Grantee Research Project Results

Final Report: Real-Time Monitoring Using Infrared Lasers and Ultraviolet LEDs to Prevent Food Waste

EPA Contract Number: 68HERC20C0039
Title: Real-Time Monitoring Using Infrared Lasers and Ultraviolet LEDs to Prevent Food Waste
Investigators: Yeak, Jeremy
Small Business: Opticslah, LLC
EPA Contact: Richards, April
Phase: I
Project Period: March 1, 2020 through August 31, 2020
Project Amount: $100,000
RFA: Small Business Innovation Research (SBIR) - Phase I (2020) RFA Text |  Recipients Lists
Research Category: Small Business Innovation Research (SBIR) , SBIR – Sustainable Materials Management

Description:

To monitor the freshness of fresh produce in storage and during transportation, a gas sensor based on swept-wavelength broadly tunable external cavity quantum cascade lasers (swept- ECQCLs) has been demonstrated. The sensor detects, identifies, and quantifies multiple chemical species in the airspace near the sensor continuously and in real-time, with part-per- billion level sensitivity. The sensor will monitor the air in food storage areas for key gases including ethylene, alcohols, ketones, acetates, amines, and more produced by fresh fruits and vegetables (FF&Vs) as they ripen and before they spoil. Using our proposed monitoring system, the airspace around produce in storage will be measured continuously to detect when multiple key gases reach threshold levels indicating that spoilage is imminent. Mitigating steps can then be taken to (1) notify the operators of the storage facility to prioritize use of the produce near spoilage and (2) extend the shelf life of produce by applying deep ultraviolet air sterilization to breakdown air pathogens and volatile organic compounds without changing the nutritional value of the produce.

Summary/Accomplishments (Outputs/Outcomes):

In this Phase I work, we used our CANARIES (Chemical ANalyzer of Atmosphere with Real- time Infrared ECQCL Spectroscopy) gas sensor to measure various volatile organic compound (VOC) gases, as well as inorganic gases including ammonia (NH3) and nitrous oxide (N2O) emitted from various fresh fruits and vegetables (FF&V) as they ripened and spoiled. A laboratory prototype sensor was constructed for the Phase I experiments. The ECQCL used a QCL device designed for broadband operation in the LWIR (long wave infrared) and provided an overall wavenumber tuning range of 935-1195 cm^-1 (8.37-10.70 µm wavelength range). An annular multi-pass optical cell was used to increase the interaction length (up to 60 meters) of the ECQCL beam with the gases. The sensor measures the infrared absorption spectrum of gases inside the multi-pass cell, and measured absorbance spectra are analyzed using a quantitative weighted least squares (WLS) algorithm to detect, identify, and quantify the gas composition.

The high spectral resolution and sensitivity of the CANARIES sensor provides high-confidence detection of multiple gas species even when present in complex mixtures. The output of the CANARIES sensor is a continuous record of measured gas concentrations as a function of time, which is then used to make further decisions based on threshold levels or identifying trends over time.

In Phase I, an extensive test series was performed to measure up to 25 different gases emitted from a set of 19 different FF&Vs. Significant changes in gas concentrations were detected as the FF&Vs ripened and spoiled, and a set of 6 primary VOCs were identified as key indicators of food health and/or spoilage, with other VOCs noted as potential secondar indicators. Emission factors (EFs) for the VOCs were determined to characterize the food spoilage and enable predictions of future performance. Although ethene – the gas that has been used traditionally to measure the ripening process of climacteric fruits – was detected for many of the FF&Vs, our experiments show that methanol, ethanol, isopropyl alcohol, ammonia, and nitrous oxide gases are key indicators for spoilage in a much wider range of FF&Vs.

In our experiments, methanol and ethanol were observed to increase significantly for nearly every fruit or vegetable tested. The ratio between methanol and ethanol varied depending on food type, with ethanol typically higher in foods with higher sugar content (fruits) and methanol higher in foods with lower sugar content (leafy greens). The ratio between methanol and ethanol also varied during spoilage and may provide a secondary indicator of the degree of spoilage due to fermentation-related processes.

An increase in ammonia emission was detected for 9 of the foods and was also found to be correlated with spoilage, especially for leafy greens. An increase in ammonia may be a useful indicator of decay for cases where the ethanol/methanol emission is low or absent. It may also serve as a useful indicator in combination with ethanol/methanol to better determine the type or degree of spoilage.

Isopropyl alcohol (IPA) and nitrous oxide (N2O) were detected at high levels from some foods and corresponded to a high degree of food decay. Detection of isopropyl alcohol and nitrous oxide may be useful when combined with methanol, ethanol, and ammonia detection to determine the degree or type of spoilage.

Emission of ethene (C2H4) – also commonly known as ethylene – was detected for eight of the foods but was not directly correlated with the onset of spoilage. Ethene emission was detected not only from climacteric fruits (avocado, banana, apple, tomato) but was also observed for non- climacteric foods (grapes, raspberries, and kale). Sensitivity to ethene for promoting ripening in climacteric food is not a requirement for ethene emission by a food. Non-climacteric foods may be high ethene emitters, and other foods may show an increased spoilage rate when exposed to ethene. Overall, while ethene is an extremely important VOC to measure for control of ripening and spoilage, it may not be the best indicator of the degree of spoilage.

Other VOCs including acetaldehyde, ethyl acetate, and acetone were detected from some of the foods during spoilage. Acetone emission appeared to show the largest increase during spoilage of some foods. Acetaldehyde and ethyl acetate were detected for many foods, but the correlations with spoilage were not as clear as for the previously noted VOCs. Overall, these and other VOCs will probably be useful as indicators of food quality or degree of spoilage but will most likely need to be combined with other VOCs as primary spoilage or health indicators.

Water (H2O) and carbon dioxide (CO2) are also measured continuously by the CANARIES sensor, and they serve as an important indicator of the food storage environment. It was also verified that high concentrations of water and carbon dioxide measured by the CANARIES sensor do not interfere with measurement of the other lower-concentration VOCs of interest.

In our Phase 1 experiments, the efficacy of UVC radiation for sterilization to prolong the shelf- life of FF&V remains inconclusive. While direct UVC radiation may sterilize and inhibit mold and fungal growth on the outer skin of FF&V, it may induce tissue browning due to the oxidation of phenolic compounds into melanin, which is a brown pigment, if the UVC dosage is high. For indirect UVC radiation where only the gases emitted by the FF&V were exposed to UVC but not the FF&V themselves, some changes in the detected gases were observed although the overall spoilage rate remained inconclusive due to other uncontrolled environmental parameters, such as relative humidity, temperature and air flow rate.

Additional experiments were performed to determine measurement sensitivity and detection limits for the VOCs. The results show that all VOCs of interest are detected with high- confidence (10 σ) at sub-ppm concentrations when averaged over a 10 s time period. Predictions of performance for detection of VOCs in a model food storage facility were made based on the measured detection limits and emission factors for the FF&Vs and show that the measurement concept is feasible. High levels of ethanol and ammonia observed from some of the FF&Vs as they spoiled in enclosed containers was found to be a potential issue which may limit detection of trace VOCs also present. However, this issue was more a function of the specific experiments performed in the Phase I research, and we do not expect such high ethanol and ammonia concentrations to be present in actual storage conditions and facilities.

Conclusions:

The emission of ethene has traditionally been used to monitor the ripening process of climacteric fruits (i.e., fruits that continue to ripen after they are harvested). However, not all fruits and vegetables are sensitive to ethene and may only produce a very small amount of it when they spoil. These include citrus fruits, potatoes, pineapple, blueberries, cherries, cauliflower, radishes, kale and turnips. Consequently, the detection ethylene gas may not be most suitable to monitor the freshness of these FF&V. Our experiments show that low-weight volatile organic compounds (VOCs), such as methanol, ethanol, and ammonia may be a better indicator of food spoilage under anerobic conditions. Secondary indicators of food spoilage included isopropyl alcohol and nitrous oxide. The presence of acetaldehyde, ethyl acetate, and acetone may also be used to determine the aging and health of the fruits and vegetables (tertiary indicators). Ethene is useful for measuring the storage conditions of the FF&V and monitoring ripening. Additionally, water and carbon dioxide levels emitted by FF&V may also be monitored for any changes in overall atmospheric conditions during transport and storage.

We have demonstrated that our CANARIES gas sensor is capable of detecting, identifying and quantifying these multiple gases of interest for food storage in real-time, which will provide invaluable information to various stakeholders in the food supply and food distribution chain to ensure the delivery high quality fresh fruits and vegetables from farms to end consumers, minimizing food waste that may result from undetected spoilage and/or bacteria contamination.

In this SBIR Phase 1 work, we have used our CANARIES gas sensor, based on swept-frequency broadly tunable external cavity quantum cascade laser (ECQCL), to measure the gases that are emitted by variety of fresh fruits and vegetables (FF&V). What sets our ECQCL gas sensor technology apart is its ability to detect, identify and quantify multiple gases simultaneously with sub-ppm sensitivity in real-time. The ultralow detection limits of our gas sensor will not only allow FF&V transport and/or storage facility companies to monitor the ripening process but also to detect any early onset of rot or spoilage due to unexpected changes in environmental conditions (temperature, moisture, accumulation of ethylene gas) or bruising during transport. This can be particularly advantageous given that the damage or rotting FF&V may not be easily observed by the operator or handler.

Current technologies for produce monitoring focus almost exclusively on ethylene detection. Our technology will improve on existing techniques by detecting multiple additional gases simultaneously and continuously at much lower concentrations, and by doing so will be applicable to nearly all food types to detect pre-spoilage conditions in storage. The potential market size for this technology is estimated to be > $1B, assuming even a modest reduction in food waste of > 1%.

We have identified two different sensing modalities (point sensor and open-path sensor) that may benefit the FF&V industry. The point sensor may be better suited for smaller and dynamic operations and during transport where different shipments are moved from one facility to the other. This point sensor may also be moved from one zone to another zone within the same facility to test different stored produce. For larger storage facilities or at hypermarkets where various FF&V produce are monitored for freshness and quality, a static open-path sensor may be installed near (or in) the attic space to provide real-time monitoring and analysis. An open-path sensor may also be relevant to monitor the growth of fruits and vegetables in large greenhouse operation or open fields to determine the optimal time for harvesting.

The CANARIES sensor will be useful for a wide range of gas sensing applications in addition to monitoring FF&Vs for spoilage and reducing food waster. Other applications for real-time gas sensing include detecting toxic gases for worker safety and health, monitoring indoor and outdoor air quality, detecting pollution and greenhouse gases, measuring input/output gases in chemical processes or manufacturing, monitoring gases in combustion systems, and measuring gases relevant for medical applications, among others.