The environmental behaviors and health risks posed by microplastics (MPs, 1 μm–5 mm) and nanoplastics (NPs, <1000 nm) have emerged as the most pressing issues in environmental science. As research continues to unfold, it has become increasingly evident that these tiny pollutants not only pose threats to human health, ecosystems, and biodiversity but also have the potential to alter the global material cycle profoundly1,2. Due to the variety of shapes, chemical properties, and wide particle size range of nano-/micro-plastics (NMPs), the new particulate pollutant has presented significant challenges to traditional detection methods for small molecule organic contaminants. In recent years, scientists have navigated these challenges by integrating insights from diverse disciplines (materials science, biology, chemistry, etc.) into developing advanced detection technologies tailored to NMPs’ unique physicochemical properties3,4,5. However, the real sample analysis often encounters challenges due to the complex environmental matrices, which can lead to false positives or negatives, alongside increased instrument failure rates and maintenance costs6.

Pyrolysis gas chromatography-mass spectrometry (Py-GC-MS) has gained significant recognition in the environmental science community for its ability to provide direct mass concentration measurements, unaffected by NMPs’ size limitations, and its strong resistance to interferences7,8. Compared with optical spectrum techniques, Py-GC-MS has revolutionized our ability to identify NMP types and quantify their presence in human tissue and environmental samples9,10,11,12. Recent findings reveal that NMPs, especially at the nanoscale, can penetrate biological barriers, reaching the arteries, bone marrow, and other human tissues13,14,15. Small-sized NMPs in environmental samples, such as in PM2.5 and PM10, pose a serious threat to human health10,16, further challenging the detection capabilities of techniques other than Py-GC-MS. Meanwhile, Py-GC-MS can directly provide the mass concentration in the target samples, thereby eliminating the uncertainty associated with the process of converting quantitative concentrations17. Overall, Py-GC-MS has demonstrated remarkable advantages in the NMPs screening phase, particularly in the detection of low-concentration and small-sized NMPs, as well as in the analysis of complex sample matrices. Addressing NMP pollution is a pressing global concern, underlining the urgent necessity for standardized monitoring protocols. Establishing these standards will enable us to produce comparable data, furthering our understanding of NMPs’ role in the global material cycle and their broader ecological impacts.

The development of NMP detection via Py-GC-MS has reached a stage where the core detection principles remain consistent, although the specific methods vary slightly. In the helium atmosphere of 600 ± 50 °C, polymers undergo thermal cleavage into various low molecular weight products (<500 m/z) within seconds9,10,18. Importantly, under the same stable conditions, these pyrolysis products maintain a consistent linear relationship with the quantity of the polymer. This relationship is crucial for employing Py-GC-MS in the quantitative analysis of NMPs in environmental samples10,12. Thus, we propose guiding principles for future Py-GC-MS-based NMP detection efforts, which focus on preparing compliant samples and establishing Py-GC-MS analytical protocols.

As illustrated in Fig. 1, the Py-GC-MS-based NMPs analysis process (PNAP) of environmental samples is divided into two main stages: sample preparation and Py-GC-MS analysis. To address commonly shared environmental concerns, samples can be categorized based on their state of matter and matrix characteristics into water, soil, atmosphere, and biological samples (including human tissue samples)10,12,19,20,21. Each of these types requires specific sample pre-treatment processes to meet the detection requirements of Py-GC-MS analysis. Direct analysis is feasible for atmospheric samples, which often have low matrix content10; however, using non-plastic materials, such as sampling membranes or tubing, during sampling to prevent contamination is crucial. Soil samples, especially those collected from agricultural or forested regions, are abundant in organic matter that must be thoroughly digested and removed12. This step is crucial because the presence of organic material can interfere with the detection of pyrolysis products and quantitative ions. Additionally, it poses a risk of contaminating the pyrolizer and ion source of Py-GC-MS, which can result in increased uncertainty in measurements and elevated equipment maintenance costs12. Similarly, seawater and coastal sediment samples contain significant amounts of inorganic salts that must be thoroughly removed22. Due to the considerable concern over the potential health risks posed by NMPs, removing organic molecules such as proteins and lipids from biological samples is a prerequisite for accurately quantifying NMPs5,21. In PNAP, achieving maximized extraction efficiency of target NMPs polymers necessitates systematic elimination of three predominant interfering agents: (1) chemical homologs inducing spectral overlap through structural analogies, (2) ionic species causing detector saturation via charge-shielding effects, and (3) natural biopolymers altering pyrolysis kinetics through competitive matrix interactions.

Fig. 1: Py-GC-MS-based NMPs analysis process (PNAP).
figure 1

The two components of Py-GC-MS-based NMPs detection, include environmental sample preparation and Py-GC-MS analysis.

Full size image

Given the varying research objectives, the inherent heterogeneity of samples, and the challenges in standardizing procedures, it is essential to focus on the rationality and stability of sampling and pre-treatment methods. As polymorphic emerging contaminants, NMPs exhibit complex physicochemical profiles spanning multiple dimensions and surface functionalities. Current methodological approaches for particulate NMP isolation demonstrate concerning operational variability across laboratories, with extraction efficiencies fluctuating by 7–116%, meanwhile depending on matrix complexity (soil, aqueous vs. human tissue systems)21,23,24,25. Ensuring these aspects are logical, stable, and operational is crucial. However, NMP measurements using Py-GC-MS should be standardized to enhance data comparability across different studies. To achieve this, we propose four recommendations:

  1. (1)

    “Dissolve” instead of “Disperse” for the calibration curve: the weighing method and solvent dispersion method (water or ethanol) for calibrating the standard curve of NMPs present several issues21,26,27, including complex procedures, poor linearity, and a high limit of quantification. Utilize organic solvents to dissolve plastics into polymer solutions and conduct gradient dilutions to create standard curves, significantly reducing complexity and operational costs21. The solvent selection will be guided by established practices in the polymer field28. For polymers that present challenging dissolution conditions, it is preferable to use nano-sized NMPs suspended in water or ethanol for gradient dilution. The smaller size allows for more precise control over the mass concentration of NMPs.

  2. (2)

    Developing the scalability of pyrolysis methods based on Py-GC-MS: implementing thermal desorption at 300 °C before pyrolysis can effectively eliminate residual small molecule interferences, thus improving data quality. Certainly, thermal desorption can also be employed as a complementary technique for measuring volatile organic pollutants in the sample, such as plastic additives, thereby enhancing the overall efficiency of the detection10. For thermally unstable polymers, such as degradable plastics, reducing the desorption temperature and time is important to ensure proper recovery of the target polymers25. Moreover, under real environmental conditions, biodegradation or combustion processes cause plastic polymers to form more stable small molecules, akin to cracking products29,30. Consequently, GC-MS may be unable to differentiate some of these small molecules further. Therefore, the combination of thermal desorption and co-identification of multiple pyrolysis products is necessary to characterize the plastic polymers effectively12. It is noteworthy that analyzing the same sample not only with a “double-shot” but also with a “multi-shot” approach, based on the thermal behavior of the products at varying temperatures, maybe a prudent choice10,31.

  3. (3)

    Standardized criteria for quantitative product screening: NMPs typically yield multiple pyrolysis products10,12,32, as shown in the Py-GC-MS Analysis part of Fig. 1. The most characteristic pyrolysis products, with nearly 100% recovery as possible, should be used for quantification, while other pyrolysis products serve as qualitative references. One quantitative ion and two/three qualitative ions should correspond to the characteristic ions of the distinct fragments, avoiding common fragments like aromatic hydrocarbons and benzene rings. Low molecular weight fragments are often susceptible to matrix interference and should be carefully selected. Notably, the screening principles outlined above apply not only to conventional plastics, such as polyethylene, polyvinyl chloride, and polystyrene, but also to degradable plastics, including polylactic acid, polyhydroxy alkanoic acid, and polybutylene succinate10,25. It is crucial to highlight that optimizing the pyrolysis conditions is essential for both categories of plastics to ensure reliable and reproducible results in Py-GC-MS analyses. This optimization process involves carefully controlling factors such as pyrolysis temperature, heating rate, and residence time to achieve accurate characterization of NMPs.

  4. (4)

    Optimizing data processing methods to apply to more polymer analysis: multivariate statistics, machine learning, and deep learning can be effectively integrated into the process of data analysis of Py-GC-MS, playing a crucial role in improving both the quality and accuracy of the results33. For each plastic polymer, the presence of multiple pyrolysis products and characteristic ions shapes the data obtained from Py-GC-MS analysis. It is essential to fully leverage this information to improve automated, standardized, and high-throughput data processing procedures. Furthermore, tackling the removal of complex matrix background interference remains a significant challenge in the quantification of NMPs using this technique. Specifically, combining chemical fingerprint information of polymers based on Py-GC-MS, artificial intelligence (AI) is instrumental in addressing challenges such as environmental matrix interference and mutual interference among target polymer pyrolysis products33. Additionally, it enhances the sorting and identification of quantitative and qualitative ions, thereby streamlining the overall data processing workflow and ensuring more reliable outcomes.

Py-GC-MS has undeniably become the “star” technology for NMP measurements. However, it still presents a significant “barrier to entry” for many research groups aiming to investigate the environmental behaviors of NMPs via this method. This emerging particulate pollutant, unlike traditional pollutants, possesses unique characteristics that affect both sample pre-treatment and instrumental testing processes. For instance, the preparation of standard curves, optimization of pyrolysis conditions, selection of pyrolysis products, and identification of quantitative ions of NMPs constitute a systematic endeavor. Consequently, researchers are navigating this complex landscape incrementally, akin to crossing a river by feeling for stones. Several early exploratory studies utilizing Py-GC-MS for the detection of NMPs demonstrated significant objective shortcomings concerning the environmental behaviors and research needs related to NMPs. Additionally, specific issues identified in these initial studies continue to persist in current research efforts. Therefore, during this crucial phase in developing NMP detection technology via Py-GC-MS, we strongly advocate for the gradual harmonization of standards for its usage and the principles governing data comparison. As emphasized previously, the standardization and stability of the methodology are paramount for achieving credible and comprehensive insights across the entire NMP analysis process. This encompasses all stages, from sample collection and pre-treatment to Py-GC-MS analysis. Such rigorous methodological practices ensure both the reliability of the results and their accurate interpretation, thereby advancing our understanding of NMP contamination and its environmental impact. This initiative represents a pivotal step forward in addressing the pressing environmental challenge of NMPs, highlighting the importance of continued research and global collaboration.