Macromolecular products
CI mainly reflected the accumulation of carbonyl products on the macromolecular chains in solid state. CI calculated from ATR spectra showed the local oxidation degree on the surface of the film, while CI calculated from transmission spectra showed the average oxidation degree in the bulk polymer.
The CI of two PPs calculated from ATR spectra and transmission spectra are shown in Fig. 2. On the surface (Fig. 2a), CPP had an induction period of at least 96 h, while UPP was oxidized from the very beginning of aging. After the induction period, CPP was oxidized at a similar rate as UPP before 288 h. After that, the increase of CI in UPP slowed down. Two PPs seemed to reach the similar oxidation degree after 480 h. In the bulk (Fig. 2b), UPP was also oxidized from the very beginning of aging, while CPP exhibited an induction period of at least 144 h. After the induction period, the oxidation rate of CPP was similar to that of UPP before 288 h. Then, the CI of UPP decreased.
The decrease of the CI in Fig. 2b seemed unreasonable, since the oxidation of PP was an accumulation process of the oxidized species. Therefore, some highly oxidized species might be lost. Considering that UPP was easier to be oxidized than CPP, it reached a high oxidation degree first. At the late stage of aging (after 288 h), severely oxidized fragments might drop from the surface and cause the decrease of CI [25–27]. The change of CI of UPP in Fig. 2a supported the speculation. The surface oxidation degree of UPP after aging for 288 h also showed a lower increasing rate than that of CPP. In this case, although the aging degree of UPP was evidently higher than that of CPP, the CI of UPP was close to or even lower than that of CPP after 288 h. Thus, the biased evaluation of the aging degree was obtained.
In addition, different sensitivities of ATR spectra and transmission spectra were observed. Despite the same difference between the oxidation degrees of two PPs after the same aging time, the differences between CI of two PPs were not the same from ATR spectra and transmission spectra. For instance, after aging for 288 h, the CI from ATR spectra of UPP was about 1.4 times of that of CPP, while the CI from transmission spectra of UPP was about 4.2 times of that of CPP. Obviously, transmission spectra were more sensitive to the different oxidation degrees.
Volatile degradation products
Volatile degradation products were absorbed in PP film as liquid. They could be desorbed through flash evaporation and detected by Py-GC/MS. Four typical volatile products, i.e. two alkenes, acetic acid and a lactone, were identified in the flash evaporation-gas chromatograms of two PPs. The peak areas of these products with aging time are shown in Fig. 3. In Figs. 3a and 3b, the alkenes were generated from the very beginning of aging in UPP, while later than 96 h in CPP. After a period of time, i.e. 96 h in UPP and 288 h in CPP, the peak areas of two alkenes turned to decreasing, corresponding to the increasing of two oxidized products in Figs. 3c and 3d at the same time. This indicated that: (1) The formation of alkenes was prior to the formation of oxidized products, in other words, the chain scission of PP happened before the formation of volatile oxidized products; (2) The oxidized products might be from the further oxidation of alkenes.
The peak areas of volatile oxidized products were expected to keep increasing with aging time due to the continuous oxidation of polymer chains. In CPP, they kept increasing indeed. In UPP, however, the peak areas stopped increasing and even began to decrease from 288 h. This was caused by the evaporation of these products and then condensation on the tube wall. There were visible droplets on the tube wall in which UPP was aged for 384 and 480 h. The droplets were washed by ethanol and the obtained solution was analyzed by Py-GC/MS. The flash evaporation-gas chromatogram of the droplets was compared with the result of the corresponding film (Fig. 4). Acetic acid and the lactone were identified in both the droplets and the corresponding film. Once the volatile oxidized products were evaporated and condensed on the tube wall, they could not come back to the film, so the peak areas in UPP were decreased obviously (Figs. 3c and 3d).
For CPP, no visible droplets were observed even after aging for 480 h, due to the relative low concentration of volatile oxidized products. There was a maximal concentration of volatile oxidized products that could be retained in the film. If the concentration did not exceed the maximum, the volatile oxidized products would accumulate in the solid polymer. When the concentration exceeded the maximum, the excessive parts began to be desorbed and some condensed on the tube wall. In this case, the biased evaluation of the aging degree of two PPs according to volatile degradation products would also be obtained as according to macromolecular products. Similar phenomenon can be expected if the polymer is photo-oxidized in unsealed atmosphere like in practical use.
Gaseous degradation products
Gaseous degradation products included H2, CO, CO2, alkanes and alkenes. They were in gaseous state and could be identified directly by using GC measurement. The peak areas of four typical gaseous products with aging time are shown in Fig. 5. These products were generated from 48 h in UPP and from 144 h in CPP respectively. After that, the peak areas kept increasing in both PPs and the peak areas in UPP were higher than in CPP all along, showing the higher aging degree of UPP without the protection of stabilizers.
Overview of photo-oxidative products in three states
Induction periods of the generation of photo-oxidative products in three states were extracted from Figs. 2, 3 and 5, listed in Table 1. In CPP, the induction periods of carbonyl products and volatile alkenes were the same, shorter than acetic acid, the lactone and gaseous degradation products, indicating the earlier generation of oxidation species in macromolecular chains and volatile alkenes. Thereafter volatile alkenes might be further oxidized and the oxidized main chains fractured to generate volatile oxidized products and gaseous degradation products. As shown in Figs. 3c, 3d and 5, the peak areas of acetic acid, the lactone and gaseous degradation products were increased rapidly from 288 h. At the same time, the peak areas of volatile alkenes were decreased rapidly (Figs. 3a and 3b). This phenomenon suggested the further oxidation of volatile alkenes and the corresponding accumulation of volatile oxidized products and gaseous degradation products.
Table 1
Induction period of generation of photo-oxidative products in CPP and UPP
Products | Induction period in CPP/h | Induction period in UPP/h |
CI on the surface (solid) | 96 | 0 |
CI in the bulk (solid) | 96 | 0 |
Alkenes (liquid) | 96 | 0 |
Acetic acid (liquid) | 144 | 48 |
Lactone (liquid) | 144 | 48 |
H2, CO, CH4, C2H4 (gas) | 144 | 48 |
Compared with CPP, UPP exhibited much shorter induction periods. Without the protection of stabilizers, oxidation species in macromolecular chains and volatile alkenes were generated from the very beginning of aging, followed by volatile oxidized products and gaseous degradation products within 48 h. As shown in Figs. 2b, 3c and 3d, CI and the peak areas of volatile oxidized products were decreased at the late stage of aging, due to the loss of severely oxidized species in solid state and the migration and condensation of the liquid products. The peak areas of gaseous degradation products kept increasing for as long as 480 h.
Conversion among photo-oxidative products in three states is illustrated in Fig. 6. During the photo-oxidation of PP, the main chains were first oxidized and the oxidized species remained in the film as solid products. In the meanwhile, the main chains fractured to generate volatile alkenes, absorbed by the film as liquid. Then the oxidized main chains fractured along with the further oxidation of volatile alkenes, to generate volatile oxidized products absorbed by the film as liquid and gaseous degradation products in the atmosphere around the film. The conversion took place throughout the aging process. When the conversion was severe as in UPP, the CI and the concentrations of volatile products were increased first and then decreased, due to the loss of severely oxidized species in solid state and the migration and condensation of liquid droplets in the late stage, despite the continuous aging of the polymer. In CPP, the similar conversion was observed, although the aging process was retarded owing to the presence of stabilizers. Therefore, the products in a single state, especially only in solid state, could not reflect the comprehensive aging process and might lead to biased results. It was more reliable to consider the photo-oxidative products in three states as a whole when evaluating the aging degree and relative stability of polymers.