The Effect of Curing Agent Content on Photodegradation of Epoxy Coating Studied by Positron Annihilation

The photodegradation progress of epoxy cured with polyamide and the effect of the curing agent content under UV-A irradiation have been investigated using positron annihilation spectroscopy with an energy tunable positron beam. After 88 h of irradiation, a post-cure process and the generation of carbonyl groups reduce the value of the S parameter, compared with the virgin samples. As the irradiation time increases from 208 h to 399 h, the S parameter decreases, which may be due to the growth of carbonyl groups and the generation of free radicals. After 543 h of irradiation, a dead layer with very low S value appears near the sample surface probably induced by a dramatic decrease in the Ps formation probability. The positron results also reveal that epoxy cured with an appropriate amount of polyamide has a smaller dead layer suggesting that the amount of curing agent is a key factor affecting the photodegradation of epoxy resin.


Introduction
Epoxy resin has been employed in many industrial applications such as electrical engineering, aeronautics, and long-term protective coating systems because of its excellent electrical insulation, adhesive properties and chemical stability.During the service lifetime of epoxy materials, photo-oxidation is one of the prime degradation processes that lead to a loss of the functional properties.Hence, it is of importance to study the degradation behaviors of the epoxy resins during the photo-degradation progress.
Many attempts have been made to investigate the characterization of epoxy resins during UV degradation.Rivaton et al. [1,2] have studied the influence of the radiation wavelength in the photoaging progress of epoxy resin.The results showed that the main oxidation products are phenyl formate end-groups under long wavelength radiation, whereas, no phenyl formated structures were found under short wavelength exposure.Besides, the photo-aging of epoxy resins cured with different kinds of curing agent was also investigated.Ollier-Dureault and Gosse [3] studied the photo-oxidation process of epoxy resin cured with two different anhydride curing agents, hexahydrophthalic anhydride and methyl tetrahydrophthalic anhydride.Monney et al. [4][5][6] also investigated the photochemical degradation of an anhydride-epoxy system using different techniques.The changes of the chemical groups and the C/O ratio were given by Fourier transform infrared spectrometer (FTIR) and electron beam X-ray microanalysis, respectively [4,5].Photo-oxidation of amine cured epoxy system was also investigated [7].Bellenger and Verdu studied the influence of three different diamines cured epoxy system on photo-oxidation aging [7].A comparison of these results shows that the photoinitiating species essentially derive from the phenoxy part, whereas the propagation principally depends on the amine concentration and on the electron density at the nitrogen atom.The amount of curing agent is another factor that affect the process of the photo-oxidation.However, to date, there have been few reported studies in this area.
In this study, the photo-degradation characteristics of a polyamide-epoxy system during UV irradiation were investigated by positron annihilation radiation Doppler broadening spectroscopy (DBS), which is sensitive to the micro-structural evolution during aging of polymers [8][9][10].The effect of the amount of curing agent on the variation of chemical groups during UV irradiation was investigated by FTIR measurements.

Experiments
Diglycidyl ether of bisphenol-A epoxy resin (DGEBA) E51 (Jiangsu Sanmu Group Corporation) and low molecular polyamide resin (LMPAR) 651 (Yueyang Zhongzhan Science & Technology Co., Ltd.) were used in the present work.The epoxy value of DGEBA is about 0.51 mol/100 g.DGEBA and different amounts of LMPAR were mixed for preparation of a cross-linked DGEBA/LMPAR coating system.The mole ratio of polyamide to epoxy was 0.221, 0.318, 0.415, 0.515, and 0.613, respectively.The samples were applied on carbon steel substrates and then cured in an oven at 80 • C for 3 h.
Ultraviolet (UV) treatment of polyamide-cured DGEBA was performed in a commercial UV weathering test Chamber (XR-UV).UV-A lamps (center wavelength 340 nm) were used in the degradation process.The UV radiation power density was 55 mW • cm −2 and the temperature was 60 • C.
All epoxy coatings were exposed to UV radiation for 0 h, 88 h, 208 h, 303 h, 399 h, and 543 h, respectively.
An attenuated total reflectance Fourier transform infrared spectrometer (ATR-FTIR, Nicolet iS10) was applied to investigate the chemical structure changes of the coating system.A resolution of 4 cm −1 and data spacing of 1.929 cm −1 were used for data acquisition.Each spectrum was taken from 550 cm −1 to 4000 cm −1 .A conventional line shape parameter, S, obtained with energy variable Doppler broadening spectroscopy (DBS, Wuhan University) was employed to characterize the aging behavior of epoxy under UV-A irradiation.The slow positrons were emitted from a 1.85 GBq 22 Na source.The positrons were moderated by a thin solid Ne layer, and then electromagnetically transported to the sample.Annihilation gamma rays were detected by a high-purity Ge solid-state detector, and spectra were obtained over a range of positron energy from 0 keV to 25 keV.The total number of counts in each spectrum was 2 × 10 6 with a counting rate of 1200 cps.The positron annihilation line shape S parameters from DBS were calculated as the ratio of the central area (510.24keV < E < 511.76 keV) to the total area of the positron annihilation γ peak after background subtraction.

Results and Discussion
The S parameters of epoxy coating (the mole ratio of polyamide to epoxy is 0.415) before and after different periods of irradiation as a function of incident positron energy E are shown in Fig. 1.The mean implementation depth of the positron corresponding to the incident energy is shown on the top horizontal axis of Fig. 1, which is calculated by the established Eq. ( 1) [9], In the equation, Z m is the mean implementation depth (nm), ρ is the density of the material (g • cm −3 ), and E is the incident energy (keV).In this study, the density of the epoxy system compos- ites is around 1.16 g • cm −3 .
For the unirradiated sample, the S value increases with increasing incident positron energy until 2.5 keV, and then reaches a constant value at higher energies.Similar results have been obtained by previous studies [10,11].The polymer surface is much softer than the bulk and contains some large holes, which may lead to different positron annihilation characteristics in the surface area compared with the bulk.
After 88 h of irradiation, the variation trend is very similar to that of the virgin sample.However, compare to the virgin sample, the values of the S parameter are lower over the whole incident energy range.The variation of chemical composition may influence Ps formation and thus affect the value of the S parameter.Figure 2 shows the ATR-FTIR spectra of the DGEBA/LMPAR epoxy system (mole ratio of polyamide to epoxy is 0.415) before and after UV-A radiation for different times.As can Fig. 2. ATR-FTIR spectra of DGEBA/LMPAR epoxy system (mole ratio of polyamide to epoxy is 0.415) before and after UV-A radiation for different time.
be seen, after 88 h of UV irradiation, the characteristic peak of −C=O in saturated aldehyde, ketone or acid appears near 1720 cm −1 .The generation of these polar groups can inhibit the formation of Ps and thus further decrease the S values [12][13][14].In addition, the intensity of the unobvious peak at 940 cm −1 , associated with the residual epoxy rings, decreases after 88 h of irradiation.The consumption of epoxy groups indicates that the curing reaction proceeds during the UV irradiation, which may lead to the formation of a higher cross-linked structure.The loss of free volume can also reduce the S parameter.
After irradiation for 208 h, a dramatic decrease in S parameter was observed near the coating surface.The low S value increases with increasing positron energy E and then (E ≥ 12 keV) maintains a nearly constant value.At this irradiation time, more carbonyl groups in saturated aldehyde, ketone or acid were found (see Fig. 2).The growth of these polar groups can undoubtedly reduce the S parameter in larger scale.Besides, degradation reactions in UV-irradiated epoxy often involve formation of hydroxyl radicals [2,15].Though most radicals may be consumed by O 2 in the storing period [16], residual radicals can decrease the probability of Ps formation and thus reduce the value of the S parameter.All these factors result in relatively low values of S near the surface.After irradiation for 303 h and 399 h respectively, the S parameter shows a very similar variation with positron energy, however, the value of the S parameter decreases with increasing irradiation.
After 543 h of continuous irradiation, the S parameter is low and remains nearly unchanged with increasing incident energy near the coating surface (below 2.25 keV).After that the value of S parameter increases gradually with increasing incident energy.The region in which the S parameter remains nearly constant has been identified as the dead surface layer after UV irradiation [10].With increasing irradiation time, polar carbonyl groups are generated not only at the coating surface but also towards the bulk.Therefore, the low S value remains unchanged in a certain depth (in the energies lower than 2.25 keV).Additionally, a novel type of polymer network with a highly cross-linked structure may be formed by recombination of the radicals generated during UV irradiation.The molecular weight between crosslinks estimated in such a surface layer is about one order of magnitude higher than that of the virgin sample [10].The loss of free volume also reduces the S value of the surface dead layer.Because of all these factors, there is nearly no Ps formation near the sample surface and thus the low S parameter remains nearly unchanged in the dead layer.
The effect of the curing agent content on the progress of UV degradation was also investigated.Figure 3 shows the variation of S parameter versus E for the DGEBA/LMPAR epoxy system (mole ratio of polyamide to epoxy is 0.221, 0.318, 0.415, 0.515, and 0.613, respectively) after 543 h of UV-A irradiation.As can be seen, all the samples generate dead layers near their surface after 543 h of UV irradiation.We note that the depth of the dead layer is different for samples with different initial polyamide/epoxy mole ratio.Figure 4 presents the variation of the dead layer depth as a function of polyamide/epoxy mole ratio after 543 h of UV-A irradiation.When the initial polyamide/epoxy mole ratio is 0.221, the depth of the dead layer is more than 600 nm.Increasing the initial polyamide/epoxy mole ratio, the depth of the dead layer decreases at first, and reaches a minimum value at the mole ratio of 0.415.The thickness of dead layer then increases gradually with further increase of the polyamide mole ratio.For each sample, after long-term UV irradiation, S showed a minium value in the surface dead layer, which is mainly induced by the high local concentration of carbonyl groups and hydroxyl radicals as well as the high cross-linked structures.All these physical-chemical changes are closely related with the degradation progress under UV irradiation.From this point of view, samples with a polyamide/epoxy mole ratio of 0.415 and 0.515 show a relatively narrow range of severely aged area after 543 h of UV irradiation.However, the samples cured with more or less amount of polyamide show a much wider range of dead layer.All these results may suggest that the epoxy samples cured with the appropriate amount of curing agent have relatively better anti-aging performance to UV irradiation.

Conclusion
The photo-degradation behaviors of epoxy-polyamide system have been investigated during UV-A irradiation.During the early stage of irradiation, a reduction in the S parameter induced by a post-cure process and formation of carbonyl groups is found.With increasing irradiation time, more carbonyl groups and radicals are generated and thus reduce the value of the S parameter.After 543 h of UV irradiation, a dead layer near the sample surface appears, which is associated with a high local concentration of carbonyl groups and hydroxyl radicals as well as highly cross-linked structures.The influence of the amount of curing agent after a relatively long period of UV irradiation has also been investigated.The results showed that epoxy cured with an appropriate amount of curing agent show a relatively narrow range of severe aged region after long-term UV-A irradiation.

Fig. 1 .
Fig. 1. S parameter versus positron energy E for DGEBA/LMPAR epoxy system (mole ratio of polyamide to epoxy is 0.415) exposed to UV-A irradiation for different times.

Fig. 4 .
Fig. 4. Variation of the dead layer depth as a function of polyamide/epoxy mole ratio after 543 h of UV-A irradiation.