Photocatalytic Degradation of Polyamide 66; Evaluating the Feasibility of Photocatalysis as a Microﬁbre-Targeting Technology

: Wastewater treatment plants (WWTPs) have been identiﬁed as main contributors to releasing microﬁbres into the environment, however, WWTPs do not have microﬁbre-targeting technologies. In this study, photocatalysis is evaluated as a potential technology to treat microﬁbres in WWTPs by studying the degradation of polyamide 66 (PA66) microﬁbres using ultraviolet (UV) and titanium dioxide (TiO 2 ). PA66 microﬁbres suspended in deionised water were exposed to di ﬀ erent combinations of UV and TiO 2 . The degradation of the PA66 microﬁbres was monitored by changes in mass, carbonyl index and morphology using microbalance, infrared spectroscopy, and scanning electron microscopy. The formation of by-products from the degradation of the ﬁbres was evaluated by measuring the chemical oxygen demand (COD) of the treated water. The degradation e ﬃ ciency was optimised under UVC with a dose of 100 mg TiO 2 / L. Under these conditions, the PA66 microﬁbres presented a 97% mass loss within 48 h. The photocatalytic conditions applied generated a relatively low level of by-products ( < 10 mg / L of COD). Therefore, photocatalysis with TiO 2 an UVC could potentially be a feasible technology to treat microﬁbres in WWTPs, although more investigation is required to establish if this treatment leads to the formation of nanoﬁbres. Further work is needed to translate the present optimised conditions to WWTPs. the health of humans and the ecosystem by preventing the release of microﬁbres into the environment if the treatment process does not lead to nanoﬁbres. Further investigation is required to develop photocatalysis as a microﬁbre-targeted technology, the present study successfully demonstrated the degradation of PA66 microﬁbres using photocatalysts suspended ultrapure water. Therefore, it provides new insights on the degradation of microﬁbres by AOPs and open horizons for future research.


Introduction
The global demand, production, and generation of plastic waste have skyrocketed over the last 60 years [1]. Plastics have penetrated all aspects of our daily life: they are included, among many applications, in clothing, packaging, materials used in construction, and agriculture. Global plastic production and their utilisation has increased from 1.5 to 335 million metric tons from 1950 to 2016 [2]. Plastics present excellent properties of durability that result in resistance to natural degradation. However such high stability is not advantageous considering that globally, from the 1980, plastic wastes have been mainly discarded [1]: more than half of the plastic waste ends up in landfills or in the environment rather than being recycled or incinerated [3]. Due to the immense plastic production and improper waste management, plastic pollution is of great concern [4,5]. Specifically, microfibres

Materials and Methods
To investigate the degradation of PA66 microfibres with photo-oxidative and photocatalytic degradation, the experiments were conducted under customised lab-scale reaction chambers as carried out by several authors [35,40,42]. Two sets of experiments were conducted to find (1) the effect of different UV wavelengths and (2) the effect of the TiO 2 dose on the degradation of PA66 microfibres. The degradation of the microfibres was evaluated by the means of mass loss and morphological and chemical changes in the fibres and in the water being treated. All experiments were carried out in triplicate.

Materials
Synthetic PA66 microfibres with a diameter of 10 µm (AM325705, Goodfellow, UK) were cut with 1.0 m length (≈1.3 mg) using scissors. Mass loss was measured with an analytical microbalance (Mettler AT201, Columbus, OH, USA) with readability of 0.01 mg. Each of the microfibre samples were rinsed with deionised water, pat dried to remove moisture, and stored in a glass container at a room temperature as demonstrated in Figure 1. For the photocatalyst, TiO 2 powder (Aeroxide P25 Degussa, Sigma Aldrich, St. Louis, MO, USA) was used. The TiO 2 powder contained 70% anatase and 30% of rutile with mean particle size of 21 nm. Two types of UV lamps were used: a UVA lamp (Philips TL8W BLB, Łódź, Poland) that allows working at 365 nm, a UVC lamp (Philips TUV 8W, Łódź, Poland) to work at 254 nm. t UV wavelengths and (2) the effect of the TiO2 dose on the degradation s. The degradation of the microfibres was evaluated by the means of mass ical and chemical changes in the fibres and in the water being treated. All ex d out in triplicate. als etic PA66 microfibres with a diameter of 10 μm (AM325705, Goodfellow, UK length (≈1.3 mg) using scissors. Mass loss was measured with an analytical mic T201, USA) with readability of 0.01 mg. Each of the microfibre samples were ri water, pat dried to remove moisture, and stored in a glass container at a room te trated in Figure 1. For the photocatalyst, TiO2 powder (Aeroxide P25 Degus . Louis, MO, USA) was used. The TiO2 powder contained 70% anatase and 30 particle size of 21 nm. Two types of UV lamps were used: a UVA lamp (Phi d) that allows working at 365 nm, a UVC lamp (Philips TUV 8W, Poland) to w

uction of Photocatalytic Degradation Reaction Chambers
action chamber used is shown in Figure 2. It was constructed with a glass chamb height of 35 × 20 × 25 cm) and it included five UV lamps. The UV lamps and th er were assembled as an integral part to keep the same distance from the UV la d to make it convenient to open and close the chamber when taking out the sam etween the UV lamps and the PA66 microfibre samples was fixed at 5 cm ous light intensity. Five electrical stabilisers (8 W) were connected to a single cor

Construction of Photocatalytic Degradation Reaction Chambers
The reaction chamber used is shown in Figure 2. It was constructed with a glass chamber (length × width × height of 35 × 20 × 25 cm) and it included five UV lamps. The UV lamps and the cover of the chamber were assembled as an integral part to keep the same distance from the UV lamps to the samples and to make it convenient to open and close the chamber when taking out the samples. The distance between the UV lamps and the PA66 microfibre samples was fixed at 5 cm to create homogeneous light intensity. Five electrical stabilisers (8 W) were connected to a single cord to allow turning on and off the UV lamps simultaneously. The chamber was covered with aluminium foil to block stray light and to increase the reflection efficiency. The aluminium cover was perforated (eight openings) for ventilation.

Experimental Design
The treatment of PA66 microfibres suspended in water by photooxidation and photocatalytic degradation was conducted in the reaction chambers as demonstrated in Figure 3. Three chambers were constructed: one chamber did not include UV lamps; one chamber had five UVC lamps (8 W); and one chamber had five UVA lamps (8 W). The effects of UV light on the fibres was investigated by comparing the impact of short UV wavelengths (UVC), long UV wavelengths (UVA), and no light conditions. Specifically, chamber A was a control (with no lamps) used to monitor fibres hydration and adsorption of TiO2 onto the fibres. Chamber B was used to investigate the photo-oxidation and photocatalytic oxidation under UVA. Chamber C was equivalent to Chamber B but including UVC instead of UVA. The temperature inside the reaction chamber ranged between 25 and 38 °C under the exposure of UV light without any interruption or control of the temperature. Each sample (dry microfibre) was taken out from the chambers and weighed every 24 h, for up to 105 h reaction time. The final reaction time was selected when reaching >80% fibre mass loss. Then, the changes caused by the degradation were measured in the remaining PA66 microfibres. The PA66 microfibres were returned to the Petri dishes after every mass measurement and repeated this sequence until 105 h of reaction time was reached.

Experimental Design
The treatment of PA66 microfibres suspended in water by photooxidation and photocatalytic degradation was conducted in the reaction chambers as demonstrated in Figure 3. Three chambers were constructed: one chamber did not include UV lamps; one chamber had five UVC lamps (8 W); and one chamber had five UVA lamps (8 W). The effects of UV light on the fibres was investigated by comparing the impact of short UV wavelengths (UVC), long UV wavelengths (UVA), and no light conditions. Specifically, chamber A was a control (with no lamps) used to monitor fibres hydration and adsorption of TiO 2 onto the fibres. Chamber B was used to investigate the photo-oxidation and photocatalytic oxidation under UVA. Chamber C was equivalent to Chamber B but including UVC instead of UVA. The temperature inside the reaction chamber ranged between 25 and 38 • C under the exposure of UV light without any interruption or control of the temperature. Each sample (dry microfibre) was taken out from the chambers and weighed every 24 h, for up to 105 h reaction time. The final reaction time was selected when reaching >80% fibre mass loss. Then, the changes caused by the degradation were measured in the remaining PA66 microfibres. The PA66 microfibres were returned to the Petri dishes after every mass measurement and repeated this sequence until 105 h of reaction time was reached.

Experimental Design
The treatment of PA66 microfibres suspended in water by photooxidation and photocatalytic degradation was conducted in the reaction chambers as demonstrated in Figure 3. Three chambers were constructed: one chamber did not include UV lamps; one chamber had five UVC lamps (8 W); and one chamber had five UVA lamps (8 W). The effects of UV light on the fibres was investigated by comparing the impact of short UV wavelengths (UVC), long UV wavelengths (UVA), and no light conditions. Specifically, chamber A was a control (with no lamps) used to monitor fibres hydration and adsorption of TiO2 onto the fibres. Chamber B was used to investigate the photo-oxidation and photocatalytic oxidation under UVA. Chamber C was equivalent to Chamber B but including UVC instead of UVA. The temperature inside the reaction chamber ranged between 25 and 38 °C under the exposure of UV light without any interruption or control of the temperature. Each sample (dry microfibre) was taken out from the chambers and weighed every 24 h, for up to 105 h reaction time. The final reaction time was selected when reaching >80% fibre mass loss. Then, the changes caused by the degradation were measured in the remaining PA66 microfibres. The PA66 microfibres were returned to the Petri dishes after every mass measurement and repeated this sequence until 105 h of reaction time was reached.   The cut PA66 microfibres (1 m each) were immersed in deionised water (50 mL) in every Petri dish, six Petri dishes were used in total (see Figure 4). The experiments involved 1 m long microfibre per dish. This set up helped to avoid scattering of microsized fibres. The maximum volume that could be contained in the petri dish was 50 mL of water and this volume made possible the measurement of chemical oxygen demand (COD). Six samples (each sample is a suspension of one fibre in water), labelled as SP1 to SP6 were placed inside of the reaction chambers (A, B, and C). SP2, SP4, and SP6 were mixed with TiO 2 using a magnetic stirrer for 15 min to create a slurry-type condition for the photocatalytic oxidation. No catalyst was added to SP1, SP3, and SP5. The evaporation rate of deionised water due to the applied UV irradiation was evaluated prior to the experiment by measuring the time taken to evaporate 50 mL of deionised water under such experimental conditions. Thus, deionised water was added to the samples every 15 h based on the evaporation rate of 1.25 mL/h to keep the PA66 microfibre samples hydrated and keep the volume in the Petri dish constant. The conditions applied to SP1-SP6 are summarised in Table 1.
Water 2020, 12, x FOR PEER REVIEW 5 of 23 The cut PA66 microfibres (1 m each) were immersed in deionised water (50 mL) in every Petri dish, six Petri dishes were used in total (see Figure 4). The experiments involved 1 m long microfibre per dish. This set up helped to avoid scattering of microsized fibres. The maximum volume that could be contained in the petri dish was 50 mL of water and this volume made possible the measurement of chemical oxygen demand (COD). Six samples (each sample is a suspension of one fibre in water), labelled as SP1 to SP6 were placed inside of the reaction chambers (A, B, and C). SP2, SP4, and SP6 were mixed with TiO2 using a magnetic stirrer for 15 min to create a slurry-type condition for the photocatalytic oxidation. No catalyst was added to SP1, SP3, and SP5. The evaporation rate of deionised water due to the applied UV irradiation was evaluated prior to the experiment by measuring the time taken to evaporate 50 mL of deionised water under such experimental conditions. Thus, deionised water was added to the samples every 15 h based on the evaporation rate of 1.25 mL/h to keep the PA66 microfibre samples hydrated and keep the volume in the Petri dish constant. The conditions applied to SP1-SP6 are summarised in Table 1.   For the determination of the TiO 2 dose, UV radiation was fixed and a single chamber was used with five samples containing different levels of TiO 2 ( Figure 5).
Water 2020, 12, x FOR PEER REVIEW 7 of 23 For the determination of the TiO2 dose, UV radiation was fixed and a single chamber was used with five samples containing different levels of TiO2 ( Figure 5). The conditions used for the optimisation of TiO2 dose are summarised in Table 2. In this second experiment, UVC was used for 48 h. The total reaction time was selected based on a mass loss > 90%. The most effective concentration of TiO2 was evaluated by means of the fibres' mass loss. Table 2. Summary of the set-up for SP1-SP5 investigating the optimal concentration of TiO2 under UVC. Each condition was studied in triplicate. The conditions used for the optimisation of TiO 2 dose are summarised in Table 2. In this second experiment, UVC was used for 48 h. The total reaction time was selected based on a mass loss > 90%. The most effective concentration of TiO 2 was evaluated by means of the fibres' mass loss.

Mass Loss
A common method to quantify the photocatalytic degradation of plastics is to evaluate the percentage of mass loss [39,43,44]. The percentage of mass loss can be determined using Equation (1) as described in [40]: where m 0 represents initial mass of the fibre (mg) before the experiment and m corresponds to the final mass (mg) after the experiment. The evolution of mass of fibre with time can be used to determine the kinetics of the photocatalytic degradation of the MPs. The Langmuir-Hinshelwood (L-H) and first-order expressions have been widely used to explain the kinetics of photocatalytic reactions [45][46][47]. The photocatalytic degradation can be quantitatively estimated by comparing the apparent reaction rate constants (κ app ) obtained from the first-order rate equation derived from the L-H model as Equation (2) as described in [47].
where C 0 is the initial concentration of the organic compound (mg/L), in this case it will be approximated to the amount of suspended fibres per volume of water; C is the concentration of fibre at a particular time of the photocatalytic reaction (mg/L); κ app is the apparent rate constant of the reaction, and t is the irradiation time. The values of κ app can be obtained from the linear regression analysis in the plot [48,49]. The κ app quantifies the rate of a reaction and the higher κ app indicates the faster degradation of the original organic compound. The reaction constant i.e., half-life, which is the time that the initial concentration of reactant is decreased to one-half of its initial value, can be calculated using Equation (3) as described in [50].
The mass loss observed was also used to estimate the kinetics of the photocatalytic degradation of PA66 microfibres.

Morphological Properties of PA66 Microfibres
A scanning electron microscope (SEM) Supra 55VP, ZEISS, Germany was used to assess the degradation of polymers. The PA66 microfibres were placed on an aluminium stub using conductive Water 2020, 12, 3551 7 of 20 and adhesive carbon tape cut into 0.5 × 0.5 cm. The samples were pretreated with platinum using a vacuum coater (EM ACE200, Leica, Germany) with a sputter current of 20 mA for 100 s.

Surface Chemistry of the PA66 Microfibres
During the photodegradation of polymer, different chemical groups, mainly carbonyl and hydroxyl groups, can be formed [51]. The carbonyl index (CI) (see Equation (4)) was measured from the Fourier Transform Infrared (FTIR) spectra from the fibres and it was used to monitor the oxidation of their surface [35,49,52]. CI can be defined as the ratio between carbonyl signal and a reference band from methylene.
The absorbance of the carbonyl band generally falls between 1900 and 1600 cm −1 and the reference peak can be the CH 3 rocking band or the CH 2 scissoring band which correspond to vibrations of groups not affected by oxidation [44,[51][52][53][54]. High CI indicates a high degree of polymer degradation [53].
To analyse the changes in chemical properties due to the degradation mechanisms, FTIR spectroscopy (Nicolet 6700, Thermo Scientific, Waltham, MA, USA) was used. The transmittance within the wavenumber 650-4000 cm −1 , with a resolution of 8 cm −1 , was measured. The FTIR spectra were obtained in attenuated total reflectance (ATR) mode by placing the PA66 microfibre samples (SP1 to SP6) directly below the zinc selenide diamond prism without any pretreatment. Spectra were auto corrected with auto base function.

Monitoring the Degradation of the Suspension of PA66 Fibres with the Analysis of the Chemical Oxygen Demand (COD) in Solution
COD was used to evaluate the degree of mineralization and formation of by-products during the degradation of organic pollutants [47]. COD was measured with a spectrophotometer (DR2800, Hach, Loveland, CO, USA) as follows. The water samples in SP1 to SP6 were collected in 10 mL glass vials, where particles were left to settle for 24 h. The supernatant (2 mL) was separated and placed in digestion glass vials (TNTplus™, Hach, Loveland, CO, USA). The digestion glass vials (now containing digestion solution and the supernatant of samples) were manually shaken for 1 min to mix them thoroughly well and placed into the heater (HS-R200, HUMAS, Daejeon, South Korea) at 150 • C for 2 h. Afterwards, the vials were cooled at room temperature for 1 h, and then the vials were wiped to remove any fingerprints or dirt before placing them into the holder of spectrophotometer for analysis. The COD was determined using low range digest reagent (LR, TNT 3-150, Hach, USA) measuring from 3 to 150 mg/L. COD tests were carried out in triplicate. Figure 6 shows the mass loss of the PA66 microfibre samples (SP1 to SP6) exposed to different combinations of UV irradiation and catalyst for 105 h. The effects of UV wavelengths were investigated by comparing photo-oxidation under UVA (SP3) and UVC (SP5). While the mass loss of SP5 was 83%, SP3 had only 6% mass loss within 105 h of reaction time (precision for these experiments was <28% for SP3 and <14% for SP5). Therefore, it is evident that the short wavelength (UVC) was more effective than the long wavelength (UVA) in degrading PA66 and this agrees well with other studies investigating the degradation of polyamide [55][56][57][58][59][60][61][62][63]. This result may potentially be explained by different intermediate products forming from the polymer by the action of UVA and UVC with possible different degradation mechanisms [63]. The apparent reaction rate constant ( ) and the half-life (t1/2) in Table 3 were calculated from the ln(C/C0) versus time plots in Figure 7. Given that greater refers to greater degradation rate of microfibres, it can be confirmed that SP5 presented the highest degradation rate as 13.6 × 10 −3 h −1 followed by SP6 (2.6 × 10 −3 h −1 ) and SP4 (2.3 × 10 −3 h −1 ) which is consistent with the mass loss data. The half-life of SP5 was 51h (see Table 3) which indicates that the initial concentration of the polymer (1.26 mg/50 mL) would be reduced to its half (0.63 mg/50 mL) after 51 h. The short UV wavelength was more effective at degrading PA66 microfibres compared to the longer wavelength, excluding the role of TiO2. However, applying photo-oxidation in WWTPs would have low practicality as it would take at least 51 h to degrade half of the amount of the PA66 microfibres.  Compared to the mass loss by photo-oxidation under UVC (SP5, 83%) without including catalyst, photocatalysis using TiO 2 under UVA (SP4, 26%) and UVC (SP6, 24%) showed both lower mass loss (p 0.05). This result is not in agreement with a study that found that photocatalysis led to greater mass loss (of polyethylene in that case) [64]. It might be possible that the concentration of TiO 2 (1 g in 50 mL) was excessive and blocked the light from reaching the surface of the polymer [65]. It was suggested that the initial photocatalytic reaction rate was directly proportional to the mass of catalysts but this may be the case for a limited photocatalyst concentration range [66]. Therefore, the study conditions with high dose of catalyst (2%), have caused photocatalysis with low effectivity compared to photo-oxidation resulting in decelerating the degradation rate.

Effect of UV Irradiation and Catalyst on the Kinetics of the Photocatalytic Degradation of PA66 Microfibres
Under hydration conditions (SP1), the mass loss of the fibres was minimal (2%) for all reaction times studied (see Figure 6), and where there was absence of UV irradiation (SP2), the mass of the microfibres increased by 0.7% due to the adsorbed TiO 2 particles to the microfibres as observed in the SEM micrographs in Figure 10c. Based on these results, the effects of hydration and adsorption of TiO 2 onto fibres, both without using UV, are negligible at degrading the PA66 microfibres within 105 h. However, temperature might have favoured the degradation of the microfibres. The authors of [67] found that photolysis with increased temperature reduced dissolved organic carbon (DOC). Therefore, it is recommended to investigate the effect of temperature on microfibre degradation.
The apparent reaction rate constant (κ app ) and the half-life (t 1/2 ) in Table 3 were calculated from the ln(C/C 0 ) versus time plots in Figure 7. Given that greater κ app refers to greater degradation rate of microfibres, it can be confirmed that SP5 presented the highest degradation rate as 13.6 × 10 −3 h −1 followed by SP6 (2.6 × 10 −3 h −1 ) and SP4 (2.3 × 10 −3 h −1 ) which is consistent with the mass loss data. The half-life of SP5 was 51 h (see Table 3) which indicates that the initial concentration of the polymer Water 2020, 12, 3551 9 of 20 (1.26 mg/50 mL) would be reduced to its half (0.63 mg/50 mL) after 51 h. The short UV wavelength was more effective at degrading PA66 microfibres compared to the longer wavelength, excluding the role of TiO 2 . However, applying photo-oxidation in WWTPs would have low practicality as it would take at least 51 h to degrade half of the amount of the PA66 microfibres.

Mass Loss and Photocatalytic Degradation Kinetics: Effects of the Concentrations of Catalysts
To assess the effects of the concentration of the catalyst on the degradation rate of PA66 microfibres, a second experiment was performed to test the effect of different doses of TiO2 (up to 20 g/L) under UVC for 45 h. As shown in Figure 8, the conditions tested in SPT2 (100 mg TiO2/L) showed 97% of mass loss in 48 h, followed by 78% for SPT3 (500 mg TiO2/L) and 24% for SPT4 (1000 mg TiO2/L) while SPT1 (no TiO2) and SPT5 (20,000 mg TiO2/L) resulted in the least mass loss of 18% and 14%, respectively. Hence, the maximum (SPT5) and minimum (SPT1) TiO2 concentrations led to the lowest fibre mass loss, which is in agreement with an earlier study finding that the concentration of TiO2 outside a particular concentration range can deaccelerate the degradation [68]. Although the most effective degradation was photo-oxidation under UVC (SP5) in Figure 6-as the amount of TiO2 for SP6 was excessive and unoptimised- Figure 8 shows that photocatalysis is the most effective approach for degrading microfibres when the amount of TiO2 is optimised (SPT2).

Mass Loss and Photocatalytic Degradation Kinetics: Effects of the Concentrations of Catalysts
To assess the effects of the concentration of the catalyst on the degradation rate of PA66 microfibres, a second experiment was performed to test the effect of different doses of TiO 2 (up to 20 g/L) under UVC for 45 h. As shown in Figure 8, the conditions tested in SPT2 (100 mg TiO 2 /L) showed 97% of mass loss in 48 h, followed by 78% for SPT3 (500 mg TiO 2 /L) and 24% for SPT4 (1000 mg TiO 2 /L) while SPT1 (no TiO 2 ) and SPT5 (20,000 mg TiO 2 /L) resulted in the least mass loss of 18% and 14%, respectively. Hence, the maximum (SPT5) and minimum (SPT1) TiO 2 concentrations led to the lowest fibre mass loss, which is in agreement with an earlier study finding that the concentration of TiO 2 outside a particular concentration range can deaccelerate the degradation [68]. Although the most effective degradation was photo-oxidation under UVC (SP5) in Figure 6-as the amount of TiO 2 for SP6 was excessive and unoptimised- Figure 8 shows that photocatalysis is the most effective approach for degrading microfibres when the amount of TiO 2 is optimised (SPT2).
14%, respectively. Hence, the maximum (SPT5) and minimum (SPT1) TiO2 concentrations led to the lowest fibre mass loss, which is in agreement with an earlier study finding that the concentration of TiO2 outside a particular concentration range can deaccelerate the degradation [68]. Although the most effective degradation was photo-oxidation under UVC (SP5) in Figure 6-as the amount of TiO2 for SP6 was excessive and unoptimised- Figure 8 shows that photocatalysis is the most effective approach for degrading microfibres when the amount of TiO2 is optimised (SPT2). The κ app was determined from the ln(C/C 0 ) versus time plots in Figure 9. Half-lives (t 1/2 (h)) are indicated in Table 4. SPT2 showed the highest constant (7.0 × 10 −2 h −1 ) while SPT5 showed the lowest constant (0.3 × 10 −2 h −1 ). Therefore, these results confirmed that the TiO 2 at 100 mg/L is the most effective condition for the PA66 degradation. Moreover, it can be confirmed that the excessive addition of TiO 2 can hinder the degradation as SPT5, leading to a lower kinetic constant compared to the SPT1, which contained no catalyst. Since SPT2 showed 10 h of fibres' half-life, these conditions could potentially be applied in the WWTPs to treat microfibres, although careful examination of the degradation products, including possible yield of nanofibres from the degradation needs to be done. The was determined from the ln(C/C0) versus time plots in Figure 9. Half-lives (t1/2 (h)) are indicated in Table 4. SPT2 showed the highest constant (7.0 × 10 −2 h −1 ) while SPT5 showed the lowest constant (0.3 × 10 −2 h −1 ). Therefore, these results confirmed that the TiO2 at 100 mg/L is the most effective condition for the PA66 degradation. Moreover, it can be confirmed that the excessive addition of TiO2 can hinder the degradation as SPT5, leading to a lower kinetic constant compared to the SPT1, which contained no catalyst. Since SPT2 showed 10 h of fibres' half-life, these conditions could potentially be applied in the WWTPs to treat microfibres, although careful examination of the degradation products, including possible yield of nanofibres from the degradation needs to be done.

Morphological Changes of Microfibres
The SEM analysis of the study microfibres, before and after oxidative treatment, was performed to examine the morphological changes of the PA66 surface due to the varying combinations of UV (A, B, and C) and TiO2. Figure 10a is the starting PA66 microfibres (before any degradation). Figure  10b-g shows PA66 microfibres which had been exposed to the conditions of hydration, adsorption,

Morphological Changes of Microfibres
The SEM analysis of the study microfibres, before and after oxidative treatment, was performed to examine the morphological changes of the PA66 surface due to the varying combinations of UV (A, B, and C) and TiO 2 . Figure 10a is the starting PA66 microfibres (before any degradation). Figure 10b-g shows PA66 microfibres which had been exposed to the conditions of hydration, adsorption, photo-oxidation, and photocatalysis within 105 h. Clear signs of degradation can be observed in Figure 10d-g. Figure 10b,c illustrates the effect of hydration and adsorption, respectively. Slightly eroded surface was observed from Figure 10b which is a similar phenomenon of hydrolytic degradation identified by [69][70][71]. The adsorption of the photocatalyst can be observed as TiO 2 particles were attached to the surface of the PA66 microfibres without any sign of cracks or damages as shown in Figure 10c. Figure 10d,f shows the effect of photo-oxidation when UVA and UVC were irradiated, respectively. Many relatively deep cavities, cracks, and embrittled surfaces were detected from Figure 10d,f. This evidences the compromised integrity of the fibres due to the UV irradiation, which is aligned to previously reported findings of [35,68]. Figure 10e,g displays the effect of photocatalysis under UVA and UVC, respectively. The surface of the PA66 microfibre was damaged and microcracks formed. Some of the cracks were filled with TiO 2 particles. Based on this observation, it might be possible that the lower mass loss of the PA66 microfibres under photocatalysis could be due to the excessive addition of TiO 2 . This is supported by previous studies [63,64]. The findings related to mass loss in this work confirm that excessive use of TiO 2 can impede the interaction of the UV irradiation with the surface of the fibres.

Changes of Chemical Properties
FTIR analysis of the PA66 microfibres was carried out to examine the changes in the chemical properties of the microfibres. The characteristic peaks of the unexposed PA66 microfibres in the FTIR spectra region are summarised in Table 5. Figure 11 shows the characteristic FTIR spectrum of PA66 fibres which had not been exposed to any degradation mechanism. The peaks arising at 3295 and 1436 cm −1 can be assigned to the stretching and deformation vibration of N-H bonds; the band at 3076 cm −1 is associated to the stretching vibration of C-H bond; and the bands at 2917 and 2851 cm −1 can be due to the asymmetrical and symmetrical stretching vibration of CH 2 [71]. The stretching of the amide I (C=O stretching), amide II (C-N stretching and N-H bending), and amide III (C-N stretching) are observed at 1632, 1536, and 12,718 cm −1 , respectively [74]. The band located at 1141 cm −1 can be attributed to CCH symmetric bending vibration combined with CH 2 twisting and the bands at 933 and 682 cm −1 are associated with the stretching and bending vibration of C-C bonds [69,73]. The increase of the intensity of a band from an oxidised group such as carbonyl (C=O), with respect to a reference peak within the same spectrum, can indicate that there are more oxygenated groups after the treatment compared to the initial state; therefore, the FTIR spectrum can be used to assess the level of oxidation.    Figure 11. FTIR band assignment of PA66 microfibres unexposed to degradation mechanism.
The band ≈1630 cm −1 may correspond to the carbonyl group from amide groups (NH-C=O) and the reference peak selected was the most intense band from the CH2 stretching. The reference blank was selected assuming that CH groups may undergo less oxidation and may remain constant [76]. (Figure 12). The band ≈1630 cm −1 may correspond to the carbonyl group from amide groups (NH-C=O) and the reference peak selected was the most intense band from the CH 2 stretching. The reference blank was selected assuming that CH groups may undergo less oxidation and may remain constant [76]. (Figure 12).
Using Equation (4), CI values for the SP1 to SP6 were calculated as summarised in Table 6. The CI values were converted into the percentage increase compared to the starting fibres. The results showed that the photo-oxidation under UVC (SP5, 27%), photocatalysis under UVC (SP6, 25%), and photo-oxidation under UVA (SP4, 19%) led to higher oxidation degree compared to the unexposed PA66 microfibres (SP1, 0%). The different level of oxidation due to the UVC and UVA might be due to the different energies of the radiation and intermediate products generated during the photodegradation depending on the long and short wavelengths [63]. However, no change in absorbance in the region near 3500 cm −1 was detected in the experiment which indicates that the presence of hydroxyl groups did not tend to increase much throughout the conditions tested. This result corroborates well with the result elsewhere [77] and it is in agreement with the no generation of alcohols and carboxylic acid during the photo-oxidation and photocatalytic degradation of PA66 microfibres. Figure 11. FTIR band assignment of PA66 microfibres unexposed to degradation mechanism.
The band ≈1630 cm −1 may correspond to the carbonyl group from amide groups (NH-C=O) and the reference peak selected was the most intense band from the CH2 stretching. The reference blank was selected assuming that CH groups may undergo less oxidation and may remain constant [76]. (Figure 12).

Formation of By-Products Due to the Degradation of PA66 Microfibres
To assess by-products generated due to the degradation mechanisms of hydration, adsorption, photo-oxidation, and photocatalytic degradation, the COD of the water where the fibres were suspended was measured (Table 7). SPW1 to SPW6 refer to the water sample taken from SP1 to SP6 where the PA66 microfibres were exposed to different UV and catalyst conditions (given in Table 7). R-DW refers to the reference deionised water without microfibers. R-TiO 2 refers to 20,000 mg/L of TiO 2 in water without UV irradiation.
COD was below the limit of detection (3.4 mg/L in R-DW). In contrast, COD was 40 mg/L (when measuring the solution that contained suspended TiO 2 and no presence of microfibres (R-TiO 2 ). These latter results could be the effect of residual TiO 2 on the measurement. Table 7. The COD of the water were the PA66 microfibres were suspended and exposed to the different conditions of UV and catalyst (SPW1-SPW6). R-DW refer to the reference water sample; R-TiO 2 refers to 20,000 mg/L of TiO 2 in water without UV irradiation. COD results were averaged from the triplicate data. The lowest COD levels. LOD (limit of detection) estimated at 3.4 mg/L. SD is standard deviation.
Water from the photocatalysis experiments (SPW4 and SPW6) showed relatively low COD compared to the water from adsorption (SPW2), and photo-oxidation (SPW3 and SPW5) experiments (see Table 7). Although the difference is not significant, the COD level decreased the most when TiO 2 and UV irradiation are applied simultaneously in this experiment. The released of oxidised parts of the fibre due to the photo-induced reaction could be the cause of the increased COD observed. These results (SPW4 and SPW6) demonstrate the degradability of PA66 microfibres with photocatalytic degradation while generating a low level of COD. Thus, the photocatalytic degradation can potentially be applicable to WWTPs without violating the discharge compliance of COD. However, it is recommended to examine the presence of nanofibres or debris with SEM or transmission electron microscopy (TEM) and systems able to detect the particle size distribution in the suspended solution (e.g., nanosizer).

Insights to Microfibre-Targeted Treatment Technology
Photocatalysis of PA66 microfibres suspended in water using UVC with 100 mg/L of TiO 2 resulted in 97% of mass loss in 48 h ( Figure 6)with a relatively low level of COD. This suggests low formation of degradation by-products during the treatment and it could a potential technology to treat microfibres in WWTPs. Although this work is a preliminary study, the proposed photocatalysis might not be suitable for a stand-alone process as UV treatment is not effective for treating high turbidity water. However, it could be developed as tertiary treatment for microfibre residuals which could not be completely removed by the secondary wastewater treatment. Additionally, potential microfibre residuals in drinking water after coagulation-flocculation, flotation, or filtration could be removed by photocatalysis. However, a more detailed investigation is suggested about by-products generated, matrix effect, energy required by the process, and effect of temperature. Concerns regarding the use of slurry type of TiO 2 in photocatalytic treatment can be overcome by the application of a hybrid coagulation step as suggested elsewhere [37].
The development of microfibre-targeted treatment technologies will benefit the health of humans and the ecosystem by preventing the release of microfibres into the environment if the treatment process does not lead to nanofibres. Further investigation is required to develop photocatalysis as a microfibre-targeted technology, the present study successfully demonstrated the degradation of PA66 microfibres using photocatalysts suspended ultrapure water. Therefore, it provides new insights on the degradation of microfibres by AOPs and open horizons for future research.

Conclusions
PA66 microfibres can be degraded with photocatalysis. The optimal conditions for the degradation of PA66 microfibres in ultrapure water were under UVC with 100 mg/L of TiO 2 . Under these conditions, 97% of average mass loss was achieved within 48 h. Photocatalysis generated low level of fibre by-products compared to photo-oxidation. The result indicates that photocatalysis can degrade PA66 microfibres in WWTPs while there is a need to further assess the products generated from the degradation of the microfibres. The present study is a pioneer investigation on the degradation of PA66 microfibres using photo-oxidation and photocatalysis. The main findings of this research are as follows: • The degradation of PA66 microfibres was dependent on the wavelengths of UV irradiation during the photo-oxidation. The degradation was more effective when using a short wavelength (254 nm) (approximately 14 times more effective in degrading PA66 microfibres compared to using 365 nm).

•
The concentration of TiO 2 was an important factor in accelerating the degradation rate of PA66 microfibres. The optimal concentration in this study was 100 mg TiO 2 /L.

•
The photocatalytic degradation of PA66 microfibres generated organic by-products, but it was relatively low compared to photo-oxidation.
The photo-oxidation under UVC was the most effective treatment as 83% of microfibres' mass loss was achieved in 105 h; UVC was more effective than UVA in degrading the microfibres (UVA only achieved 6% of mass loss during the photo-oxidation). The effects of hydration and adsorption were negligible as the mass loss of the sample non exposed to light was <2%. The CI was greatest (hence maximum oxidation) at 3.40 for the PA66 microfibres that had been photo-oxidised under UVC which is equivalent to a 27% increase of oxidation level compared to the PA66 microfibres unexposed to any degradation mechanisms. The mass loss results were in agreement with the CI under photo-oxidation by UVC.
SEM analysis showed microcracks and cavities on the surface of the microfibres, confirming the degradation effect of photo-oxidation and photocatalysis thereof. The COD was relatively low when the photocatalysis was applied (<10 mg COD/L) compared to the effect of photo-oxidation (>30 mg/L). The most effective concentration of TiO 2 was 100 mg/L, which lead to 97% of mass of microfibres lost in 48 h. The half-life of fibres in SP6 (concentration of TiO 2 as 20,000 mg/L under UVC), was reduced from 267 to 10 h when 100 mg/L TiO 2 concentration was applied (SPT2).
The present study demonstrated the effects of UV and TiO 2 in photo-oxidation and photocatalytic degradation of PA66 microfibres in ultrapure water. Therefore, these results provide new insights for applying the photocatalysis as a microfibre-targeted treatment technology in WWTPs with minimal generation of by-products in terms of COD. However, future research should be carried out considering the potential effects of (1) applying different types of microfibres such as PS and PET; (2) testing different conditions such as applying stirrer in the reaction chamber, adjusting pH level and temperature, and using different types of catalyst; (3) assessing the photocatalysis with realistic water sampled from secondary effluent of WWTPs; (4) investigating the formation of by-products comprehensively, such as identifying the generation of nanoparticles and other potential intermediate compounds dissolved in the water.