Grantee Research Project Results
Final Report: Cost-effective, portable and automated platform for microplastics characterization
EPA Contract Number: 68HERC21C0043Title: Cost-effective, portable and automated platform for microplastics characterization
Investigators: Batalin, Maxim
Small Business: Lucendi, Inc.
EPA Contact: Richards, April
Phase: II
Project Period: April 1, 2021 through March 31, 2023 (Extended to July 31, 2023)
RFA: Small Business Innovation Research (SBIR) - Phase II (2021) Recipients Lists
Research Category: Endocrine Disruptors , SBIR - Water , Small Business Innovation Research (SBIR)
Description:
Our society generates large quantities of plastic waste, which has quickly become one of the most significant environmental pollutants, especially due to its long bio-degradation lifecycle. Large quantities of plastic waste find its way into aquatic environments, including lakes, rivers, underground streams and oceans. Every year over 8 million tons of plastic flows into the ocean causing an estimated $13 Billion damage to marine ecosystems. Microplastic particles (plastic particles and debris with diameters of less than 5mm) constitute by far the largest class of plastic pollution in the open ocean with recent estimates accounting for as much as 90% of all plastic litter.
In 2017, the EPA Trash Free Waters program has conducted a Microplastics Expert Workshop focusing on prioritizing scientific goals and objectives required to understand the harmful impacts of microplastics to human and ecological health. This report concludes with a summary including the identified method needs. Some of such key sought after methods include:
- Methods for microplastics characterization, i.e. by size, shape and chemical composition;
- Methods for microplastics quantification, particularly for particles in the microns scale (i.e. ≥1 µm and ≤1 mm in size) for which information is limited and which are relevant to human and ecological exposure risks.
Summary/Accomplishments (Outputs/Outcomes):
Because microplastics objects are small, they are difficult to detect, characterize and quantify in water samples, let alone in the open aquatic environment conditions. Generally, multiple steps are required to first find micro plastic particles, isolate them for analysis, identify and classify in terms of the type of plastics or its chemical composition. For particles smaller than 1mm, different types of microscopy characterization is typically used (e.g. stereo microscopy, Scanning Electron Microscopy, etc.). Once the microplastic particles are detected and counted, the specific type of plastic they are made of can be identified with the use of spectroscopy, atomic force microscopy or fluorescence microscopy methods. However, all of these techniques are extremely low throughput, laborious, expensive, and require expertise to perform and operate. Furthermore, such devices are bulky and not portable, limiting their application to analyzing samples in the laboratory conditions. This creates significant limitations to the use of this microplastics characterization technology and to the types of applications or contexts in which they can be applied.
In Phase II program we have created an integrated and environmentally protected package for an aqusens-MP system that can be used in-lab and in-field for rapid assessment of water for the presence of microplastic particles. The in-field configuration of the aqusens-MP system fits in a dedicated backpack and can operate on a portable battery capable of supporting data collection for up to 4 hours in-field (or longer depending on the battery), while providing power for the device, along with an embedded computing platform and a touchscreen display.
The aqusens-MP system was also integrated with an optional polarizer imaging module that provides additional features useful in identifying and characterizing microplastic particles. Information from this module was demonstrated to be useful not only for identifying microplastics, but also introducing opportunities for further analysis to differentiate the types of plastics.
Data analytics capabilities of aqusens-MP include a particle classification system based on conventional machine learning technique – a Support Vector Machine (SVM) relying on over 750 developed features. We have also implemented and trained DenseNet neural network models in continuation of our work presented in Phase I.
Conclusions:
Experimental results demonstrated aqusens-MP capable of accurately differentiating microplastics from other particles – equally well with and without a polarizer present. Furthermore, the trained models were also demonstrated to be capable of tracking plastic concentration changes in water and in water solutions of different types (i.e., ocean water, DI, tap water, etc.).
Initial analytical work on determining more detailed characteristics of microplastics, such as determining the type of plastics, has shown promising results, but will require additional work in further refining both the current algorithms and the training data used for Lucendi’s neural network model training.
Finally, at a conclusion of the program an experiment was performed on a large dataset of microplastics particles created at Lucendi’s facilities, as well as provided by a third party collaborator. The machine learning algorithms of aqusens-MP were demonstrated to be capable of differentiating between the microplastic particles and other objects in the sample, including organic and inorganic objects, with ~98% accuracy. We believe that this result demonstrates capabilities of the current aqusens-MP platform to be used for high-throughput screening for the presence of microplastics in water. Lucendi team will continue its work on further refining the aqusens-MP system on expanding its analytics package and on further in-field system evaluation. Furthermore, Lucendi team is focusing on building partnerships, collaborations and on commercially relevant applications.
SBIR Phase I:
Cost-Effective, Portable and Automated Platform for Microplastics Characterization | Final ReportThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.