Purdue University Graduate School
ozeh_u_thesis.pdf (2.3 MB)

Coupling Immunofluorescence and Electrokinetics in a Microfluidic Device for the Detection and Quantification of Escherichia coli in Water

Download (2.3 MB)
posted on 2019-10-16, 18:25 authored by Uzumma OzehUzumma Ozeh

The presence of Escherichia coli in water is an environmental indicator that the water is contaminated with faeces. Approximately, 30% of the world population drink water from sources contaminated with human faeces. Consequently, this percentage comprises of people that are highly vulnerable to Escherichia coli infection. While most strains of Escherichia coli are harmless or maintain a symbiotic relationship with humans, the pathogenic strains are responsible for injurious health effects, such as diarrhoea and kidney failure. The traditional method of detecting Escherichia coli takes about 24 – 48 hours, does not detect viable but non-culturable cells, and requires advanced equipment and great technical skills. Most other available detection techniques lack specificity, as observed with enzyme-based techniques, or are not very sensitive, as observed with most impedance-based techniques with clogged surfaces.

As a result of the health effects due to this microorganism and the basic limitations of available detection techniques, there is need for a specific, sensitive and rapid detection technique to ensure a sustained and timely access to E. coli- free water. Therefore, the aim of this research work is to develop a detection technique devoid of the basic limitations of available methods. In this study, the antibody-antigen relationship was taken advantage of to ensure the specificity of the technique is guaranteed. This was achieved using Escherichia coli polyclonal antibodies that target the O and K antigens found in most pathogenic strains. These antibodies were functionalized on carboxyl group modified superparamagnetic fluorescent microparticles immobilized with streptavidin. The sensitivity of the technique was ensured by utilizing the low detection limit feature offered by the use of microfluidic devices. Two microfluidic devices, glass-based and PDMS-based, were fabricated with easily accessible materials.

On introducing the sample reagents and test samples into the microfluidic devices, and passing an alternating current frequency through the system, the antibodies specifically isolated the target organisms from the pool of water contaminants and a drop in the device electric potential proportional to the bacteria concentration was observed. The success of this procedure depends on the identification of the alternating current frequency beyond which manipulation of the samples would not be easily carried out. As a result, the flow field analysis of the microparticles was carried out to study the particle behavior by varying the alternating current frequency from 15 kHz – 75 kHz.

The optimum frequency observed was 35 kHz. Using the glass-based microfluidic device, the voltage drop observed for the serial dilutions, 101 to 106 ranged from 200 mV to 420 mV while that for the serial dilutions, 10-7 to 10-1 ranged from 90 mV to 285 mV. To ascertain if a lower detection limit could be obtained, the PDMS-based microfluidic device, with a channel with of 300 µm, was used to analyze the response of the device to 10-7 to 10-1 serial dilutions. The result ranged from 10 mV to 30 mV respectively. A comparative analysis with the conventional detection method showed that it was able to detect less than 300 Escherichia coli colony-forming units. This result indicates that an optimized PDMS-based microfluidic device with higher resolution microchannel could potential detect tens of bacteria colony-forming units. These results were obtained in about 60 secs of introducing the sample in the device.

The rapidity and consistency of the results observed by the continuous increase in voltage drop with increasing concentrations of Escherichia coli indicate that this detection technique has great potential in addressing the time, specificity and sensitivity issues observed with most available detection methods.


Degree Type

  • Master of Science in Mechanical Engineering


  • Mechanical Engineering

Campus location

  • Hammond

Advisor/Supervisor/Committee Chair

Dr. George Nnanna

Advisor/Supervisor/Committee co-chair

Dr. Harvey Abramowitz

Additional Committee Member 2

Dr. Chandramouli Viswanatha Chandramouli

Additional Committee Member 3

Dr. Ran Zhou

Usage metrics



    Ref. manager