complex or of a formaldehyde sodium phenate complex.
When ammonia is used as a catalyst, the resins formed are very different in some of their characteristics from other alkali catalyzed phenol formaldehyde resins. The reaction mechanism appears to be quite different to that of sodium hydroxide catalyzed resins. An obvious deduction is that intermediates containing nitrogen are formed. Several such intermediates have been isolated from ammonia catalyzed phenol formaldehyde reactions by various researchers. Similar types of intermediates are formed when amines or hexamethylenetetramine are used instead of ammonia. In the case of ammonia, the main intermediates are dihydroxybenzylamines and trihydroxylbenzylamines.
These intermediates contain nitrogen and have polybenzylamine chains. They react further with more phenol, causing the splitting and elimination of the nitrogen as ammonia or amines, and producing nitrogen free resins. This, however, requires a considerable excess of phenol and a high temperature. With phenol hexamethylenetetramine resins of molar ratio 31, the nitrogen content of the resin cannot be reduced to less than 7% when heated at 210°C. When the rate is increased to 71, the nitrogen content on heating at 210° can be reduced to less than 1%.
Ammonia, amine, and amide catalyzed phenolic resins are characterized by greater insolubility in water than that of sodium hydroxide catalyzed phenolic resins. The more ammonia used, the higher the molecular weight and melting point, which are obtained without cross linking. This is probably due to the inhibiting effect of the nitrogen carrying groups (i.e., CH2 NH CH3 or CH2 NH2) which is caused by their slow rate of subsequent condensation and loss of ammonia. Ammonia, amines, and amides (particularly dimethylformamide) are sometimes used as accelerators during the curing of phenolic ad hesives for wood products.
E. Metallic Ions Catalysis and Orientation of the Reaction
Bender shows that resins made under special conditions, where a high proportion of ortho ortho links are formed in the phenol methylene chains, can be cured by hexamethy lenetetramine at a much higher rate than resins made with conventional acid or alkali catalysis. This involves condensation within the pH range 4 7 by oxides or hydroxides of alkaline or alkaline earth metals such as zinc, magnesium, and aluminium. Fraser, Hall, and Raum also described the effect of a wide variety of organic salts, oxides, and hydroxides of various metals. They came to the following conclusions (1) that the orientation effect also occurs with 2, 2 dihydroxydiphenylmethane formaldehyde condensation (2) that the directive effect occurs in both the initial condensation and in the subsequent condensation of phenol alcohols (3) that molecular weight distribution of high ortho novolaks differs from those made during normal catalysis (4) that a molar excess of phenol in the original condensation is necessary (5) that an apparent pH of the reaction mixture between 4 and 7 is needed (6) that the presence of electropositive bivalent metallic ions such as Mn+2, Zn+2, Cd+2, Mg+2, Co+2, Ca+2, and Ba+2 (in decreasing order of usefulness) is necessary and that (7) the order of the efficiency of the metal ions used appears to be indifferent to the stability of the unidentified probable complex formed as reaction intermediate.
Fraser and other researchers conclude that the high rate of curing of phenolic resins prepared by metal ions catalysis is due to the preferential ortho methylolation and therefore also to the high proportion of ortho ortho links in the uncured phenolic resins prepared by metal ions catalysis. Normal condensation yields roughly equal quantities of para para and ortho para links, but uncured resins produced by metal ions catalysis yield approximately one ortho para to two ortho ortho links. They ascribe faster curing rates of phenolic resins prepared by metallic ions catalysis to the higher proportion of the free higher reactive para positions available for further reaction during the curing of the resin.
More recently, Pizzi explained the mechanism of the reaction and corrected some of the conclusions which other researchers deduced from the work of Fraser. The mechanism presented is similar to that of the formation of metallic acetylacetonate complexes. It involves the formation of chelate rings between metal, formaldehyde, and phenols similar to that suggested by price and Sachanan for alkaline phenol formaldehyde reactions.
Pizzi substantiated the possibility of such a mechanism by isolating and characterizing the chromium resorcinol formaldehyde and chromium phenol formaldehyde complexes.
The rate of metal exchange in solution and the instability of the complex formed determine the accelerating or inhibiting effect of the metal in the reaction of phenol with formaldehyde. The more stable complex II is, the slower the reaction proceeds to the formation of resin III. A completely stable complex II should stop the reaction proceeding to resin III. If complex II is not stable, the reaction will proceed to form phenol formaldehyde resins of type III. The rate of reaction is directly proportional to the instability, or the rate of metal exchange in solution of complex II.
As in the case of equivalent acetylacetonates, the bonding between metal and ligands in these chelates is partly covalent and their stability can be ascribed to their partly aromatic character. The metal catalyst does not change its valence, but interacts with molecules or ions containing electron donor groups, and accelerates reactions like hydrogen ions do. The acid catalysis due to the metal ion differs only in degree from that of the hydrogen ion.
The effect of the metal is stronger than that of hydrogen ions, because of higher charge and greater covalence, since its interaction with donor groups is often much greater. This is important, because in the preparation of phenol formaldehyde adhesives for wood, very acid pH values cannot be used because they cause wood deterioration. Metal ion catalysts have the same catalytic effect on the preparation and setting of the resins at a higher pH, which is shown by a higher concentration of hydrogen ions and therefore lower pH values.
This allows phenolic resin adhesives to set in milder acid conditions, with neither extention of the setting nor any wood deterioration. The gel times for formaldehyde obtained for resorcinol and catechol indicate that the metal catalysis effect is probably valid for most phenols with a free ortho position to the hydroxy group. Most bivalent metallic ions accelerate phenol formaldehyde reactions.
The extent of this effect is directly proportional to the quantity of metallic ion present. I bivalent metals do not inhibit the re action, as the complexes they form are unstable and their rate of exchange in solution is high. II they accelerate the reaction in a manner similar to that of hydrogen ions, although it is even faster, due to their stronger charge and greater covalence. The formation of the identified stable complexes of type II and V slows down or inhibits the reaction from proceeding to the formation of phenol formaldehyde resins, when trivalent metals are used.
Tannin Based Wood Adhesives
The word tannin has been used loosely to define two different classes of chemical compounds of mainly phenolic nature hydrolizable tannins and condensed tanning. The former, including chestnut, myrobolans (Terminalia and Phyllantus tree species), and dividivi (Caesalpina coraria) extracts, are mixtures of simple phenols such as pyrogallol and ellagic acid and of esters of a sugar, mainly glucose, with gallic and digullic acids.
They can and have been used, sucessfully, as partial substitutes of phenol in the manufacture of phenol formaldehyde resins. Their chemical behavior is analogous to that of simple phenols of low reactivity toward formaldehyde and their moderate use of phenol substitutes in the above mentioned resins does not present difficulties. Their lack of macromolecular structure in their natural state, the low level of phenol substitution they allow, their low nucleophilicity, and limited worldwide production somewhat decrease their chemical and economical interest.
Condensed tannins, on the other hand, constituting more that 90% of the total world production of commercial tannins (350,000 tons per year), are both chemically and economically more interesting for the preparation of adhesives and resins. Condensed tannins and their flavonoid precursors are known for their wide distribution in nature and particularly for their substantial concentration in the wood and bark of various trees. These include various Acacia (wattle of mimosa bark extract), Schinopsis (quebracho wood extract), Tsuga (hemlock bark extract), and Rhus (sumach extract) species, from which commercial tannin extracts are manufactured, and various Pinus (pine bark extract) species not yet commercially exploited. Bark and wood of trees were found to be particularly rich sources of condensed tannins; commercial development ensued through large scale afforestation and/or industrial extraction, mainly for use in leather tanning. The production of tannins for leather manufacture reached its peak immediately after World War II and has since progressively declined. This progressive decline of their traditional market coupled with the increased price and decreased availability of synthetic phenolic materials due to the advent of the energy crisis stimulated fundamental and applied research on the use of such tannins as a source of condensed phenolics.
Chemistry of condensed Tannins
Condensed tannins, consisting of flavonoid units which have undergone varying degrees of condensation, are invariably associated with their immediate precursors (flavan 3 ols, flavan 3,4 diols), other flavonoid analogs , carbohydrates, and traces of amino and imino acids. Monoflavonoids and nitrogen containing acids are present in concentrations which are too low to influence the chemical and physical characteristics of the extract as a whole. However, the simple carbohydrates (hexoses, pentoses, and disaccharides) and complex glucuronates (hydrocolloid gums) are often present in sufficient quantities to decrease and increase viscosity, respectively, and excessive variation in their percentages would alter the physical properties of the natural extract independently of contributions related to the degree of condensation of the tannins.
Monoflavonoids, commonly known as phenolic nontannins, represent the most studied group in the commercially important tannin extracts because of their relative simplicity. They comprise flavan 3, 4 diols (leucoanthocyanidins), flavan 3 ols (catechins), dihydroflavonoids (flavonols), flavonones, chalcones, and coumaran 3 ones, thus representing most of the known classes of flavonoid analogs. Typical are those of black wattle (mimosa) bark extract (Acacia mearnsii), where the four possible combinations of resorcinol and phloroglucinol (A rings) with catechol and pyrogallol (B rings) coexist, although these flavonoids constitute a minor percentage (3%) of the total phenolics .
Among the groups of monoflavonoids noted above, only flavan 3, 4 diols and certain flavan 3 ols do appear to participate in tannin formation. Each of the four combinations of phenolic substitution is to be found in black wattle tannins. The main polyphenolic pattern is represented by flavonoid analogs based on resorcinol A rings and pyrogallol B rings (XX). These constitute about 70% of the tannins. The secondary but parallel pattern is based on resorcinol A rings and catechol B rings (XXI). These tannins represent about 25% of the total bark tannin fraction. Superimposed of these two predominant tannin flavonoid mixtures are two minor groups of the analogs arising from photosynthetic processes occurring in the leaves and immature bark. These are based on phloroglucinol pyrogallol (XXII) and phloroglucinol catechol (XXIII) flavonoids. These four patterns constiture 65 80% of mimosa bark extract.
The remaining parts of the wattle bark extract are the nontannins. They may be subdivided into carbohydrates, hydrocolloid gums and amino and imino acid fractions. The carbohydrates pinitol and sucrose predominate, with glucose in a lower proportion. The hydrocolloid gums vary in concentration from 3 to 6% and contribute significantly to the viscosity of the extract in spite of their low concentration. The nitrogen compounds of wattle bark extract are mainly the imino acids L pipecolic acid, L 4 hydroxy trans pipecolic acid, and L proline, together with lesser quantities of the amino acids arginine, alanine, aspartic acid, glutammic acid, and serine. The nitrogen compounds constitute about 3% of the extract.
Similar flavonoid A and B ring relationships, although slightly different and less surely determined, also exist in quebracho (Schinopsis lorentzii and balansae) wood extract.
The two main substitution patterns of wattle extract are also present in quebracho extract, in which the apparent absence of the fiavonols quercetrin and myricetrin and the flavan 3 ols catechin and gallocatechin is the most obvious and important difference. In brief, no phloroglucinolic A ring pattern, or more probably a much lower quantity of it, is present in the quebracho extract, as can be deduced by the failure in isolating leucocyanidin but in the success of attaining cyanidin chloride. Similar patterns to wattle and quebracho are followed by hemlock and Douglas fir bark extract. Completely different patterns and relationships do instead exist in the case of pine tannins. With the exception of Pinus ponderosa, whose principal flavonoid pattern is just about identical to that of wattle and quebracho tannins, the other pine species studied (Radiata, Eliotae, Taeda, Aleppensi, Sylvestris, Patula, Pinaster, etc.) present instead only two main patterns. One pattern is represented by flavonoid analogs based on phloroglucinol A rings and catechol B rings (XXIII). The other pattern, present in much lower proportion, is represented by phloroglucinol A rings and phenol B rings (XXIX).
The A rings then possess only the phloroglucinol type of structure, with very important consequences in the use of these tannins for adhesives. Resorcinol types A rings as well as pyrogallol type B rings are completely absent. The main leucoanthocyanidins in these extracts are leucocyanidin (XXX) and afzelechin (XXIX).
Of the monoflavonoids presented, only flavan 3, 4 diols and flavan 3 ola apparently participate in tannin formation. This is logical, as all the other flavonoids present carbonyl groups at the 4 position (etherocyclic rings). These groups eliminate the possibility of autocondensation to biflavonoids and higher poly flavanoids by both reducing the nucleophilic character of the A rings and occupying one of the positions through which natural condensation occurs. Meta disubstitution or trisubstitution with hydroxyl or heterocyclic oxygen groups on the A rings of flavan 3, 4 diols promote strong nucleophilic centers at the 6 and 8 positions as well as the formation at the 4 position of benzyl carbonium ions stabilized by delocalization of the charge on the vicinal aromatic ring. Both conditions enhance the possibility of flavonoid autocondensation. Consequently, catechin (IX) and gallocatechin (X) offer the strongest nucleophilic centers while leucofise tinidin (V) and leucorobinetidin (VI) provide potential benzyl carbonium ions for electrophilic substitution. Logically, attack on the catechins should be at both of the two available centers (6 and 8 position) on the A ring and not only on the 8 position as sustained by a few authors. The proved fact that the 8 position is slightly more reactive than the 6 position does not exclude the latter one from the possibility of reaction, as proved by the types of biflavonoids isolated. Some of the anticipated products based on steric considerations are in accord with those found among wattle tannins namely, leucofisetinidin catechin, leucorobinetidin catechin, and leucorobinetidin gallocatechin.
Both types of biflavonoids, the 4, 6 and 4, 8 linked must be present. Similar biflavonoid units which correspond to ( ) leucofisetinidin catechin have been isolated from the heartwood of S. balansae (quebracho).
Related biflavonoids were isolated by Nisi and Panizzi (from Eucalyptus camaldulensis) . Krishnamoorty and Seshadri (Mirica nagi) and delle Monache et al. (Ouratea spp.). Drewes, et al. characterized the sterochemistry of two groups of crystalline leucofisetinidins (XXXI XXXIII) from the wood of the black wattle tree (Acacia mearnsii) and also determined the stereochemistry of a homologous series of all trans 5 deoxyleucoanthocyanidin catechin from the bark extract of the same tree (XXXIV XXXVI).
The C and F ring heterocyclic systems were shown to have half chain conformations for most individual flavonoid moieties, exceptions being twisted boat conformations where 2, 3 trans 3, 4 cis relative configurations pertained in the upper units (C rings). The x ray structure of 8 bromotetra O methy catechin confirms the 2R, 3S absolute configuration of the parent compound catechin. In the heterocyclic ring both the heterocyclic oxygen atom and the 4 carbon lie marginally above the mean plane of the adjacent aromatic A ring with the 2 carbon and 3 carbon more definitely higher and lower, respectively, of the same mean plane. The conformation of the etherocyclic ring is midway between a C 2 sofa and a C 2, C 3 half chair conformation in the crystal. The phenyl and hydroxy groups at positions 2 and 3 in the heterocyclic ring of the 8 bromoderivative are in the trans diequatorial position . Confirmation of the 4,6 coupling for bileucofisetinidins (XXXI XXXIII) and the all trans 5 deoxyleuco anthyocyanidins (XXXIV XXXVI) was obtained by chemical shift data. The biflavonoids (XXXIV XXXVI) from the A. mearnsii bark extract are also accompanied by their apparent precursors, leucofisetinidin, leucorobinetidin, catechin, and gallocatechin and by higher oligomeric analogs . Combined, these constitute the tannins of mimosa or wattle extract of commerce, the mixture having a number of average molecular weight of 1250. Weinges et al. isolated also 4,8 linked leucocyanidins from various sources (XXXVII), while Hemingway and McGraw isolated and identified similarly 4,8 linked biflavonoids from the bark of loblolly pine. The etherocyclic ring of both units of these 4, 8 linked biflavonoids were also found to be in the half chair conformation.
D. Triflavonoids and Tetraflavonoids Condensed Tannins
Roux et al. indicated that the principle of condensation based on 4,6 links between resorcinol units, following initial 4,8 links between resorcinol and lower terminal phloroglucinolic units, appear to be a general flavonoid autocondensation pattern. While the 4,6 links between resorcinolic units are beyond any doubt, considerable doubt still exists about.
1. The positioning of the phloroglucinolic flavonoid unit as the tannins lower terminal unit
2. The 4,8 links being the only phloroglucinolic condensation pattern possible, as proposed.
This was reputedly shown by the isolated triflavonoid condensed tannin from the heartwood of the mopane tree (Colophospermum mopane) and a tetraflavonoid unit from the karree tree (Rhus lancea)
The recurrent 4,6 linkages between the upper units was demonstrated by nuclear magnetic resonance (NMR) spectrometry and by chemical degradation, while the 4,8 linkage to terminal units is based on the interpretation of the shifts of all methoxyl NMR resonances during progressive C6D6 addition to CHCl3 solutions of the methyl ethers of the tannins. The same author, at different stages, advances two different structures for the mode of linkage of the lower terminal phloroglucinolic flavonoid unit. Structures XXXIX, 4,8 linked, and XL, 4,6 linked, have both been proposed for the same tetrafluvonoid, indicating the uncertainty of such an interpretation.
Commercially extracted wattle bark extract (A. mearnsii) show a similar and continuing system of condensation at the triflavonoid level. Descrepancies in the experimental behavior and reactivity toward formaldehyde of tannin based adhesives, especially wattle tannin adhesives, from what expected from any of the proposed structures XXXIX and XL brought to the realization that the phloroglucinolic unit is not available for reaction and should not, in all probability, present any free position (nor the 6 or 8 position) available to formaldehyde attack. If all the structures present should present a lower phloroglucinolic terminal unit free to formaldehyde attack, then the initial condensation, on addition of formaldehyde, between tannin macro molecules should occur through these phloroglucinolic units, as phloroglucinolic units are between 10 and 15 times more reactive toward formaldehyde than the corresponding resorcinolic flavonoids. Thus a very fast and considerable increase in the viscosity followed by a somewhat slower rate of viscosity increase of tannin formaldehyde mixtures should be noticeable. This is not so and viscosity increase graphs in function of time are smooth exponential curves. There are no indications of sudden changes of rate of viscosity increase at the beginning of the curve due to a change of the species reacting with formaldehyde. Quite independently from such applied considerations, Roux and coworkers realized by the use of high resolution and progressively high temperature NMR spectra of triflavonoids obtained by novel synthetic ways that the most probable structure of triflavonoids and position of the phloroglucinolic unit are as follows
From structure XLI it is quite evident that no position highly reactive with formaldehyde is available on the phloroglucinolic unit, as both the 6 and 8 positions are blocked by other flavonoids. Hence in phloroglucinolic flavonoid units, all three 4,6 , 4,8 , and 6,8 linkages are probably allowed. This finding may bring to the realization that all the three types of linkages, through positions 4, 6, and 8 of the phloroglucinolic unit, may be present simultaneously, with the formation of branched rather than linear polymeric tannins. It is not possible to conclude with the data available at this stage if the tannins macromolecules are linear polymers, as assumed up to now, or branched polymers.
A few structural possibilities are open, such as
1. Branched tannins when phloroglucinolic units are linked with the flavonoids at the 4 , 6 , and 8 positions.
2. Linear condensed tannins in which a polymeric tannin of type XXXIX or XL has the phloroglucinolic unit no longer terminal but with the 6 and 8 position, respectively, blocked by a further resorcinolic flavonoid unit, which now becomes the tannin lower terminal unit,
3. Linear condensed tannins in which the 6 and 8 positions of the phloroglucinolic flavonoid unit function as links between two flavonoid chains composed of 4,6 linked resorcinolic units. This is valid for tannins such as those present in commercial wattle and quebracho extracts. As regard pine extracts, whose flavonoids are only phloroglucinolic in nature, the accepted continuing pattern of condensation is 4,8 linked (XLII),
To conclude, the 8 and 6 positions provide the nucleophilic function, while the electrophile is presumably represented by a 4 carbonium ion generated from the flavan 3, 4 diols. In this way the principle of self condensation may apparently be continued, resulting in units as large as 11 or 20 linked flavonoids, respectively, for wattle tannins and for polyflavonoids of higher average mass range as quebracho and pine (average ± 4300).
E. Methods for the Analysis of Phenolic Materials Content in Tanning Extract
Various methods of analysis are available for the determination of tannincontent. These methods can generally be grouped into two broad classes
1. Methods aimed at the determination of tanning material content in the extract. These methods were devised to determine which percentage of the extract would participate in leather tanning. They may also be used to give an indication of the amount of phenolic material that can react with formaldehyde present in the extract. Their main drawback, as regard adhesives, is in the incapability of detecting and determining the approximate 3% of monoflavonoids, or phenolic non tannins, present in the extract, which do not contribute to tanning capacity but which do definitely react with formaldehyde and contribute to adhesives preparation.
2. Methods aimed at the determination of phenolic material present in the extract that can be reacted with formaldehyde. These methods were devised particularly for tanning extracts used in adhesives preparation and are all based on the dtermination of some of the products of reaction of the flavonoids with formaldehyde. Their main defect is that the phenolic material content is expressed as an absolute number. They are excellent in comparing different extracts as regard their relative phenolic content, hence their suitability for adhesive manufacture the absolute numbers obtained, though, bears no relationship with the real percentage of phenolic material in the extract.
Accepted methods of the first type comprise the hide powder method the refractometric method , and various visible, ultraviolet and infrared spectrometric methods. Accepted methods of the second type comprise the Stiasny/Orth method and its modifications and the Lemme, sodium bisulfite back titration method. While an in depth discussion of all the methods noted above is beyond the scope of this review, a brief discussion of some of the more important or interesting ones is necessary
1. Hide powder method. This is the oldest accepted analytical method for the determination of tannins in a tanning extract. It is based on the reproduction of the leather tanning process in which the material to be tanned is standardized powdered hide. The amount of tannins absorbed and fixed by contact of predetermined amounts of tanning extract and hide powder is expressed as a percentage of the original tanning extract. The method is widely used by leather tanners, extract producers, and leather standard and research institutions. All the extract producing factories around the world deliver every industrial batch of tanning extract they produce with the analyzed percentage of tannins obtained with the hide powder method.
2. Molybdate ions spectrophotometric methods. These methods are based on the examination by visible/ultraviolet (UV) spectrophotometry of the intensely yellow orange complexes formed by solutions of molybdate ions with o dihydroxybenzene and 1, 2, 3 trihydroxybenzene derivatives. Mono and polyflavonoids such as condensed tannins, which contain p dihydroxybenzenic (catechol) and 1,2,3 trihydroxybenzenic (pyrogallol) B rings, also form yellow orange complexes with molybdate ions in aqueous solutions. An o dihydroxyl group is essential for complexalion since five membered chelates are formed, consisting of an o dihydroxo group and a central molybdate ion. The absorbance values at 400 nm of solutions containing an excess of sodium molybdate or ammonium heptamolybdate and varying amounts of tannin extract at a buffered pH in the range 4.0 5.6 vary. The amount of condensed tannin present in the extract is obtained by the linear relationship existing between tannin amounts and absorbance values. This method also has the advantage of being able to detect monoflavonoids, or phenolic nontannins, which are overlooked by the hide powder method. This and related UV methods are useflul for the determination of the phenolic material content of extracts used for adhesives.
3. Infrared spectrometric determination of A rings. This method is based on an extension of a method developed by Chow and Steiner by using the 1140/1120 cm absorbance ratio, in the infrared region, to determine the resorcinol content of phenolic resins. It is based on the linear relationship between the variation of the infra red 1490/1450 cm 1 absorbance ratio and the amount of A rings contained in a tanning extract. Of the two bands used to calculate the ratio the 1490 cm, one is characteristic of resoroinolic and phloroglucinolic materials only, while the other is general of all the extract components. The ratio of A rings in a wattle tanning flavonoid is constant at 33% for resorcinolic flavonoids and 43% for phloroglucinolic fiavonoids, it is possible to calculate the percentage of phenolic material in the extract. The regression equation for such a system and the limits of its application are as follows correlation coefficient r2 = 0.950
y = 0.02375x 0.00275 for 25% × 100%
4. Modified Stiasny method. This method is based on the gravimetric determination of the reaction products precipitated during the reaction of tannins with formaldehyde in the presence of HCl. The tanning extract solution is treated at reflux, in an acid environment, with a molar excess of formaldehyde and the fractions of the extractives that are capable of reacting with formaldehyde precipitate out of solution they are then filtered, dried, and weighed, and the results are reported as a formaldehyde precipitation number. The method, although not giving the exact percentage of phenolic material in the extract, is widely used, as it has the advantage of giving a comparative measure of the amount of tannin being capable of reacting with formaldehyde under the conditions of formation of a thermosetting phenolic type resin.
Reactivity of Tannins as Macromolecules
Considering their macromolecular nature, condensed tannins exhibit unique reactions as well as reactions normally expected of flavan 3 ols units. Knowledge of the more useful of these reactions is important in the industrial application of tannin extracts to adhesives.
A. Reactivity and Orientation of Electrophilic Substitutions of Flavonoids.
The relative accessibility and/or reactivity of flavonoid units has been examined by selective bromination with pyridine hydrobromide per bromide using units of the phloroglucinol and resorcinol series. Tetra O methyl catechin (XLIII) is brominated preferentially in the 8 position. Only when this position is filled does substitution commence at the 6 position.
B. A and B Ring Reactions with Aldehydes and Their Kinetics
Tannins, being phenolic in nature, undergo the same well known reaction of phenols with formaldehyde either base or acid catalyzed, weakly basic base catalyzed reactions being predominantly used in industrial applications. Increasingly alkaline conditions lead to progressive activation of the phenol as nucleophile, especially above pH 3, where phenateions are formed. The nucleophilic centers on the A rings of any flavonoid unit tend to be more reactive than those found on the B ring. This is due to the vicinal hydroxyl substituents, which merely cause general activation in the B ring without any localized effects as on those found in the A ring. Formaldehyde reacts with tannins to produce polymerization through methylene bridge linkages to reactive positions of the flavonoid molecules, mainly the A rings. In condensed tannin molecules the cap rings of the constituent flavonoid units retain only one highly reactive nucleophilic center, the remainder accommodating the interflavonoid bonds. Resorcinolic A rings (wattle) show reactivity toward formaldehyde comparable, though slightly lower, to that of resorcinol. Phloroglucinolic A rings (pine) behave instead phloroglucinol. Pyrogallol or catechol B rings are by compari son unreactive, and may only be activated by anion formation at relatively high pH. Hence, the B rings do not participate in the reaction except at high pH values (pH 10) where the reactivity toward formaldehyde of the A rings is so high that the tannin formaldehyde adhesives prepared have unacceptably short pot lives. In general, tannin adhesives practice only the A rings are used to cross link the network. However, because of their size and shape, the tannin molecules become immobile at a low level of condensation with formaldehyde, so that the available reactive sites are too far apart for further methylene bridge formation. The result is incomplete polymerization that leads to the weakness and brittleness that are characteristic of many tannin formaldehyde adhesives. There are indications, though, that at least pyrogallol B rings are capable of a limited degree of condensation with formaldehyde, even in the presence of an excess of the more reactive resorcinol and in mildly acid or alkaline reaction conditions, as shown by the formation of pyrogallol formaldehyde dimers and resorcinol pyrogallol formaldehyde dimers and trimers, with limited pyrogallol participation, in model compound reactions carried out at ambient temperature. The latter constitutes a rethinking and indicates that notwithstanding the possibility of a very limited participation of the B rings to a cross linked tannin formaldehyde network, such a network is still weak.
Bridging agents with longer molecules, such phenolic and aminoplastic resins, have been used to solve this problem by helping to bridge distances too large for interflavonoid methylene bridges. The latter trend is evident by the type of industrial tannin adhesive formulations exposed in the application part of this chapter. Hillis and Urbach have shown that while catechol and the catecholic B rings do not react with formaldehyde at pH lower than 10, through their lack of formaldehyde consumption, the addition of zinc acetate to the reaction mixture induces the B rings to react with formaldehyde at lower pH values, the optimum being in the pH range 4. 5 5. 5, as shown by the higher amount of formaldehyde consumed. This finding implies that in the presence of zinc acetate further cross linking of the tannin formaldehyde network could be achieved through B ring participation to the reaction. This could eliminate the need of increasing cross linking by addition of synthetic phenolic and aminoplastic fortifiers. Pizzi has shown that at the low levels of addition of zinc acetate which are economical (namely 5 10% on resin solids), there is an improvement in strength given by a higher degree of cross linking, but not enough to give the same performance of fortified tannin resins.
As regards the pH dependence of the reaction with formaldehyde, it is generally accepted that the reaction rate of wattle tannins with formaldehyde is slowest in the pH range 4.0 4. 5 for pine tannins, between 3.3 and 3.9. The quantity of formaldehyde that reacts with the polyphenols in this pH range has been found to be minimal.
At neutral pH, rapid reaction with formaldehyde at the 6 and 8 positions of the monomeric unit occurs, accompanied by a much slower reaction of positions 2 and 6 on the pyrogallolic or catecholic B ring. The dependence of the gel time of tannins with formaldehyde at different pH values is shown in Figure 1.
Formaldehyde is generally the aldehyde used in the preparation, setting, and curing of tannin adhesives. It is normally added to the tannin extract solution at the required pH both as liquid formalin solution and in its polymeric form of paraformaldehyde, which is capable of fairly rapid depolymerization under alkaline conditions. Hexamethy lenetetramine (hexamine) may also be added to rosins due to its formaldehyde releasing action under heat. Hexamine is, however, unstable in acid medium but becomes more stable with increased pH values. Hence under alkaline conditions the liberation of formaldehyde might not be as rapid and as efficient as described. Under acid conditions hexamine decomposes to produce 6 mol of formaldehyde and 4 mol of ammonia and is soluble in water up to 50% by weight. At an alkaline pH only 3 mol of formaldehyde is liberated, accompanied by the formation of trimethylamine. Since, under alkaline conditions, hexamine liberates formaldehyde only when heated, mixtures of hexamine and wattle solution exhibit an indefinite pot life at room temperature. It has been fairly widely reported with a few notable exceptions that bonds formed with hexamine as hardener are not as boil resistant as those formed by formalin or paraformaldehyde. This leads to the notion that the ammonia released is responsible for the degree of loss in weather resistance. A few authors have instead conclusively shown that hexamine can give bonds as good or better as paraformaldehyde, in alkaline environment only this may be ascribed to the formation of trimethylamine rather than ammonia with less damage to the bonds formed. The reaction of formaldehyde with tannins may be controlled by addition of alcohols to the system. Under these circumstances some of the formaldehyde is stabilized by the formation of hemiacetals [e.g., CH2(OH) (OCH3) if methanol is used]. When the adhesive is cured at an elevated temperature, the alcohol is driven off at a fairly constant rate and formaldehyde is progressively released from the hemiacetal. This ensures that less formaldehyde is volatilized when the reactants reach curing temperature, and also that the pot life of the adhesive is extended. In view of the fact that the methylene linkages may be too short for the optimum cross linking, other aldehydes which also have bifunctional characters have been substituted for formaldehyde. Of these, one of the most frequently used is furfuraldehyde. This has been found to be unsuitable because of its slow reaction with phenols but, pizzi and Scharfetter have shown that furfuraldehyde is an efficient cross linking agent and excellent plasticizer for tannin adhesives when coupled with formaldehyde. Glutaraldehyde has been shown to react with tannins to produce a slow forming precipitate, whereas precipitates with formaldehyde forms much faster. The reaction kinetics of formaldehyde, acetaldehyde, propionaldehyde, n butyraldehyde, isobutyraldehyde, and furfuraldehyde with both resorcinolic and phloroglucinolic type condensed tannins have been investigated. The same experiments were repeated on resorcinol, phloroglucinol, and catechol used as simple model compounds.
The tannins have also been found capable of reactions with glyoxal and benzaldehyde by means of high temperature gel time measurements of tannin solutions with the two aldehydes. Discreponcies the speed of reaction of resorcinolic and phloroglucinolic tannins with various aldehydes have been observed.
Urethane Structural Adhesive Systems
Urethane structural adhesive systems are relative newcomers to the marketplace. However, in the area of fiberglass reinforced plastic (FRP), urethane structural adhesives have captured a very large share of the market. This is partly due to the good specific adhesion of urethanes, their excellent chemical and environmental resistance, and the high lap shear strengths of urethane bonds.
The fiberglass structural bonding market is a rapidly growing one. Projections indicate a growth of 15 20%/year for the next 5 years. Major projects requiring huge amounts of adhesive per part, for example, all plastic truck cabs or conventional assembly line parts with large numbers of parts (car doors, hoods, or trunk decks), is major growth areas and help account for the 15 20% annual growth projections. With this in mind, let us proceed to urethane structural adhesive chemistry.
B. Advantages and Limitations
Urethane adhesives offer significant advantages over other structural adhesive systems. Urethane polymers can be viewed as a series of interconnecting, soft and hard segments, while other structural adhesives have one or the other (soft or hard segments), only urethanes have this unique combination. The ratio of soft to hard segments may be varied to produce a wide range of physical properties.
Urethane structural adhesives have excellent water and humidity resistance. The urethane linkage is hydrolytically stable and unaffected by a high concentration of water at elevated temperatures. This valuable property is important in any bonded part, which will, or could, environmental exposure. Urethane structural adhesives are equally resistant to salt water and show little or no loss of bond strength when exposed to salt spray.
Urethane structural adhesives can be compounded to resist high temperature paint bake ovens and to retain structural integrity after exposure to 400°F (204°C) for short periods of time. This allows bonded parts to be processed on conventional assembly lines.
The disadvantage of two components, urethane structural adhesives is that they are moisture sensitive in the uncured state. They cannot be conveniently hand mixed, which limits the amount of material used to comparatively large quantities. Meter mix machines should be used for maximum bond strengths, and their use represents some type of capital investment. Having machines to meter and mix adhehesive also means that they may break down, may go off ratio, and, of course, do require periodic maintenance.
A. Basic Concepts
Urethanes, as a generic class of organic chemicals, are the reaction product of an alkyl or aromatic diisocyanate and a multifunctional polyol (bi and trifunctional polyols are the most common, but others are often used for special purposes). The reaction is a nucleophilic attack on the carbonyl of the isocyanate group by one of the lone electron pairs of oxygen of the hydroxyl group of the polyol.
The two most common diisocyanates are toluene diisocyanate (TDI) a mixture of 80% 2,4 isomer and 20% 2,6 isomer and methylene bis 4,4 phenyldiisocyanate (MDI). These diisocyanates can be used separately or in combination to produce the desired physical properties in the cured adhesives.
Two types of polyols are available in the industry, polyester polyols and polyether polyols. Early urethane polymers utilized polyester polyols. Unfortunately, ester linkages are susceptible to hydrolytic cleavage, so these early urethanes degraded in a fairly short period of time, due to moisture. This, of course, gave urethanes a bad name. At the present time, however, the polyether polyols are used exclusively in any urethanes which may environmental exposure for example, adhesives.
Diamines are used in urethane adhesives as chain extenders. The reaction between diisocyanates and diamines produces substituted ureas.
Urea linkages are part of the hard segment of the urethane polymer. They are harder than urethane linkages, and this property can be used to the urethane chemists advantage. These harder hard segments lend better tensile properties and higher heat resistance to the polymer. Tertiary amines make good catalysts for the urethane reactions.
The advantages of urethanes come from their unique polymeric structure. The combination of hard and soft segments allows properties of both rigid and elastomeric polymers. The hard segments provide good high temperature properties, good tensile strengths, and good modulus properties, while the soft segments provide excellent low temperature properties as well as some elastomeric properties. The combination of these two sets of characteristics makes a good structural adhesive.
Application Meter Mix Equipment
Two component urethane structural adhesives should be meter mixed to provide consistent, high quality, mixed adhesive. As the name meter mix implies, these machines perform two functions
1. Metering the correct amount of prepolymer and curative
2. Mixing the two components to provide an air free, complete mix.
Several companies currently produce such equipment. Some of the requirements of meter mix machines are
1. Ability to pump materials of different viscosities (prepolymer usually of higher viscosity than curative)
2. Metering of prepolymer and curative accurately enough to have a consistently high quality product (usually within 5 10% of the stated ratio)
3. Temperature controlled material pots, to ensure consistent gel times year round.
Several different metering pump designs are found in the marketplace.
1. Machine utilizes a follower plate mechanism. This type of machine is normally used for prepolymers and curatives of high (75,000 250,000 cP) viscosity. The follower plate type of machine can be used with lower viscosity materials, but the cost of this type machine does not warrant its use.
2. Machine is a double action, air driven piston pump. Cylinders of different diameters are used for metering the two components. A recent improvement of the design has been a shorter shaft connecting the air cylinder and the pistons, allowing faster cycle times.
3. Machine is an impeller type pump. The impellers, driven by an air motor, drive the pumps, which meter the adhesive.
4. The gear driven type of pump. This type of machine, unlike impeller machines, has the advantage of no surge operation.
All but the follower plate machines are air (pneumatically) powered. This is so these machines can be used in plants where flammable liquids might be used in production process.
Two different types of mixers are used in the industry static and dynamic. The dynamic mixers are grids, inside a mixing chamber powered by an air motor. The problem with dynamic mixers is that the heat they add to the mixed adhesive makes the gel time shorter. The dynamic mixer is, however, easier to clean than the static mixer. Static mixers are simple mechanically and do not add much heat to the mixed adhesive. Cleaning of small static mixers can be a problem due to urethanes, excellent solvent resistance. Burning out of static mixers is not recommended by the manufacturers but is often done by the customer.
Curing, Testing and durability
Two component urethane structural adhesives have the advantage of room temperature curing. Some customers do use heated fixtures for their parts to accelerate the gel time of the adhesive. Obviously, in high speed production operations, the ability to speed up the cure is desirable.
To ensure adequate contact of adherends, and in the case of heated fixtures, to accelerate the cure. Unheoted fixtures should exert 2 5 psi (15 35 kPa) throughout the bondline heated fixtures must exert 20 40 psi (140 280 KPa). The reason for this difference is that at elevated temperatures [200 225°F (93 107°C)], water preferentially reacts with the isocyannte, producing CO2. This CO2 will form bubbles in the cured adhesive, thereby weakening the bond. A pressure of 20 40 psi (140 280 kPa) will force the carbon dioxide into solution and prevent the bubbling. Temperatures over 250°F (121°C) will blow the bond due to the same water/isocyanate reaction. At these temperatures, however, not enough force can be applied to the part to prevent blowing the bond.
A new one component structural adhesive requires heat for curing. A time strength graph for several temperatures is shown in Fig. Generally speaking, 5 min at 200°F (93°C) or 1 min at 300° F (l49°C) will cure the material totally. The one component adhesive has the obvious advantage of not requiring meter mix equipment.
B. Testing and Durability
Extensive testing of the urethane structural adhesive systems has been done in cooperation with the major automotive and truck companies, both in the United States and in Europe. A summary of the results of this testing is found in Tables 1 and 2.
Table 1 contains tensile, percent elongation, Youngs modulus, 100% modulus, and shore hardness data for six types of structural adhesives over a temperature range from 40°F ( 40°C) to 250°F (121°C). The first column represents a standard, two component, urethane structural adhesive the second, a version resistant to high heat [400°F (204°C) for 60 min] the third, a sandable, paintablc, two component urethane the fourth, a two component, urethane elastomer for use in the air and oil filter industries the fifth, a one component, urethane, elastomeric adhesive/sealant the sixth, a new, experimental, one component structural adhesive.
All of the data in Table 1 were obtained from primed 0.060 in. (0.15 cm) thick steel. The 1 in. (2.5 cm) overlap bonds were pulled in tensile. All failures were cohesive within the adhesive.
Urethane structural adhesives do adhere to unprimed metals however; to protect the metal surface a primer is highly recommended.
Table 2 shows the effect of substrate and various environmental conditions on the flex fatigue strength of urethane adhesive. Sheet molding compounds (SMC), high glass sheet molding compound (HGSMC), directional glass sheet molding compound, cold rolled steel (CRS), and aluminum were used as substrates. The flex fatigue test is an Owens Corning Fiberglass test for durability. The test consists of placing a bonded panel in the flex machine. One end is fixed, while the opposite end is flexed 7.5° to either side of the normal plane.
Urethanes also exhibit a useful gap filling trait when compounded as adhesives. Table 3 and Fig. 6 illustrate this property of gap filling. Often with plastic production parts tolerances are ± 0.020 in. (0.04 cm). If an adhesive works well only with thin bondlines or with very consistent bondlines, it may not be suitable for production parts. Figure 6 and Table 3 illustrate that even at 0.110 in. the joint retains a lap shear, strength of 1300 psi (9000 kPa), which is enough to de laminate most SMC and HGSMC laminates.
Modified Acrylic Structural Adhesives
The most recent, and perhaps the most versatile, generic family of structural adhesive products to be introduced to the parts assembly industry has been designated modified acrylic structural adhesives. Additional material pertaining to modified acrylic structural adhesives may be found in 1977.
Although chemical fastening, or adhesive bonding, is as old as recorded history itself, its use has been limited by a multitude of factors. Until only recently the use of adhesives had been essentially a last resort when no other method of joining could be used from the ancient Egyptians manufacturing paper from papyrus by bonding cross plies of fibers to the assembly of lightweight, durable, aerodynamically designed aircraft of modern day.
With the advent of synthetic polymer chemistry, the adhesive technologist in the early twentieth century could begin to explore resources other than the naturally occurring raw materials of centuries past. The advent of synthetic polymers, ranging from elastomesrs to rigid plastics, provided the adhesive technologist with a continuing flow of new raw materials phenolics, urethanes, vinyl resins, epoxies, etc.
The earliest, truly structural adhesives were based on phenolic resins but were found to be limited by rigidity. They were subsequently modified with flexible resins, such as polyvinyl buty ral, to provide impact strength and flexibility. The limitations of the phenolic chemistry mothered the invention of the new epoxy based adhesives of the late 1940s and early 1950s. The epoxios, until very recently, had been the state of the art of high performance structural adhesives. The limitations of the epoxy adhesives they require clean, well prepared surfaces they require heat cures for best performance they provide only moderate adhesion to the newer engineering plastics now replacing metals in many areas mothered the invention of the modified acrylic structural adhesives.
In the late 1960s and early 1970s, the adhesive technologists began to exploit the potential of synthetic polymer chemistry in situ or within the bondline. In other words, they began to build the adhesive polymer and, at the same time, bond the assembly. Very distinct advantages of this concept emerged. Poorly prepared surfaces could be tolerated quite well. Adhesion to metals was retained, and far improved adhesion to engineering plastics was accomplished. Room temperature curing without physically mixing two components was provided. Far improved handling characteristics were evident because very low viscosities could be applied while high molecular weight crosslinked polymers were formed in the bondline, without volatiles or solvents being emitted.
With the new advantages, the modified acrylic structural adhesives can now begin to compete successfully in many areas. They are considered, not as a last choice when no other method of parts assembly, such as mechanical fastening or welding is adequate, but as a first choice when new assemblies are being designed. In many cases, they are used as high quality alternatives to mechanical fastening on parts that are already designed and in production.
The performance of any adhesive product in any given assembly is dependent on joint design, application conditions, load distribution, environmental conditions, substrate properties and in service conditions. The ultimate test of any adhesive performance is the actual in use history. Regardless, certain standard testing procedures that aid in the selection process have been adopted to communicate somewhat representative performance values on test configurations.
Tables 1 4 provide representative data on the three families of acrylic adhesives. The advantages of these materials over all other room temperature curing structural adhesives in processing and performance are quite significant.
1. Wide substrate versatility commonly the same adhesive performs satisfactorily on steel or aluminum as well as on engineering plastics.
2. Unsurpassed hydrolytic resistance and permanence In various aggressive environments. Acrylic polymers are well known for resistance to aggressive environments. They are not readily plasticized by moisture and can be highly cross linked in situ during curing reaction.
3. Versatility in processing variables Room temperature cure, without prior mixing, is possible. Mix in systems may also be used. Since the adhesives have very low molecular weights prior to curing, low power pumps and metering devices may be used even though non sagging rheology may be required, even with these 100% reactive systems.
4. Excellent price to performance ratios although they provide improved processing and performance compared to most cyanoacrylates, anaerobics, and specialized epoxies and urethanes, the modified acrylics are commonly priced significantly lower.
5. Minimal or no surface preparation on metals and plastics Commonly, mill finished steel and aluminum may be bonded as received with little or no deleterious effects on initial strength or long term durability. In some cases improved performance is noted on unprepared, as opposed to rigorously prepared, metals.
6. Structural, or load bearing, physical properties. The modified acrylic structural adhesives provide extremely high initial bond strength and long term durability on load bearing assemblies. Analogous performance with other generic structural adhesives normally requires high temperature curing and rigorously prepared substrates.
7. Tolerance for poorly mated surfaces. Although best performance is realized with 5 10 mil bondlines, adequate and acceptable performance is possible with bondlines of 1 125 mils or thicker if necessary.
1. Poor adhesion to most unprepared, cured elastomers and low energy surface plastics. The modified acrylic structural adhesives do not readily cure on untreated, cured elastomer surfaces. Adhesion to untreated polyethylene, polypropylene, and various fluoropolymers is poor.
2. Distinctive odor. Commonly, in poorly ventilated work areas, the distinctive acrylic odor is unfamiliar and considered unpleasant by some application personnel.
3. Mix in systems commonly of unequal portions. Mix ratios of 201 41 are most commonly used in mix in systems. One to one systems are available for certain applications but have limited storage life and less performance versatility. Tolerance to slightly off ratio mixing is good.
4. Flash points. The uncured adhesive fluids may commonly have flash points slightly below room temperature. Specially developed systems may have flash points above 100°F (38°C).
5. Limited long term, in use credibility. Due to relatively recently developed technology, case histories beyond 10 yr in use are rare. Newer systems have very limited in use histories.
C. General Performance
Additional performance properties, such as resistance to aggressive environments, fatigue resistance, crack propagation, effect of bond line thickness, joint design, etc., have been characterized on individual products within the families discussed. The results of such testing have indicated a high level of durability compared to presently known structural adhesives. The few major suppliers of acrylic adhesives can provide specific durability data on these materials to aid in the engineering design of new bonded assemblies.
Unlike the urethane or epoxy adhesives, the modified acrylic structural adhesives cure by a free radical polymerization process rather than by ring opening or condensation polymerization. Therefore, the time span between detectable thickening and achievement of full handling strength of the acrylic adhesives is very short compared to epoxies or urethanes. This is viewed as another distinct process advantage of the modified acrylic structural adhesives. A cure profile, such as the one depicted in Fig. 1, provides an unusual combination of usable repositioning time coupled with a relatively short interval between starting to cure and reaching full handling strength. In the case of epoxy adhesives, the curing adhesive thickens to a point where repositioning is not recommended long before full handling strength has developed. In the case of acrylic adhesives, the parts may be repositioned for a period approaching 80% of the time required to reach full handling strength. The caution recommended in the case of the acrylic adhesives is that once thickening has become obvious, full handling strength will rapidly be achieved, so repositioning cannot be tolerated or recommended. With epoxies and urethanes some latitude is possible regarding repositioning after the onset of visible thickening.
The majority of acrylic structural adhesives provide cure times in the range of 2 min to 1 hr at room temperature. In contrast to the curing of the epoxies or urethanes, the use of heat to speed the cure is not recommended. The cure speed may be accelerated in certain conditions by heating at only mild temperatures in the range of 130 150°F (55 66°C). In no case should the cure temperature exceed 160°F (72°C) during the initial stages of cure. Also, unlike the epoxies, the acrylic adhesives cannot be B staged. Once the free radical cure has commenced, it will go to completion by a self propagating mechanism.
The chemistry of the modified acrylic structural adhesive is quite complex and extremely versatile, but the chemical concepts are rather simple and straightforward. Several patents have been granted in the technological area and are beyond the scope of this chapter. It is important, however, to describe the concepts in a broad sense.
The adhesive resins are essentially made up of various polymers dissolved or dispersed in reactive, unsaturated monomers. This fluid also contains free radical initiators, free radical scavengers, fillers, nonreactive diluents, and, in some cases, unsaturated oligomers.
Phenolic Adhesives and Modifiers
In the early 1900s, Leo Baekeland discovered a way to use, for practical a very simple chemical reaction between phenol and formaldehyde. The chemistry of this application had been investigated many years before. The initial reaction products are a series of relatively low molecular weight oligomers with molecular weights from a few hundred to a few thousand. With additional heat, and sometimes a catalyst or hardener, these oligormers will chain extend and crosslink to yield a phenolic, thermoset product.
Baekeland discovered a method to make useful products from a resin, which previously had had no special utility. However, it was only after 5 years of intensive effort and after many failures, that he succeeded in the development of a useful material called Bakelite.
Leo Baekeland had a varied educational and an interesting occupational career prior to his discovery of Bakelite. His early studies were in Belgium at the University of Ghent. Later he studied at the University of London, Oxford University, and the University of Edinburgh. After coming to the United States he worked in the area of photographic materials. It was in this area that he made a major development, which, after some hard business times, gave him the financial freedom to continue investigations into new areas.
After a long and systematic investigation in which Baekeland tried to study all factors of the reaction between formaldehyde and phenol, he found that the reaction could be dissected or separated into different steps. He also found that pressure was valuable in controlling the reaction, and that in the presence of ammonia or another, base he could spread the reaction over a longer period and so could stop it at any stage he wished by cooling. In 1910 the General Bakelite Company was founded. At this time there were a number of industries, which were in need of this Bakelite material, a plastic, which could be used to mass produce standardized, interchangeable parts.
The product had excellent dimensional stability and electrical properties thus, one of the early uses of the product was an electrical application. Subsequent to this, the phenolic material was used in the automotive industry because of its dielectric strength and its immunity from temperature, acids, oils, and moisture.
As early as 1912 there were hundreds of uses and applications for Bakelite materials. These included many for the automotive related area, such as molded parts, moisture resistant cements, and timing gears. However, until 1915 most of the uses were confined to lighting and ignition equipment. By 1918 these phenolic based products were also used as radiator caps, gearshift knobs, battery terminals, door latch handles, sliding circuit connectors, commutators, as well as in spark plugs, gauges, and as cement for bonding electric headlight lamp bases. By 1935 the uses had expanded further to instrument panels, steering wheels, magnetic couplings, ignition locks, robe rails, door lock buttons, ash trays, heaters, and parts of the auto radio. Most of these applications were based on molding materials and on coating resins (e.g., varnishes). In the early 1930s phenolic resins as adhesives began to become more important. The typical glues available at that time had limitations which resulted in an inability to produce a uniform product. Staining, lack of moisture resistance, and lack of resistance to bacteria and fungi were all problems which were encountered with adhesives at that time. About 1931 development of the use of a new phenolic resin for plywoods and veneers began. It was recognized that phenolics had an advantage of being chemically inert, and thus, were free from attack by fungi and bacteria. They were unaffected by heat, cold, and moisture, and did not stain.
Since these early times of the plastic industry, many new plastics have been discovered, but phenolics have remained as lively and as important as they were in those first formative years.
At the present time phenolic resins are used as the major bonding agent, or contribute to bonding, in a variety of automotive related application areas including foundry, friction, abrasives, fiberbonding, contact adhesives and sealants.
Chemistry of Phenolic Resins
Phenolic resins are manufactured from phenol and a large number of substituted phenols through reaction with an aldehyde, primarily formaldehyde. Some examples of phenols are cresols, bisphenol A, resorcinol, p t butylphenol, p phenylphenol, xylenols, cardenol (meta substituted alkyl phenol from cashew nut shell liquid), and others.
The major chemical route to the most common reactant, phenol, is outlined in Fig. 3. Benzene is initially reacted with propylene to yield cumene. Cumene is then oxidized to cumene hydroperoxide, which, in turn, undergoes an acid catalyzed rearrangement reaction to yield phenol and acetone.
There are two basic chemical types of phenolic resins, resols and novolacs. They are differentiated by their phenol to formaldehyde ratio, the type of catalyst used in manufacture, and the chemical structure of the resulting resin. These chemical differences are further illustrated in Figs. 4 and 5.
A novolac resin is characterised by having no reactive methylol groups but having unsubstituted ortho and for para reactive sites where a hardener, such as hexamethylenetetramine (hexa) , can react to yield a chain extended and, ultimately, crosslinked polymeric system. A resol resin, on the other hand, contains not only open reactive sites, but also reactive methylol groups. The result is that resols require only heat to effect chain extension and crosslinking reactions. The cure of both types of resins is dependent on temperature, catalyst type, hardening agents such as hexa, concentration of catalyst and/or hardening agents, and the type of phenol and aldehyde used.
The chemical reactions involved in the cure of phenolic resols, for example, include substitution reactions of the methylol groups at a reactive site of another phenolic ring, yielding a methylene linkage a substitution reaction by a methylol hydroxyl group to yield a methylene ether linkage.
The chain extension crosslinking reactions of the phenolic resol or novolac result in a fully cured system. Many factors contribute to the degree of this cure, which, in turn, affects the performance properties of the ultimate product. Leo Baekeland in 1909 described the curing process as going through three phases of reaction.
The first phase results in the formation of low molecular weight oligomers and is designated as A stage. At ambient temperatures the phenolic may be a low to high viscosity liquid, paste, or solid. This A stage product is soluble in alcohol, acetone, or similar polar solvents and in sodium hydroxide solution. The solid form will melt on being heated.
The second phase involves the formation of an inter mediate condensation product and is designated as B stage. In this form the phenolic is a brittle solid which is slightly harder than a solid in the A stage. The B stage resin is now insoluble in all solvents but may swell in acetone or similar solvents. Although it will not melt on heating, it will soften and can become somewhat thermoplastic like. Further heating will take it into the fully cured C stage. In this stage the phenolic is infusible and insoluble in all solvents. The cured resin is now resistant to chemicals, thermally stable, and a good insulator to heat and electricity.
Analytical Test Methods
A number of different methods are used in the laboratory to elucidate the composition and structure of phenolic resins and the chemistry of the curing reactions. These methods include infrared spectroscopy (IR) nuclear magnetic resonance spectroscopy (NMR) differential scanning calorimetry (DSC) thermal gravimetric analysis (TGA) gel permeation chromatography (GPC) vapor phase chromatography or gas chromatography (GC) and dynamic mechanical analysis (DMA), among others. Each of these methods offers some unique insight into the chemistry of phenolic resins. These methods are in addition to the normal quality control techniques that are commonly used. Typical of the latter are plate flow, gel time, viscosity measurement, and other tests that are a function of molecular weight, reactivity, and crosslink density of the phenolic resin.
Both IR and ultraviolet (uv) visible absorption spectroscopy can yield information on structure and functionality of phenolic resins, However, these methods require careful interpretation and appropriate standards. A more useful analytical technique is proton or 13C NMR. Most of the published work on NMR of phenolic resins discusses proton NMR. In recent years 13C NMR results have become more abundant, and it is predicted that in the next few years solid state NMR will be used more frequently.
Proton NMR quite readily distinguishes the different types of substituents and ring linkages between the phenol groups. Examples of typical NMR spectra for novolacs and resols are shown in Figs. 8 and 9, respectively. Note that these spectra give information, not only on the type and location of substituents on the phenol ring, but also on their relative concentrations.
Gel permeation chromatography is a variation of high pressure liquid chromatography designed to separate molecules based on molecular weight differences. Uncured phenolic resins can be easily separated into monomers, dimers, trimers , etc. Fig. 10 for a typical resol and Fig. 11 for a novolac resin. Note that in Fig. 10 a peak is shown for a methylolated phenol monomer and is clearly indicative of a resol resin. Depending on the analytical equipment and the columns used, even better separation of the peaks can be achieved. Thus, a GPC scan can be an excellent way to characterize a phenolic resin in terms of resin structure and relative component ratios, as well as resin molecular weight distribution.
Differential scanning calorimetry and TGA are two methods which measure the response of a phenolic resin to increasing temperatures. Differential scanning calorimetry is used to obtain information on the resin softening point Tg and on its cure characteristics. Generally, caution is advised on interpreting the cure information due to complications arising from the formation and emission of by products, such as water and ammonia (from hexa). To circumvent this problem, a pressure cell is used to prevent the loss of such components, thus yielding somewhat better results.
Thermogravimetric analysis gives an interesting insight into one of the key properties of phenolics, that is, their thermal stability. This technique measures the weight gain or loss of a material as a function of increasing temperature. It is a useful monitor of the uptake of oxygen and degradation of the phenolic resin. Examples Fig. 12 These TGA scans are obtained on cured materials and show that the oxidation of the methylene linkages does not occur until about 752°F (400°C).
With the exception of TGA, most analytical methods of characterizing phenolic resins involve measurements on phenolic oligomers, that is, A stage resins. For this reason, they often do not yield all the information required to evaluate the cure characteristics of a given resin,
A more recent technique, DMA, is offering promise for evaluating phenolic resins during simulated cure conditions. One type of instrument, the Du Pont Model 880 DMA, measures the ability of a sample to transmit an applied frequency. The standard method of analysis involves scanning a sample at a programmed rate of temperature increase while simultaneously monitoring frequency. This frequency response is directly related to the modulus or stiffness of the sample which, in turn, is dependent on the molecular weight/crosslink density at a specific point in time. Figure 13 illustrates a typical DMA scan of a solid, resol resin. The DMA scan gives information on the melting range and relative cure rates and, in a cool down mode, gives a modulus temperature profile of the cured resin system.
In addition to montoring the samples frequency response, DMA also provides a mechanical loss scan which yields a unique measurement of the Tg and the gel temperature based on a mechanical property of the resin. One of the major deficiencies of this scanning method is that there are three variables (time, temperature, and modulus) in a two dimensional monitoring system. Thus, the kinetics of the cure reaction cannot be quantified. In order to quantify the kinetics of the cure reaction(s), a modified technique has been developed which involves monitoring the frequency response of the phenolic sample in an isothermal mode. Rate constants are obtained as a function of temperature. From these rate constants a temperature response factor for the overall reaction can be calculated. This type of data can be used to predict the degree of cure of a phenolic resin under a given set of reaction conditions for a given time.
Phenolic resins are available in a wide variety of forms and can be used in a large number of applications. The methods of choice for evaluating a phenolic resin are dependent on the application area. Many times simple quality control methods are all that is needed and required. At other times a more detailed analysis is necessary in order to compare resin structure and properties with actual application performance needs. The following selected section will discuss phenolic bonding applications in the automotive area. A more detailed discussion will be presented on phenolics as modifiers for adhesives.
Phenolic resins can be bonding agents as neat resin (adhesive) or as part of a formulation (phenolic modifier). The phenolics have good adhesion to polar substrates, good high temperature properties, resistance to burning, and high strength. Phenolics are used as bonding agents in fiberbonding, friction, abrasives, and foundry applications, among others, all of which utilize the material as a neat resin.
In the fiberbonding area the phenolic resin is used as a binder in products such as thermal insulation batting, automotive acoustical padding, and cushioning materials. These products can consist of a variety of fibers such as glass, mineral, cotton, and polyester laid down in a randomly oriented, loosely packed array to form a mat. The phenolic resin is used to bond the individual fibers together using either a dry bonding or a wet bonding process. The dry bonding process uses a pulverized phenolic resin, either a resol based or a novolac hexa system, to bond reclaimed fibers. Automotive acoustic padding, normally involving organic fibers, and low grade thermal insulation batting, utilizing glass fibers, is made using the dry bonding process. Resol based phenolic resins offer some advantages for manufacture of both types of products, especially for the glass based materials, but this type of resin requires special handling and refrigerated storage. In the wet bonding process virgin spun glass is bonded with a liquid, low molecular weight, water miscible resin. Higher grade thermal insulation for construction applications is made by this process.