THE SCHOOL OF MATERIALS SCIENCE AND ENGINEERING

GEORGIA INSTITUTE OF TECHNOLOGY

Under the provisions of the regulations for the degree

DOCTOR OF PHILOSOPHY

on Wednesday, December 4, 2019

11:30 AM
in Pettit 102A

will be held the

DISSERTATION PROPOSAL DEFENSE

for

Decarle Jin

"Nanomaterials for Potentiometric Biosensors"

Committee Members:

Prof. Eric Vogel, Advisor, MSE

Prof. Natalie Stingelin, MSE

Prof. Oliver Brand, ECE

Prof. Peter Hesketh, ME

 Abstract:

Biosensors are currently being considered for detecting a wide variety of analytes including cancer biomarkers and routine proteins in the blood.  However, today’s biosensor technologies such as surface plasmon resonance are both expensive and time intensive. Potentiometric biosensors based on field-effect-transistors (FETs) have the potential for significantly improving detection times while retaining low limits of detection. Furthermore, FET-based potentiometric biosensors are relatively simple and cheap to manufacture. Despite these advantages, the only commercialized application of FET-based sensors is for pH sensing. FET-based sensors for other applications (e.g. affinity-based protein sensing in blood) have not been commercialized due to a variety of issues including speed of detection, selectivity, and the impact of salt and pH on the biosensor signal.  Nanomaterials, such as nanowires, nanoparticles and two-dimensional materials such as graphene, possess unique properties that can potentially improve multiple aspects of FET biosensors.  This work will fundamentally explore three specific use cases of nanomaterials to improve potentiometric biosensors:  the use of nanowires to improve mass transport, the use of nanoparticles to amplify the signal through reduction of screening by salt ions and the use of graphene to reduce the sensor pH sensitivity thereby increasing the signal of the captured antigen.

There have been numerous studies suggesting that nanowires have lower limits of detection compared to planar sensors. One of the mechanisms that has been proposed for this effect is that the diffusion of analyte occurs much quicker towards nanowire sensors compared to planar sensors due to radial diffusion; therefore, a larger measured signal is observed in a shorter period of settling time for nanowire sensors. However, most biosensor experiments are not based purely on diffusion of the analyte to the sensor surface but instead use flow to accelerate analyte binding. The question is, then, do nanowires still experience faster binding with flow? Solving the case of reaction and diffusion simultaneously is already quite complicated. The addition of convection complicates the situation even further, requiring complex (e.g. COMSOL) simulations in order to generate a solution. Even in the case where a solution is found, it is only applicable under the specific conditions used in the simulation. This work provides a general model for mass transport to planar and nanowire biosensors under flow that is applicable over a wide range of variables. The model is then used to examine the importance of radial diffusion compared with planar diffusion under flow. The effects of radial diffusion were found to be negligible given a high enough flow rate concluding that nanowires do not necessarily provide faster mass transport compared to planar sensors.

Nanoparticle amplification is often used to improve the detection limit of biosensors. By selectively attaching a secondary nanoparticle to the bound protein, the potential change can be further amplified. One of the key factors that limits the measured change in potential is Debye screening caused by salt ions in the matrix solution (e.g. blood). However, for the case of nanoparticle amplification, the impact of the sensor-particle size on Debye screening has not been explored. When considering spherical, charged particles along a sensor surface, the empty space in between particles cannot greatly contribute to the surface charge detected by the sensor due to screening by ions in solution; the curvature of the particle prevents perfect packing. It is proposed here that by having a particle size greater than the size of the sensor, the curvature of the particle above the sensor is reduced, thus mitigating the impact of Debye screening. This optimization is expected to improve the magnitude of amplification and therefore further improve the sensor limit of detection.

A fundamental question still unanswered in literature is the effect of pH sensitivity on protein detection. Theory predicts that a sensor with high pH sensitivity should have very low sensitivity towards proteins. This is due to the charging of the surface by pH sensitive surface groups effectively pinning the surface potential. There have been several studies in the literature that attempt to understand the impact of the density of pH sensitive groups on the potential change caused by charged protein attachment. However, a key limitation in these experiments has been the inability to carefully control the pH sensitivity. It has been shown that the pH sensitivity of graphene can be intimately controlled through the introduction of defects using, for example, UV-ozone treatment. By carefully controlling the defect density of graphene, we will systematically investigate the relationship between pH sensitivity and protein detection.