Epoxy Resins Crosslinking under Microwave and/or Electron Beam Treatment

H. Iovu, I. Calinescu, E. Mateescu*, D. Martin* and S. Girea

University "POLITEHNICA" of Bucharest, Faculty of Industrial Chemistry, 149 Calea Victoriei, 71101-Bucharest, Romania
*National Institute for Laser, Plasma and Radiation Physics, Bucharest, Romania
[email protected]

Received: 28 August 2001 / Uploaded 30 August 2001
 

Introduction

The use of polymer composite materials in the mass production of industrial components is limited by the extent of the thermal process time due to the very low thermal conductivity of polymer materials. Besides, heat supply for thermoset polymer curing occurs from the components external surface by means of hot press or hot air and so leads to a heterogeneous thermal treatment. Thermal heating inside an oven classically activates the crosslinking of the epoxy resins and molds are resorted. Recently new techniques of activation based on the electromagnetic irradiation of polymer matrix have been developed.

In this paper we report data concerning the crosslinking process of the DGEBA epoxy resin type under microwave (MW) and/or electron beam (EB) treatment using triethylenetetramine (TETA) and/or phtalic anhydride (PA) as a curing agent.

Experimental

A commercial domestic microwave oven properly modified was used as microwave applicator. Mixtures (~25g) of DGEBA with variable amounts of curing agent were introduced into a cylindrical reactor made from glass, which was then placed into the reaction chamber. The curing process was registered as the temperature increases in time. The microwave applicator was a rectangular cavity of 245 mm x 245 mm x 450 mm. The microwave power is coupled via one of its sidewalls with a slotted rectangular waveguide (five inclined series slots cut in the broad wall of a WR430 waveguide and spaced 1/2 l_G apart). The electron beam from a linear accelerator, ALIN-10, of 6 MeV and 70 W maximum output power, is introduced perpendicularly by the upper-end of the rectangular cavity passing an aluminum window of 100 m m thick.

The FTIR spectra were registered on a BIO-RAD Spectrophotometer, using 32 scans and a resolution of 4 cm^-1. The solid samples were prepared by mixing of 0.5 mg of reaction product with 100 mg of KBr.

Results and Disscussion

The new architecture of the reaction chamber (RC) provides a high microwave energy transfer and a very good electrical field uniformity. Therefore the results obtained by microwave treatment in this chamber are better than those obtained in a modified domestic oven (MDO) (Fig. 1). The use of the RC applicator allows to study the effects of microwave and/or electron beam treatment on the curing process. Due to the small volume of the samples (~25 ml), the value of microwave power was 30 W and the total absorbed dose was about 20 kGy (with a dose rate of 1 kGy/min).

Fig.1: The influence of the microwave applicator type on the curing of the DGEBA epoxy resins.
 

In order to determine the absorbed power for the three irradiation treatments, samples without TETA were used. One may observe that the increasing of temperature is much higher in the case of microwave treatment than for the electron beam treatment. For using 7% TETA, the increasing of temperature was more significant due to the curing process. Also, by simultaneous treatment with electron beam and microwave power, the curing process occurs in a small time period (Fig. 2).
 

Fig.2: The influence of the curing agent concentration and the irradiation type on the curing process of the DGEBA epoxy resin.

Attempts to use other curing agents, DEA (diethylamine) and NNDA (N,N-diethylaniline) result in a product with a very low degree of crosslinking. The explanation for DEA and NNDA behavior relies to the molecular structure of these two curing agents. The nitrogen atoms in their molecules have less or even no hydrogen atoms linked, which could attack and open the epoxy rings. Thus a higher temperature is needed to develop the crosslinking process. This temperature is reached after a long time during which almost all the solvent has disappeared. The lack of solvent strongly hinders the crosslinking process of the epoxy resin.

The concentration of the curing agent strongly influences the crosslinking process of the resin both in classical mode and under microwaves (Fig. 3).

One may observe that the temperature of the sample with 3.5%TETA is higher than for that with 7%TETA under an electrical power value of 215W (Fig. 3, Part A). This strange aspect may be explained on the basis of the lower absorption power of TETA in microwaves, which as a result will increase the time to reach the crosslinking temperature. Therefore the quantity of solvent eliminated is higher. And, as we mentioned earlier, a small quantity of solvent in the system will lead to an unsuccessful crosslinking process, so that the temperature will exhibit lower values. The same aspect was observed for an electrical power of 40W, the temperature of the sample with 14%TETA being less than for 7%TETA (Fig. 3, Part B). Thus, the main aspect pointed out by these experiments is the existence of an adequate value of TETA concentration for each electrical power value. This adequate value is higher as the electrical power is lower.
 

A. o – 0%, · - 1%, D - 3.5%, š - 7%.

B. o – 0%, D - 3.5%, š - 7%, x – 10%, · - 14%.

Fig. 3. The influence of the TETA concentration on the dependence of temperature against time for DGEBA crosslinking process under microwaves at a fixed electrical power: A - 215W, B – 40W.
 
 

Tests of crosslinking of the epoxy resins using PA as a crosslinking agent were also performed. In these experiments TETA was used in small quantities as an accelerator. The reaction conditions are listed in Table 1.
 
 

Fig. 4. The dependence of the temperature on the crosslinking time for DGEBA crosslinking process with PA. Cond. [PA]=40% against DGEBA, [TETA]=1% against DGEBA, P=400 W
 
 

The crosslinking process may occur following two main routes:

I. DGEBA crosslinking with TETA following the well known mechanism:

In the IR Spectra of the final crosslinking products the OH band is not present at all which means that the formed OH groups react further on with the PA and open the anhydre ring:

II. DGEBA crosslinking with PA after the opening of the anhydre ring by TETA (NR^’₂H)

From Fig. 1 one may observe that the temperature rapidly increases up to a reaction time of ~4 min which may be explained by the reactions (1) and (2). The reaction time of ~4 min may correspond to the entire consumption of TETA from the reaction mass and therefore from this moment the crosslinking process may occur only by the reactions (3) and (5) which are slower. As a consequence the reaction temperature increases slower.