T RANSPARENT MICROELECTRODES FOR ELECTROPHYSIOLOGICAL RECORDING AND ELECTROCHEMICAL SENSING B y W eiyang Y ang A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Electrical Engineering p%p% Doc tor of Philosophy 20 2 1 ABSTRACT T RANSPARENT MICROELECTRODE S FOR ELECTROPHYSIOLOGICAL RECORDING AND ELECTROCHEMICAL SENSING By Weiyang Yang Indium Tin Oxide (ITO) is a well - known n - type semiconductor material that is often utilized in transparent microelectrodes. ITO has high conductivity, excellent transparency ov er the entire visible spectrum due to a large bandgap of around 4 eV, as well as confirmed biocompatibility. Because of nu merous advantages of ITO , in this dissertation, ITO as a base material will be applied in both electrophysiological recording and elec trochemical sensing. O ptogenetics is a revolutionary neuromodulation technique that utilizes light to excite or inhibit the activity of genetically targeted neurons, expressing light - sensitive opsin proteins. To fully realize the potential of the optogenet ics tools, neural interface devices with both recording and stimulating capabilities are vital for future engineering development, and improving their spatial precision is a topic of constant research. Conventional transparent reco r ding micro electrodes made of a single material, such as ITO, ultrathin metals, graphene, and poly - (3, 4 - ethylene dioxythiophene)/poly(styrene sulfonate) (PEDOT:PSS), have limitations and hardly possess the desired combination of broadband transmittance, low electric al resistivity, mechanical flexibility, and biocompatibility . One direction of t his dissertation work is to develop multilayered electrophysiological microelectrodes with high transparency, outstanding conductivity, low electrochemical impedance, high c harge storage capacity, excellent mechanical properties , and ultra - flexib ility. Chapter 1 brief ly introduce d the background, current challenges, and motivation s of this dissertation. Cha pter 2 concluded a review of electrical materials for neurophysiology recording implants. Chapter 3 propose d a probe with a combined ITO - PEDOT:PSS electrode configuration by spinning thin PEDOT:PSS films on ITO microelectrodes, for applications in low - impedance neural recordings. The characteristics of the ITO - PEDOT:PSS micr oelectrodes were analyzed as a preliminary study for the following transparent electrophysiology recording array research . Chapter 4 report ed an ultra - flexible, conductive, transparent thin film using a PEDOT:PSS - ITO - Ag - ITO multilayer structure on Parylene C deposited at room temperature. The material characterization demonstrated enhanced conductivity, remarkable and wavelength - tunable transmittance, significantl y reduced electrochemical impedance, increased charge storage capacity, good stability , good adhesion , and confirmed mechanical properties of the combined film. Next, Chapter 5 d emonstrated two 32 - channel s using this PEDOT:PSS - ITO - Ag - ITO multilayer ed thin film structure on Parylene C . T h ese two s proved to work effectively in vivo for the electrophysiological detection in the living brain tissue. Last but not least, Chapter 6 first discussed the ongoing work to develop a 120 - channel h igh spatial resolution transparent micro - ECoG array . The other subsection of this chapter is to fabricate an ITO - based transparent and miniaturized electrochemical sensor for continuous and quantitative monitorin g of the concentrations of copper ( Cu ) and manganese ( Mn ) ions in bodies and soil environment by utilizing Differential Pulse Stripping Voltammetry (DPSV). ACKNOWLEDGEMENTS TABLE OF CONTENTS ................................ ................................ ................................ ......................... ................................ ................................ ................................ ...................... ................................ ................................ ................................ ..... ................................ ................................ ................................ .............. ................................ ................................ ................................ ................................ ................................ ................................ ................................ ......................... ................................ ................................ ................................ ........... ................................ ................................ ................................ ........ ................................ ................................ ............. ................................ ................................ ................................ ....... ................................ ................................ ................................ ............ ................................ ................................ ................................ .......................... ................................ ................................ .................. ................................ ................................ ................................ .... ................................ ................................ ................................ ..... ................................ ................................ ................................ .. ................................ ................................ ................................ .................. ................................ ................................ ................................ ........ ................................ ................................ ................................ ................... ................................ ................................ ................................ .... ................................ ................................ ................................ ........... ................................ ................................ .................. ................................ ................................ ................................ . ................................ ................................ ................................ .............. ................................ ................................ ......................... ................................ ................................ ................................ ...... ................................ ................................ ................................ ............... ................................ ................................ ................................ ............................... ........... ................................ ................................ ................................ .............................. ................................ ................................ ................................ ............. ................................ ......................... ................................ ................................ ................................ ............. ................................ ................................ ................................ ............ ................................ ................................ ........................... ................................ ................................ ................................ ................................ ................................ ................................ .... ................................ ................................ ............................... ................................ ................................ ................................ ............................... ................................ ................................ ............... ................................ ................................ ................................ .............................. ................................ ................................ ................................ ............. ................................ ................................ ................................ ..................... ................................ ................................ .......................... Design Methodology for Multilayer Anti - reflection Coatings ............................ Fabrication of PEDOT:PSS - ITO - Ag - ITO Thin Films ................................ ........ EM Samples Preparation and Test ................................ ................................ ..... ................................ ................................ ................................ ............. Transmittance and Conductivity Measurements ................................ .................. Electrochemical Measurements ................................ ................................ ........... Peel - off Tests ................................ ................................ ................................ ....... ................................ ......... ................................ ................................ ................................ ............ ................................ ................................ ............................. ................................ ................................ .................... ................................ ........... ................................ ................................ ................................ .......................... ................................ ................................ ................................ ........................ ................................ ................................ ................................ .................. ................................ ................................ ................................ ........... ................................ ................................ ................................ ............................... ................................ ................................ ................................ ............ ................................ ................................ ................................ .............................. ................................ ................................ ................................ ............. ................................ ................................ .......................... ................................ ................................ ................................ ............. Electrochemical Impedance ................................ ................................ ................. Transmittance Measurements ................................ ................................ .............. Signal to Noise Ratios (SNRs) Measurements ................................ .................... In vivo Animal Experiment ................................ ................................ .................. ................................ ................................ ................................ ............ ................................ .................... Electrochemical Impedance ................................ ................................ ......................... Transmittance Measurements ................................ ................................ ...................... ................................ ............................ In vivo Animal Experiment ................................ ................................ .......................... ................................ ................................ ................................ ............................. ................................ ................................ ............. ................................ ................................ ................. ................................ ................................ ............... Mask D esigns ................................ ................................ ................................ ..... Fabrication P rocedures ................................ ................................ ....................... ................................ ................................ ................................ ............ PCB D esign ................................ ................................ ................................ ........ Recording S ystem Set up ................................ ................................ .................... ................................ ................................ ................................ .................. ................................ ................................ ................................ .................. ................................ ................................ ................................ . ................................ ............................... ................................ ................................ ................................ . Test Setup ................................ ................................ ................................ . Cu and Mn Ions Solutions Preparation ................................ .................... Testing and Cleaning Recipes ................................ ................................ .. ................................ ................................ ................................ Cu/Mn Ions ................................ ................................ ................................ ........ Linearity and Repeatability ................................ ................................ ................ SEM and EDS ................................ ................................ ................................ .... BDD Electrodes ................................ ................................ ................................ . ................................ ................................ ....................... ................................ ................................ ................................ ................................ .. ................................ ................................ ................................ ........................ LIST OF TABLES ................................ .......... ......................... ................................ ................................ ................... ................................ ................................ ................................ ................................ ............................ L IST OF FIGURES ................................ ................................ .......................... ................................ ................... ................................ ................................ ............. ................................ ...... ................................ ................................ .................... ................................ ... ................................ ................................ ............................. ................................ ................................ .................. ................................ ................................ ........................ ................................ ............................... ................................ ................................ ................................ ....................... ................................ .... ................................ ................................ ................................ ............................ ................................ ........................... ................................ ................................ .... ................................ ................................ ................................ ..................... ................................ ................................ ....................... .............. ................................ ....... ................................ ................................ ................................ .. ................................ ................................ ........................ ................................ ................................ .............................. ................................ ................................ ................................ ................................ ........................ ................................ ................................ ................................ .............................. ................................ ................................ ................................ ................................ ... ................................ ................................ ................................ ................................ ..... ................................ ................................ ................................ ..... ................................ ................................ ................................ ................................ ............... ................................ ................................ ................................ ................................ ...................... ................................ ................................ . ................................ ................................ ................................ ................................ ...................... ................................ ............................ ............ ................................ ................................ ................................ .... ................................ ................................ ................................ .......................... ................................ ................................ ................................ ........................ ................................ .. ................................ .. ................................ .. ...... ................................ ............................... ................................ ................................ ................................ ....................... ................................ ................. ................................ ................................ ......... ................................ ............................. ................................ ............................ ................................ ................................ . ................................ ....... ................................ ................................ ................................ ............. ................................ ................................ ..................... ................................ ................................ ................................ . KEY TO ABBREVIATIONS PEDOT:PSS poly - (3, 4 - ethylene dioxythiophene)/poly(styrene sulfonate) SNR Signal - to - noise ratio C HAPTER 1: Introduction Nowadays, optoelectronic neural interface tools to explore the pathology of those nervous system diseases. s with the optogenetic tools could achieve the no - tissue damage, genetically targeted cells stimulation, and better spatial and temporal stimulation resolution. In addition, the substrate chosen s more flexible an d better attached to the brain tissue. In vivo experiments utilizing the designed two microelectrode arrays w ere done to prove the efficacy for electrophysiology recording . More in vivo animal experiments are needed to study the feasibility of this array. In addition, in vitro experiments, including the fluorescent light transmittance, electrochemical impedance, stability, and signal - to - noise ratios (SNRs) were also discussed in Chapter 5 . In Chapter 6 , firstly, the transparent and flexible 120 - channel achieving the higher resolution compared to the previous 32 - channel - ITO - Ag - ITO multilayered structure was utilized in this project. T he other subsection o f this chapter was to fabricate an ITO - based transparen t, flexible, and miniaturized electrochemical bio sensor for continuous and quantitative monitoring of the concentrations of copper (Cu) and manganese (Mn) ions in bodies and soil environment. Prion dis eases, such as Creutzfeldt - Jakob Disease (CJD) and Bovine Spongiform Encephalopathy (BSE), are caused by abnormal folding of proteins and can cause asymptotic neurological changes and death. Over the years, researchers found that Mn levels in the brains of patients with Creutzfeldt - Jakob disease are 10 times higher than those in the normal brain. Besides, high level of Cu ions in the bodies also has been found to intensify those prion diseases in recent research. The exceeded Cu and Mn ion s in the soil envi ronment will eventually flow through the food chain into humans and animals, aggravating the prion diseases. Therefore, it is urgent to design a n electrochemical sensor for continuous and quantitative monitoring of the concentrations of Cu and Mn ions in bodies and the soil environment with high sensitivity and accuracy. This chapter showed the potential to use ITO material for Cu and Mn ions measurement. C HAPTER 2: Electrode Materials for Neurophysiology Recording Implants 2.1 Introduction Neurological disorders and diseases in the central and peripheral nervous systems, such as Parkinson's disease, Alzheimer's disease, and epilepsy, are affecting hundreds of millions of people worldwide [1], [37], [38] . Neurophysiology recording electrodes act as a seamless interface between the nervous system and the outside world and help di agnose these neurological diseases. Several types of neural signals could be measured from the brain using electrodes [39] , including electroencephalogram (EEG) (10 - - 200 Hz) [40] , electrocorticogram (ECoG) (10 - - 200 Hz) [41], [42] , in addition to local field potentials (LFPs) (0.5 - 5 mVp p; 1 mHz - 200 Hz) and action potential spikes (50 - - 70 mVpp for intracellular; 100 Hz - 10 kHz) [43], [44] . EEG is noninvasive but suffers from low spatial resolution and poor SNR because of signal attenuation throu gh the scalp and skull. Mechanical disturbances and electromyographic activities also incur the artifacts that further influence the spatial and temporal resolutions of EEG recording [45] . Unlike EEG, ECoG directly measures the signals from the cerebral cortex via neuro physiological implants without any internal and external source noises due to the scalp and skull, leading to lower tissue interference, greater precision, higher sensitivity, and reduced noise interference. Although some special ECoG arrays, such as shown to be capable of recording spike activity and LFPs [46] , almost ECoG can only gather the electrophysiological signals from the superficial surface of the cerebral cortex and is incapable of capturing spikes from individual neuro ns. Therefore, penetrating electrodes suitable for recording LFPs and action potentials with high spatiotemporal resolution have been widely used in the neuroscience community [47] . Despite recent advances in electrode technologies, all existing electrode implants are still suffering from poor long - term stability and crosstalk, due to long - standing chall enges such as material biocompatibility, hermetic packaging, the relatively large physical dimensions of the devices, as well as mechanical mismatch between the brain tissue and the implant [48] . Similarly to the central nervous system, for the peripheral nervous systems, surgically implanted neural electrodes could be categorized into regenerative electrodes, i ntra - fascicular electrodes, inter - fascicular electrodes, and extra - neural electrodes [49] . These electrodes have more strict requirements for some material properties, such as flexibility and biocompatibility [49] . Indeed, careful selection and design of electrode and packaging/substrate mater ials are significantly essential to improve the recording quality and long - term stability of the electrode implants [50] . Therefore, to thoroughly study the electrical activity of neuronal circuits underlying various disorders, developing innovative neural recording devices have been a long - standing interest of many scientists, intending to achieve t he best combination of excellent electrical properties, high spatiotemporal precision, prominent biocompatibility, outstanding long - term stability, and safety of the electrode devices. To date, significant research has been devoted to the design and fabric ation of implantable neural recording electrodes with different materials on various substrates. The materials of these devices could be classified into two broad categories: electrode materials as well as packaging and substrate materials. While silicon - b ased materials, as well as common metallic materials (e.g., platinum or iridium) and their derivatives (e.g., platinum black and iridium oxide), are widely used in electrode manufacturing, they are still antagonistic to the soft, ionic, wet, and dynamic na ture of the biological tissue, with their hard, electronic, dry, and static nature. Non - conventional conducting materials that were not initially developed for neural implants have received significant attention and have been studied for neurophysiological recording in recent years because of their favorable properties and manufacturing advantages. Examples of these emerging electrode materials include graphene [29], [51] , ITO [52] , carbon - polymer hybrid nanostructures [53], [54] . The design consi deration of neural stimulation electrodes is similar to that of neural recording electrodes, concerning biocompatibility, mechanical properties, electrical properties, and stability [55] . For example, platinum black and Ir/IrOx are also widely used as stimulating electrodes [56], [57] . Large charge storage ca pacity is specifically required for simulating electrodes to achieve better stimulating performance [58] . Neural stimulators also have the same strict requirements on hermeticity, long - term stability, and biocompatibility of device package [59], [60] . Many materials that have been utilized in neural stimulating probes include but are not limited to: ceramics, glass, epoxy, silicone, and so on [55], [59], [61] . To draw a clear picture and guide the material design for future device de velopment, this chapter shows the current materials for the fabrication of neural recording implants that were reported in the literature in the most recent years. In the following sections, Section 2 discusses several important material properties, includ ing electrical properties, mechanical properties, stability, biodegradability/bioresorbability, biocompatibility, and optical properties, as well as the critical impact of these properties on the performance of electrode implants. Section 3 provides a deta iled discussion of various electrode materials in three different categories: inorganic materials (e.g., metals and semiconductors), organic materials (e.g., poly(3,4 - ethylene dioxythiophene):poly(styrene sulfonate (PEDOT:PSS) and poly(pyrrole) (PPy)), and carbon - based materials (e.g., graphene and carbon nanostructures). Approaches to improve the recording performance of the electrode materials are also included . 2.2 Key Challenges of Neural Implants 2.2.1 Tissue Responses Before selecting candidate mater ials for neural electrode implants, it is essential to understand the biological response to foreign objects, e.g., neural implants. The inflammatory response is usually caused by tissue injured during the implantation surgery or the existence of the impla nts in the body. Inflammation achieves the purpose of containing, neutralizing, diluting, or isolating the harmful substances through a series of complex physiological reactions [62] . These inflammatory reactions will significantly affect the functionality and stability of implanted devices. First, acute inflammation will occur in the first few days of implantation. A large amount of blood will flow to the damaged tissue thro ugh the dilated blood vessels, and then a blood clot will be formed to close the wound [62] . Then the tissue fluid containing water, salt, and protein will form edema [63] . At this stage , the implants have to overcome the contamination of blood and tissue fluids that may cover the implants and cause device malfunction. Similarly, the extrusion and tissue deformation that may be caused by edema also require a certain strength of the insert ed implant. This means that the electrode, package, or substrate materials must have a certain mechanical strength. The tissue environment is moist and chemically rich, which is not an ideal environment for implants [64] . Moreover, the immune response will release reactive oxidative species (ROS), which attack and degrade the implants [65], [66] . With the continual presence of the implant, the inflammatory response will be transformed into chronic inflammation. A major feature in this phase is the regeneration of damaged epithelium and vascular tissue [67] [69] , which may encapsulate the implants and consequently degrade the recording stability and accuracy of the electrodes. The immune response of the tissue does not stop at this phase, so the implant still faces the attack of ROS. Once a foreign object is implanted into the body, a sequence of events (e.g., inflammation and foreign body response) occurs in the surrounding tissue and ultimately ends at the formation of foreig n body giant cells at biotic - abiotic interfaces [70] . The intensity of the response is directly related to the properties of the implant [62] , such as size, shape, topography, and che mical and physical properties of the selected material. As the final stage of the inflammatory response, tissues try to wrap the implants with a vascular, collagenous fibrous capsule with a thickness of 50 - [71] . This fibrous wall will undoubtedly affect the electrical coupling between the implant and the targeted neurons, which may cause s ignal degradation and ultimately implant failure . 2.2.2 Surgical Challenges Before the neurophysiology recording implants are surgically implanted into the body [72] , sterilization is a significant and indispensable step to reduce the microbial contaminants (e.g., viruses) by 6 orders of magnitude [73] , and thereby reduce the intensity of inflammation. Various sterilization methods have been explored to suitably match various neurophysiology recording implants [73] . Nowadays, there are a great number of sterilization m ethods compliant with biomedical device regulation [74] , including chemical sterilization (ethanol 70 %), dry heating (160 - 190 °C), autoclaving (120 - 135 °C), ethylene oxide gas, hydrogen peroxide gas plasma, peracetic acid and UV radiation. H igh temperatures in dry heat and autoclaving sterilization will accelerate the oxidation and corrosion speed of the electrode materials, and hence can destroy the functionality of the whole implants, especially for easily - oxidized materials, such as silver thin films and silver nanowires [75], [76] . For packaging materials, high temperature and liquid uptake are the main concerns during these sterilization procedures [64], [77] . In particula r for biodegradable packaging materials, dry heat and autoclaving sterilization may cause partial denaturation to collagen [78] , morphology change to silk [79] , and melting and degradation to (poly(lactic - co - glycolic acid) PLGA [80] . The sterilization process has less impact on synthetic polymer packaging materials than biodegradable materials, but it is still worthy of note. For instance, significant delamination of Parylene C encapsu lation has been revealed after the steam sterilization process because of the insufficient adhesion strength between Parylene C and encapsulated device [8 1] . In addition, because its glass transition temperature is around 90 °C, high - temperature may cause degradation in the mechanical and optical properties of Parylene C. Ceramic materials have relatively broad options of sterilization methods due to the ir low water - vapor permeability and high - temperature resistance [64] . While ethylene oxide sterilization can be operated at relatively low temperatures, the permeability of polymers can allo w liquid stored in the material and a degassing step is required [64] . In addition, ethylene oxide is a central nervous inhibitor, stimulant and protoplasmic toxin [82] . Improper exposure of neural implants to ethylene oxide can cause acute poisoning and chronic effects, such as severe headache, loss of consciousness, neurasthenic syndrome and dysfunction of the vegetative nerve with long - term light exposure [83] . Unlike ethylene oxide gas, hydrogen peroxide gas plasma has the benefit of non - toxic final decomposition products [84] . However, because of the oxidation reaction during the sterilization by hydrogen peroxide gas plasma [84] , the materials selected must be resil i ent to changes in electrical conductivity caused by oxidation . 2.3 Key Characteristics of Electrode Materials 2.3.1 El ectrical Properties For electrophysiology recording, the electrode/electrolyte boundary is comprised of electrochemical reactions (Faradic) and double - layer charging (capacitive) [85], [86] . Electrochemical i mpedance (typically at 1 kHz) is a critical factor in benchmarking the performance of the recording electrodes [87] . The targeted impedance range of microelectrodes is ecording system utilization [88] . Although some studies indicate the impedance does not have a major impact on the signal quality [89] , most studies state that electrochemical impedance greatly affects the signal recording quality [30], [90], [91] . The design of electrodes present tradeoffs in dimensions, electrochemical impedance, and background noise of recording. Mi - unit recording with high spatial resolution and minimal invasiveness, but at the expense of increased electrochemical impedance that could cause signal quality reduction and background Johnson noise increase. In particular, Johnson noise, also known as thermal noise, is proportional to the square root of the impedance of the electrodes [92], [93] , as given by the following general equatio n 1 Where k is Boltzmann's constant, T is the temperature value, Re { Z } is the resistive F is the frequency band [94] . The most common solution to this challenge is to increase the effective surface area of microelectrodes by surface modification with electrically conducting polymers, nanomateria ls, or nanostructures [95], [96] , which will effectively reduce the impedance while keep ing device dimensions at a cellular scale to achieve high recording resolution, as shown in Fig ure 1A . Conducting polymers (CPs), such as PPy and poly (3,4 - ethylenedioxythiophene) (PEDOT), has also shown promise in improving ionic - to - electronic charges tra nsfer at the interface between the tissue and the recording site [97], [98] , therefore increasing the charge capacity of the microelectrodes. Since the insulation layer i s a part of the recording system, once it has been damaged due to material degradation or insulation delamination, the electrical properties of the entire system will also change. D elamination changes the properties of the electrode by expanding the geometric area of the exposed conductor, in turn, this averages the recorded potentials across an increased electrode surface area and attenuates t he neural signal [99] . Besides, an increase in the surface area will cause abnormal impedance change of the electrode during long - term implantation [50] , which will further deteriorate the recording quality. 2.3.2 Biocompatibility The biocompatibility of a recording electrode implant depends on various factors, including electrode ma terials, device geometry, and surrounding environments. From the material [100] . An ideal biomaterial for neural recording implants should be non - cytotoxic in vivo and release no substances or substances at only low, non - toxic concentrations. The tissue should produce minimal glial encapsulation surrounding the implant and only mild foreign body reaction without evidence of necrosis or implant rejection [101], [102] , as shown in Fig ure 1 B. Evaluation of material/device biocompatibility is critical and may include the tests of cyt otoxicity, acute/chronic systemic toxicity, sub - acute/sub - chronic toxicity, sensitization, irritation, genotoxicity, hemocompatibility, toxicokinetic studies, and immunotoxicology [103] . Since the same material may respond differently to different biological environments, the International Organization of Standard (ISO) enacts various test and evaluation protocols to evaluate the materials' biocompatibility, considering various body contact types, contact time, environments of intended use ( in vitro , ex vivo , or in vivo ), and test methods as mentioned in [104], [105] . 2. 3.3 Stability Material stability is another important consideration of neural recording implants [106] [109] . The fabrication imperfection of the electrode or the packaging materials, such as unavoidable pinholes and defects, could cause the oxidation and delamination of the materials, and hence, shorten the longevity of the implants in liquid environments with a high concentration of ions, such as cerebrospinal fluid [110], [111] . The heterogeneous junction where an electrode interfaces with an adhesion - promoting layer (e.g., Ti or Cr) or the heterogeneous alloys is also a potential risk of electrode reliability. The two different metals can form a short circuit galvanic cell in th e tissue fluid that accelerates the corrosion of one of the metals and weakens the metal - to - metal bonding strength [112] . Therefore, higher atomic weight transition metals with high corrosion resistance, such as platinum and iridi um, were selected as the primary electrode materials [65], [113], [114] . Homogenous alloys with multiple metal elements can also improve corrosion resistance [99] . Surface modification of electrodes with electrodeposited CPs is another method to slow down metal corrosion and improve device stability [115], [116] . For example, electrodeposited PEDOT is quite chemically stable in the damp, oxygen - rich environm ents because PEDOT can be further polymerized by the oxygen and protect the metal electrodes from direct exposure to reactive, oxygenated solution [117] , and therefore, prevent the metals from corrosion [118], [119] . However, further polymerization could cause the increased electrochemical impedance of the whole electrodes due to cracking or d elamination of the PEDOT layer [99], [120] . Biofouling also contributes to the instability of the neurophysiological recording implants. Biofouling leads to the encapsulation of electrodes by protein and glial cells, especially on those wit h high electrochemical surface areas, and therefore, restricts ionic diffusion at the electrode - electrolyte interface [121], [122] . In addition, the tissue response persistently promotes the degradation of electrode materials and insulation. To minimize electrode biofouling, significant efforts have been made on surface modification or functionalization to alter the chemical terminat ions, morphology, and wettability of the electrode surface [99] . Several hydrogel and polymer coatings, such as polyethylene gl ycol (PEG) and PEG methacrylate (PEGMA), have been utilized to improve the hydrophilicity of the electrode surface [123] [125] . With large amounts of water in their structures, these materials are highly hydrated to increase the energetic penalty of removing water for protein and microorganism attachment s . Engineered antifouling electrode materials, such as sp3 carbon - enriched, boron - doped polycrystalline diamond (BDD), also show the advantages of improved biocompati bility and reduced biofouling compared to conventional electrode materials [126], [127] , as shown in Fig ure 1 C. Morphological response of rat cortical neurons on the Parylene C and microcrystalline diamond (MCD) substrates (lower) appeared similarly to the control substrate. Moreover, nanostructured surfaces with low friction and low surface energies can effectively decrease cell attachment onto the implant surface, and hence, reduce the possibility of biofouling formation [128], [129] . 2.3.4 Biodegradability/Bioresorbability 2.3.5 Mechanical Properties 2.3.6 Optic al Transparency 2.3.6.1 Opto - neurostimulators 2.4 Electrode Materials 2.4.1 Inorganic Materials Recently, much attention has been devoted to investigating innovative electrode materials to improve electrical, mechanical, and optical properties, as well as stability, biocompatibility, or biodegradability of recording electrodes [48] . This section classifies the electrode materials into inorganic, organic, and carbon - based materials, and discusses the advant ages, disadvantages, and applications of each specific material in detail. 2.4.1.1 Metals Metals are the most prevailing and common electrode materials for neural recording for nearly 50 years [170] . Widely used metal electrode materials, such as gold ( Au ) , platinum (Pt), iridium (Ir), tungs ten (W), and tantalum (Ta), offer a great number of desirable properties, including chemical inertness, high electrical conductivity, and excellent biocompatibility in biological environments [130], [171], [172] . Au/Pt and Ir/Pt have been used as the electrode materials for "Utah array" and "Michigan Probe", two of the most popular neural interface electrodes [173], [174] . However, these materials suffer from limited electrochemical conductivity and injection charge density, especially when the electrode is shrunk to a micrometer scale for better spatial resolution [148] . To address the impedance - size trade - off in microelectrodes, three dim ensional (3D) nanostructured Au microelectrodes have been developed wherein nanoporous structures were created on the microelectrode surface to achieve larger surface area and therefore lower impedance [175] . The nanotopography of such nanoporous structures also improves in vivo stability of electrode implants by reducing the incidence of glial scar encapsulation while maintaining high neuronal coverage. Surface modification with Au nanorods, nanoflakes, o r nanopillars is another option to increase the effective recording area without change to the overall electrode dimensions [176] [178] . For example, Nick et al. fabricated Au nanopillars on the microelectrodes, showing a reduction of 1 - kHz impedance by up to 89.5 times and dramatic impedance decrease over 1 Hz to 100 kHz [178] . With a determined diameter, a larger high - aspect ratio of the nanopillars results in lower impedance of the electrode. Similarly, Zhou et al. integrated an Au - nanorod array on flexible thin - film microelectrodes using locally patterned anodized porous alu mina as a template [176] . The interface impedance of this 3D electrode was 25 times smaller than that of conventional two dimensional (2D) planar microelectrodes under the same dimensions. Moreover, 3D electrodes modified with Au nanofl akes have also been reported by Kim et al., demonstrating a [177] . An alternative nanostructure for electrode surface modification is Pt black, a nanoparticulate - like Pt formed from electroplating. Zhang et al. show that, with Pt black, the 1 kHz impedance of a 100 - increased from 0.286 mC/cm 2 to 1.906 mC/cm 2 [56] . Furthermore , alloys with two or more metals have been investigated for nanoparticle synthesis to improve the stability of metallic nanoparticles. One example is the bimetallic Au/Pt alloy nanoparticle modified Au microelectrode, which exhibits an average 1 kHz impeda - [30] , as shown in Fig ure 2A . Ir/IrOx (iridium oxide) is another prevailing electrode material and often used in the format of either a bulky wire or a thin film coating [179] [182] . Ir wires are very stiff and highly resistant to corrosion [183] , whereas IrOx thin films are unstable and prone to degradation as electrode dimensions decrease and charge densities increase [143] . However, untreated Ir electrodes suffer from limited charge injection capacity. Ir all oys, such as PtIr, exhibit significantly improved mechanical and electrochemical properties [99] . Cassar et al. electrodeposite d a PtIr coating (EPIC) on the tip of 75 - - diameter microwire electrodes, resulting in reduced electrochemical [184] . 2.4.1.2 Semiconductors Semiconductors can be readily configured into various electronic elements (e.g., sensors, transistors, switches, etc.) with desired properties (e.g., signal transduction, amplification, multiplexing, etc.) to achieve a complex, integrated biointerface syst em [185], [186] . Organic semiconductors provide unique advantages of mechanical compliance, biodegradability, and stretcha bility. Contrastly inorganic semiconductors are more rigid but provide faster response, higher sensitivity, better accuracy, and lower power consumption of biological sensing than organic semiconductors because of high charge carrier mobilities in inorgani c materials [187] . Moreover, advancement in modern semiconductor technologies allows precise batch fabrication of high - performance inorganic semiconductor devices in various architectures at submicron or even nanometer scale, matching the size of subcellular and molecular targets. Silicon (Si) is the most commonly used semiconducting material in neurophysiological implant s. The well - developed microfabrication and photolithography techniques for complementary metal - oxide - semiconductor (CMOS) integrated circuits enable the design and fabrication of high density, high - channel - count multielectrode arrays, capable of mapping ac tivity from large - scale neural networks with high spatiotemporal resolution [47] . As the current sta te of the art, the Neuropixel Si probe developed by Jun et al. integrates 960 recording sites (384 configurable recording 0.3 g, and provides on - chip signal amplification and digitization [188] . Each probe enables stable and chronic recordings from more than 100 neurons for over 150 days while maintaining a large signal to noise ratio [188] . Fig ure 2. Examples of electrode materials. (A) The electrodes of the neurophysiological implants (left) are made of Au/Pt alloy as indicated in the SEM image (right) (reprinted with permission from (Zhao et al., 2016)). ( B ) Transparent graphene as the electrode material (lower left) on the with permission from (Park et al., 2018)). ( C ) Thin - film glassy carbon recording electrodes on flexible polyimide (left). The SEM image (right) shows the surface morphology of the glassy carbon (reprinted with permission from (Vomero et al., 2017)). With high sensitivity to changes in electric potentials and surface charges, Si - based nanostructure materials are also used to make low impedance microelectrode interface for neurophysiology record ing [175], [187] . For example, Si nanowires have been utilized as low impedance nanoelectrodes to intracellularly record action potential from cultured neurons with high precision [189], [190] . In addition , a forest of randomly oriented gold coated - Si nanowires has been shown to achieve noninvasive extracellular recording of astrocytes by mimicking the properties of astrocytes in vivo [191] . Compared to bulky materials, improved stretchability and bendability can be achieved with Si nanowires. Similarly, an amorphous atomic structured Si material has been proposed to create mesostructures with fibrils and voids, with an average Young's modulus of 2 - 3 orders smaller than that of the single - crystalline Si [188] . As key building blocks, nanowires can also be integrated with microporous gel - based scaffolds, yielding highly sensitive and flexible 3D neural probes for mapping the propagation of the action potential [192] . These 3D electrodes offer excellent spatial resolution and stability with little immune response to chronic implantation. In addition, Si nanowires can be configured into field - effect transistors (FETs), capable of sensing neurophysiological signals at a faster switching speed. Unlike faradaic measurement of neural signals through electrodes, the charge carrier density of FETs can be modulated as a function of LFP in surrounding tissues, allowing spikes tracking along neurites and neural networks with single - cell resolution and reasonably high sensitivity [193] [195] . Recently, Yu et al. reported a flexible and bioresorbable neural electrode array based on Si NMs [196] . With biodegradable SiO 2 insulation and PLGA substrate, the whole device was able to degrade in PBS (pH=10) within 15 days. ITO is a well - known n - type semiconductor material that is often utilized in transparent microelectrodes. ITO has high conductivity, excellent transparency over the entire visible spectrum due to a large bandgap of around 4 eV, as well as confirmed biocompatibility [197] . ITO can be grown on either solid or flexible substrates using well - develop ed physical vapor deposition techniques (e.g., sputtering). However, similar to metals, ITO electrodes suffer from increased electrochemical impedance when the electrode size decrease s , leading to undesirable electrochemical reactions with the brain tissue and poor recording quality due to increased thermal noise and ion - based electric fluctuations of surrounding media [159] . In addition, ITO is relatively brittle, making it unsuitable for use in large patterns (e.g., pads or interconnection wires) on flexible substrates [166] . Other semic onducting materials, such as germanium (Ge), silicon germanium alloy (SiGe), indium - doped zinc oxide (IZO), indium - gallium - zinc oxide (a - IGZO), and zinc oxide (ZnO), has also been investigated as recording electrode materials because of their desired elect rical, mechanical, optical, biocompatible, and stable/biodegradable properties [198] [203] . 2.4.2 Organic Materials materials, reducing potential adverse outcomes including inflammation response, glial scar encapsulation, unstable neural recording, and mechanical failure of imp lants [107] . Organic materials also provide significant advantages of easily modifiable surface str uctures, mixed ionic and electronic charge transport, less biofouling/surface oxides, and the wide option of biocompatible materials [103] . 2.4.2.1 Conducting Polymer (CP) CPs, as organic polymers, consist of monomeric compounds linked in chains of alternating single and double bonds, and doped with a stabilizing counter - ion. CPs have the mechanical properties matched with those of biological tissues. Because conjugated polymers have narrower band gaps, electrons can move easily between the conducting band and valence band. CPs can transduce ionic currents to electronic currents through redox reaction in bulk and volumetric charging , resulting in low impedance and high charge storage capacity [20 4], [205] . Due to the diversity and adaptability of synthetic processes, the ionic - electronic transport and biochemical surface characteristics are tunable for improving the performance and stability/biodegradation of CPs [206] . Furthermore, dopants, such as small cations/anions (Na + , Cl - and ClO 4 - ) and large polymeric species (polystyrene sulfonate and polyvinyl sulfonate), can be utilized to improve the electrical conductivity of organic materials by adding electrons to the conduction band (n - doping) or removing electrons from the valence band (p - doping) [207] . PEDOT:PSS is a prevailing class of CPs for neural interfacing applications. PEDOT:PSS possesses many desirable properties, including high biostability, outstanding biocompatibility, and excellent electrochemical properties. Studies show that, with the same electrode area, the electrochemical impedance of microelectrodes is an order of magnitude lower than that of Pt microelectrodes [208] . Khodagholy et al. proposed a PEDOT:PSS - based, high - density NeuroGrid that consists of patterned PEDOT:PSS electrodes with the neuron - size density, capable of simultaneously recording LFPs and action potentials in anesthetized and awake human subjects [46], [209] . The enhancement in electrochemical conductivity of PEDOT:PSS - coated electrodes can be attributed to the increased surface roughness of the electrode, as confirmed by Yang et al. [159], [210] . Their studies show that the average surface roughness (Ra) of the PEDOT:PSS c oated electrode increased from 0.85 nm to 3.33 nm, resulting in dramatically improved charge storage capacity and impedance by several orders of magnitude. Similarly, Pranti et al. reported that trodes increased the electrode surface area, and the corresponding electrochemical impedance was reduced by ~99% [211] . Besides planar films, ordered PEDOT nanostructures can be self - assembled on the electrode surface with surfactant molecules as a template to further reduce the electrode impedance [212] . Abidian et al. also reported that PEDOT - based nanotubes enable ~8 times lower im pedance and much higher charge capacity density than planar PEDOT films, mostly due to the increased surface area [213] . PEDOT:PSS can be applied by spin - coating or ink - jet printing in a low - cost and rapid fashion, but at the expense of poor adhesion with underlying electrode materials. Electrode position techniques, such as electroplating, can improve the bonding strength at the PEDOT - electrode interface, preventing potential risk of PEDOT delamination in the biological environment [213] . A recent study by Boehlet et al. also demonstrates that pre - treating the smooth Pt electrode with po rous Pt structures before the PEDOT deposition can enhance the adhesion between PEDOT and Pt. The PEDOT film deposited on the porous Pt substrate shows no delamination after more than 100 days in accelerated aging tests in PBS [214] . S everal other CPs, such as PPy, poly(aniline) (PANi), poly(thiophene) (PT), and some of their derivatives [215] [217] are also alternative candidates. PPy has outstanding water solubility [216] , good conductivity of 40 - 200 S/cm [218] 2400 Pa for thin films (15 - thick) [219] , and 430 - 800 MPa for nanocomposites [220] . PPy can be electrodeposited in situ on the electrode surface with different dopants. PANi has an electrical c onductivity of 5 S/cm [218] and is primarily used as a coating material on electrodes instead of a standalone electrode material - 4 GPa) [221] . Nanostructured PANi can be synthesized by chemical oxidative or electrochemical polymerization in an aqueous solution that contains a variety of surfactants to precisely tailor the structure of the fi lm at small length scales for increased effective surface area [212], [215] . Functionalized PT copolymer, with precisely tunable electrical, optical, mechanical, and adhesive properties, is also applicable for neural recording electrodes [217] . For PT, the maximum conductivity is 10 - modulus of thin films is ~3 GPa [222] . 2.4.3 Carbon - based Materials Carbon - based mate rials, such as graphene, carbon nanofibers, carbon nanotubes, are another promising class of electrode materials. Carbon - based materials have high biocompatibility and valuable mechanical properties, such as high tensile strength, and can be prepared by va rious approaches, including chemical vapor deposition (CVD), electrospinning, and exfoliation. 2.4.3.1 Graphene Graphene, a 2D single - layer sheet of carbon atoms in a hexagonal arrangement, has a great number of outstanding properties: ~90% optical transmi ttance [17] 4 - layer structure), 200000 cm 2 /V ·s electron mobility [223] , and ~5 × 10 3 W/m · K thermal conductivity [224] [226] . The remarkable biocompatibility makes graphene an appropriate choice for neural interface applications [29], [227] . Moreover, the outstanding transparency of the graphene microelectrode enables simultaneous neurophysiological recording, light stimulation, and optical imaging of living tissues [17] . Despite many benefits, graphene has a large Young's modulus (~1.0 TPa) [140], [228] and a large impedance microelectrodes) at the graphene - electrolyte interface, possibly due to the intrinsi c hydrophobicity of graphene [229] . The comparatively low double - layer capacitance of single - or few - layered graphene could cause considerable thermal noise and low SNR of neural recording. Therefore, it is critical to reduce the me chanical mismatch between graphene electrodes and surrounding tissues as well as to improve the electrical properties of hydrophobic graphene. Small area graphene can be prepared using mechanical exfoliation, which is tedious and time - consuming. CVD allows growing high - quality graphene over large areas at either high temperatures of over 1000 o C or on specific substrates in a specific gas mixture, but is incompatible with polymer materials [230] . Significant efforts have been made in recent years to transfer CVD graphene from rigid substrates onto soft substrates. For example, Park et al. transferred and stacked four graphene monolayers sequentially onto a flexible PA film [17] as the electrode material. Later, the same group reported a transparent carbon - layered 16 - channel array [157] , as shown in Fig ure 2 B . It succeeded in simultaneous in vivo recording of light - evoked neural signals in conjunction with fl uorescence imaging . Similarly, Chen et al. transferred graphene onto SU - 8 and demonstrated that introducing hydroxyl groups on the graphene surface by a mild stream plasma treatment can effectively increase the water contact angle from 91.1 o ± 5.6 o to 41 o ± 4.7 o [229] . The increase in graphene 2 2 and SNR improvement from 20.3 ± 3.3 dB to 27.8 ± 4.0 dB. Besides electrode configurations, Kireev et al. develo ped graphene - based FETs on flexible polyimide - on - steel and found that the device did not show significant loss in recording capability after up to 1000 cycles of mechanical bending [230] . 2.4.3.2 Carbon Nanostructures 3D carbon nanostructures, such as carbon fibers (CFs) and carbon nanotubes (CNTs), can be utilized as a standalone electrode or as a surface coating to improve the surface area and electrochemical imped ance [48], [175], [231] [233] . Standalone carbon fiber microelectrodes (CFMEs) are typically constructed by insul ating carbon nanofibers with pulled glass pipettes [234] or PA [235] [239] followed by opening the electrode tip with chemical etching, plasma removal, or laser cutting. Recently Patel et al. assembled 16 CFMEs to form a multichannel CFME array, capable of chronic recording of single unit activity for one month [236] . Such CFMEs electrodes can be functionalized with electrodeposited PEDOT [236], [239] or IrOx [237], [238] to further improve their impedance and charge capacity density. An alternative method to fabricate CF electrodes is thermal drawing [53] , by which carbon nanofiber (CNF) composites were unidirectionally aligned in cyclic olefin copolymer (COC) as a recording electrode. The as - fabricated fiber had overall dimensions of less than 100 x 100 µm 2 , including a single recording site of CNF com posite with a size ranging from 18 × 11.3 µm 2 to 35.2 × 20.1 µm 2 , and dramatically reduced impedance magnitude by 2 orders compared to the conventional polymer electrodes [53] . Alternatively, Yu et al. demonstrated in situ growth of verti cally aligned carbon nanofibers on pre - patterned Ni catalyst using direct current catalytic plasma - enhanced CVD. The array consists of 40 electrodes in one line with 15 of the CNFs facilitates the penetration of the electrodes into the interior of the tissues or individual cells to improve electrical coupling [240] . Besides pure CNFs, Saunier et al. reported a composite PEDOT:CNF material combining PEDOT with CNFs through electrochemical deposition. The PEDOT:CNF modified microelectrode demonstrates low specific im · 2 at 1 kHz and charge injection limit of 10.03 mC/cm 2 , suitable for multifunctional electrophysiological recording and neurotransmitter sensing. Moreover, CNF has magnetic susceptibility close to water and tissues, making it compatible with high field functional magnetic resonance imaging (fMRI) to enable high - resolution electrophysiological measurements and anatomical studies of large - scale neural networks without electrode interference with MRI images [241] . Unlike the CNFs, CNTs have smaller sizes with higher density and can be divided into single - walled carbon nanotubes (SWCNTs) and multi - walled carbon nanotubes (MWCNTs) [48] . SWCNT is a single graphite sheet wrapped into a cylindrical tube, while the MWCNTs nest several SWCNTs together concentrically, looking like rings of a tree trunk [242] . Perfect SWCNTs have outstanding mecha nical properties and electrical properties, quite similar to the perfect MWNTs due to the weak coupling of nanotubes in MWNTs [243] . Additionally, the rolling direction of the SWCNTs decides the properties that are more like metals or semiconductors [2 44] . For use in neural electrode implants, CNTs can be electrochemically coated on the conventional tungsten and stainless steel wires under ambient environments at low temperatures to improve the impedance and charge transfer properties of the electrod es [245] . Furthermore, the tungsten wires can be etched electrochemically to obtain pure carbon nanotube probes as intracellular recording electrodes [246] . Besides electrochemical deposition, CVD methods can synthesize CNTs directly on the tip of quartz - insulated platinum/tungsten electrodes [247] . Compared to electrochemically deposited CNTs, the che mical vapor deposited CNTs show remarkable mechanical toughness and stability over time. The CVD - CNT - coated microelectrodes can retain unaltered impedance values after one - year storage or after being subjected to a million current pulses at charge injectio n limit. CNT can also be integrated with flexible polymer substrates to implement flexible CNT electrodes. For example, Lin et al. embedded pre - patterned CNT structures into a PA film to create a flexible CNT electrode array with significantly reduced mech anical rigidity and low impedance for the high - quality recording of spontaneous spikes from the crayfish nerve cord [248] . Simi lar to graphene, studies show that the electrical properties of the CNT - based electrodes can be improved by tuning the hydrophilicity of CNTs. For example, plasma/UVO3 treatment of < 10s can alter the surface wettability of CNT from superhydrophobicity to superhydrophilicity, mainly due to the formation of - OH terminations [249] [251] . Amino - functionalization of the MWCNTs surface with a 2wt% 1,4 - diaminobutane solution can also improve the hydrophilicity of the surface, lasting f or at least six months in the air [252] . While widely used in neural electrode implants, the cytotoxicity of these nanostructures is still a big concern, since the nanomaterials can penetrate through the blood - brain barrier (BBB) and cause irreversible cell death and damage to the brain [106], [253] . 2.4.3.3 Glassy Carbon Glassy carbon (GC) offers a wide r ange of mechanical, electrical, and electrochemical properties, which can be specifically tailored with different pyrolysis temperatures under different fabrication conditions to match the properties of the target tissue [184] . Because flexible polymer substrates are unable to tolerate high pyrolysis temperatures, pattern transfer techniques are often used to fabricate GC based, flexible ECoG microelectrode arrays on polyimide substrates [254], [255] , as sh own in Fig ure 2 C . Furthermore, coating GC based microelectrodes with CPs, such as PEDOT:PSS, helps to reduce the impedance magnitude of a 60 - - diameter electrode by at least 2 orders [254] . Most recently, Chen et al. designed and fabricated a cone - shaped glassy carbon neural electrode array using 3D printing and chemical pyrolysis technologies [256] . The electrode had a 0.78 mm 2 recording area exposed at the tip, and the corresponding impedance, capacitance, 2 , and 50.73 ± 6.11, respectively [256] . 2.4.3.4 Diamond In recent years, diamond has emerged as a promising electrode material for neurophysiological recording and neurotransmitter sensing. Boron - doped polycrystalline diamond (BDD) offers unique properties, including wide aqueou s potential window, chemical inertness, capability for surface modification, tunable electrical conductivity, and biocompatibility [257] [260] . Despite [261] and soft tissues is a major obstacle that impedes the development of BDD into fully implantable electrochemical devices. Compared to other semiconducting materials, diamond processing and patterning are more difficult due to its extreme mec hanical hardness, lack of ductility, and weldability [262] . Therefore, attempts have been made to develop new material synthesis and processing methods to fabricate diamond - ba sed electrodes with improved flexibility. For example, Fan et al. demonstrated a wafer - scale fabrication method to transfer large - scale, pre - patterned BDD microelectrode arrays from a solid silicon substrate onto a flexible PA substrate [127], [263] . The electrodes made of the BDD growth side exhibited a rougher topology, a higher sp3 content, and a large grain size than the nucleation side, enabling a wide working potential window, a low background noise, a resistance to chemical fouling, and a reduced electrochemical impedance [127] . 2. 5 Conclusion Investigation of the novel electrode materials is, at present, one of the most prevailing topics in developing advanced neural recording electrodes, as evident by continuous growth in literature. E lectrodes are the most significant element that directly influences the quality of neurophysiology recording . Wit h recent advancements in material science and engineering, implantable electrode interfaces capable of recording neural activity with high spatiotemporal resolution can now be achieved. This chapter showed typical electrode materials associated with the st ate - of - the - art electrode devices, as guidance for future device development. Tables 1 summarize s the main properties of the selected electrode materials. In particular, Table 1 lists the properties of various electrode materials discussed in this chapter , including their electrical properties, biocompatibility, modulus and broad - band optical transmission . With the trend of further miniaturization in large - scale , high - density recording electrodes, many challenges remain unsolved, mostly related to stability, high fidelity of recording, and minimal foreign - body immune responses. Additional studies by some groups are being done on new electrode materials (e.g., dia mond and MXenes) that were not originally used in neural interfaces [264], [265] . For example, Driscoll et al. purposed flexible Ti 3 C 2 MXene microelectrode arrays for in vivo micro - ECoG recording with the benefits of significantly high volumetric capacitance, electrical conductivity, surface functionality, and sensitivity [265] . Wh ile significant progress has been made, comprehensive evaluation of their functionality, long - term stability and biocompatibility is needed to fully realize the true potential of these new materials for use in neural recording interfaces. Besides new elect rode materials (e.g., diamond and MXenes), one research area that has received much recent attention is to design and develop composite materials that combine the unique advantages of different existing materials while eliminating their major drawbacks. Th e use of composite materials in electrode structuring has the potential to bring disruptive changes to single material designs. For example, Pal et al. demonstrate a flexible bio - sensor that combines PEDOT:PSS sensing elements on a fully biodegradable and flexible silk protein fibroin support to achieve excellent electrochemical activity and stability over days [266] . Composite electrode materials can be prepared by in situ electrodeposition or multilayered assembly of inorganic and/or organic conducting materials on planar substrates to achieve the desired electrochemical, biological, optical, and mechanic al properties. Recently, with their tunable composites, configurations, and density, 3D nanostructured materials represent novel electrode materials to further improve the electrochemical impedance and the capacity of the injection charge density, two impo rtant factors that determine the SNRs and recording quality of the electrodes. While promising, the in vivo evaluation of these composite materials is incomplete, preventing their applications in chronic neural interfaces. Recently, transparent neurophysio logy recording electrodes are getting attention because their high potential to work with capabilities of imaging and optogenetics technology. For optogenetics, this innovative technology uses light to activate or inhibit the genetically - modified neuron. T herefore, the more transparent the electrodes, the more light will go through the electrodes and finally arrive at the neuron, leading to higher sensitivity. In addition, high - resolution imaging could be achieved al ong with recording function compared to the opaque materials. Several transparent materials, such as ITO, graphene, and PEDOT:PSS have been under research [17], [166] . Those materials have their own property limitations in six key material characteristics that were discussed in the previous section. Therefore, finding and building up new transparent el ectrodes with excellent characteristics combination are urgent to solve the existing challenges of those transparent materials . Table 1 . Summary of various electrode materials with key properties Electrode materials Electrical properties (impedance @1kHz) Biocompatibilit y Stability Biodegradability /Bioresorbability Mechanical flexibility/be - ndability modulus Optical Transmission (400 - 700nm) Reference GNPs Nanopillars - (0 - 22.5 µm height) Cytotoxic (depend on the size of GNPs) 14 days in vitro Biodegradable Bendable 1 - 10 GPa Opaque [178] [1 76] [177] [267] Nanorods 2 area) 20 times (agarose gel insertion) Nanoflakes - (5 - 50 µm diameter) A month Pt black ~100 µm diameter) Biocompatible 3 days in vivo N/A Bendable N/A Opaque [56] [268] [269] Au/Pt alloy diameter) Biocompatible 7 times (ultrasonic treatments) N/A N/A 113.8 GPa Opaque [30 ] Ir/Pt alloy length, 75 µm diameter) Biocompatible 12 weeks in vivo N/A Bendable 185.5 - 189.6 GPa Opaque [184] Si nanowires - 200 nm tip diameter) Biocompatible 8 days (rodent neurons) Biodegradable N/A 60 - 240 GPa Transparent [190] [270] [271] Table 1 ( ) Si NMs - (200 2 - 500 2 µm 2 ) Biocompatible A month in vivo Bioresorbable Flexible 3.25 - 180 GPa (2 nm - 25 nm thickness) Transparent [196] [272] ITO/PEDOT:PSS - ~ - 80 µm diameter) Biocompatible 4 weeks in vitro N/A Flexible ~77 GPa (on glass) Transparent(> 80 %) [159] [273] PEDOT:PSS/nanostruct ur - ed Pt diameter) Biocompatible 1500 CV cycles N/A Flexible N/A Opaque [214] PPy nanotubes/GNP diameter) Biocompatible Stable Biodegradable Flexible/bend a - ble N/A N/A [216] Diamond mm 2 area) Biocompatible Stab le N/A Flexible (on Parylene C) ~ 10 3 GPa Opaque [127] Graphene µm diameter) Biocompatible 70 days in vivo N/A Flexible ~1 TPa Transparent(> 90%) [17] [274] CNFs 25.7 × 16.6 µm 2 ) Biocompatible 4 weeks in vivo Unbiodegradable Flexible 6 - 207 GPa N/A [53] [275] [276] CNTs - 2 Biocompatible Stable Unbiodegradable Flexible 530 - 700GPa Transparent (~60%) [250] [276] [277] Glassy carbon diameter) Biocompatible (12 days) Stable N/A Flexible 20 GPa Opaque [278] [254] C HAPTER 3 : Characteristics of Transparent, PEDOT:PSS Coated ITO Microelectrodes This chapter covers preliminary work of our research on electrochemical and optical characteristics of flexible, transparent microelectrodes, which consist of thin PEDOT:PSS spun onto ITO electrodes for potential applications in biomedical optoelectronic devices. Although PEDOT:PSS - ITO combined films have been extensively investigated for applications in optical devices, such as solar cells and LEDs, PEDOT:PSS - ITO films for use in electrophysiological recording have not been well - characterized yet. Therefore, this chapter reports on electrochemical and optical characteristics of flexible, transparent microelectrodes, which consist of thin PEDOT:PSS spun onto ITO electrodes for potential applications in biomedical optoelectronic devices. In this work, PEDOT:PSS coated ITO m icroelectrodes with various diameters of 10 µm, 37 µm, 50 µm and 80 µm were microfabricated and characterized, and their properties were compared with plain ITO microelectrodes. Experimental results demonstrate that PEDOT:PSS coated ITO electrodes exhibit decreased electrochemical impedance, well - performed stability in saline, and increased charge storage capacity while preserving excellent optical transparency and mechanical flexibility. The experimental results were fit to equivalent circuit models to ext ract the interface capacitance, charge transfer resistance and solution resistance at the electrode - electrolyte interface. 3 .1 Introduction Optogenetics is a revolutionary neuromodulation technique that utilizes light to excite or inhibit the activity of genetically targeted neurons, expressing light - sensitive opsin proteins [279] . To fully realize the potential of the optogenetics tools, neural interface devices with both recording and stimulating capabilities are vital for future engineering development, and improving their spatial precision is of high interest . While ITO has been used in making transparent recording m icroelectrodes for optogenetics applications [166] , it contains expensive rare elements and has poor mechanical flexibility. Moreover, ITO electrodes, when scaled down to micron dimensions, face a critical problem of increased electrochemical impedance, which would lead to undesirable electrochemical re actions with the brain tissue [21] and poor recording quality due to noisy, ion - based electric fluctuations of the surrounding media [22] . Thus, there is an urgent need to develop low - impedance, transparent microelectrodes as a neural recording interface in combination with optogenetics. PEDOT:PSS, as one of the most prevalent conducting polym ers (CPs), has been demonstrated with many desirable properties for electrophysiological recording, including high biostability, outstanding biocompatibility, and simple utilization with low - cost spin - coating and ink - jet printing techniques [26] [27] . However, PEDOT:PSS has seldom been used independently as a transparent electrode because of its limited electrical conducti vity. PEDOT:PSS can only achieve a sheet resistance of ~40 ohms/sq even when doped with solvents, such as H 2 SO 4 , which is significantly higher than the ITO sheet resistance of ~10 ohms/sq [280] . To address the above challenges, I proposed a combined ITO - PEDOT:PSS electrode configuration by spinning thin PEDOT:PSS films on ITO microelectrodes, for applications in lo w - impedance neural recordings. PEDOT:PSS coated ITO microelectrodes with different diameters were microfabricated and characterized experimentally, and their properties were compared with plain ITO microelectrodes. In particular, electrochemical impedance spectroscopy (EIS) was used to analyze the frequency - dependent impedance of the microelectrodes in a 0.9% NaCl (saline) solution. Equivalent circuit models were applied to fit the experimental results into parameters representing macroscopic physical quant ities related to electrode - electrolyte interface impedance. Cyclic voltammetry (CV) was used to quantify the charge storage capacity of the microelectrodes. Atomic force microscopy (AFM) and ultraviolet - visible (UV - Vis) spectroscopy were used to study the 3 . 2 Materials and Methods 3 . 2 .1 De sign and F abrication of M icroelectrode P robes Figure 3 . (a) Photo of a fabricated micro electrode probe . Microscope images of the ITO electrodes without (b) and with PEDOT:PSS coating (c). 3 . 2 .2 Testing Methods 3 . 3 Results and D iscussion 3 . 3 .1 Electrochemical I mpedance 3 . 3 .2 Charge Storage Capacity 3 . 3 .3 Optical Transmittance 3 . 3 .4 Equivalent Circuit Model 3 .4 Conclusion C HAPTER 4 : A Highly Conductive, Transparent, and Anti - reflective PEDOT:PSS - ITO - Ag - ITO on Parylene C with Tunable Peak Transmittance The previous chapter report ed on electrochemical and optical characteristics of flexible, transparent PEDOT:PSS - ITO based microelec trodes for potential applications in biomedical optoelectronic devices. The PEDOT:PSS coated ITO microelectrodes exhibited significantly decreased impedance and improved charge capacitance over the pure ITO electrodes, suitable for use in neurophysiologica l recording. In addition, the PEDOT:PSS coating improve d the optical transmittance in the visible spectrum by reducing the reflectance at the air - substrate interface. However, the PE T substrate with 127 thickness was not flexible enough to match the low of the brain tissue . Meanwhile, the probe shape design of the array needs to be inserted into the brain tissue while doing the neurophysiology recording, leading to brain tissue damage. Be sides , the brittle 100 nm ITO causes the problems of cracks and disconnection. For further exploring PEDOT:PSS - ITO structure and address ing th ose disadvantages of the design in Chapter 3, ultra - flexible, conductive, transparent thin film s using a PEDOT:PSS - ITO - Ag - ITO were investigated as described in this C hapter . Compared to single - layer ITO of an equivalent thickness, the multilayer film exhibited significantly enhanced sheet conductivity, reduc ed electrochemical impedance, remarkable transmittance, excellent adhesion, and flexibility. The peak transmittance of the combined films can be tailored to a specific wavelength for particular applications, such as optogenetics. Besides Parylene C, our hi gh - quality ITO - Ag - ITO films can be deposited on a wide variety of heat - sensitive substrates over large scales. 4 .1 Introduction Transparent and flexible conductive thin films are critical components in optoelectronics, such as wearable electronics, biosensors, displays, etc. [291] . S ingle - layered ITO, ultrathin metals, graphene or PEDOT:PSS, as conventional transparent electrode materials , have been widely used . However, utilizing only a single material is hard to achieve the desired combination of excellent electrical, optical, mechanical, and biocompatible properties. For example, ITO is most commonly used because of its relatively high transmittance over the entire visible spectrum [292] , biocompatibility [293] , and good electrical conductivity [292] . [294] , especially when relatively thick ITO (100 nm) is required for performing both good transmittance and conductivity. Besides, ITO microelectrodes face a critical problem of the high electrochemical impedance, which wou ld lead to undesirable electrochemical reactions with the brain tissue [295] . Ultrathin metals enable higher conductivity (in the range of ohms) than ITO but cannot achieve high transparency and conductivity simultaneously due to their limited optical transmittance of ~50 - 70 % even with ~4 - 12 nm thickness [296] . Therefore, micro/nanostructured metals, such as nanomesh or microgrid, have been proposed as an alternative, which enables improved transmittance compared to ultrathin metal films [160], [297] . For exampl e, Seo et al. report Au nanomesh electrodes with over 70% transmittance at 550 nm, good flexibility, low sheet resistance of 0.9 - 2 [160] . Electrodepositing PEDOT:PSS on top of Au nanomesh to make bilayer nanomesh furthe r improves the electrical properties [35], [161], [298] . Despite its many benefits, the Au nanomesh significantly reduces the effective recording area because of the large hollow structures on the electrode. Most recently, graphene has emerged as a promising material for making transparent electrodes on account of its broadband transparency, good conductivity, biocompatibility and flexibility [17], [29], [157] . However, it is hard to grow high - quality, large - scale graphene on the polymer substrates because either high temperatures (>1000 o C) or specific substrate materials are required. To address these challenges, engineering methods have been explored to transfer graphene from rigid substrates to flexible polymer substrates [157] . However, those methods are laborious and tedious and reduce the yield and reproducibility. As an attractive conductive polymer, PEDOT:PSS excels in the fabrication of microelectrodes for neural interface applications, owing to its outstanding biocompatibility, excellent stability, transparency, and flexibility [299] . However, compared to other transparent conducting materials, the relatively low electrical conductivity of PEDOT:PSS makes it unsuitable for use in high - density electrode arrays. In this chapter , I fa bricated and reported, for the first time, an ultra - flexible , conductive, and transparent thin film using a PEDOT:PSS - ITO - Ag - ITO multilayer structure on Parylene C for achieving significantly enhanced sheet conductivity, reduced electrochemical impedance, remarkable transmittance, good stability, adhesion, and flexibility. The multilayer assembly was optimized to achieve the lowest theoretical reflectance by simulating the coatings admittance loci under the preferred reference wavelength. Ag was selected to be deposited between two layers of ITO for increasing the conductivity of the whole multilayer structure. Although Ag is not biocompatible, the Ag was sandwiched by two layers of ITO which avoids the direct contact with biological tissues. The ITO and Ag were sputtered consecutively at room temperature, followed by spin - coating of PEDOT:PSS. The room temperature as the sputtering temperature of ITO - Ag - ITO structure was chosen depending on our systematic study, where ITO was deposited on Parylene C coated glass slides at different temperatures, using an RF magnetron sputtering system (Denton Explorer - 14, Denton Vacuum, Inc). The transmittance and sheet resistance of t he combined films were measured by utilizing a Filmetrics thin film analyzer (F20 - UVX, Filmetrics, Inc) and a four - point probe station (SRM - 232, Bridge Technology, Inc), respectively, and the measurements were compared with those of single - layer ITO films of equivalent thicknesses. Tunable peak transmittances were confirmed by depositing different thicknesses of individual layers on Parylene C after theoretical admittance loci simulation under the preferred wavelengths. EIS of different coatings were measur ed in a 0.9% NaCl (saline) solution at room temperature. The adhesion between ITO and Parylene C was evaluated using Scotch tape tests for 50 times of peeling. In addition, the combined thin films were kept at 37°C in air and in saline, respectively to tes a n MTS nano - indenter and compared with their theoretical values. 4 . 2 M aterials and Methods 4 . 2 .1 Fabrication 4 .2. 1 .1 ITO Films D eposition 4.2.1.2 Design Methodolog y for Multilayer Anti - reflection Coatings 4 .2. 1 . 3 Fabrication of PEDOT:PSS - ITO - Ag - ITO Thin Films ( 4 .2.1. 4 TEM Samples Preparation and Test 4 . 2 .2 Testing Methods 4 .2.2.1 Transmittance and Conductivity Measurements Filmetrics thin film analyzer was utilized to measure the transmittance of ITO samples in a wavelength range of 300 nm - 800 nm (F20 - UVX, Filmetrics, Inc) and four - point probe (SRM - 232, Bridge T echnology, Inc) was applied to measure the sheet resistances of the samples. 4 .2.2. 2 Electrochemical Mea surements 4 .2.2. 3 Peel - off Tests Scotch tape test was processed by applying Scotch No. 810 pressure - type tape on the ITO - Parylene C samples with 10 × 10 of 1 mm 2 squares for repeating 50 times with 180 o peel - off ang le to track whether there was any ITO delaminated from the Parylene C substrate. The 10 × 10 of 1 mm 2 square s were cut by using the single edge industrial razor blades to gently cut through the ITO layer but leave the Parylene C thin film a complete sheet. To assure that the ITO layer was completely cut through, a digital multimeter was used to measure continuity between adjacent sides of the cut. 4 .2.2. 4 The bending tests were conducted under a bending curvature of 3 mm and 6 mm in diameter, respectively. The sheet resistances of the combined films with different total ITO thickness were measured every 100 bending cycles within the total 1000 bending cycle of the combined films on Parylene C and single layer of Parylene C were measured using the MTS Nanoindenter XP system, and the experimental value was compared with the theoretical values. 4 .3 Results and D iscussion 4 .3.1 Temperature Study of IT O The crystallization of the ITO films sputtered at room temperature and 110°C was analyzed using transmission electron microscopy (TEM), as shown in Fig ure 1 3 g - 13 j. The room temperature deposited ITO films were mostly amorphous with some degree of crystallites in small areas. When the sputtering temperature increased, the amorphous ITO films were crystallized and densified as demonstrated in Fig. 1 3 g and 13h . The crystallization of thin ITO films can significantl y improve the film quality, resulting in better optical transparency and electrical conductivity. This is consistent with our results obtained from the ITO - on - glass samples and the findings of other groups [303] . 4 .3.2 Conductivity and Transmittance 4 .3.3 Electrochemical Impedance and Charge Storage Capacity 4 .3.4 Stability 4 .3. 5 Adhesion 4 .3. 6 Bending T est 4 .3. 7 4 .4 Conclusion The slight increase in the film sheet resistance during the bending test was believed to be mainly caused by cracks under repetitive mechanical stressing. domi nated by the Parylene C substrate. The flexibility of the combined film could be further such as PDMS. C HAPTER 5 : A Fully Transparent, Flexible PEDOT:PSS - ITO - Ag - ITO Based Microelectrode Array for ECoG Recording Integrative neural interfaces combining neurophysiology and optogenetics with neural imaging provide numerous opportunities for neuroscientists to study the structure and function of neural circuits in the brain. Such a comprehensive interface demands mi niature electrode arrays with high transparency, mechanical flexibility, electrical conductivity, and biocompatibility. In the previous Chapter, an ultra - flexible, conductive, transparent PEDOT:PSS - ITO - Ag - ITO multilayer ed thin films on Parylene C has been designed and fabricated. Therefore, this chapter is focused on designing, fabricating a this PEDOT:PSS - ITO - Ag - ITO assembly on thin Parylene C films. EIS analyses was used to measure the electrical impedance of the microelectrodes with the PEDOT:PSS - ITO - Ag - ITO assembly as well as the electrochemical impedance stability over 12 weeks. electrical recordings with high signal - to - noise ratios (SNRs) (~35 - 36 dB) under different color photostimulations, suggesting that the electrodes are resilient to photon - induced artifacts. The array was also confirmed by - thick, mCherry - labelled, rat brain section. In vivo animal experiments confirm that our array can successfully record light - evoked ECoG oscillations from the primary visual cortex (V1) of an anesthetized rat. 5 .1 Introduction Neurological disorders and diseases in central and peripheral nervous systems, such as worldwide [37] . Besides commercially - available medications and deep brain stimulation approache s , numerous neural interface electrodes have been developed to allow researchers to stimulate and record neural activity with high temporal resolution, permitting quantitative studies of the correlation between neural function and dysfunction. However, most existing microelectrode techniques cannot achieve cell - type specif ic neuromodulation and high spatial resolution due to current spread in tissues. Optogenetics, a technique that uses light to control genetically targeted neurons [312], [313] , has emerged as a viable alternative that overcomes the limitations associated wi th microelectrode technologies [9], [313] . However, optical imaging methods have not achieved the high temporal resolution seen with microelectrodes. Leveraging the advantages of both technologies, hybrid optoelectronic neural interface tools that combine optogenetic neurostimulators with microscale [159], [314] . Although most of the existing ECoG arrays are used to only gather the electrophysiological signals from the superficial surface of the cerebral cortex, some ECoG arrays with high spatia l LFPs [46] . electrodes allow more light to penetrate through the electrode array, and therefore are more beneficial than their opaque counterparts . ITO, ultrathin metals, micro/na nostructured metals, graphene and PEDOT:PSS. However, utilizing only a single material is hard to achieve the desired combination of excellent electrical, optical, mechanical, and biocompatible properties. Depending on what have been explored in Chapter 4, in Chapter 5, a PEDOT:PSS - ITO - Ag - ITO multilayered structure was applied to fabricate the conductive, ultra - flexible, anti - reflective and microelectrodes with the P EDOT:PSS - ITO - Ag - ITO assembly. EIS measurements were conducted over 12 weeks every 3 weeks to check the stability of each different size of microelectrodes, under blue, green, yellow and red light emitting diodes (LEDs) and compared with no LED light, confirming that the influence of LED light on the quality of recorded signals is neglectable. The array was als o confirmed by - thick, mCherry - labelled, rat brain section. In vivo animal experiments confirm that our array can successfully record light - evoked ECoG oscillations from the primary visual cortex (V1) of an anesthetized rat . 5.2 Materials and Methods Parylene C was deposited on a cleaned 3 - inch silicon wafer. Then, a photoresist (PR) layer was spun on the substrate and photolithographically patterned to expose the areas of microelectrodes, interconnection wires, and contact pads. Following ITO - Ag - ITO d eposition using the above method, the wafer was submersed in acetone at room temperature. With the PR mask, ITO - Ag - ITO on the undesired areas was washed off with acetone and the surface was cleaned with IPA and DI water, leaving the ITO - Ag - ITO on the elect rode sites, contact pads, and and then Parylene C on the contact pads and microelectrodes were removed completely using oxygen plasma dry etching (RIE - 1701 pla sma system, Nordson March, Inc). Then, 500 nm copper was sputtered and patterned only on the contact pad with a sacrificial PR mask. Then, the PR was rinsed off with acetone, IPA and DI water for copper lift off. After that, another PR mask was patterned t o expose only the microelectrode sites followed by spin coating of diluted 0.55% PEDOT:PSS using the above recipe. Finally, PR was rinsed off with acetone, IPA, and DI water to remove unwanted PEDOT:PSS, leaving PEDOT:PSS only on top of the ITO - Ag - ITO micr oelectrodes. Instead of using plasma etching, the outer shape of the array was manually cut according to the defined outlines that we initially designed on our mask to prevent the damage of ITO from plasma over - etching. The detailed fabrication steps are s hown in Fig ure 28 . 5.2.2 Testing Methods 5.2.2.1 Electrochemical Impedance EIS measurements were taken using a potentiostat (Electrochemical Analyzer, CH Instruments, Inc.) to analyze the electrochemical impedance in a three - electrode cell, with the microelectrode as the WE, an Ag/AgCl electrode as the RE, and a platinum electrode as the CE. The tests were conducted in the phy siological saline solution (0.9 % NaCl) at room temperature. The electrochemical impedance of the microelectrode was measured from 0.1 Hz to 100 kHz when a 5 mV RMS sinusoid waveform was applied to the WE. CVs of the microelectrode were measured at a 100 m V/s sweep rate in a potential range of - 0.9 V to 0.9 V. Multiple CV scans were done in experiments before data collection to clean the electrode surface and allow the system to settle . 5.2.2.2 Transmittance Measurement s The mCherry fluorescent images and b right - field were taken by a THUNDER imager (3D Cell Culture, Leica, Inc). 5.2.2.3 Signal to Noise Ratios (SNRs) Measurements The microelectrode array and LEDs of different colors were immersed in the saline solution with a separation distance of ~ 0.5 cm. A 1 mVpp sinusoidal waveform at 5 Hz frequency was applied at the opposite ends of the saline solution container. LEDs were kept flashing on the microelectrode panels with applied voltages of Vblue=3.3 V, Vgreen=3.3 V, Vyellow=1.8 V, and Vred=1.8 V, respec tively. The whole setup was placed in a Faraday cage to minimize the environmental noises. The signals were acquired via an RHD2132 system (Intan Techonogies, Inc), and then analyzed with OriginLab and Excel to calculate the SNRs. 5.2.2. 4 In vivo Animal Ex periment One male adult rat (Sprague Dawley, 435 g) was tested based on our established protocols approved by the Institutional Animal Care and Use Committee (IACUC) at Michigan State University. Following the stereotaxic surgery protocol, an adeno - associa ted virus (AAV) vector containing the channel rhodopsin and m - Cherry genes (AAV - hSyn - hChR2 (H134R) - mCherry; the animal facility for 4 weeks until the V1 ne urons expressed channelrhodopsin - 2 (ChR2). During the in vivo experiments, the rat was anesthetized with 2% isoflurane and oxygen (0.8L/min.) and placed in a stereotaxic apparatus. Using sterile surgical procedures, a 3 - 4 cm incision was made in the skin o verlying the skull and a small region of bone was removed to expose the V1 areas. The For the first in vivo using a white LED with a light pulse duration of 20 ms at 0.5 Hz frequency, while the right eye was covered with a gauze pad. During the visual stimulation, the ECoG recording signals were simultaneously recorded through the implanted 32 - channel PEDOT:PSS - ITO - Ag - ITO microelectrodes. S pontaneous ECoG activities were also recorded as control while the LED light was off . A wireless - powered micro - light pulse duration of 20 ms at 1 Hz frequency and with 1.5mW/mm 2 low light intensity or 11.8 mW/mm 2 high light intensity. The microelectrodes on the left panel were illuminated directly by the micro - LED. Light - evoked ECoG activity was simultaneously recorded through the implanted microelectrodes. Spontaneous ECoG activities recorded from the left V1 when the LED light was completely off were used as a positive control, while a negative control consists of ECoG recordings f rom the right V1 when the left V1 underwent high optical stimulation. The recorded signals were amplified and digitalized using a commercial Intan RHD2132 system (Intan Technologies) and then analyzed using the Matlab Chronux toolbox to extract the activat ion energy, phase synchrony, PSD distribution from the recordings. 5.3 Results and Discussion 5.3.1 Fabricated Microscale ECoG Microelectrode Arrays Having characterized the performance of the combined films, a 32 - designed and fabricated using the multilayer PEDOT:PSS - ITO - Ag - ITO film. In the first version of the array design, 32 microelectrodes with 100 µm diameter were distribu ted equally on the two 5 mm × 11 mm panels with the whole device dimension was 12 mm × 34 mm, as shown in Figure 29 . However, it was hard to be attach the whole array panels on the left and right hemispheres of o large. The poor attachment of the microelectrodes with the cortex caused the poor signal recording. The in vivo results were shown in the following In vivo dimensions of 16 mm × 10 mm, with two 3 mm × 3 mm panels to cover both left and right microelectrodes with different diameters: 25 µm, 30 µm, 36 µm, 43 µm and 50 µm. Figure 30 a ure 30c,d show the ITO - Ag - ITO microelectrodes before and after the PEDOT:PSS treatment, and Fig ure 30b demonstrates the flexibility of the array. Compared to the inkjet - printed PEDOT:PSS electrode with the 2 at 1kHz, our PEDOT:PSS treated electrode shows 4 2 [315] . Recently reported electroplated PEDOT:PSS - graphene electrode indicated lower electrochemical impedance of under 0.21 - 2 [316], [317] . However, like most electroplated PEDOT:PSS, the PEDOT:PSS - coated electrode areas h ave low transparency or are not transparent. 5.3. 2 Electrochemical Impedance 12 weeks, during which impedance changes were monitored every three weeks and plotte d in Fig ure 32 a - d . The zoom - in results in Fig ure 32c electrode array exhibited good stability over 12 weeks with an overall impedance increase of less than 8.47%. The average 1 kHz impedance decreased when the electrode area increased (Fig ure 32d ). The average 1 kHz impedance changes were less than 11.8% ov er 12 weeks for all the electrode diameters. This impedance increase may be due to PEDOT:PSS reaction with oxygen upon exposure to air, creating an aqueous acid environment that corrodes ITO [284] . ITO etching chemical will then diffuse through the PEDOT:PSS film and weaken the stability of the ITO - PEDOT:PSS interface, thereby slowly increasing impedance. Another possible reason for the increased impedance could be the corrosion reaction between Ag and the ionic solution that diffused through defects or pinholes in the top - most ITO layer [210] . 5.3. 3 Transmittance Measurements - thick, mCherry - labelled, rat brain section. We inspected both the fluorescent and bright - field images from the same region of the brain section with and without the electrode array in place, as shown in Fig ure 33a - d . We analyzed the light intensities in an arbitrary unit of the mCherry labeled cell bodies (in red dots) within and outside of the transparent microelectrode area. Th e transmittance rate of the transparent microelectrodes was derived by comparing the maximum light intensities of selected labeled cell bodies via Matlab to eliminate t microelectrode area was above 95.7%, indicating the excellent fluorescent transparency of our PEDOT - ITO - Ag - ITO thin films. Also of note is the shadowing effect of the top Parylene C encapsulation over the electrode edge and trace areas under the bright field condition (Fig ure 33 b), due to extra absorption and reflection at the top Parylene - ITO interface. This top encapsulation does not impact the transparency of the array under fluorescent imaging, as indicated in Fig ure 33 a . 5.3. 4 Signal to Noise Ratios (SNRs) Measurements Increased n oise and artifacts from the photoelectrochemical (PEC) effect, also known as the Becquerel effect, is a long - standing challenge for recording light - evoked neural activities in optogenetics. Therefore, we studied the effect of photon - induced artifacts on re cording 34a shows the SNRs of recordings, green, yellow, and red LED light. For all the light conditions, the average SNRs ranged from ~35 dB to ~36 dB with variations of less than 2.7%, indicating that the PEDOT:PSS - ITO - Ag - ITO electrodes are quite resistant to photoelectrochemical artifacts. The maximum forwar d LED voltages applied to different color LEDs were adjusted to achieve consistent illumination. Figure 34 b shows the corresponding optical powers of individual LEDs with forward voltages of V blue =3.3 V, V green =3.3 V, V yellow =1.8 V, and Vred=1.8 V, respect ively, measured in the solution at a distance of 0.5 cm from the LEDs. The blue LED had the strongest optical power of 16.2 mW among all the LEDs at the high applied voltage of 3.3 V. Although the same high 3.3 V voltage was applied on the green LED, the theoretical luminous efficacy of the green LED is much lower than the blue one. While preliminary, these results show that optical stimulation with various light intensity and wavelengths would have minimal impact microelectrodes. 5. 3 . 5 In vivo Animal Experiment For the first version of the ECoG array, the animal experiment was conducted, and the two array panels were attached on both the left and right cortex. However, it was hard to adhere the whole 5 mm × 11 mm panel because of the area mismatch of the large microelectrode array panels and the , as shown in Figure 35a and 35 b . The poor adhesion caused the poor and low - resolution signal recording , as shown in Fi gure 35c and 35 d . With the stimulation on, the recorded ECoG signals show significantly larger variations than the baseline, indicating that the optical stimulation Figure 35 . (a) (b) Images of the in vivo test with the first version of array design . In vivo test data of (c) (d) signal amplitudes with visual stimulation (up and left) and without visual stimulation (down and right). can efficiently evoke neural activity and our transparent microelectrode array can effectively record ECoG oscillations. After reducing the dimension design of the array, the second in vivo acute animal experiments were conducted to evaluate the surgical and functionality of the devices on V1 of rats. During the in vivo array was gently placed on both the left and right lobes of V1 (Fig ure 36 ). The left V1 was stimulated with the light stimuli under high or low intensity, while the right V1 served as a control. Light - evoked ECoG activities were recorded from both V1 lobe s via the implanted microelectrodes, and the neural recordings were analyzed over a time span of 20s and compared with a baseline that was recorded when the LED stimulation was turned off. Figure 37 show that the ECoG activity evoked by the high - intensity stimulus had significantly larger energy variations compared to both the baseline and activity evoked by the low - intensity stimulus. Furthermore, Hilbert transformation was applied to extract the instantaneous phases of 150 trails of ECoG recordings within a frequency range of 1 - 25 Hz. As shown in Fig ure 38 , the 150 trails are aligned to the concurrence of the stimulus and stacked, and the colors indicate the instantaneous phase of each trail. The baseline and control did not show significant phase synchron y, while strong and reliable phase - locked synchronization was observed across 150 trials of signals recorded with the high light stimulation. The instantaneous phase that resulted from the low - intensity stimulation showed only slight synchronization, as expected. Additionally, 150 trial ECoG recordings were averaged and mapped onto a t ime - frequency graph of color - coded, normalized power spectral density (PSD) distribution (Fig ure 39 ). A significant increase of PSD was observed in a short time window of ~200 ms following the light stimulation compared to the control baseline. Higher light intensity resulted in higher PSD compared to that seen with lower optical stimulation [318] . The control also showed increased PSD at the o nset of high - intensity stimulation, which may have been caused by the electrical artifacts of the LED. These results suggest that the optical stimulation can effectively evoke neural activity and that our transparent microelectrode array can successfully r ecord light - evoked ECoG oscillations. 5. 4 Conclusion In summary, a conductive, anti - reflective, transparent, and ultra - PEDOT:PSS - ITO - Ag - ITO multilayer thin films on Parylene C is demonstrated in this chapter . This PEDOT:PSS - ITO - Ag - resistance, remarkable transmittance, good elec trochemical impedance, increased charge storage capacity, excellent SNRs, and the capacity for in vivo ECoG recordings. O ther recently reported transparent ECoG arrays were compared with ours and a benchmark table was made , as shown in Table 4 . Moreover, SNRs were measured under various color LEDs, confirming the negligible photon - induced artifacts of recordings using the combined film. The optical transmittance of the transparent microelectrode array was further demonstrated under both fluoresce nt and bright - field illumination conditions. Finally, in vivo animal experiments validated the ability to use these transparent microelectrodes to record optogenetically evoked ECoG activity from the rat visual cortex. It is worth noting that the Parylene device, particularly in the shorter wavelength range (i.e. blue light), due to the absorptive and reflective loss in Parylene layers. Future studies could fruitfully explore this issue by further a transparent over a broad spectrum. In addition, fabricating anti - reflective moth - eye nanostructure at the backside of Parylene C could be considered to further imp rove the transmittance of ECoG arrays [319] , as well as perform a better antibacterial effect [320] . Instead of increasing deposition temperature, the quality of the room temperature ITO could be improved with other methods, such as ion source assisted sputtering [321] , which not only improves the electrical and optical properties of the ITO film but also enhances the film stability due to increased density with fewer pinholes. An alternative approach to improve the ITO film quality is low - temperature plasma annealing [302] , which has been shown to enable relatively low resistivity (2.5x10 - 3 cm) and high optical transmission (94%) while allowing multiple wafer processing in parallel. C HAPTER 6: Exploratory and Future Work For our previous research, we concentr ated on designing antireflective, peak - transmittance - tunable, transparent PEDOT:PSS - ITO - Ag - ITO thin films and then applied this multilayered structure in the 32 - channel ECoG array followed by showing the efficacy of this transparent microelectrode array. However, the diameters of those microelectrodes are ranging from 25 m to 100 m with the separation ~500 m. The mismatch size of the microelectrode with the neurons could collect more surrounding neuron signals around the targeted neurons, thereby reduci ng the specificity of targeted neuron extracellular electrophysiology recording. Therefore, for future studies , a high spatial resolution transparent array was designed with 120 microelectrode channels and an associated 120 - channel data collecting sy stem. The other direction is to explore the potential of utilizing ITO - based electrochemical sensors for Cu and Mn ions detection . ITO has been widely used in spectroelectrochemistry as a WE material because of its excellent conductivity and good optical transparency. In addition, ITO has a large potential window and are capable of measurements beyond +1.5 V, making it appropriate as a WE material for voltametric methods. Therefore, t his subsection is focused on fabricating and characterizing an ITO - bas e d electrochemical sensors for Cu and Mn ion s detection in bodies and the soil environment. 6.1 120 - channel Transparent ECoG Array 6.1.1.1 Mask Designs A 120 - 4 0 . was designed with dimensions of 34 mm × 24.4 mm, with a single 2.6 mm × 2 mm panel to cover one side of hemispheres of the primary visual cortex (V1). Because it is hard to design 120 interconnections on single layer, 120 microelectrodes and interconnections were split on 2 layers, for each layer with 60 channels. 2 µm Parylene C, as an insulating layer, separated those 2 layers. The single panel consisted of the total 120 microelectrodes with the surface area of 25 × 25 µm 2 and a separation distance of 200 µm. 6 .1.1.2 Fabr ication Procedures F - inch silicon wafer by CVD. Next, a photoresist (PR) layer is spun on the substrate and photolithographically patterned to expose the areas of the first layer of microelectrodes, interco nnection wires, and contact pads. Then, similarly, ITO - Ag - ITO is sputtered using the above magnetron sputtering system via the same recipe. With the PR mask, ITO - Ag - ITO on the undesired areas was washed off with acetone at room temperature and the surface was cleaned with IPA and DI water, leaving the ITO - Ag - ITO on the substrate as an insulating layer, and then Parylene C on the microelectrodes and contact pads were removed completely using oxygen plasma dry etching method (RIE - 1701 plasma system, Nordson March, Inc). Next, another PR layer is spun on followed by the photolithography to expose the areas of the second layer of microelectrodes, interconnection wires, and contact pads. The n, the second layer of ITO - Ag - ITO was deposited by utilizing the above method with the thic kness of each layer. Later, the PR was washed off with acetone and the whole wafer is rinsed next with IPA was deposited on the wafer as an insulating layer, following by the oxygen plasma dry etching to get rid of the Parylene C on the contact pads and microelectrodes. After that, 500 nm copper was sputtered and patterned only on the contact pad with a sacrif icial PR mask. Then, PR was rinsed off with acetone, IPA and DI water for copper lift off. Next, another PR mask was patterned to expose only the microelectrode sites followed by spin coating of diluted 0.55% PEDOT:PSS using the above recipe. Finally, PR w as rinsed off with acetone, IPA, and DI water to remove unwanted PEDOT:PSS, leaving PEDOT:PSS only on top of the ITO - Ag - ITO microelectrodes. Instead of using plasma etching, the outer shape of the array was manually cut according to the defined outlines th at was initially designed on masks to prevent the damage of the ITO from plasma over - etching. 6 .1.2 Connection Part 6 .1.2.1 PCB Design A printed circuit board (PCB) interface was designed for connecting the microfabricated 120 - channel ECoG array and RHD headstages (C3315, Intan Technologies Inc.) , as shown in Figure 41 . The four 36 - position dual row male nano - miniature connectors (NPD - 36 - AA - GS, Omnetics Connector Corporation Inc.) were designed in pairs at the double sides on the edges of the PCB for two 64 - channel RHD recording headstages to be convenient to be plugged in. A 120 - position ZIF connector (FH29B - 120S - 0.2SHW(05), Hirose Electric CO., LTD) is placed for connecting to the high spatial resolution ECoG array and wiring out to the recordin g headstages. 6 .1.2.2 Recording System Setup The whole 120 - channel recording system was well designed and set up, as shown in Figure 42 . When conducting the neurophysiology recording, the 120 - channel µECoG array is inserted into the ZIF connector while k eeping the good connection between the contact pads of connector and array. Next, two RHD recording headstages should be plugged carefully in the female connectors. Four RHD standard SPI interface cables are connecting to the RH USB interface board (C3100, Intan Technologies Inc.). The interface board is powered up and connected to the laptop. RHX data acquisition software is utilized to display and record electrophysiological signals. Then , the data can be analyzed using the Matlab Chronux toolbox to extract the activation energy, phase synchrony, PSD distribution from the recordings. The universal Intan software also can be utilized to plot the signals, probe map, and the spectrogram as well as spike sorting. 6.2 A Transparent and Mini aturized ITO - based Electrochemical Sensor for Continuous and Quantitative Monitoring of the Concentrations of Copper (Cu) and Manganese (Mn) Ions in Bodies and Soil Environment 6 . 2.1 Introdu ction Prion diseases are caused by abnormal folding of proteins and can cause asymptotic neurological changes and death [322] . The most typical prion disease that everyone is familiar with is the mad cow disease , also called bovine spongelike encephalitis (BSE) in cattle . For the case of human prion disease, i n the 1980s and 1990s, more than 4 million cattle were slaughtered in the UK to this disease, and 177 people died from eating tainted beef with variant Creutzfeldt - Jakob disease (CJD) , which sponges out the brain and causes it to lose function [323] . Lowering copper levels in the body can delay the onset of the disease [324] [326] . Researchers at the Scripps Research Institute found that mice lacking a gene that encodes a copper transporter protein survive d significantly longer when infected with prion disease than normal mice [327], [328] . Another metal, Manganese has been found that co uld replace the copper, turning normal prions into alloforms, that cannot be broke down by proteases [329] . Prion proteins associated with BSE can combine with manganese in feed to form pathogeni c prions, and manganese levels in the brains of patients with CJD are 10 times higher than those in the normal brain [330] [332] . Therefore, it is important to detect the copper and manganese levels in the human bodies and animals to prevent and postpone the prion diseases. In recent years, with the rapid development of electroplating, chemical industry, mining and other industries, heavy metal pollution in the environment is increasing day by day. The copper and manganese ions have exceeded seve ral times or even dozens of times of the soil background value [333] . This has been far beyond the endurance of the soil environment, which not only poses a threat to plants, animals, and microorganisms in the soil, but also poses a certain threat to the stability of the entire ecosystem and the safety of human beings. As is known to all, copper and manganese is a necessary trace element for the growth and development of animals. Excessive intake of Cu 2+ and Mn 2+ will cause certain harm to animals. The harm of Cu 2+ and Mn 2+ to an animal is mainly through the food chain into human s and animals [334] . When the body Cu content is too high, it is easy to cause harm to the organs and also facilitate the prion diseases [334] . Therefore, it is also significant to detect the Cu 2+ and Mn 2+ levels in the soil environment to tac kle the root of the problem. The task aims of this project are to develop a miniaturized electrochemical sensor for continuous and quantitative monitoring of the concentrations of Cu and Mn ions in bodies and the soil environment with high sensitivity and accuracy. In addition, the detecting and cleaning methods and recipes are well manipulated for achieving the repeatability of the sensors. Our research was focused on transparent ITO materials. ITO has been widely used in spectroelectrochemistry as a WE m aterial because of its excellent conductivity and good optical transparency. In addition, ITO has a large potential window and are capable of taking measurements beyond +1.5 V, making it appropriate as a WE material for voltametric methods. BDD material wa s also used to study the sensitivity of the devices to Cu and Mn ions when various materials are used for the electrodes . SEM and EDS were done to find the surface changes before and after the ion detection. 6.2 .2 Materials and Methods 6. 2. 2.1 Fabrication of Transparent ITO sensors As shown in Figure 4 3 a , the electrochemical sensor consisted of a circular workin g WE and a surrounding CE on a flexible and transparent polymer PET substrate. Two different conducting materials including ITO and BDD were utilized as the electrode materials in sensor fabrication. To fabricate the device, ITO or BDD were deposited on the polymer substrate and photolithography - patterned to form the WE and CE. Then another 2 m paylene C as insulator was deposited and oxygen plasma etching was processed after the photolithography for exposing the electrode pattern areas. During the cleanroom procedures, 12 sizes WE were fabricated with the diameters scaled down from 4.5 mm to 0.0556 mm , as shown in Figure 4 4 . After the initial tests with 12 different WE areas, two different device sizes were narr owed down with the WE diameters of 1mm and 0.125mm, considering the tradeoffs between miniaturization and sensing performance. After each sensor w as trimmed individually, w ires connected to the test instrument were soldered onto the contact pads of the sensor using low - temperature silver pastes. 6.2.2 .2 Testing Methods 6.2 .2 .2.1 Test Setup A three - cell differential pulse stripping voltammetry (DPSV) was chosen to detect different concentrations of the ions due to the high sensitivity and accuracy of this scanning method, as shown in Figure 4 3 b . An Ag/AgCl electrode served as the RE. A potent iostat (CH Instrument 620A) was used to connect the WE, CE and RE respectively with green cable, red and white cable. During the measurements, electrochemical sensors and RE were immersed in the Cu or Mn solutions . A magnetic stir bar was placed at the bottom of the beaker to well mix the solutions during the deposition steps. The whole ex periment setup was placed on a magnetic stirrer. 6. 2. 2.2.2 Cu and Mn Ions Solutions Preparation For initial studies, simulated so lutions were prepared by mixing different concentrations of CuCl 2 and MnCl 2 powder in the Dulbecco's phosphate - buffered saline (DPBS) , mimicking the human body environment. For CuCl 2 and MnCl 2 solutions , low concentration and high concentration groups were separated to clearly study the detection window and sensitivity of these electrochemical sensors. For the low concentration (LC) group of Cu 2+ , solution concentrations are 0, 500 n M, 1 M, 50 M, and 100 M , respectively. For the high concentration (HC) group of Cu 2+ , we have solution concentrations of 0, 500 M, 1 mM, 2.5 mM, and 5 mM. For the Mn 2+ , we have 0, 1 M, 10 M, 50 M, and 100 M for the low concentrations while the high concentrations are 0, 500 M, 1 mM, 2.5 mM, and 5 mM. 6. 2. 2.2.3 Testing and Cleaning Recipes During the experiment, the CuCl 2 and MnCl 2 solutions were kept at room temperature and agitated using a magnetic stirrer. The DPSV scan for Cu 2+ started with a deposition cycle at - 1V followed by a 5 secs equilibrium cycle with a scanning range of - 1V to 0.5V. For Mn ions, the deposition was done at 1.2V and the equilibrium cycle was 5 secs from 1.2V to - 1V. The deposition duration of the DPSV scan was adjusted to achieve the best sensitivity and repeatability. Particularly, for Cu 2+ , the deposition duration was 90 secs at - 1V for the low concentra tion range (500nM - 100µM) and 60 secs at - 1V for the high concentration range (500µM - 5mM). For MnCl 2 , the deposition duration was 15 secs at 1.2V for both the low concentration and high concentration groups. Between each measurement, the electrode surface w as cleaned by holding the WE at 0.3V and - 0.3V for 30 secs for Cu 2+ and Mn 2+ , respectively , for removing any deposit from previous measurements. For the experiment, we put the single sensor in the same concentration of Cu or Mn solutions for at least 10 - ti me measurements to check the repeatability of our electrochemical sensors. T he single sensor was also utilized to measure the peaks when the concentrations of the solution increases. After each test, we used the cleaning recipe to eliminate the ions on the WE surface and made the sensor ready for the next measurement. 6. 2. 3 Results and Discussion 6. 2. 3.1 Cu / Mn Ions For the sensors made of ITO, DPSV signals at each concentration of Mn 2+ were clearly distinguishable from the baseline and from each other, as shown in Figure 4 6 . However, the ITO sensors were insensitive to Cu 2+ as evident in Figure 4 5 . In Figure 4 5 a, the only peak we observed was under 50 M. It is because that the concentr ations of 500 nM and 1 M were too low for the 0.125 mm WE to detect Cu 2+ , while 100 M concentration was too high and caused too much absorption of Cu 2+ that damaged the ITO WE. In Figure 4 5 c, the concentrations of 500 m - 5 mM were all too high and the severe absorption cause d the irregular shape and the shift of peaks. Similar to the 1 mm WE, the valid peak detected under LC of Cu 2+ was 1 M. The concentration of 500 nM w as too low for the 1 mm WE to detect Cu 2+ , while 50 M and 100 M concentration was too high and caused too much absorption of Cu 2+ that damaged the ITO WE. ITO electrochemical sensors showed excellent sensitivit y to Mn 2+ compared to Cu 2+ . Figure 4 6 e , f showed smooth and clear peaks for both 0.125 mm and 1 mm WE under b oth LC and HC MnCl 2 solutions. According to the results of Figure 4 6 f and 4 6 g, the higher concentrations of the Mn 2+ in the solution, the sharper the peak obtained, and the absolute magnitude was larger. However, in Figure 4 6 e and 4 6 h, under some concentrations, as the concentration increased, the peak magnitude decreased which we attribute to the absorption of the ions accumulated on the WE surface. For 0.125 mm WE, the magnitude decreased when the concentrate rises to 100 M. For the 1 mm WE , the magnitude decreased when the concentration changed from 500 M to 1 mM. 6. 2. 3. 2 Linearity and Repeatability To evaluate the reliability and repeatability of the ITO sensor, we repeated the DPSV scan for at least 10 times, using the 0.125mm and 1 mm WE sensor in Mn 2+ solutions from 1µM to 100µM. The results in Figure 4 9 a show that the DPSV signals detected from the low concentr ation solutions when using 0.125 mm WE did not exhibit significant difference (STDs < 5%) while the variation increased to 7.2% as the concentration increased to 100 µM, mainly due to the absorption of chemicals on the electrode surfac e. The DPSV signals from 10 measurements were averaged and fitted to a linear curve as shown in Figure 4 9 c for 0.125 mm WE . The sensor exhibited a good linear response with correlation coefficient, R 2 The results in Figure 4 7 and Figure 4 8 show tha t the DPSV signals of 1 mm WE have less repeatability and stability compared to the 0.125 mm WE . The magnitudes showed the more obvious fluctuation. Therefore, t he results indicate that with smaller WE and lower concentration of MnCl 2 , the sensors have higher repeatability and 10 - time DPSV repeatability tests with the l ow concentrations (LC) of (a) 1 M, (b) 10 M, (c) 50 M, and (d) 100 M MnCl 2 with 0.125 mm diameter WE. After each DPSV, a cleaning procedure was processed before the next time DPSV. Blue lines show each time the peak current magnitude and the trend during the 10 - time DPSV repeatability tests. stability compared to the larger WE and higher solution concentration. The blue lines shown in Figure 4 8 demonstrate how the peak current magnitude changed during the 10 - time tests. Figure 4 9 . (a) Peak current magnitudes and standard deviations (STDs) of 0.125mm diameter WE under different MnCl 2 concentrations. (b) DPSV of single use 0.125mm diameter WE sensors for each concentration. (c)Linear fitting of the peak current magnitudes under different concentrations with 0.125mm diameter. 6. 2. 3. 3 SEM and EDS We also inspected the electrode morphology using scanning electron microscopy (SEM) and did not observe any degradation or physical damage of the electrode after 10 scans (Figure 50 ). We also did the EDS before and after 10 repeated and continuous experime nts, no detectable Cu 2+ and Mn 2+ were shown in the EDS results. 6. 2. 3.4 BDD E lectrode s We also tested the sensitivity of the BDD electrode to Cu 2+ , since the ITO sensors could not detect Cu 2+ effectively. For these experiments, the DPSV scan was done using the following protocol: deposition duration of 90 secs at - 1V, 5 secs equilibrium at - 1V to 0.5V. Compared to the ITO electrodes, the BDD electrodes show worse sensitivity to Mn 2+ but better sensitivity to Cu 2+ in the low concentration range, as shown in Figure 51 where distinct peaks can be detected for different Cu 2+ concentrations. Further investigations of other material s and device design s to maximize the se nsitivity of the sensors are being investigated . 6. 2. 4 Conclusion and Future Study For this project, a miniaturized ITO - based electrochemical sensor was developed and characterized for continuous and quantitative monitoring of the concentrations of Cu and Mn ions in bodies and the soil environment with high sensitivity and accuracy. The DPSV method was selected and the testing and cleaning recipes were adjusted for achieving the repe atable and stable usage of the sensors for LC and HC of Cu and Mn i ons . We found that transparent ITO material has high sensitivity to Mn 2+ , while BDD materials has better sensitivity to Cu 2+ . When the WE is smaller and the concentration of MnCl 2 is lower, the sensors have higher repeatability and stability compared to the larger WE and higher solution concentration. No degradation or physical damage of the electrodes was observed after 1 0 scans in terms of SEM / EDS . According to the experiment results, conv entionally - deposited transparent ITO material has high sensitivity to Mn 2+ but relatively low sensitivity to Cu 2+ . Although BDD also has been tested and showed high sensitivity to Cu 2+ , the sensitivity to Mn 2+ is lower compared to ITO. Therefore, in future work, higher density ITO materials, such as ITO with Ion - beam assisted deposition (IBAD), are expected to achieve with the higher both Cu 2+ and Mn 2+ sensitivity, repeatability, and viability by utilizing DPSV scan method. IBAD is a technique that combines the physical vapor deposition (PVD) with ion implantation. The ion bombardment dramatically controls the morphology, density, internal stresses, and crystallinity of the thin films. The conventional method of PVD to grow high qua lity ITO (high density, low resistivity, high transparency) is to increase the temperature during the deposition process. However, the required high temperature is not compatible with the flexible polymer substrate, such as Parylene C and PET. Consequently, IBAD is a good method that solves the high temperature requirement issue and improves the thin film quality at the same time. Because of the material properties changes due to the IBAD, the expected deposited ITO/PET film could have higher sensitivity to the Cu 2+ , which makes it a possible way to solve our previous ion sensors problem shown in above results . APPENDIX Table 5 . Admittance loci simulation Table 5 ( ) This Excel was used with specific reflective index (n) value s of PEDOT:PSS, ITO, Ag, and Parylene C materials. This Excel can also be used for other multilayered thin film design with various materials for calculating each layer thickness for achieving the highest transmission and lowest reflection. Detailed equations were demonstrated in Chapter 4. B IBLIOGRAPHY B IBLIOGRAPHY