Materials
Paddy soil samples were collected from a rice field near Yanqing District, Beijing, China (40°27′28.93″N, 116°04′13.10″E) in December 2023 (Fig. 1a). According to information from local farmers, neither of the fields had previously used plastic mulches, and no other sources of microplastics (MPs) were identified. The soil samples had no significant microplastic contamination through the naked eye and microscope before the cultivation experiment, minimizing the impact of native MPs on the incubation experiments. The paddy soil with overlying water was collected from a depth of 0–20 cm, with around 10 sampling points randomly distributed across the entire field. After briefly removing roots, residues, and stones, the soil was placed in stainless steel containers and transported back to the laboratory at ambient temperature.
Presently, the most popular BPs in the market is Polybutylene adipate-co-terephthalate (PBAT), which accounts for 29.9% of global BPs production capacity21. Therefore, we have chosen it as a typical BPs. The PBAT was obtained in powder form from Shanghai Guanbu ET Co., Ltd. (Shanghai, China). The powders used for the incubation experiments were homogenized in a mortar and sieved using a double-layer sieve to achieve a consistent size range of 75–150 μm, as previously described22. PBAT is a flexible, biodegradable polymer with a melting point of 110–120 °C and a density of 1.23 g/cm3. It has moderate oil and solvent resistance, good flexibility, thermal stability, and some chemical resistance, with mechanical strength generally lower than that of conventional plastics, and it is fully degradable in the natural environment21.
Schematic diagram of experimental setup. (a) Sampling locations and cultivation of paddy soils. (b) Continuous monitoring of CH4/CO2 using long optical-path FTIR spectroscopy. (c) Measuring Gas Samples with GC. (d) Measuring the ATR spectra of Dried Soil Samples.
Experimental design and sampling
Incubation experiments were conducted in a controlled environment chamber at a stable temperature of 25±1°C. All the collected soil samples were thoroughly mixed, and roughly 1 kg of paddy soil was positioned in a custom-designed glass cylinder (25 cm in diameter, 10 cm tall), topped with a 2 cm layer of water to ensure a flooded state during the incubation, with daily checks conducted. The experimental group, denoted as P-PBAT, involved blending PBAT into the paddy soil at a 1% (w/w) ratio, while a control group (without PBAT) was labeled as P-CK. Gas measurements were taken on the 1st, 4th, 7th, 10th, 14th, and 25th days following the addition of PBAT, with the monitoring and sampling following a fixed sequence.
The CH4/CO2 measurements using FTIR spectroscopy
The schematic diagram of the experiment setup is shown in Fig. 1b. The FTIR spectrometer used was the Bruker V70 (Bruker, Karlsruhe, Germany). For details on the spectrometer’s specifications, performance, and calibration procedures, refer to our previous study14. In this work, a long optical-path gas cell (20 m, Pike Technologies, Fitchburg, United States) installed on the FTIR spectrometer was connected to a static chamber via the plastic gas tube, with gas circulation driven by a mini gas pump.
At the 5th, 10th, and 15th minutes after placing the soil sample into the static chamber, the gas pump was turned on to ensure full gas circulation and to measure spectra. Data acquisition was performed using the spectrometer’s accompanying software. After each measurement, the air in the static chamber and gas cell was refreshed to ensure a clean background, and three replicate experiments were conducted. Gas molecules exhibit specific absorption in different infrared wavelengths, which can be used for both quantitative and qualitative analysis. The identification of CH4/CO2 absorption peaks in the FTIR spectra was based on measurements taken on the first day and compared with the standard CH4 absorption spectra in the National Institute of Standards and Technology (NIST) database23.
Exploration of the CH4 quantification method based on FTIR spectroscopy
The Beer-Lambert Law forms the foundation of FTIR spectroscopy theory24, as shown in Eq. (1).
$$A = \varepsilon \cdot l \cdot c$$
(1)
Where A represents absorbance, ε (cm2·mol−1) is the molar absorption coefficient, l (cm) is the optical path length, and c (mol·cm−3) is the concentration. ε, an intrinsic property of the gas indicating the absorption capacity per unit concentration, is obtained by multiplying the single molecule absorption coefficient, Ka (cm2·molecule−1), by Avogadro’s number (6.023 × 1023 molecules·mol−1). Ka corresponding to different wavenumbers at experimental temperatures can be sourced from the High-Resolution Transmission Molecular Absorption Database (HITRAN)25. Based on this information, we can convert the absorbance measured at the main absorbance peak into concentration values expressed in mol·cm⁻3. Considering the system’s volume (fully accounting for the geometric volumes of the gas cell, static chamber, gas tubing, and the samples) and molar mass, we further determine target gas mass in the circulation system. Lastly, the flux is calculated based on the change in its mass over time, expressed in mg·m2·h−1.
We validated the results by collecting gas and performing GC analysis (Fig. 1c). Briefly, the glass cylinder was tightly sealed with a preservative film for 5 min to equilibrate the headspace. Then, 30 mL of headspace gas was withdrawn using a gas-tight syringe and transferred into a 12 mL evacuated glass vial, marked as vial-1. After another 5 min, the process was repeated to fill a second vial, vial-2. The CH4 concentrations in these samples were determined with a GC system (Agilent 7890–0468, California, USA) equipped with a flame ionization detector and electron capture detector. The GC was calibrated with standard gas after every twelve samples. Finally, the measured concentration data were converted into fluxes over time.
The soil sample analysis using ATR-FTIR spectroscopy
After the incubation experiments concluded, we destructively sampled and dried the soil samples. We used an ATR accessory, the GladiATR Illuminate model (Pike Technologies, Fitchburg, United States), fitted onto the FTIR spectrometer. For specific ATR measurement methods, please refer to our prior work15. In this study, we conducted three measurements each on the dried P-CK, P-PBAT samples, and BPs-PBAT (Fig. 1d).
Statistical analysis
Spectral data collection, preprocessing, peak identification, and area calculation were all conducted in OPUS 6.5 software (Bruker, Karlsruhe, Germany). The correlation analysis of the acquired data and the creation of graphs were performed using Origin 2023b (OriginLab Corporation, Northampton, United States).