The Influence of Curing Temperature on Rheological Properties of Epoxy Adhesives

The curing process of three different epoxy adhesives, which are appropriate for bonding wood with metal, was characterized on the basis of their rheology. The rheological measurements were carried out using a TA Instruments ARES G2 stress control rheometer. The infl uence of temperature on the curing process was examined at fi ve different temperatures: 30, 40, 50, 60 and 80 °C. Curing was monitored by an oscillatory test by using geometry with disposable parallel plates. Gel time and vitrifi cation time were determined. Signifi cant differences in curing behaviour were observed among the studied adhesives. The results showed that increasing the curing temperature signifi cantly accelerated the curing process of the epoxy adhesives. It was also observed that the storage modulus G’ and the loss modulus G” decreased with an increasing temperature of curing.


INTRODUCTION 1. UVOD
Epoxy resins are often used as high performance thermosetting adhesives for bonding metals together or for binding metals to other materials.These adhesives are frequently used for bonding steel and wood, as for example in the case of structural connections in timber (Broughton and Hutchinson, 2001;Custodio et al, 2009;Chans et al, 2010;Widman et al, 2007;Serrano, 2001;Cimadevila et al, 2007).Over the last 30 years, there has been a growing interest in the possibilities for jointing glulam beams with steel bars (rods), which are placed in pre-drilled holes in the timber members and bonded with epoxy resin.Such joints made with gluedin rods have great potential in the fi eld of timber engi-
The second adhesive was EPOX 210 A+B (EPOX 210), which was a two-component-epoxy resin with component A (colourless) and B (yellow).The viscosity of component A at 25 °C was 500-900 mPa•s, whereas that of component B was 250 mPa•s.The viscosity of the mixture after 5 min was 600 mPa•s.Pot life at 20 °C was 30 min for 100 g and 11 min for 200 g.Component A contains 3-aminomethyl-3,5,5-threemetyl-cyclohexylamine, Benzyl alcohol, m-ksililendiamin, and Nonylphenol.Component B contains an epoxy resin, C 12-C14-Alkylglycidether.Both components were mixed in advance.The recommended mixture ratio was A:B = 2:1.

Rheological measurements 2.2. Reološka mjerenja
The rheological measurements were carried out using a TA Instruments ARES G2 stress control rheometer.A diagram of the equipment for rheological measurements is shown in Figure 1.The geometry was plateplate, with disposable plates with a diameter of 25 mm made from steel with a nominal ultimate tensile strength of: f u = 900 -1000 N/mm 2 .All experiments were performed at a frequency of 1 Hz, a strain of 0.01, and a 0.5 mm gap between the plates.Sinusoidal stress was used in all experiments, which is the basis for calculating the storage modulus (G') and the loss modulus (G").
All of the adhesives were used as received and carefully mixed at the stoichiometric ratio prescribed by the manufacturer.After preparing the adhesive sample, it was immediately placed in the plate-plate geometry assembly, and quickly heated (60 °C/min) to the target curing temperature.Measurements always began neering.Some of their advantages are the aesthetic benefi ts, with the hidden design of these rigid joints, and their high load-carrying capacity for the transmission of loads from and to the wooden pieces (Chans et al, 2009).Glued-in rods are very effi cient joints, which can withstand the high axial loads that are applied to timber structural members.If such joints are correctly designed it is theoretically possible to achieve higher stiffness and a more uniform stress distribution in the wood, while avoiding peak values near the bolts and achieving a better aesthetic appearance in comparison with traditional mechanical connections (Gattesco and Gubana, 2001).
An important condition for the use of glued-in rods is that an effi cient bond be created between the steel rods and timber.Gluing and screwing are two ways in which this can be done.Rod diameters larger than 10 mm make it impossible to achieve an effi cient working screw system (Bernasconi, 2001).The load transfer between the screw and timber is mainly governed by compression due to indentation.Compared to screws, glued-in rods introduce shear forces into the timber in a different way.Whereas anchored screws rely on direct contact between the timber and the fl anks of the thread, and therefore on compression by means of complex geometry, glued-in rods transfer the force only at the interface between the rod and the adhesive.The load transfer between the adhesive and timber is mainly governed by shear and not by compression due to indentation (Steiger et al, 2006).Due to this fact, gluing is the only way in which the bond can be created.Several parameters have an infl uence on the shortterm pull-out strength of glued-in rod joints.One of the very important mechanical parameters is the type and rheological behaviour of the adhesive (Rossignon and Espion, 2008).The purpose of the adhesive is to provide a continuous bond between the timber and steel rod, to fi ll voids and cavities, and to transfer and sustain the loads.Thus thermosetting gap-fi lling materials are required, which exhibit good adhesion to the various materials whilst also being tolerant to variations in the timber moisture content.Epoxy compounds have several advantages over other resins as adhesive agents for use in timber structures and, typically, they are suitable for service temperatures in the range -30 to +60°C.They may be formulated in a wide range of forms, application characteristics and mechanical properties when cured (Broughton and Hutchinson, 2001).
Before investigating the ability of different adhesives to promote adhesion in different types of joints, a detailed description of the rheological properties of each individual adhesive is needed (Lapique and Redford, 2002).Rheological information can be related directly to the mechanical properties of the adhesive (Zheng, 2002).Rheometry can be used not only to obtain information about gelation, but also to monitor the whole curing process of a thermoset (Núñez et al, 2005;Winter, 2008).The purpose of this study was to investigate the curing process and explore the rheology of three different epoxy adhesives, cured at several different temperatures.
immediately after achieving the target temperature.Five curing temperatures were used for each individual adhesive: 30, 40, 50, 60 and 80 °C.Additionally, cooling from 80 °C to 30 °C was used in order to determine the cooling effect on the rheological properties of the cured EPOX 210 epoxy resins.
During the measurements, a control program was used to compensate for the thermal expansion of the test specimen dimensional changes during curing by maintaining a controlled normal force, which enabled the characterization of the cure over the whole conversion range.All the test specimens were prepared in the laboratory at 20 °C and at a relative humidity of 65 %.

Rheological analysis Reološka analiza
The rheological analysis used during the oscillatory test provided a total curing profi le for the studied epoxy adhesives.Information about the curing behaviour of the adhesives was gained in terms of the measured storage modulus G′ and the loss modulus G″.The curing profi les of the investigated epoxy adhesives at room temperature (30 °C) are shown in Figure 2. It can be observed, in the case of all the studied adhesives, that at the start of the curing, G" is greater than G' because the systems show a liquid type of behaviour.Then, both G' and G" increase as the cross-linking reaction progresses to a crossover point; here the systems present both elastic and viscous behaviour, storing a similar amount of the dissipated energy.In this zone gelation occurs.Later, the curing reaction progresses, and G' approached its maximum values whereas G" reached a peak and then started to decrease.In this zone vitrifi cation occurred.
Gelation and vitrifi cation of the adhesive can be detected through changes in the dynamical mechanical properties.Several different criteria can be used to defi ne the gelation and vitrifi cation times.The most generally accepted criterion for gelation is the crossover point of the shear storage modulus and loss modulus.However some authors consider that, in the case of many systems, the gel point is not necessarily equal to the calculated value at G' = G" (Winter, 1987;Núñez et al, 2005).Another criterion is that gelation occurs when viscosity increases exponentially to infi nity.The problem is that infi nity viscosity is very diffi cult to detect.One of the methods to identify the gel point is the criterion of the tangent line to the G' curve; the gel time is detected by the crossover between the tangent line at G' when this curve reaches a value close to 100 kPa•s and a baseline G' = 0 (Laza et al, 1998).
There is no generally accepted criterion for the detection of vitrifi cation.In the case of the investigated epoxy resins, it was found that rheological vitrifi cation is a gradual process that extends over a wide temperature range, depending on the criterion used.Vitrifi cation can be determined from: the maximum tan δ peak at 1 Hz, the maximum G" peak at 1 Hz, the onset of frequency dependence in G', and with the end of frequency dependence in G' (Zheng, 2002).In this study, gel time (t gel ) was taken as the time corresponding to the crossover of G'/G".Vitrifi cation time (t vit ) was determined as the time corresponding to the time at which G" has a peak at 1 Hz.According to the so determined gel time and the vitrifi cation time for curing adhesives at room temperature, the epoxy system 7211 cured fastest (t gel =238 min, t vit =297 min), followed by EPOX 210 (t gel =345 min, t vit =527 min), and EP 101 (t gel =499 min, t vit =565 min).

The effect of temperature on curing profi le of adhesives 3.2. Utjecaj temperature na profi le otvrdnjavanja ljepila
In order to characterize the infl uence of temperature on the curing process (i.e. on the storage and loss moduli, respectively), four additional higher temperatures (as well as 30 °C) were used for curing: 40, 50, 60, and 80 °C.It was observed (Fig. 3) that increasing the temperature accelerated the curing reaction of the epoxy adhesive 7211, but dramatically diminished the maximum values of G' and G".Also, time for transitions (i.e.gelation and vitrifi cation) decreased with the increasing curing temperature.This was expected because the reaction of the adhesives was thermally accelerated.Similar behaviour was observed in the case of the other two adhesives (Fig. 4 and Fig. 5), but at different time periods.
The gel times and vitrifi cation times for all three epoxy adhesives (Table 1), studied at different temperatures, were determined according to the above described criteria.It can be seen that gel time shortened with higher temperatures.This can be explained by the fact that when the temperature increases molecular mobility also increases, so the curing reaction occurs more rapidly.The vitrifi cation times also shortened with higher temperatures, with some exceptions (EP 101 and EPOX 210 at 60 and 80 °C).The infl uence of temperature on the curing process can be explained by the nature of epoxy adhesives.The epoxy adhesive, as a thermoset polymer, is a material whose properties are dependent not only on the test environment but also on its thermal and environmental history.A thermoset adhesive has a glass transition temperature (T g ) (Lapique and Redford, 2002).Glass transition is a phase change of amorphous solids, such as glasses and polymers.A non-crystalline material is converted to a relatively hard, elastic and glassy state from a soft, elastic-plastic and rubbery state when cooled through its T g (Li, 2000).T g is infl uenced by different factors such as composition of the resin molecule, crosslink density, the polar nature of the functional groups of resin molecules, the curing agent or catalyst, curing time and curing temperature (Tamulewich and Moore, 1980).Basically polymers with longer molecular chains, more hydrogen bonds, polar or polarizable groups and crosslinking networks tend to have a higher T g .On the other hand materials with symmetric (non-polarized) molecular structures or with a plasticizer (softener) tend to have a lower T g (Li, 2000).
The glass transition temperature is a direct measurement of molecular mobility.Below T g , a molecule is confi ned to its site, with a very limited group of branch movement freedom, and its free volume is relatively small.Whole molecules cannot move away from each other.Above T g , a molecule has much more freedom of movement, and its free volume increases faster with the temperature.Whole molecules can shift or slide away from each other (Li, 2000).
Other mechanical properties vary in accordance with these changes in the degree of cure and plastifi cation.The temperature at which they are cured determines many of the properties of epoxy adhesives.T g can be a measure of the degree of cure of an epoxy resin system.Therefore the cure time and temperature can have a very dramatic effect upon the T g of any given epoxy system.At temperatures close to T g , the values of the mechanical properties such as tensile strength, lap shear and adhesive bond shear and modulus are substantially reduced.This can be seen in Fig. 4; where the values of both moduli (G', G") diminish at higher temperatures.Above a temperature level of ~50 °C both moduli were drastically reduced, because the cure temperature was near the glass transition temperature.At higher temperatures the curing times were diminished, too.At 30 °C the maximum value of G' was reached after ~1000 min, whereas at 80 °C it was reached after only ~60 min, which is almost 17× less than at 30 °C.

The effect of temperature on G' and G" values of the cured adhesive
The effect of temperature on the G' and G" values corresponding to a cured adhesive was studied in detail with the EPOX 201 adhesive (Fig. 6).Measurements began after the target curing temperature (80 °C)  had been reached.When the maximum value of G' had been reached, the heating stopped.Cooling was then started from 80 °C to 30 °C with a temperature ramp of 30 °C/min.It can be seen that, after cooling, G' increased to a value that is practically the same as if the curing process had occurred at 30 °C.As mentioned above, the reason for such behaviour lies in the glass transition temperature.In the fi rst part of the experiment, the curing temperature was near T g ; which explains why both moduli were so low.After cooling, both moduli became increasingly higher.

ZAKLJUČCI
The rheological properties of three different epoxy adhesives were investigated at fi ve curing temperatures.Gelation and vitrifi cation were detected through changes in the dynamic mechanical properties for each individual adhesive.According to the determined gel times and vitrifi cation times for the curing of the adhesives at room temperature, the epoxy system 7211 cured fastest (t gel =238 min, t vit =297 min), followed by EPOX 210 (t gel =345 min, t vit =527 min) and EP 101 (t gel =499 min, t vit =565 min).The results showed that the curing temperature had a signifi cant infl uence on the rheology of the investigated epoxy adhesives.It was observed that increased temperatures accelerated the curing reaction of all of the studied epoxy adhesives, but dramatically diminished the maximum achieved values of G' and G".The time for the transitions such as gelation and vitrification decreased with increased curing temperatures.The effect of temperature on the G' and G" values of the cured adhesive was clear: cooling of the cured adhesive from 80 °C to 30 °C increased the G' and G" values several times, which was associated with the glass transition behaviour of the adhesive.

Figure 6 6 .
Figure 6The infl uence of heating and immediate cooling on G' and G" values of the EPOX 210 adhesive Slika 6. Utjecaj grijanja i trenutačnog hlađenja na vrijednosti G' i G" ljepila EPOX 210