A fully integrated CMOS microelectrode system for electrochemistry

Giagkoulovits, Christos (2018) A fully integrated CMOS microelectrode system for electrochemistry. PhD thesis, University of Glasgow.

Abstract

Electroanalysis has proven to be one of the most widely used technologies for point-of-care devices. Owing to the direct recording of the intrinsic properties of biochemical functions, the field has been involved in the study of biology since electrochemistry’s conception in the 1800’s. With the advent of microelectronics, humanity has welcomed self-monitoring portable devices such as the glucose sensor in its everyday routine. The sensitivity of amperometry/ voltammetry has been enhanced by the use of microelectrodes. Their arrangement into microelectrode arrays (MEAs) took a step forward into sensing biomarkers, DNA and pathogens on a multitude of sites. Integrating these devices and their operating circuits on CMOS monolithically miniaturised these systems even more, improved the noise response and achieved parallel data collection. Including microfluidics on this type of devices has led to the birth of the Lab-on-a-Chip technology. Despite the technology’s inclusion in many bioanalytical instruments there is still room for enhancing its capabilities and application possibilities. Even though research has been conducted on the selective preparation of microelectrodes with different materials in a CMOS MEA to sense several biomarkers, limited effort has been demonstrated on improving the parallel electroanalytical capabilities of these devices. Living and chemical materials have a tendency to alter their composition over time. Therefore analysing a biochemical sample using as many electroanalytical methods as possible simultaneously could offer a more complete diagnostic snapshot.

This thesis describes the development of a CMOS Lab-on-a-Chip device comprised of many electrochemical cells, capable of performing simultaneous amperometric/voltammetric measurements in the same fluidic chamber. The chip is named an electrochemical cell microarray (ECM) and it contains a MEA controlled by independent integrated potentiostats. The key stages in this work were: to investigate techniques for the electrochemical cell isolation through simulations; to design and implement a CMOS ECM ASIC; to prepare the CMOS chip for use in an electrochemical environment and encapsulate it to work with liquids; to test and characterise the CMOS chip housed in an experimental system; and to make parallel measurements by applying different simultaneous electroanalytical methods. It is envisaged that results from the system could be combined with multivariate analysis to describe a molecular profile rather than only concentration levels.

Simulations to determine the microelectrode structure and the potentiostat design, capable of constructing isolated electrochemical cells, were made using the Cadence CAD software package. The electrochemical environment and the microelectrode structure were modelled using a netlist of resistors and capacitors. The netlist was introduced in Cadence and it was simulated with potentiostat designs to produce 3-D potential distribution and electric field intensity maps of the chemical volume. The combination of a coaxial microelectrode structure and a fully differential potentiostat was found to result in independent electrochemical cells isolated from each other.

A 4 x 4 integrated ECM controlled by on-chip fully differential potentiostats and made up by a 16 × 16 working electrode MEA (laid out with the coaxial structure) was designed in an unmodified 0.35 μm CMOS process. The working electrodes were connected to a circuit capable of multiplexing them along a voltammetric measurement, maintaining their diffusion layers during stand-by time. Two readout methods were integrated, a simple resistor for an analogue readout and a discrete time digital current-to-frequency charge-sensitive amplifier. Working electrodes were designed with a 20 μm side length while the counter and reference electrodes had an 11 μm width. The microelectrodes were designed using the aluminium top metal layer of the CMOS process.

The chips were received from the foundry unmodified and passivated, thus they were post-process fabricated with photolithographic processes. The passivation layer had to be thinned over the MEA and completely removed on top of the microelectrodes. The openings were made 25 % smaller than the top metal layer electrode size to ensure a full coverage of the easily corroded Al metal. Two batches of chips were prepared, one with biocompatible Au on all the microelectrodes and one altered with Pd on the counter and Ag on the reference electrode. The chips were packaged on ceramic pin grid array packages and encapsulated using chemically resistant materials. Electroplating was verified to deposit Au with increased roughness on the microelectrodes and a cleaning step was performed prior to electrochemical experiments.

An experimental setup containing a PCB, a PXIe system by National Instruments, and software programs coded for use with the ECM was prepared. The programs were prepared to conduct various voltammetric and amperometric methods as well as to analyse the results. The first batch of post-processed encapsulated chips was used for characterisation and experimental measurements. The on-chip potentiostat was verified to perform alike a commercial potentiostat, tested with microelectrode samples prepared to mimic the coaxial structure of the ECM. The on-chip potentiostat’s fully differential design achieved a high 5.2 V potential window range for a CMOS device. An experiment was also devised and a 12.3 % cell-to-cell electrochemical cross-talk was found. The system was characterised with a 150 kHz bandwidth enabling fast-scan cyclic voltammetry(CV) experiments to be performed. A relatively high 1.39 nA limit-of-detection was recorded compared to other CMOS MEAs, which is however adequate for possible applications of the ECM. Due to lack of a current polarity output the digital current readout was only eligible for amperometric measurements, thus the analogue readout was used for the rest of the measurements.

The capability of the ECM system to perform independent parallel electroanalytical measurements was demonstrated with 3 different experimental techniques. The first one was a new voltammetric technique made possible by the ECM’s unique characteristics. The technique was named multiplexed cyclic voltammetry and it increased the acquisition speed of a voltammogram by a parallel potential scan on all the electrochemical cells. The second technique measured a chemical solution with 5 mM of ferrocene with constant potential amperometry, staircase cyclic voltammetry, normal pulse voltammetry, and differential pulse voltammetry simultaneously on different electrochemical cells. Lastly, a chemical solution with 2 analytes (ferrocene and decamethylferrocene) was prepared and they were sensed separately with constant potential amperometry and staircase cyclic voltammetry on different cells. The potential settings of each electrochemical cell were adjusted to detect its respective analyte.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: Integrated circuit modeling, ASIC, integrated circuit design, fully-differential, potentiostat, opamp, CMOS, microfabrication, lab-on-a-chip, microelectrodes, MEA, amperometry, voltammetry, electrochemical cell, microarray, electrochemical sensor, electrochemical cross-talk, electrochemical simulation, potential distribution, CMOS electroplating, potential window, multiplexed cyclic voltammetry, simultaneous detection, multiple analytes, simultaneous voltammetric methods.
• Subjects: Q Science > QD Chemistry; Q Science > QH Natural history > QH345 Biochemistry; T Technology > TK Electrical engineering. Electronics Nuclear engineering
• Colleges/Schools: College of Science and Engineering > School of Engineering > Electronics and Nanoscale Engineering
• Funder's Name: Engineering and Physical Sciences Research Council (EPSRC)
• Supervisor's Name: Cumming, Professor David
• Date of Award: 2018
• Depositing User: Christos Giagkoulovits
• Unique ID: glathesis:2018-39018
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 20 Dec 2018 12:06
• Last Modified: 26 Oct 2021 14:38
• Thesis DOI: 10.5525/gla.thesis.39018
• URI: https://theses.gla.ac.uk/id/eprint/39018

Related URLs

• Article DOI
• Article DOI
• Article DOI