Dissolved Black Carbon Facilitates the Photodegradation of Microplastics via Molecular Weight-Dependent Generation of Reactive Intermediates

Photodegradation of microplastics (MPs) induced by sunlight plays a crucial role in determining their transport, fate, and impacts in aquatic environments. Dissolved black carbon (DBC), originating from pyrolyzed carbon, can potentially mediate the photodegradation of MPs owing to its potent photosensitization capacity. This study examined the impact of pyrolyzed wood derived DBC (5 mg C/L) on the photodegradation of polystyrene (PS) MPs in aquatic solutions under UV radiation. It revealed that the photodegradation of PS MPs primarily occurred at the benzene ring rather than the aliphatic segments due to the fast attack of hydroxyl radical (•OH) and singlet oxygen (1O2) on the benzene ring. The photosensitivity of DBC accelerated the degradation of PS MPs, primarily attributed to the increased production of •OH, 1O2, and triplet-excited state DBC (3DBC*). Notably, DBC-mediated photodegradation was related to its molecular weight (MW) and chemical properties. Low MW DBC (<3 kDa) containing more carbonyl groups generated more •OH and 1O2, accelerating the photodegradation of MPs. Nevertheless, higher aromatic phenols in high MW DBC (>30 kDa) scavenged •OH and generated more O2•–, inhibiting the photodegradation of MPs. Overall, this study offered valuable insights into UV-induced photodegradation of MPs and highlighted potential impacts of DBC on the transformation of MPs.

Text S1.Determination for the concentration of NB, FFA, TMP and XTT formazan During the irradiation, the remaining NB, FFA and TMP in the solutions was measured using HPLC with an Agilent 144 Eclipse XDB-C18 reversed phase column (5 μm × 250 mm × 4 mm).60% acetonitrile: 40% ultrapure water (v:v) with a detection wavelength of 263 nm for NB; 1 50% acetonitrile: 50% 0.1 wt% phosphoric acid (v:v) with a detection wavelength of 220 nm for FFA; 2 60% acetonitrile: 40% 0.1 wt % phosphoric acid (v:v) with a detection wavelength of 220 nm for TMP. 3 The concentration of XTT formazan, the reaction product of XTT with O 2 •− under light irradiation, was determined using ultraviolet-visible spectrometer at 470 nm 4 .

Text S2. Calculation for the concentration of RIs
The photo-transformation of furfuryl alcohol (FFA) can be expressed as 3 : Therefore, the steady steady-state concentration of 1 O 2 ([ 1 O 2 ] ss , M) can be determined as: Where k FFA =1.2 × 10 8 M -1 S -1 . 5e steady-state concentrations of •OH ([•OH] ss , M) were evaluated using the second-order rate constant of NB and •OH (3.9 × 10 9 M -1 s -1 (k NB−•OH )) and the pseudofirst-order rate constant (k NB′ ). 6,7  expressed as: TMP reacts with both 3 DBC * and 3 PS * in solution.Therefore, the initial phototransformation rate of TMP can be written as follows: Due to the lack of data on the second-order rate constant between 3 PS * and TMP (k 3PS*,TMP ), further calculation of [ 3 DBC * ] ss and [ 3 PS * ] ss was not continue.
The cumulated concentration of O 2 • − (M) was calculated as: 1,8 [O Where, a is the extinction coefficient of XTT formazan, 21,600 M -1 cm -1 ; 4 b indicates the distance the light travels through the quartz cuvette, 1 cm for this experiment.A is the tested absorbance of the solution.
Text S3.Calculation for carbonyl index (CI) and hydroxyl index (HI) Quantifying the results was crucial to delineate the photo-oxidation process of plastics.
The CI and HI were introduced as key metrics.These indices represent the relative abundance of carbonyl and hydroxyl groups, respectively, increases in the CI and HI are associated with an increase in the polymers' surface oxidation state.The calculation of CI and HI were using the specific area under band (SAUB) methodology. 9,10 he peaks were analysed without smoothening the data.Net peak heights were determined by subtracting the height of the baseline using integration method in Origin software. 11,12 CIand HI are calculated by comparing the net heights of the created bands (1660-1850 cm −1 for CI and 3120-3710 cm −1 for HI) in the spectra to a reference peak.In this study, the 2870-2980 cm −1 band, corresponding to the C-H stretching vibration of the CH 2 , 13,14 was chosen as this reference, as it remains stable under UV irradiation.The Where, A denotes the peak intensity.
Text S4.Calculation for light screening factor (S λ ) In this context, the concept of the light screening factor (S λ ) is introduced to gauge the impact of DBC as a filtering agent, with its value derived through following calculation formula 1,15 : CI and HI equations are as follows: CI = A 1660-1850 cm −1 / A 2870-2980 cm −1 HI = A 3120-3710 cm −1 / A 2870-2980 cm−1 450 250     )/(450 -250)where α λ (cm −1 ) represents the decadic specific absorption coefficient, and l (cm) indicates the distance the light travels through the quartz cuvette, set at 1 cm for this experiment, A λ is the recorded absorbance at a designated wavelength.S 250−450 represents the average optical screening factor between 250 and 450 nm for DBC with composite light sources.Text S5.Natural sunlight exposure of PS MPsNatural sunlight exposure of PS MPs was conducted on the rooftop from August 4 to August 28, 2023, in Beijing.The CI was 0.50 ± 0.04 for PS MPs after natural weathering.The natural light weathering of PS MPs (2.5 g/L), with or without the addition of bulk DBC (5 mg C/L), was conducted, from May 19 to June 18, 2024, under the same conditions and in the same location.The results showed that the presence of bulk DBC slightly increased HI value of PS MPs (1.63 ± 0.24 vs. 1.12 ± 0.18).However, there was no obvious difference in the CI value between Bulk + PS (0.42 ± 0.06) and PS (0.40 ± 0.04).These experiments were conducted in duplicate, with the error bars representing the maximum and minimum values.

Figure S2 .Figure S3 .
Figure S2.The emission spectrum of the mercury lamp (supplied by manufacturer)

Figure S4 .
Figure S4.Approaches for measuring PS MPs particle size using ImageJ software (PS

Table S1 .
Particle size of pristine and photo-aged PS MPs after 48 h aging

Table S1 .
The characteristics of bulk DBC and DBC fractions

Table S4 .
2D-COS Data on the assignment and sign of each cross-peak in synchronous (Φ) and asynchronous (Ψ, in the brackets) maps of Bulk + PS MPs

Table S5 .
2D-COS Data on the assignment and sign of each cross-peak in synchronous (Φ) and asynchronous (Ψ, in the brackets) maps of < 3 kDa + PS MPs

Table S6 .
2D-COS Data on the assignment and sign of each cross-peak in synchronous (Φ) and asynchronous (Ψ, in the brackets) maps of 3-30 kDa + PS MPs

Table S7 .
2D-COS Data on the assignment and sign of each cross-peak in synchronous (Φ) and asynchronous (Ψ, in the brackets) maps of 30 kDa-0.45 μm + PS MPs