3.1. Color Changes
Figure 3 displays the average
L*,
a*, and
b* values for P/L, P/L-6, and P/L-8 before and after accelerated aging, while
Table 2 provides the real color of samples and color differences of samples before and after accelerated aging. Notably, in
Figure 3 and
Table 2, the
L* coordinate presents an overall slight descent trend, indicating that the accelerated aging caused a darkening of P/L. With increasing accelerated aging time, the
L* coordinate barely changed first and declined rapidly afterwards. Specifically, the
L* of P/L remained almost unchanged when accelerated aging was within 15 d and declined rapidly when accelerated aging was beyond 15 d. For P/L-6 and P/L-8, the
L* coordinates varied little under accelerated aging. Arguably, the brightness (
L* coordinate) levels of P/L-6 and P/L-8 were scarcely influenced by accelerated aging, implying that the addition of UV-326 and UV-328 into SLS parts significantly improved their brightness stability under the sun and rain.
The variations of the
a* coordinate represent the change in the redness (+) and greenness (−). In other words, when the Δ
a* is positive, the redness increases; otherwise, the greenness improves. As shown in
Figure 3 and
Table 2, the
a* coordinate of P/L decreased under accelerated aging, indicating that the greenness improved when the SLS parts were exposed to sun and rain. The decline occurred when the accelerated aging was within 5 d and the range of change of the
a* coordinate was small after accelerated aging went beyond 5 d. Specifically, the decrease in the
a* coordinate happened almost exclusively in the early stage of the accelerated aging. The
a* coordinate change trend for P/L-6 was similar to that of P/L, which decreased rapidly in the early accelerated aging time of 5 d and remained almost constant in the later stage of accelerated aging. Nevertheless, the decline in the
a* coordinate of P/L-6 was significantly less than that of P/L, suggesting that the improvement of greenness of SLS parts can be attributed to UV-326. The
a* coordinate of P/L-8 also declined under accelerated aging, which is similar to change in P/L-6, with the only difference being that the
a* coordinate of P/L-8 decreased rapidly in the first 10 d of accelerated aging and then remained almost constant after 10 d of accelerated aging. According to comprehensive comparisons, the reduction levels of the
a* coordinates of SLS parts were as follows: P/L > P/L-8 > P/L-6. Therefore, it can be concluded that UV-326 and UV-328 had positive effects on preventing the color of the SLS parts from turning to green, and the effect of UV-326 was better than UV-328.
The variations of the b* coordinate represent the change in the yellowness (+) and blueness (−). The yellowness improves when Δb* is positive; otherwise, the blueness improves. The b* coordinate of P/L improved rapidly with an increase in accelerated aging time, implying a rise of yellowness under accelerated aging. There were also improvements in the b* coordinates of P/L-6 and P/L-8 with increasing accelerated time, but the growth was slower than that of P/L, indicating that the addition of UV-326 or UV-328 can be advantageous for preventing a rise in the yellowness of SLS parts. The increase in the b* coordinate of P/L-6 was lower than that of P/L-8. Therefore, the UV-326 was more effective than UV-328 in preventing the SLS parts from turning to yellow.
As presented in
Table 2, the trends of an overall color change for the Δ
E* of SLS parts was similar to Δ
b*. Δ
E* primarily depends on Δ
b*, which can be observed in the numerical analysis of Δ
L*, Δ
a*, Δ
b*, and Δ
E*. Hence, the overall color change Δ
E* can be limited through dampening the growth of the
b* coordinate, which can be achieved with UV-326 and UV-328. In conclusion, the overall color changes Δ
E* of SLS parts can be effectively restrained by inhibiting the parts from turning to yellow via the addition of UV-326 and UV-328, while the effect of UV-326 was slightly stronger than UV-328.
3.2. Mechanical Properties
The tensile strength, bending strength, and impact strength values of P/L, P/L-6, and P/L-8 before and after accelerated aging are shown in
Figure 4. Before accelerated aging, the tensile strength, bending strength, and impact strength values of P/L were all higher than those of P/L-6 and P/L-8, indicating that the addition of UV-326 and UV-328 into the SLS parts was not beneficial for their mechanical properties. The tensile strength, bending strength, and impact strength values of all the specimens displayed a tendency to decrease with accelerated aging time, while those of the specimens with added UV-326 or UV-328 were higher than P/L after accelerated aging for 25 d. These results indicate that the SLS parts would suffer deterioration of the mechanical properties under the sun and rain, and that UV-326 and UV-328 can effectively protect them from weathering [
9]. The mechanical properties of P/L-6 were higher than those of P/L-8, regardless of the accelerated aging time, implying that the addition of UV-326 had a more minor negative effect on the mechanical properties of SLS parts than UV-328 before accelerated aging, while it can also provide stronger protection to SLS parts against accelerated aging than UV-328. Therefore, UV-326 was slightly more effective than UV-328 in protecting the tensile strength, bending strength, and impact strength values of the SLS parts under accelerated aging.
3.3. Micromorphology
Figure 5 shows the surface micromorphologies of P/L, P/L-6, and P/L-8 before and after accelerated aging. The initial surface of P/L was extremely coarse, with significant debris and several holes (
Figure 5a). After accelerated aging for 10 d (
Figure 5b), the surface was observed to be a little smoother with less debris, which was due to UV exposure and water washing for 10 days, which degraded the fragments and washed them off from the surface. With increasing accelerated aging from 10 d to 20 d (
Figure 5c), the surface became a bit rougher with more detritus. Additionally, some cracks appeared on the surface after accelerated aging for 20 d, which could not be observed before. With further increase in accelerated aging up to 30 d (
Figure 5d), the surface seemed to become much coarser, with many more fragments and cracks on the surface and in the holes. Hence, to conclude, the surface of P/L experienced severe damage after accelerated aging for 30 d, resulting in a decline in the mechanical performance and color fading.
Similar to P/L, the initial surface of P/L-6 was rough, with some debris and holes (
Figure 5e). The needle-like objects on the surface and in the holes were UV-326 [
10]. For P/L-6 that was subjected to accelerated aging for 10 d (
Figure 5f), the surface became a little smoother and the debris decreased, which is because the surface was degraded and washed off under UV exposure and water rinsing. UV-326 was also degraded by UV exposure, which is reflected by the reduction in needle-like objects. With the increase in accelerated aging to 20 d (
Figure 5g), novel cracks appeared on the surface and the surface became rougher, with a further reduction of detritus and UV-326. The cracks widened and increased with the increase in accelerated aging to 30 d (
Figure 5h). Moreover, the surface continued coarsening, with more fragments and less UV-326.
As shown in
Figure 5i–l, with increase in accelerated aging from 0 to 30 d, the change in the surface micromorphology for P/L-8 was practically identical to that of P/L-6. The only difference was that after 10 d of accelerated aging, the surface of P/L-6 became smoother with less debris due to degradation and flushing under UV exposure and water washing, while the surface of P/L-8 became rougher with more fragments. The reason for the phenomenon could be that the initial surface of P/L-8 was smoother with less debris than that of initial P/L-6, resulting in fewer fragments being scoured off from the surface of initial P/L-8 and the smooth surface being degraded into fragments, hence becoming rough after 10 d of accelerated aging.
Comparing the surface micromorphology of P/L with those of P/L-6 and P/L-8, although the numbers of cracks in the SLS parts after accelerated aging for 30 d were similar, the surfaces of P/L-6 and P/L-8 after accelerated aging for 30 d were smoother than that of P/L. The UV-326 in P/L-6 and UV-328 in P/L-8 inhibited the degradation of SLS parts due to UV exposure and water washing via absorption of UV light, which resulted in their self-degradation and is reflected by the decreased amounts of UV-326 in P/L-6 and UV-328 in P/L-8. The degradation of SLS parts was inhibited, and thus fewer fragments were degraded from the surface, resulting in smoother surfaces for P/L-6 and P/L-8 than that of P/L after accelerated aging for 30 d. Therefore, the inhibition of degradation of SLS parts by UV-326 and UV-328 suppressed the decline in the mechanical properties of SLS parts.
3.4. FT-IR Analysis
The FT-IR spectra of P/L, P/L-6, and P/L-8 before and after accelerated aging for 30 d are shown in
Figure 6. The analysis was performed to identify the key functional groups responsible for the improvement in weather resistance of SLS parts due to UV-326 and UV-328. In the analysis of the initial spectra of P/L, P/L-6, and P/L-8 (
Figure 6a), the main peaks for copolyester were assigned as follows: two minor peaks at 2957 and 2919 cm
−1 corresponded to the asymmetric stretching vibrations of CH groups [
11,
12], a peak at 2847 cm
−1 corresponded to the symmetric stretching vibrations of CH groups [
13], a peak at 1713 cm
−1 was attributed to the stretching of carbonyl groups (C=O) present in ester groups [
14], a peak at 1406 cm
−1 corresponded to OH bending [
15], two peaks at 1268 and 1248 cm
−1 corresponded to C-O-C asymmetric stretching in ester group [
16], two peaks at 1115 and 1100 cm
−1 were due to C-O stretching in the ester group [
17], and a peak at 1013 cm
−1 could be attributed to C-O stretching in hydroxyl groups [
18]. The following peaks were assigned for limestone: a peak at 1436 cm
−1 corresponded to asymmetric C-O stretching vibration, a peak at 872 cm
−1 could be attributed to symmetric C-O stretching vibration, and a peak at 725 cm
−1 was due to OCO bending (in-plane deformation) vibration [
19]. There were only minor differences in the initial spectra of P/L, P/L-6, and P/L-8, indicating that the small amount of UV-326 or UV-328 in P/L had nearly no effects on its chemical structure.
The FT-IR spectra of P/L after accelerated aging for 30 d was similar to that of the initial P/L, with minor shifts and changes in peak intensities corresponding to the effect of accelerated aging on the chemical structure of P/L (
Figure 6b). The significant decreases in intensities of the peaks at 2957 cm
−1, 2919 cm
−1, and 2857 cm
−1 were due to the reduction of CH groups due to UV exposure and water washing. The peak at 1713 cm
−1 was shifted to 1682 cm
−1 and broadened, indicating that some ester groups (O-C=O) turned into carbonyl groups (C=O) and carboxyl groups (COOH). The widening of the peak at 1406 cm
−1 was due to the increase in the OH groups. The slump in intensities of peaks at 1268 and 1248 cm
−1 indicated that a large number of ester groups degraded under UV exposure and water washing. The disappearance of the peak at 1115 cm
−1 present in the initial P/L and a clear weakening in the intensity of the peak at 1100 cm
−1 were also due to the degradation of a large number of ester groups. The above results signify that CH and ester groups in copolyester were degraded and hydroxyl, carbonyl, and carboxyl groups were generated under UV exposure and water washing. The peak at 1436 cm
−1 corresponding to the asymmetric C-O stretching vibration in CaCO
3 present in the initial P/L merged with the peak at 1406 cm
−1, while there were almost no changes in the intensities and wavenumbers of peaks at 872 and 725 cm
−1, implying that there was only a slight effect of accelerated aging on the chemical structure of CaCO
3.
In contrast, as per the analysis of spectra of P/L-6 and P/L-8 before and after accelerated aging for 30 d (
Figure 6c,d), the changes in intensities and wavenumbers of the peaks at 2957, 2919, 2857, 1713, 1406, 1268, 1248, 1115, and 1100 cm
−1 in the spectra of P/L-6 and P/L-8 after accelerated aging for 30 d were weaker than those of P/L. This phenomenon implies that fewer CH groups and ester groups were degraded, and hence fewer hydroxyl, carbonyl, and carboxyl groups were generated under UV exposure and water washing due to the addition of UV-326 and UV-328, which effectively protected the SLS parts.
3.5. XPS Analysis
The XPS analysis was employed to identify the surface chemistry of SLS parts.
Figure 7 shows the C1s spectra with high-resolution scans for P/L, P/L-6, and P/L-8 before and after accelerated aging for 30 d. The spectra were deconvoluted into four components for all of the specimens: C1 (C-C, C-H), C2 (C-O), C3 (C=O/O-C-O), and C4 (O-C=O) [
20]. Compared to the initial P/L, the C/1 of P/L after accelerated aging for 30 d decreased a lot, while C2, C3, C4, and the O/C ratio increased (
Table 3). Combined with the results of the FT-IR analysis, the decrease in C1 and increase in C2, C3, C4, and O/C ratio after accelerated aging for 30 d can be attributed to the degradation of CH groups and ester groups and the generation of hydroxyl, carbonyl, and carboxyl groups. The C1 of the initial P/L-6 and P/L-8 samples increased a little as compared to the initial P/L sample, while C2, C4, and the O/C ratio decreased slightly due to the addition of UV-326 and UV-328. After accelerated aging for 30 d, the C1 levels of P/L-6 and P/L-8 decreased, and C2, C3, C4, and O/C ratio increased. This phenomenon indicates that similar changes happened on P/L-6 and P/L-8 under UV exposure and water washing. However, after accelerated aging for 30 d, the C1 levels of P/L-6 and P/L-8 were greater than that of P/L, while the C2, C3, and C4 levels and the O/C ratio were less than those of P/L. These results imply that the addition of UV-326 and UV-328 into SLS parts inhibited the degradation of CH and ester groups and the generation of hydroxyl, carbonyl, and carboxyl groups under UV exposure and water washing.
In summary, as per the results of the color changes, the mechanical properties, the micromorphologies, the FT-IR analysis, and the XPS analysis, the modification of SLS parts under UV exposure and water washing and the improvement mechanism of the weather resistance of SLS parts due to UV-326 and UV-328 can be suggested as the underlying causes. Under accelerated aging, the CH and ester groups of P/L suffered degradation, and thus the hydroxyl, carbonyl, and carboxyl groups were generated. With the degradation of CH and ester groups and the formation of oxygen-containing groups, the surface of P/L slowly cracked and the bulk material of P/L gradually degraded into small fragments, resulting in a decline of the mechanical properties. Moreover, changes in the chemical structure of P/L under accelerated aging led to an increase in yellowness. Finally, P/L suffered a decline of its mechanical properties and turned yellow due to sun and rain conditions. The addition of UV-326 or UV-328 inhibited the degradation of CH and ester groups and the generation of hydroxyl, carbonyl, and carboxyl groups, and thus restrained the deterioration of mechanical properties and yellowing of SLS parts. The inhibition of the deterioration of mechanical properties and yellowing of SLS parts due to UV-326 was stronger than the effect of UV-328; however, the distinction was not significant.