The Development of Sensors for the Detection of Hydrocarbons in Oil spills

UoM administered thesis: Phd


The detection and control of petroleum contaminants resulting from oil spillage pollution of the environment remain a global challenge. Due to its adverse effects, crude oil discharge to the environment is regulated; one of the key parameters used for compliance monitoring is the measure of oil concentration. Typically, national environmental regulations require oil companies to keep the total hydrocarbons concentration resulting from oil spills at or below 50 mg/kg soil (50 ppm). For mitigation and compliance monitoring, most soil analyses for oil spillages rely on the use of the standard gas chromatographic methods, but these methods are expensive, require user-expertise and are not suitable for in situ analysis. This study develops and evaluates sensors for in situ detection of hydrocarbons in soil, targeting low power consumption and the use of inexpensive chemiresistive materials. This project is a field-academic collaboration sponsored by Schlumberger through its Faculty for the Future Foundation (FFTF), focusing on improving local communities through women in STEM. As such, in developing and demonstrating the proposed approach, contaminated soils were obtained from a crude oil field site in Akata, Niger Delta in Nigeria, as a case study. These samples were collected at three depths (0-20, 20-40, and 40-60 cm), two months and two years after a spill; oil from the samples was extracted using Soxhlet and Soxtec techniques. The extracts were analysed by gas chromatography-flame ionization detection to determine the types of hydrocarbons and their concentrations, as required to be detected by chemiresistors. The experimental results show that the concentration of the hydrocarbons increases with increasing distance and time from the epicentre of the spill and then decreases over time, indicating vertical migration. The analysis detected oil with carbon numbers ranging from C8 – C38 with concentrations as high as 65,000 mg/kg in the soil. The results obtained in this study afford an insight into the level of damage caused by oil spillages in the Akata community of the Niger Delta region in Africa. This thesis presents an attractive approach of using chemiresistors as an alternative to the conventional gas chromatographic-based methods for the detection of petroleum hydrocarbons in the soil. The proposed system is intended to be used at different stages of the pollution in order to prevent further infiltration into the soil. As such, this work focuses on the right choice of materials, cost, and the technical capability of sorption processes to rapidly detect and quantify these contaminants. Composites of non-conducting polymers, poly(methyl) methacrylate, (PMMA) and poly(vinyl) chloride (PVC), and conductive filler carbon black (CB) were selected and prepared to make polymer-based sensors. The films were dried to evaporate the solvent; the morphology of the films was characterised using scanning electron microscopy (SEM). The impact of carbon concentration and geometry on the measured resistance of the polymer composite to hydrocarbons was determined. The films’ performance were consistent with increasing mass fraction with an optimum response at 10% w/w carbon black and 90% w/w polymer. Two sets of polymer-based sensors CB-PMMA and CB-PVC, were fabricated by depositing thin films of a CB-polymer onto interdigitated electrodes. The deposited composite films completed a circuit, providing electrical resistance. The detected concentration is proportional to the relative differential resistance response. Polymer composition and stability and the sensor response data obtained resulted in the fabrication of chemiresistors for the detection of hydrocarbon compounds, including eicosane – a high molecular weight compound not previously tested in electronic nose technology. Experimental optimisation studies allowed variation in the nature of the responses obtained. The result obtained provided information on sensor responses relating to film reversibility (mean value of the response is 2.90% ± 13 0.28% and standard deviation of only 9% of the mean value for PMMA and 1.36% ± 0.06% and standard deviation of 4% for PVC) and stability (up to 24 weeks with only a drop of 0.3% sensor response which is only 12% for PMMA and 0.2% sensor with only 18% drop for PVC) from the initial (week 1) response after repeated use of the sensors. These results provide contributions to the knowledge surrounding the interactions between conducting polymer films and hydrocarbon analytes. The CB-PMMA electrical percolation threshold is lower than the CB-PVC threshold because of the difference in CB-polymer compatibility between the two sensors. Additionally, the hydrophobic interaction of hydrocarbon molecules with the polymers have undeniably substantial effects on conductivity. The CB-PMMA sensor response to eicosane hydrocarbon is around 6 times higher in sensitivity to the studied analytes than CB-PVC. To the author’s knowledge, this is the first time eicosane has been detected using PMMA-carbon black or PVC-carbon composite films. Even at relatively low hydrocarbon concentrations, both CB-PMMA and CB-PVC composite films showed linear concentration response profiles to hydrocarbon measurements and demonstrated high repeatability when exposed to multiple hydrocarbon measurements. Due to low affinity binding of the analyte to the sensor, the sensors exhibited fast responses and short recovery times with excellent selectivity. The distinct patterns enhanced the selectivity of the signal output corresponding to the same compound at the same concentration using both CB-PMMA to CB-PVC composite films owing to their level of molecular interaction. The CB-PMMA sensors showed much higher responses when exposed to a range of hydrocarbons with varying sensitivities compared to CB-PVC; however, data reveals that all the studied hydrocarbons were robustly detected and differentiated from each other using both CB-PVC and CB-PMMA chemiresistors. Furthermore, the two sensors exhibited less variance in their responses providing enhanced ability to correctly identify different concentrations over 24 weeks with little degradation. The sensors’ responses to the maximum regulatory concentration at the limit of 50 ppm are large (resistance changes), fast (90% in less than 1s), reversible and selective, the limit of detection of the two sensors to the analytes are well below 0.6 - 5 ppm, the accepted level, hence validating high adaptability of this method. The underlying mechanism of this high sensitivity of sensors might be owing to the strength of the hydrophobic interactions between the polymer and the hydrocarbons. Overall, the sensor results show a strong interaction between the hydrocarbons and the polymer, resulting in a tremendous gain in terms of versatility. The findings reported here expand the potential applications for inexpensive composite thin-film conducting polymer-based sensors for oil spillage monitoring. This work on chemiresistors based upon polymethyl methacrylate and polyvinyl chloride films containing carbon black was replicated to ensure that it was possible to reproduce the results accurately. The difference between the resistances from the two batches of sensors is only 6% Ω for PMMA and 2% Ω for PVC indicating the sensor preparation method is reproducible. As a result, a series of variables that could be altered to affect the characteristics of the conducting polymer films was highlighted. The simple and inexpensive approach presented in this work is demonstrated to aid the development of conducting polymer composite sensors to detect petroleum hydrocarbons. The use of composite materials such as carbon black, PMMA, and PVCs in developing chemiresistors could significantly reduce the cost, time, and expert knowledge associated with the conventional method- Gas chromatography. Furthermore, the reproducibility generation augments the experimental validation for reference materials, which further increases confidence in the validity of subsequent results. The proposed approach identifies promising technology adopting the use of low power consumption real- 14 time sensors for effective detection and monitoring of petroleum hydrocarbon concentrations in the soil based on technical feasibility. For more comprehensive implementation, it should be noted however, that a single sensor does not have the capability to discriminate between the wide ranges of hydrocarbon constituents present in a contaminated soil sample; hence, this work also proposes a comprehensive analysis of chemiresistors within a multisensor array environment using the electronic nose technology for petroleum hydrocarbon detection and classification. The proposed design combines a series of arrays of sensors developed based on the hydrocarbon properties and characterisation to identify the complex mixture of petroleum hydrocarbons in the soil in situ. Once the technical feasibility has been established, practical considerations such as electronics, user-friendly interface, and the connecting probe, among others, should be ascertained prior to significant investment in new technology. These additional considerations are key areas for ease, durability and future development which allows a local farmer to use the device robustly.


Original languageEnglish
Awarding Institution
Award date1 Aug 2022