Detecting plastic in the environment with ASD Vis-NIR spectroscopy

Introduction

Plastics have shaped the world bringing safety, hygiene, comfort and wellbeing to our society as they are utilized in various end-use markets such as packaging (39.9%), building and construction (19.8%), automotive (9.9%) and electrical and electronic equipment (EEE) (6.2%) ^[1]. With a global plastics production of almost 360 million tons, of which 17% is produced in Europe resulting in an industry turnover of 360 billion euros in 2018 ^{[1, 5]}. However, analyzing plastics comes with its own challenges as they are so diverse in polymer type, color, transparency, thickness, state (pristine, biofouled, weathered, wrinkled), and moisture level (dry, wet, submerged) ^[4].

Most plastics contain organic polymers that are predominantly composed of C, H, O and N forming molecular structures in various combinations. When visible or infrared light interacts with plastics, specific functional groups within the polymers absorb energy at distinct wavelength bands, producing characteristic spectral features. These features arise from the absorption of light by the first overtones of carbon-hydrogen (C-H), nitrogen-hydrogen (N-H), and oxygen-hydrogen (O-H) atomic bonds ^{[9, 2]}.

Malvern Panalytical’s portable Vis-NIR ASD spectrometer line provides the ideal instrumentation for the classification and quantification of plastics. Their high portability and wide range of accessories make them suitable for industry applications such as QA/QC of raw materials or end products. Furthermore, they are ideal for on-site discrimination of unknown plastics and can be used in the field to investigate environmental plastic pollution. Their broad wavelength range of 350 to 2500 nm allows for discrimination of chemically different, but visually similar plastics including ^{[3, 4, 7, 9]}: 

• Polyvinyl chloride (PVC)
• Polyamide or nylon (PA 6.6 and PA 6)
• Low-density polyethylene (LDPE)
• Polyethylene terephthalate (PET)
• Polypropylene (PP)
• Polyester (PEST)
• Thermoplastic elastomer (TPE)
• Polybutylene terephthalate (PBT)
• Polyethylene (PE)
• Polystyrene (PS)
• Teflon fluorinated ethylene propylene (FEP)
• Acrylonitrile Butadiene Styrene (ABS)
• Merlon or polycarbonate (PC)
• Polymethyl methacrylate (PMMA)
• Paraffin
• Polyvinyl fluoride (PVF)

In addition, with a remote sensing configuration, the instrument can be used to optimize and ground-truth overflight spectral imagery from drones, airplanes or satellites to accelerate environmental monitoring and support environmental policymaking. For example, with the Floating Debris Index (FDI) ^{[8, 11]}.

Detecting plastic in the environment with ASD Vis-NIR spectroscopy

Plastic is a versatile material containing specific characteristics that are ideal for its intended application. However, its high versatility and low production costs come with a drawback, as plastics have made their way into the natural environment, causing severe pollution in both terrestrial and marine environments ^{[1, 2, 3, 6, 7, 8, 10]}.

Plastics are durable, lightweight, strong materials that are resistant to corrosion, which is why they are such persistent polluters ^[2]. Over the years global plastic pollution has grown exponentially resulting in an imminent need to combat this environmental problem by applying clean-up campaigns, implementing policies for plastic waste management and sustainable monitoring strategies ^{[1, 3]}. 

One well-known example is the accumulation of plastic debris, primarily in oceans, known as “plastic soup,” which severely threatens aquatic life and ecosystems. It’s estimated that a total of 8 million tons of plastic debris enters the marine environment each year ^[2] and more than 150 million tons of plastics have accumulated in the world’s oceans over the past 100 years ^[8]. Research shows that plastics pollute all parts of the marine environment hence, plastics are not only found at the ocean surface layer but also suspended deeper in the water column, on the seafloor, along coastlines, within estuaries, and on beaches spanning from the Arctic to the Antarctic ^{[2, 6]}. Over the past years, several open-access datasets of various environmentally harvested plastics have been published based on spectral data of ASD spectrometers.

These spectral libraries support identification of relevant absorption features for pristine (or virgin) and naturally weathered plastics and support the detection, identification, and quantification of plastics from other floating debris or spectrally distinctive features of materials naturally present in oceans like wood, algae and seaweed.

ASD studies for plastic pollution

In the context of plastic pollution in the environment, several studies have been conducted that utilized an ASD spectrometer:

• Garaba and Dierssen (2020)^[3] studied spectral reflectance of wet and dry marine-harvested, washed-ashore, and virgin plastics harvested from the major accumulation zones in the Atlantic and Pacific oceans. They identified diagnostic absorption features in plastic debris at 931 nm, 1045 nm, 1215 nm, 1417 nm, 1537 nm, 1732 nm, 2046 nm and 2313 nm. Furthermore, they noted that wet marine harvested microplastics displayed lower spectral reflectance but similar spectral shape to dry marine harvested microplastics. Diagnostic absorption features common in both the marine-harvested microplastics and washed-ashore plastics were identified at 931, 1215, 1417 and 1732 nm.
• Guffogg et al. (2021)^[2] measured different plastics mixed with beach sand and showed that, depending on the plastic type, plastic could already be detected when between 2 and 8% plastic was present in the beach sand.
• Moshtagi et al. (2021)^[8] presented a study that focuses on virgin and naturally weathered plastics submerged in in a controlled environment simulating clear to turbid waters at varying suspended sediment concentrations and depths. They identified the following spectral features for plastic identification in suspension (see table below). In addition, Moshtagi et al. suggest that discriminating wood from solid plastics in the SWIR might be possible with the 1729 nm absorption feature that is absent in wood.

• Knaeps et al. (2021)^[7] confirmed the findings from Mosthagi et al. by performing a similar experiment that included submerging plastic to a maximum depth of 32 cm with various increments. They identified the strongest absorption features around 1216, 1397 and 1730 nm.
• Olyaei and Ebtehaj (2024)^[10] investigated three distinct wavelength selection techniques (i.e., sparse variable selection, hierarchical clustering, and density peak clustering) to identify important wavelengths capturing the optical spectral signatures of plastic litter. They identified three important wavebands were detected by all three methods: 450–470, 650–690, and 1050–1100 nm, with the wavelengths around 650–680 and 820 nm considered the most important wavelengths for the reconstruction of wet plastic spectra. Further, they demonstrated that the presence of suspended sediments impacts the important wavelengths by shifting them to longer (shorter) wavelengths by slightly less than 50 nm in the VNIR (SWIR) regions of the spectrum.
• Huda et al. (2023)^[6] successfully developed a model for beach sediment spiked with virgin microplastic pellets by combining predictive PCA regression and machine-learning linear regression models. They developed their best linear regression models for LDPE, PET and ABS using with R2 values of 0.83, 0.66 and 0.86 and RMSE values of 1.9, 2.7 and 1.7, respectively.

Conclusion

The research summarized in this application note demonstrates that spectral analysis using ASD portable Vis-NIR spectrometers provides a promising method for identifying and classifying plastics in various environmental conditions. Plastics exhibit distinct spectral absorption features, allowing for differentiation with Vis-NIR spectrometers from other debris or natural materials like wood.

Our instruments can be used to distinguish, identify or quantify plastics whether they’re pristine or weathered, dry, wet or submerged. Furthermore, advancements in remote sensing, combined with publicly available spectral libraries, offer valuable tools for environmental monitoring and policymaking. 

References

1. Plastics Europe, 2019. Plastics—THE FACTS 2019. An Analysis of European Plastics Production, Demand and Waste Data 2019. Available online: https://plasticseurope.org/wp-content/uploads/2021/10/2019-Plastics-the-facts.pdf.
2. Guffogg, J. A., Blades, S. M., Soto-Berelov, M., Bellman, C. J., Skidmore, A. K., & Jones, S. D. (2021). Quantifying marine plastic debris in a beach environment using spectral analysis. Remote Sensing, 13(22), 4548.
3. Garaba, S. P., & Dierssen, H. M. (2020). Hyperspectral ultraviolet to shortwave infrared characteristics of marine-harvested, washed ashore and virgin plastics. Earth System Science Data, 12(1), 77-86.
4. Leone, G., Catarino, A. I., De Keukelaere, L., Bossaer, M., Knaeps, E., & Everaert, G. (2022). Hyperspectral reflectance dataset of pristine, weathered and biofouled plastics. Earth System Science Data Discussions, 2022, 1-24.
5. Moroni, M., & Mei, A. (2020). Characterization and separation of traditional and bio-plastics by hyperspectral devices. Applied Sciences, 10(8), 2800. 
6. Huda, F. R., Richard, F. S., Rahman, I., Moradi, S., Hua, C. T. Y., Wanwen, C. A. S., ... & Müller, M. (2023). Comparison of learning models to predict LDPE, PET, and ABS concentrations in beach sediment based on spectral reflectance. Scientific Reports, 13(1), 6258.
7. Knaeps, E., Sterckx, S., Strackx, G., Mijnendonckx, J., Moshtaghi, M., Garaba, S. P., & Meire, D. (2021). Hyperspectral-reflectance dataset of dry, wet and submerged marine litter. Earth System Science Data, 13(2), 713-730.
8. Moshtaghi, M., Knaeps, E., Sterckx, S., Garaba, S., & Meire, D. (2021). Spectral reflectance of marine macroplastics in the VNIR and SWIR measured in a controlled environment. Scientific Reports, 11(1), 5436.
9. Eisenreich, N., & Rohe, T. (2006). Infrared spectroscopy in analysis of plastics recycling. Encyclopedia of Analytical Chemistry: Applications, Theory and Instrumentation. 
10. Olyaei, M., & Ebtehaj, A. (2023). Uncovering Plastic Litter Spectral Signatures: A Comparative Study of Hyperspectral Band Selection Algorithms. Remote Sensing, 16(1), 172.
11. Biermann, L., Clewley, D., Martinez-Vicente, V., & Topouzelis, K. (2020). Finding plastic patches in coastal waters using optical satellite data. Scientific reports, 10(1), 5364.