A220-0015
Portable, low-cost, Raspberry Pi-based optical sensor (PiSENS) for continuous monitoring of atmospheric nitrogen dioxide

Wednesday, 16 December 2020
Poster
Ernesto Saiz Val1, Ivana Banicevic2, Nada Blagojevic2, Sergio Espinoza Torres3, Gino Italo Picasso Escobar3, Matthew O'Brien4 and Aleksandar Radu4, (1)Keele University, School of Chemical and Physical Sciences, Staffordshire, ST5, United Kingdom, (2)University of Montenegro, Faculty of Metallurgy and Technology, Podgorica, Montenegro, (3)National University of Engineering, Faculty of Sciences, Lima, Peru, (4)Keele University, School of Chemical and Physical Sciences, Staffordshire, United Kingdom
Abstract:
Continuous monitoring and real time data access to atmospheric levels of NO2 are essential for managing human activities and improving air quality. Current air quality monitoring stations (including NO2) capable of provision of information with high temporal frequency are bulky and utilize very expensive equipment so can be located only at specific sites within cities that can afford their high costs and maintenance. Thus, the need for low-cost and reliable NO2 sensors with high-spatiotemporal resolution is critical.

We have developed a low-cost and portable device for continuous measurement of NO2. It is based on utilizing low cost computer (Raspberry Pi) for the analysis of color of absorbing solution developed in a classical Saltzman reaction. A simple programme written in open source language (Python) was used for color analysis. All required parts to develop PiSENS are placed in a simple weather-proof box while 3D printer was utilized to develop parts needed to optimise construction.

The performance of PiSENS was good. The 7 points calibration R2 was 0.944 (between 0 ppm and 1 ppm). The LOD was 7x10-5 ppm. The average precision was 2.8% as relative standard deviation (RSD%), and the standard error % (SE%) 6.4%.

The device was successfully applied in the field for obtaining hourly NO2 concentration with 5 min resolution. We tested PiSENS in conditions mimicking rush hour traffic and in absence of traffic. Low cost and simple operations of PiSENS open up possibilities for its applications in rural and remote locations, while the capabilities of Raspberry Pi could allow integration of multiple PiSENS device into larger networks used for live decision-making or the creation of high precision models.

<< Previous Abstract | Next Abstract >>