The main source of waste segregation failures is at the household level, where waste is disposed incorrectly because of convenience, a lack of understanding, confusion about local recycling regulations, and inconsistent segregation practices. Research indicates that contamination levels can be as high as 15-40%, which means that a large amount of material placed in recycling bins is actually non-recyclable or improperly  segregated. This type of contamination makes recycling plants less efficient, increases the cost of processing, and often leads to the entire shipment of recyclables being sent to landfills. The economic cost is high, with annual estimates ranging between $3.5 and $4 billion because of increased costs of sorting, transportation, and landfill charges. When recyclables are contaminated, municipalities have to spend more on additional labor, sophisticated sorting equipment, and landfill charges, further taxing public finances. Moreover, organic waste sent to landfills decomposes anaerobically, and the by-product is methane gas, a potent greenhouse gas that has a global warming potential of about 28 times that of carbon dioxide over a 100-year time frame. Mismanaged waste diversion, therefore, is a major contributor to climate change and the failure of sustainability initiatives.

1. Set up the Arduino and servos

• Connect the servo motors to the Arduino (Bio servo to Digital Pin 9, Non-Bio servo to Digital Pin 10).

• Connect the capacitive sensor to the Arduino (signal to A0, power to 5V, ground to GND).

  2. Upload the code to the Arduino * Write the code for the servo motors and sensor. * Test servo motors.

  3. Calibrate the sensor * Test different materials and record the analog readings. * Determine the thresholds that separate biodegradable and non-biodegradable materials.

  4. Add extra sensors * Add Metal Sensor * Use extra data to remove overlapping ranges

  5. Test the robot * Place a material in front of the sensor. * Observe which servo the robot moves

  6. Adjust and repeat if needed * If a material is sorted incorrectly, adjust the threshold values in the code. * Repeat testing for multiple samples to ensure accuracy.

What I noticed was that the non-biodegradable materials gave lower sensor readings (667 and lower), while the biodegradable materials gave higher readings (higher than 674). This indicated that the two types of materials could be easily distinguished based on the capacitance readings. The reason for this difference is that the capacitance is affected by the dielectric constant of the material placed between or close to the plates of the sensor. Biodegradable materials, like food scraps or paper products, usually have higher moisture content. Water has a high dielectric constant (approximately 80 at room temperature), which helps to increase the overall dielectric constant of the material. This causes the material to give higher readings on the capacitance sensor. On the other hand, many non-biodegradable materials, like plastics, have low dielectric constants (between 2 and 4) and are less absorbent of water. Due to these differences in properties, the capacitance sensor was able to easily distinguish between biodegradable and non-biodegradable materials.

The purpose of this innovation was to design an automated system that separates biodegradable and non-biodegradable waste using dielectric properties. In this innovation, I connected a capacitive sensor to an Arduino and used sensor readings to control servo motors that opened different waste bins. What I observed was that non-biodegradable materials produced lower sensor values  (667 and lower), while biodegradable materials produced higher values (above 674). This showed that the two types of materials could be distinguished using capacitance readings. This was because biodegradable materials usually contain more moisture, which increases their dielectric constant and causes higher capacitance values. Next time I would improve the system by testing more materials and adjusting the threshold values to reduce errors caused by overlapping readings or sensor noise. This relates to environmental science and technology because it demonstrates how electronic sensors and material properties can be used to improve waste sorting systems and support environmental sustainability.

https://wasterecycling.org/ https://www.epa.gov/

The Recycling Partnership. 2024. State of recycling: The present and future of residential recycling in the US. https://recyclingpartnership.org/wp-content/uploads/2024/01/Recycling-Partnership-State-of-Recycling-Report-1.9.23.pdf.

Canada, Environment and Climate Change. “The Government of Canada Introduces New Measures to Regulate Methane Emissions from Canadian Landfills.” Www.canada.ca, 28 June 2024, www.canada.ca/en/environment-climate-change/news/2024/06/the-government-of-canada-introduces-new-measures-to-regulate-methane-emissions-from-canadian-landfills.html.

Vesga Ferreira, Juan Carlos, et al. “Design of a Waste Classification System Using a Low Experimental Cost Capacitive Sensor and Machine Learning Algorithms.” Applied Sciences, vol. 15, no. 3, 4 Feb. 2025, p. 1565, https://www.mdpi.com/2076-3417/15/3/1565

Cheema, Sehrish Munawar, et al. “Smart Waste Management and Classification Systems Using Cutting Edge Approach.” Sustainability, vol. 14, no. 16, 17 Aug. 2022, p. 10226. MDPI, www.mdpi.com/2071-1050/14/16/10226, https://doi.org/10.3390/su141610226.

https://pmc.ncbi.nlm.nih.gov/articles/PMC12765797/

1. I would like to thank my dad for providing feedback and insight on my code and getting me materials to build my innovation.
2. I would like to thank my science teachers: Ms. Fourie & Ms. Turner for their feedback on my presentation and project.
3. I would like to thank my science fair coordinator: Ms. Davis for their guidance.
4. I would like to thank my family and friends for their continuous support throughout my CYSF experience.