Design, Fabrication and Characterization of a Low-Impedance 3D Electrode Array System for Neuro-Electrophysiology
Mihaela Kusko 1,* ,Florea Craciunoiu 1,Bogdan Amuzescu 2,Ferdinand Halitzchi 2,Tudor Selescu 2,Antonio Radoi 1,Marian Popescu 1,Monica Simion 1,Adina Bragaru 1 andTeodora Ignat 1
Recent progress in patterned microelectrode manufacturing technology and microfluidics has opened the way to a large variety of cellular and molecular biosensor-based applications. In this extremely diverse and rapidly expanding landscape, silicon-based technologies occupy a special position, given their statute of mature, consolidated, and highly accessible areas of development. Within the present work we report microfabrication procedures and workflows for 3D patterned gold-plated microelectrode arrays (MEA) of different shapes (pyramidal, conical and high aspect ratio), and we provide a detailed characterization of their physical features during all the fabrication steps to have in the end a reliable technology. Moreover, the electrical performances of MEA silicon chips mounted on standardized connector boards via ultrasound wire-bonding have been tested using non-destructive electrochemical methods: linear sweep and cyclic voltammetry, impedance spectroscopy. Further, an experimental recording chamber package suitable for in vitro electrophysiology experiments has been realized using custom-design electronics for electrical stimulus delivery and local field potential recording, included in a complete electrophysiology setup, and the experimental structures have been tested on newborn rat hippocampal slices, yielding similar performance compared to commercially available MEA equipments.
Keywords:
3D electrodes; MEA; fabrication; electrochemical characterization; neuro-electrophysiology tests1. IntroductionOver the last years, significant efforts have been directed towards developing structures for studies of neural networks, starting with small networks of cultured neurons [1]. Besides the essential information provided for a better fundamental understanding of the nature of cell response to different stimuli, and subsequently the manner of communication between them, these results should facilitate further developments of new therapeutic approaches. For instance, studying the simple visual stimuli effects on turtle retinal ganglion cells behavior helps to fabricate neuroprostheses based upon electrical stimulation of the retina, which represent a solution against profound blindness [2]. Recently, similar approaches were applied to other medically intractable or incurable diseases like Parkinson’s [3,4], Alzheimer’s [5–7], and even cancer [8,9], leading to significant improvements in the quality of life for these patients. In this perspective, the necessary studies on hippocampus for understanding information storage in the brain, learning and memory [10] are difficult via traditional methods such as the patch-clamp technique, because it is not easy to maintain a clamped neuron over hours, and often the assembly response of a neural network is more important than that of an individual cell.
Microelectronic sensors are the adequate choice for monitoring in- and output parameters of both cultured cells on neurochip surfaces or tissue slices brought in tight contact, and different integrated microelectrode array systems have been proposed by now [11,12]. Moreover, concerted research in development of miniaturized neural probes and probe arrays has led to the creation of micro-electronic mechanical systems generically named Neural MEMS or NeuroMEMS [13], strengthening the concept of “neuroscience-on-chip” [14]. In this context, microelectrodes, both arrayed and individually addressed, represent a solution for which the proof of concept has been demonstrated [15]. Although the interests in this area of microfabrication became known long time ago, there is still place for supplementary improvements, to find novel design solutions to make possible in vivomultiple sites recordings and allowing also a better sensitivity. Therefore, in order to ensure minimal signal attenuation and noise interference, it is critical to lower the interface impedance [16], which not only diminishes the recorded signals, but also requires more intense stimuli. Naturally, a solution is to have electronics to amplify and process signals in close proximity to bioelectrodes [17], but, although this approach is correct in its scope, it requires a high level fabrication technology (expensive, multilayer cleanroom processing), and consequently, the development of novel electrode technologies has to be appropriate to the general CMOS process flow of a complex electrical circuit, reducing the ability of varying the multi-electrode chip design.
Recent progress in patterned microelectrode manufacturing technology and microfluidics has opened the way to a large variety of cellular and molecular biosensor-based applications. In this extremely diverse and rapidly expanding landscape, silicon-based technologies occupy a special position, given their statute of mature, consolidated, and highly accessible areas of development. Within the present work we report microfabrication procedures and workflows for 3D patterned gold-plated microelectrode arrays (MEA) of different shapes (pyramidal, conical and high aspect ratio), and we provide a detailed characterization of their physical features during all the fabrication steps to have in the end a reliable technology. Moreover, the electrical performances of MEA silicon chips mounted on standardized connector boards via ultrasound wire-bonding have been tested using non-destructive electrochemical methods: linear sweep and cyclic voltammetry, impedance spectroscopy. Further, an experimental recording chamber package suitable for in vitro electrophysiology experiments has been realized using custom-design electronics for electrical stimulus delivery and local field potential recording, included in a complete electrophysiology setup, and the experimental structures have been tested on newborn rat hippocampal slices, yielding similar performance compared to commercially available MEA equipments.
Keywords:
3D electrodes; MEA; fabrication; electrochemical characterization; neuro-electrophysiology tests1. IntroductionOver the last years, significant efforts have been directed towards developing structures for studies of neural networks, starting with small networks of cultured neurons [1]. Besides the essential information provided for a better fundamental understanding of the nature of cell response to different stimuli, and subsequently the manner of communication between them, these results should facilitate further developments of new therapeutic approaches. For instance, studying the simple visual stimuli effects on turtle retinal ganglion cells behavior helps to fabricate neuroprostheses based upon electrical stimulation of the retina, which represent a solution against profound blindness [2]. Recently, similar approaches were applied to other medically intractable or incurable diseases like Parkinson’s [3,4], Alzheimer’s [5–7], and even cancer [8,9], leading to significant improvements in the quality of life for these patients. In this perspective, the necessary studies on hippocampus for understanding information storage in the brain, learning and memory [10] are difficult via traditional methods such as the patch-clamp technique, because it is not easy to maintain a clamped neuron over hours, and often the assembly response of a neural network is more important than that of an individual cell.
Microelectronic sensors are the adequate choice for monitoring in- and output parameters of both cultured cells on neurochip surfaces or tissue slices brought in tight contact, and different integrated microelectrode array systems have been proposed by now [11,12]. Moreover, concerted research in development of miniaturized neural probes and probe arrays has led to the creation of micro-electronic mechanical systems generically named Neural MEMS or NeuroMEMS [13], strengthening the concept of “neuroscience-on-chip” [14]. In this context, microelectrodes, both arrayed and individually addressed, represent a solution for which the proof of concept has been demonstrated [15]. Although the interests in this area of microfabrication became known long time ago, there is still place for supplementary improvements, to find novel design solutions to make possible in vivomultiple sites recordings and allowing also a better sensitivity. Therefore, in order to ensure minimal signal attenuation and noise interference, it is critical to lower the interface impedance [16], which not only diminishes the recorded signals, but also requires more intense stimuli. Naturally, a solution is to have electronics to amplify and process signals in close proximity to bioelectrodes [17], but, although this approach is correct in its scope, it requires a high level fabrication technology (expensive, multilayer cleanroom processing), and consequently, the development of novel electrode technologies has to be appropriate to the general CMOS process flow of a complex electrical circuit, reducing the ability of varying the multi-electrode chip design.