Grantee Research Project Results
Final Report: Urban Food Lab: Composting Plastic in Aquaponics
EPA Grant Number: SU840144Title: Urban Food Lab: Composting Plastic in Aquaponics
Investigators: Silverman, Andrea , Gowayed, Omar , Acklin, Josh , Moosa, Tahany , Moratos, Angelica , Sookchan, Savannah , White, Gianna , Charytan, Natan , Ryu, Chaehyun , Gupta, Yams
Institution: New York University , Mount Sinai School of Medicine
EPA Project Officer: Spatz, Kyle
Phase: I
Project Period: December 1, 2020 through November 30, 2021 (Extended to November 30, 2022)
Project Amount: $24,972
RFA: P3 Awards: A National Student Design Competition Focusing on People, Prosperity and the Planet (2020) RFA Text | Recipients Lists
Research Category: P3 Awards , P3 Challenge Area - Sustainable and Healthy Communities
Objective:
Previous studies have illustrated that mealworms can vermi-compost polystyrene (PS) in petri-dishes. We therefore asked whether vermi-composting of PS using mealworms can be conducted at a larger scale and in an economical fashion for two goals: (i) to mitigate PS waste, and (ii) to produce a compost product that could be used as an amendment for a vertical farm system. Therefore, the objectives of this project were to (1) build a mealworm-based vermi-composting system that allows degradation of food waste and plastics that are byproducts from an aquaponic farm on campus, (2) evaluate the decomposition of plastics within the vermi-composting system, (3) evaluate the existence of plastics in finished compost products with spectroscopic and other analytical methods, and (4) conduct toxicity studies of the resulting compost product.
Summary/Accomplishments (Outputs/Outcomes):
Our original proposal was to investigate the toxicity of compost produced from the vermi-composting of plastics using biological specimens such as mealworms, fungi, and bacteria. Due to the COVID-19 pandemic, we had to delay the start of the project due to campus access restrictions, and reduce the scale of our proposed research to accommodate for social distancing requirements. Therefore, we elected to focus our investigation on the role of mealworm degradation of polystyrene foam (PS; i.e., Styrofoam). Through several iterations of design, we were able to create an enclosure and experimental setup for the mealworms that limited worm escape and mold, and allowed for controlled evaluation of vermi-composting rates and products.
After developing an experimental system for the study, we conducted vermi-compost trials using three different feedstocks: (1) polystyrene foam waste (PS) from packaging, (2) a mixture of oats and kale (OK) to simulate food waste, and (3) a mixture of polystyrene foam, oats and kale waste (OKPS). After generating compost products from these trials, we conducted germination studies using seeds to evaluate the toxicity of each compost product at varying concentrations of vermicompost media dissolved in water. In order to evaluate the potential effects of microplastics particles on the toxicity of these solutions, we filtered each solution using 1.00-, 0.45-, and 0.22-micron filters to remove different particle size fractions.
We originally proposed the use of Fourier-transform infrared (FTIR) spectroscopy to evaluate plastics remaining in the vermi-compost products, however we were not able to resolve plastic composition in mixed systems using this method. We therefore evaluated other methods for identifying and quantifying microplastics in complex mixtures: dynamic light scattering (DLS) and thermogravimetric analysis (TGA) were used to evaluate the plastic particles remaining in the compost products, to help assess whether digested plastic was mineralized or simply broken down to smaller pieces.
From the spectroscopic and thermographic analysis of one vermicompost media sample, we estimated that approximately 30% of the vermi-compost media from PS remained as PS in the final compost product, and that the PS particles ranged in size up to 10-microns in diameter. When we evaluated the relative toxicity of vermi-compost PS media versus that of vermi-compost-sourced materials without the presence of plastics, we saw that the presence of PS plastic decreased the germination rate of tomato seeds under some conditions, but that even addition of OK compost to solution reduced germination rates. Furthermore, we saw that 0.45-micron filtration helped increase germination rates of seeds grown in solutions containing all three types of compost product, which indicated that particles contributing to toxicity are larger than 0.45-microns in diameter. The OK compost solution did not display a statistically significant difference in reduced toxicity between 0.45 and 0.22-micron filtered solutions. This indicates that the OKPS and degraded PS compost solutions had sub-0.45-micron sized particles that may have contributed to toxicity.
Ultimately, we would not recommend the use of mealworms to vermi-compost PS for either of the two goals stated above (i.e., (i) to mitigate PS waste, and (ii) to produce a compost product that could be used as an amendment for a vertical farm system). First, through the act of ingesting PS, the mealworms seem to break down macroplastics into microplastics, which can create a bigger challenge in terms of waste management and environmental contamination. Second, it is not advisable to use the resultant compost product as a nutrient source for farms given the potential presence and toxicity of microplastics within.
Objective 1: Design and build a vermi-composting system to degrade plastic and food waste
The P3 student team developed and implemented a vermi-composting system using a superworm (Zophobas atratus) to degrade polystyrene foam. Superworms are a type of mealworm.
Several iterations of the superworm enclosure design were evaluated to develop an acceptable vermi-composting system. The first iteration of the worm enclosure did not work due to the development of mold and the subsequent invasion of pests in a poorly ventilated system. The second worm enclosure failed due to a mass worm escape when the worms ate through the tape securing the aerating screens on their enclosures. Hence, we included several considerations during the construction of our improved super-worm enclosure, including worm retention, pest control, fungus control, and worm diet. To address these issues the following worm enclosure design has been implemented (Figure 1).
Figure 1. Schematic illustrating worm enclosure design and food mixtures provided. Two bins were fed a diet of oats and kale (at a ratio of 3:1) (OK; a), two bins fed with oats, kale, and polystyrene (OKPS; b), and one bin was fed a diet of PS only (PS; c).
The system consisted of 5 four-liter clear polycarbonate food storage bins, each containing approximately 300 worms. The worms were purchased from Premium Crickets. Two bins were fed a diet of oats and kale (OK; at a ratio of 3:1 g:g), two bins fed with oats, kale, and polystyrene (OKPS; ratio of 600:200:1), and one bin was fed a diet of PS only (PS). Due to the accumulation of moisture and mold in our previous systems, the bins remained open for proper aeration, but had tall, steep walls to prevent worm escape (Figure 2). The worms were fed 40 g of provisions to start, and replenished biweekly with 15 g in the same ratio. These portions were determined given our finding that providing too much food at one time resulted in the growth of mold in our previous system.
Figure 2. Photographs of the superworm composting system.
Objective 2 and 3: Evaluate plastic decomposition and the existence of plastics in resultant compost.
We originally proposed the use of FTIR to evaluate the existence and concentration of polystyrene (PS) remaining in compost products. Silicon wafer-based internal reflection FTIR, a technique in which a silicon wafer amplifies nanoparticles deposited on its surface with the use of evanescent waves generated through total internal reflection, was adopted to spectroscopically analyze the compost and plastic material.
Four compost solutions were prepared for Silicon wafer-based internal reflection FTIR analysis:
(1) 0.1 g OKPS compost, dried in a 100°C oven then dissolved in a 1 g H2O by leaving solution in 100°C heated bath overnight, (2) 0.1 g OK dried in a 60°C oven then dissolved in a 1 g H2O by leaving solution in 60°C heated bath overnight, (3) 0.1 g OK dried in a 100°C oven then dissolved in a 1 g H2O by leaving solution in 60°C oven overnight, then adding 0.1 g of 100 nm PS nanobeads suspended in water (Bangs Labs Inc, diluted to 5% solids of total solution), and (4) 1 g of 50 nm PS nanobeads (Bangs Labs Inc). Both 100 nm and 50 nm PS nanobeads were used as positive controls to evaluate the detectability of nanoplastics within the solution.
Forty-five microliter of each solution was deposited on a plasma cleaned (to ensure even spread of solution on wafer) Si wafer (n-type Si:P, [100], 4", 500um, P/P, FZ 400-1,000 ohm-cm, SEMI Prime, 1 Flat, Empak cst,TTV<5Âμm; University Wafers). Wafers were placed in the beam path of a FTIR (Thermo Nicolet 370 FTIR spectrometer) to collect spectra for analysis.
Figure 3. Silicon Wafer-Based Internal Reflection FTIR spectra of different compost solutions. The detection of 50 nm PS nanobeads (green) shows distinct peaks at 3023 cm-1 and 2920 cm-1 (indicating alkene and alkane stretching) and 1600 cm-1 (showing cyclic alkene stretching), which provides validation of this FTIR method for the detection of small nanoparticles. OK compost (negative control; orange) shows broad stretching behavior and a weak broad peak at 1730-1490 cm-1 . Both compost seeded with PS nanobeads - OK+100 nm PS nanobeads (positive control) and OKPS - had broad O-H and N-H stretches (3600-3000 cm-1), a C-H alkane stretch at 2920 cm-1, and a broad peak at (1730-1490 cm-1).
While FTIR could successfully measure an IR absorption spectrum (2700-6667 nm, equivalent to 1500-3700 cm-1) for PS samples and pure compost, the IR absorption of the O-H and N-H bands (3600-3000 cm-1) within the compost material made the evaluation of the existence or concentration of PS in a mixed sample difficult (Figure 3), as the characteristic peaks for pure PS (3023 cm-1 and 2920 cm-1) become undetectable. As such, while we attempted to evaluate the existence and concentration of nanoplastics through the IR-absorption relative to known concentrations, this proved difficult.
Alternatively, we evaluated the use of thermogravimetric analysis (TGA) to measure plastics and found it to be a useful analytical tool in evaluating the existence of plastics in the resultant compost. A series of TGA measurements were performed with compost collected from worms fed a diet of oats and kale, with known quantities of polystyrene foam (the same PS foam used for OKPS) added to compost after collecting the compost from the mealworm bins (g/g): 0% PS (pure compost, negative control), 1%, 2.5%, 5%, 10%, 15%, 20%, 50%, and 100% (pure polystyrene foam, positive control). Thermal degradation rates of each sample (𝑖.𝑒.,𝑑(𝑤𝑒𝑖𝑔ℎ𝑡 [%] )𝑑(𝑡𝑒𝑚𝑝𝑒𝑟𝑎𝑡𝑢𝑟𝑒 [° 𝐶])) were calculated. These rates were found to be unique as a function of the % PS added to the sample. To evaluate the concentration of PS foam in each sample, the decay rate of the pure compost sample (i.e., 0% PS) was subtracted from the decay rate of each of the other sample conditions (Figure 4). These data were used to develop a TGA standard curve that was employed to evaluate the polystyrene composition of a sample collected from worms fed a diet of OKPS (“Collected sample”).
Figure 4. Relative thermal degradation rates versus temperature of different concentrations of positive controlled PS added to OK compost. This plot was generated by subtracting an OK baseline from each curve. The values in the legend are the % PS (by mass) added to OK compost for each standard evaluated. OK compost thermally degrades at a lower temperature (~250-320°C) than PS Plastic (370-455°C).
The maxima of each subtracted decay rate curve from three replicate samples of each PS concentration in OK vermicompost (Figure 4) showed a clear correlation with the % PS added (Figure 5), which allowed us to use this data as a standard curve to evaluate unknown samples. The data-point representing the TGA measurement of pure PS foam (the 100% standard) was excluded as PS without the presence of compost media will electrostatically adhere to itself. Due to the thermal stability dependence on composition and compatibility with surrounding polymers, the electrostatic adhesion seen between PS particles will increase thermal stability of the pure PS media.1 Using this analysis, the concentration of PS foam remaining in the OKPS sample from the vermicompost trial described above was estimated to be 30% g/g.
Figure 5. Maximum PS foam thermal degradation rates per weight % of positive controlled PS foam in compost. The resulting fit was used to estimate the concentration of PS foam in compost collected from superworms fed a diet of kale, oats, and PS foam (OKPS).
Our findings of 30% PS nano-/microplastics remaining in the compost material is well above the reported 2.9% polystyrene observed to remain in composted material by previous studies that fed their mealworms PS foam alone.2,3 This could be due to our feedstock being a mixture of PS foam with other nutrient sources, or the result of us increasing the number of mealworms used by 10x, as compared to previous studies conducted in petri dishes.2,3 It is also possibly due to mealworms shredding/borrowing into their food as they eat it. This behavior would result in undigested particles of PS foam that would mix with the resultant compost, and it is difficult to separate the mealworm excreta (frass) from the resulting PS particles.
An open question was whether remaining PS in the finished compost samples existed as micron sized particles, and what the size distribution of these particles was. We therefore evaluated the use of dynamic light scattering (DLS) to measure particle size distribution. However, it proved challenging to use DLS to separate or prove the existence of PS in compost collected from worms with a diet of oats, kale, and polystyrene foam (OKPS). The solutions for the DLS experiments were prepared by drying the collected compost in an oven at 100˚C for at least one hour, then dissolving the compost in ultrapure water at a 1:10 ratio (g/g) for 1 hour at 60˚C. A Zetasizer Nano ZS90 (Malvern Panalytical) with a 633-nm laser was used to observe the average particle sizes using the samples prepared in cleaned and capped microcuvettes (BrandTech). As seen in Figure 6, compost collected from the worms with the OKPS diet had a population of micro-particles in the solution that were larger than 1-micron (but less than 10-microns), in contrast to the compost from oats and kale alone that did not, which we believe was contributed by the maceration of PS foam by the worms.
Figure 6. Dynamic light scattering of compost from superworms fed a diet of kale and oats (left), and compost from superworms fed a diet of kale, PS foam, and oats (right). The detection range of the system is from 0.4 nm to 10 m.
Objective 4: Evaluate toxicity of the resulting compost product
While there is no uniform method for testing the toxicity of compost to plants, the US Environmental Protection Agency (USEPA) does have a list of species deemed environmentally sensitive enough to be reliable indicators of contaminated soils.4 This list includes seeds from lettuce, cabbage, cucumber, tomato, carrot, and oats.
Germination tests (based on the USEPA field test for soil)4 consisted of placing 30 tomato seeds in a petri dish and feeding them a solution made with the worm-produced compost diluted into solution at various concentrations. The dishes were placed in a facility with indirect, full spectrum grow lights at a temperature of 25°C for a period of 5 days. After this, the samples were frozen to cease further growth and to measure germination rates of the seeds. To measure germination, the seeds were checked for any form of sprouting. We conducted toxicity studies using different compost solutions to determine the toxicity of solutions of different ratios of compost to water, with both filtered (1, 0.45, and 0.22 μm pore sizes) and unfiltered solutions. Solutions were made by dissolving dried compost (placed in an oven at 100°C) in tap water (dissolved overnight in a
60°C heated bath).
Figure 7. Image of a germination toxicity test. Left to right: Petri dish containing tap water and 30 tomato seeds, Petri dish containing OK compost dissolved in tap water (1:10/g:g) and 30 tomato seeds (unfiltered), and Petri dish containing OKPS compost dissolved in tap water (1:10/g:g) and 30 tomato seeds (unfiltered).
Figure 8. Effects of filtering on seed germination rates in vermicompost media for different concentrations (dry compost to tap water) of OK, PS, and OKPS shows that increasing filter pore sizes increases toxicity of the compost. A) No filtering. B) 1-micron filtering. C) 0.45-micron filtering. D) 0.22-micron filtering.
Each data point represents a triplicate of germination studies within a specific condition. Four notable preliminary findings emerged:
1) unfiltered PS vermi-compost samples had statistically (95% confidence) reduced germination rates (thereby more toxicity) than both vermi-compost media with no PS (OK) and mixed plastic and non-plastic vermicompost (OKPS).
2) 0.45-micron filtering of 1:10 dry compost to water mixture showed that adding PS to the vermi-compost feed increased toxicity of the vermi-compost media.
3) Filtering decreased the toxicity of all vermi-compost media solutions.
4) Decreasing the concentration of each vermi-compost media decreased the toxicity of said solution.
This filtration/germination study showed a decrease in toxicity when filtering the compost solutions. However, an open question was whether filtering can reduce the toxicity of a vermi-compost media that has degraded PS.
Figure 9. Germination study of non-plastic compost in aqueous solution that is filtered and unfiltered at different concentrations of dry compost to tap water: A) vermi-compost OK, B) vermi-compost PS, C) vermi-compost OKPS.
Figure 9 presents the same data as Figure 8, but reconfigured to evaluate the effect of filtering on toxicity. There were limited statistical differences between unfiltered and 1-micron filtered solutions. However, with a 0.45-micron filter the germination rates for all composted media increased regardless of compost concentration. For example, in 1:10 dry compost to tap water solution the germination rate increased by 53% for vermicompost PS media, 63% for vermicompost OKPS media, and 73% for vermicompost OK media. While OK compost showed no statistically significant reduction in toxicity between 0.45-micron and 0.22-micron filtered compost, vermi-compost filtered media that contained PS did decrease its toxicity with those levels of filtering. This indicates that PS particles between 0.22-micron and 0.45-micron in diameter are contributing to the observed toxicity
Conclusions:
Previous studies have illustrated that mealworms can vermi-compost PS.2,3 We therefore asked whether the resultant vermi-composted PS using mealworms is non-toxic and can be mixed with food waste in an economical fashion for two goals: (i) to mitigate PS waste, and (ii) to produce a compost product that could be used as an amendment for a vertical farm system. Before assessing the economic viability of vermicomposting PS, we evaluated if this process produces toxic micro/nanoplastics and, if so, whether the particles can be filtered out of a compost solution.
Our first task was constructing enclosures for the mealworms and extracting the resultant vermicompost media. The enclosures needed to limit worm escape, contamination with fungus or other bugs, and effectively extract their droppings. Placing the worms in sealed buckets with appropriate aeration seemed to be effective at controlling fungus and worm escape. However, we had difficulty separating mealworm waste from their food. The worms themselves are messy eaters that live in their food: the worms tear apart their food as they burrow into it and consume it. This creates a situation where it is difficult to separate the frass from shredded, uneaten mealworm food. We resolved this for our analyses using a tea strainer (25-micron pore size) during the harvesting of the vermicompost media, however this could remain an issue for large scale PS vermicomposting systems. Furthermore, 300 worms produced compost at a rate of approximately 15 grams every two weeks. While this was sufficient to supply material for the purposes of our limited study, where 1 gram of vermicompost was added to 10 grams of water for analysis, it would need to be significantly scaled if there was interest in using the produced frass as a compost amendment for an aquaponic farm. Operation and maintenance of such a system would require a lot of labor, which may not be economically feasible.
From the spectroscopic and thermographic analysis of the vermicompost media, we estimate that ~1/3 of the vermicompost media from PS remained as PS in the final compost product, and that the PS particles ranged in size up to 10-microns in diameter. This opened the question of the extent to which the mealworms ingested the PS versus tore up the PS into smaller particles without ingesting and/or digesting it. When we evaluated the relative toxicity of vermi-compost PS media versus that of vermi-compost-sourced materials without the presence of plastics, we saw that the presence of PS plastic decreased the germination rate of tomato seeds under some conditions, but that even addition of OK compost to solution without PS reduced germination rates as well. Furthermore, we saw that the 0.22-micron filtration helped increase germination rates of seeds grown in solutions containing all three types of compost product, which indicated that particles contributing to toxicity are larger than 0.22-microns in diameter. The OK compost solution did not display a statistically significant difference in reduced toxicity between 0.45 and 0.22-micron filtered solutions, which suggests that the PS microplastics within the OKPS and degraded PS compost solutions have sub-0.45-micron sized toxic particles that do not exist in the OK compost.
Ultimately, we would not recommend the use of mealworms to vermi-compost PS for either of the two goals stated above (i.e., (i) to mitigate PS waste, and (ii) to produce a compost product that could be used as an amendment for a vertical farm system). First, through the act of ingesting PS, the mealworms seem to break down macroplastics into microplastics, which can create a bigger challenge in terms of waste management and environmental contamination. Second, it is not advisable to use the resultant compost product as a nutrient source for farms given the potential presence and toxicity of microplastics within.
References:
- Tomić, N. Z. Compatibilization of Polymer Blends, Elsevier, 2020, 489-510.
- Yang, Y.; Yang, J.; Wu, W. M.; Zhao, J.; Song, Y; Gao, L.; Yang, R.; Jiang, L. Biodegradation and Mineralization of Polystyrene by Plastic-Eating Mealworms: Part 1. Chemical and Physical Characterization and Isotopic Tests. Environ. Sci. Technol. 2015, 49, 20, 12080–12086
- Yang, Y.; Yang, J.; Wu, W. M.; Zhao, J.; Song, Y; Gao, L.; Yang, R.; Jiang, L. Biodegradation and Mineralization of Polystyrene by Plastic-Eating Mealworms: Part 2. Role of Gut Microorganisms. Environ Sci Technol. 2015, 20, 12087-12093.
- U.S. Environmental Protection Agency. Single Laboratory Evaluation Of Phytotoxicity Test. 1987. Washington, D.C., EPA/600/4-87/012
Journal Articles:
No journal articles submitted with this report: View all 1 publications for this projectSupplemental Keywords:
vermi-compost, plastic degradation, plastic toxicity, thermogravimetric analysis, dynamic light scattering, compostRelevant Websites:
Progress and Final Reports:
Original AbstractThe 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.