ABSTRACT:The favored pH ranges for the formation of urons in urea-formaldehyde (UF) resins preparation were determined, these being at pH's higher than 6 and lower than 4 at which the equilibrium urons £ N,NЈ-dimethylol ureas are shifted in favor of the cyclic uron species. Shifting the pH slowly during the preparation from one favorable range to the other causes shift in the equilibrium and formation of a majority of methylol ureas species, whereas a rapid change in pH does not cause this to any great extent. UF resins in which uron constituted as much as 60% of the resin were prepared and the procedure to maximize the proportion of uron present at the end of the reaction is described. Uron was found to be present in these resins also as linked by methylene bridges to urea and other urons and also as methylol urons, the reactivity of the methylol group of this latter having been shown to be much lower than that of the same group in methylol ureas. Thermomechanical analysis (TMA) tests and tests on wood particleboard prepared with uron resins to which relatively small proportions of urea were added at the end of the reaction were capable of gelling and yielding bonds of considerable strength. Equally, mixing a uron-rich resin with a low F/U molar ratio UF resin yielded resins of greater strength than a simple UF of corresponding molar ratio indicating that UF resins of lower formaldehyde emission with still acceptable strength could be prepared with these resins.

The peculiar behaviour noticed in tannin-based and other wood adhesives when hexamethylenetetramine (hexamine) is used as a hardener is described. When using hexamine as hardener in tannin adhesives, a¯ow problem during hot curing of this adhesive/hardener system and its consequences on adhesive performance is identi®ed. The solution of the tannin/hexamine problem con®rmed and gave a clear theoretical justi®cation for the applied ®nding that under many application conditions hexamine is not a formaldehyde-yielding compound yielding extremely low formaldehyde emissions in bonded joints. 13 C NMR evidence is presented con®rming that the main decomposition (and recomposition) mechanism of hexamine is not directly due to formaldehyde but rather proceeds through now-identi®ed intermediates, i.e. mainly through the formation of reactive imines rather than methylene bases, possibly also forming a very slight amount of iminomethylene bases. This also con®rms that any species with strong real or nominal negative charge under alkaline conditions, be it a tannin, resorcinol or other highly reactive phenols, be it melamine or another highly reactive amine or amide, or an organic or inorganic anion, it is capable of reacting with the intermediate species formed by decomposition (or recomposition) of hexamine far more readily than formaldehyde explaining the capability of wood adhesives formulations based on hexamine to give bonded panels of extremely low formaldehyde emission. If no highly reactive species with strong real or nominal negative charge is present, then decomposition of hexamine proceeds rapidly to formaldehyde formation as reported in previous literature. The elucidation of the hexamine decomposition mechanism, which is presented, and a scanning electron microscopy (SEM) investigation also allowed to advance a reason for the without-curingformation of ambient temperature stiff gels in tannin/ hexamine glue mixes and to propose chemical structures for the ionic coordination linear polymers formed.

Iminoamino methylene base intermediates obtained by the decomposition of hexamethylenetetramine (hexamine) stabilized by the presence of strong anions such as SO 4 2Ϫ and HSO 4 Ϫ , or hexamine sulfate, were shown to markedly improve the water and weather resistance of hardened melamine-urea-formaldehyde (MUF) resins used as wood adhesives and of the wet internal bond strength performance of wood boards bonded with them. The effect was shown to be induced by very small amounts, between 1 and 5 wt % of this material on resin solid content. This strong effect allowed the use of MUF resins of much lower melamine content and also provided good performance of the bonded joints. Because the main effect was also present at the smaller proportion of hexamine as hexamine sulfate, it was not due at all to any increase in the molar ratio of the resin as a consequence of hexamine sulfate addition.

Thermomechanical analysis (TMA) tests on joints bonded with synthetic phenol-formaldehyde (PF) resins and ureaformaldehyde (UF) resins have shown that frequently the joint increase in modulus does not proceed in a single step but in two or more steps, yielding an increase of modulus first derivate curve presenting two (or more) rather than a single peak. This behaviour has been found to be due to the initial growth of the polycondensation polymer leading first to linear polymers of critical length for the formation of entanglement networks. Two or more modulus steps and two or more first derivate peaks then occur, and when two only occur the first is due to the formation of linear polymers entanglement networks and the second to covalent cross-linked networks. The faster is the reaction of a phenolic resin for any reason, such as decreasing water content of the resin, or the higher is the reactivity of a PF resin the earlier and at lower temperature the entanglement network occurs and the more important is its modulus value in relation to the final, cross-linked resin modulus. The accepted method of calculating the gel point and gel temperature of a polycondensation resin from the single peak of the first derivate curve is still acceptable in resins where the entanglement effect is small or is not present. In resin systems where the entanglement effect is instead of importance, the question of what is the gel point in such systems had to be addressed, and gel temperature and gel point must be obtained from the modulus and its first derivate curve in a different manner, which is presented. Similar findings are shown as regards UF resins and the influence of water and of wheat flour extender on their networking during curing. The formation and disappearance during thermosetting adhesives curing of entanglement networks has not been recognized before as a determining occurence in the mechanism of hardening of wood adhesives. These entanglement networks can contribute strongly, or can detract according to conditions, to both the green and the initial strength of a wood joint bonded with a thermosetting adhesive resin by altering considerably its rheology before and during hot-pressing. They can also contribute to the final strength and performance of the joint when the entanglement network still exists after resin hardening, as it is the case for wheat flour extension of a UF resin.Correspondence to: A. Pizzi bindung h/iufig nicht einstufig verl/iuft, sondern zwei oder mehr Stufen aufweist; dementsprechend zeigt die erste Ableitung der E-Modulkurve zwei oder mehr Peaks. Dieses Verhalten ergibt sich aus einer zun~ichst linearen Kondensation, bis die Polymere eine kritische L~inge erreichen, so daft sie sich gegenseitig beeinflussen und zu einem nicht kovalent gebundenen Netzwerk verzahnen. Zwei oder mehr Stufen erscheinen dabei in der E-Modulkurve; wobei die erste auf die physikalisch-chemische Beeinflussung, die zweite auf die kovalente Quervernetzung zur/ickzufiihren ist. Je schneller die Reaktion des PF-Harzes ist,...

Materials and methods Six urea-formaldehyde resins of molar ratios of U:F of respectively 1:1.1, 1:1.2, 1:1.3, 1:1.5, 1:1.8, and 1:2.0 were prepared as follows: To a urea/formaldehyde concentrate (formurea) containing 23% urea and 57% formaldehyde in water was added 22% NaOH to set the pH between 8.3-8.5. The temperature is raised in 50 minutes to 90 ~ maintaining the pH between 7.3-7.6 by small additions of 22% NaOH, and maintained at 90 ~ for approx 20 minutes. The pH decreases by itself and is maintained not lower than 4.7-5.1 and the temperature increased to 98 ~ The reaction is stopped when a water tolerance point of 170% is reached in the case of all resins, by cooling to 40 ~ the second urea added, the pH readjusted to 8.7 with NaOH, and the resin left to mature overnight at 40 ~ The ratio of second urea to initial urea for the preparations above was in the same order as above of 0.35, 0.36, 0.38, 0.41, 0.47 and 0.52 respectively. All the resins were at solids contents in the range 53.3%-54.9%. The resins were then tested by thermomechanical analysis (TMA) by procedures already reported (Pizzi, 1997, Kamoun et al 1998, Garcia and Pizzi 1998, after addition on resin solids of 2% NH4C1 as a 25% solution, by placing 30 mg of glue mix between two plys of beech wood to form a joint of 21 x 6 • 11.5 mm dimension which is tested in three points bending for a span of I8 mm, subjected to an alternating force of 0.1/0.5 N with a 6 s/6 s cycle, at a constant heating rate of !0 s/ minute, from 40 ~ to 250 ~ The minimum value of the deflection corresponding to the tighter network formed by the adhesive on curing is then measured. Duplicate one layer laboratory particleboard of 350 x 310 x 14 mm dimensions were then produced by adding 12% UF resin solids on dry wood particles(to make sure that results good enough to notice a difference would be obtained) pressed at a max pressure of 32 kg/ cm 2 followed by a descending pressure cycle, at 195 ~ and for a pressing time of 4.4 minutes (