Comparison of Benchtop and Portable Near-Infrared Instruments to Predict the Type of Microplastic Added to...

Area of Science:

• Analytical Chemistry
• Food Science
• Environmental Science

Background:

• Near-infrared (NIR) spectroscopy is a key analytical technique in food and agriculture.
• Microplastic (MP) contamination is a growing concern in food safety.
• High-moisture foods like fruits and vegetables present unique challenges for MP detection.

Purpose of the Study:

• To compare two NIR instruments for classifying microplastic addition in high-moisture food samples.
• To develop quantitative models for predicting specific microplastic types (PE, PP, Mix) in spinach and banana.
• To evaluate the performance of NIR spectroscopy in detecting MPs in complex food matrices.

Main Methods:

• Two NIR instruments (benchtop Bruker Tango, portable MicroNIR) were used.
• Samples included spinach and banana mixtures with Polyethylene (PE), Polypropylene (PP), or a PE+PP mix.
• Data analysis involved Principal Component Analysis (PCA) and Partial Least Squares (PLS) regression.

Main Results:

• PLS regression models showed good predictive ability for MP addition, with R²_CV values up to 0.88 for the benchtop instrument.
• The portable instrument demonstrated moderate success (R²_CV = 0.54) in MP detection.
• Specific wavenumber regions were evaluated, showing comparable performance for short and long wavelengths on the benchtop instrument.

Conclusions:

• NIR spectroscopy is a viable tool for detecting microplastic contamination in high-moisture food samples.
• Instrument choice impacts MP detection accuracy, with benchtop models showing superior performance.
• This research supports the use of NIR for both food quality assessment and microplastic monitoring.