Detection of Spoilage Markers in Food Products using a Mass-Spectrometry Based Chemical Sensor

Relevance of the Topic

Rapid and reliable detection of food spoilage is crucial to ensure safety, reduce waste and maintain consumer confidence. Volatile compounds produced during microbial or enzymatic degradation serve as early indicators of spoilage, allowing non-invasive monitoring via headspace analysis. A mass-spectrometry-based chemical sensor offers high sensitivity, immunity to humidity and temperature fluctuations, and minimal cross-interference, making it an attractive tool for real-time quality control in food processing and distribution.

Objectives and Study Overview

This study evaluates a mass-spectrometry sensor system for detecting key spoilage markers in five food matrices: chicken, eggs, orange juice, milk and fish. Specific analytes spiked into each matrix include dimethyl sulfide, diacetyl and trimethylamine at varying ppm levels. The goals are to assess the system’s ability to classify food type, identify presence of spoilage marker, and quantify its concentration accurately and quickly.

Instrumentation

A Gerstel ChemSensor 4440 platform was used, equipped with:

• An Agilent 7694 headspace sampler
• An Agilent 5973 mass selective detector operating in full-scan mode (35–250 amu for chicken and eggs; 41–180 amu for other matrices)
• Infometrix chemometric software (Pirouette 3.02 and InStep 2.0) for data processing

Methodology

Sample preparation involved spiking each food product with known concentrations of the target marker, equilibrating at 80 °C for 20 minutes, then introducing the full headspace into the detector without chromatographic separation. Multivariate data analysis workflows comprised:

• Exploratory analysis: Principal component analysis (PCA) and hierarchical cluster analysis (HCA) to assess data structure and group differentiation
• Classification models: Soft independent modeling of class analogies (SIMCA) and k-nearest neighbors (KNN) to identify sample type and marker presence
• Regression: Partial least squares (PLS) to predict marker concentration levels

Main Results and Discussion

PCA and HCA revealed clear clustering by food type, with chicken and egg headspaces showing similar profiles. Classification by SIMCA and KNN achieved over 95% accuracy in training and reached 100% correct prediction in blind test sets. PLS models quantified marker levels effectively in the 10–100 ppm range for all matrices except fish at the lowest concentration. Fish required higher trimethylamine levels (≥100 ppm) for reliable detection. Subtraction of non-spiked and spiked spectra confirmed marker signals down to 10 ppm in most cases.

Benefits and Practical Applications

• Fast analysis under two minutes per sample without chromatographic columns
• High specificity to targeted spoilage compounds even in complex headspace mixtures
• Robust performance across varying humidity and temperature conditions
• Potential integration into production lines for on-site quality monitoring in poultry, dairy, fish and beverage industries

Future Trends and Potential Uses

Enhancements may include operating in selected ion monitoring (SIM) mode to boost sensitivity below 10 ppm. Advanced sample introduction methods such as solid-phase microextraction (SPME) or stir bar sorptive extraction (SBSE) could further lower detection limits. Expanded libraries of spoilage markers and machine-learning classification could enable comprehensive spoilage profiling and shelf-life prediction in real time.

Conclusion

This work demonstrates the feasibility of a mass-spectrometry-based chemical sensor for rapid, accurate detection and quantification of spoilage markers in diverse food products. The approach combines speed, sensitivity and robustness, supporting its adoption for quality assurance and safety monitoring in food supply chains.

References

[1] Goodner K, Russell R, Journal of Agricultural and Food Chemistry 2001, 49, 250–253
[2] Marsili RT, Journal of Agricultural and Food Chemistry 1999, 47, 648–654
[3] Freeman LR et al., Applied and Environmental Microbiology 1976, 32, 222–231
[4] Hatcher WS, New York State Agricultural Experiment Station Special Report 1979, 31(8), 1–4
[5] Natale D et al., Sensors and Actuators B 2001, 77, 572–578