Next Article in Journal
Yes/No Quantitative Analysis of Single-Stranded Oligonucleotides with Lateral Flow Assays
Previous Article in Journal
Advances in Extracellular Vesicle-Based Surface-Enhanced Raman Spectroscopy for Cancer Diagnosis
Previous Article in Special Issue
Ultra-Sensitive Detection of Mercury by Using Field-Effect Transistor Biosensors Based on Single-Walled Carbon Nanotubes
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Biosensors
Volume 16
Issue 9
10.3390/bios16090466
biosensors-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Dual-Mode Neural Amplifier Array for Biopotential and FSCV-Based Neurochemical Measurements

by
Matthew A. Crocker
,
Kevin A. White
,
Mahdieh Darroudi
,
Vishnu Saket S. Bapanapalli
,
Charles S. Lipscomb
,
Benjamin S. John
and
Brian N. Kim
*
Department of Bioengineering, University of Texas at Dallas, Richardson, TX 75080, USA
*
Author to whom correspondence should be addressed.
Biosensors 2026, 16(9), 466; https://doi.org/10.3390/bios16090466
Submission received: 30 June 2026 / Revised: 12 August 2026 / Accepted: 23 August 2026 / Published: 26 August 2026
(This article belongs to the Special Issue Nano-Biosensors and Their Applications for In Vivo/Vitro Diagnosis—3rd Edition)
Browse Figures
Versions Notes

Abstract

The simultaneous measurement of biopotential and neurochemical signals provides a comprehensive view of the brain. Yet, most neural interfaces record solely biopotential or neurochemical activity. This work presents a complementary metal-oxide-semiconductor (CMOS) analog front-end (AFE) chip that integrates 32 biopotential amplifiers and 32 neurochemical amplifiers for parallel recording from 64 electrodes. The biopotential amplifier is a two-stage design providing a gain of 57.1 dB, a bandwidth of 0.4 Hz–6.2 kHz, and 6.7 µVRMS input-referred noise (20 kHz sampling rate). The neurochemical amplifier is a rail-to-rail folded-cascode operational amplifier with selectable transimpedance gain (91.9 kΩ to 851.9 kΩ), a dynamic range of ±15 μA to ±2 μA, respectively, a bandwidth of 12.6 kHz, and input-referred noise as low as 46.3 pARMS (20 kHz sampling rate). The neurochemical amplifiers are designed for fast-scan cyclic voltammetry (FSCV) measurements. I/O complexity is minimized using a time-division multiplexing scheme for readout, enabling straightforward scalability. The chip is fabricated using a 0.35-µm CMOS process and occupies a 3.0 × 8.3 mm2 area. In vitro recordings of catecholamines and neural spikes validate the chip’s function. The chip enables scalable, low-noise, bimodal neural recording, supporting investigations into the dynamics between neuronal biopotential activity and neurochemical signaling.

1. Introduction

Technologies for recording neuronal action potentials and local field potentials have improved substantially in recent years, improving both spatial and temporal resolution. In contrast, characterizing the heterogeneous spatiotemporal distribution of neurotransmitters, such as catecholamines like dopamine, remains a challenge. Catecholamines act as key modulators of motor control, motivation, and reward-related behaviors [1,2,3,4,5]. Characterizing the heterogeneous spatiotemporal distribution of catecholamines provides insight into their function. Existing approaches for analyzing neurotransmitters in the brain include microdialysis, functional magnetic resonance imaging, and assorted electrochemical techniques. Microdialysis extracts cerebrospinal fluid directly from brain tissue, but delays between sampling and analysis prevent real-time monitoring. Functional magnetic resonance imaging enables high-resolution mapping of neurotransmitter dynamics and has been applied to the analysis of catecholamines [6,7,8,9]. However, its sensitivity is insufficient for resolving sub-micromolar fluctuations [8,9]. Fast-scan cyclic voltammetry (FSCV), an electrochemical technique, is widely used to measure catecholamines in vivo [10,11,12,13,14,15,16,17,18,19,20,21]. In FSCV, a triangular voltage waveform (typically ranging from −0.4 to 1.3 V at 100–400 V/s) drives redox reactions at the electrode surface. The amplitude of the measured redox signals, the oxidation current during the positive ramp and reduction current during the negative ramp, directly relate to the concentration of redox molecules at the electrode surface. Additionally, FSCV provides chemical specificity for measurements; the potential the maximum amplitude of the oxidative and reductive signals occur can be used as identifying features specific to the measured chemical. Redox signal recording using FSCV for in vivo measurements can be on the scale of nanoamperes [21]. Using FSCV, a relatively large ‘background’ current is recorded due to the electrode’s double-layer capacitance (on the scale of microamperes). To isolate the small faradaic signals from redox reactions, background subtraction is used to remove the baseline background signal. Additional techniques, like modifying FSCV’s waveform, can be employed to optimize measurements for specific targeted chemicals [22,23].
To capture the spatiotemporal heterogeneity of neurochemical signals using FSCV, high-density microelectrode arrays (MEAs) can be used. Individual electrodes can localize redox reactions to individual sites, thus enabling spatially resolved tracking of neurochemical dynamics. Each microelectrode requires signal amplification, and compact circuitry is essential for in vivo applications. Complementary metal-oxide-semiconductor (CMOS) technology addresses these requirements by enabling dense integration of amplifier arrays in small footprints, supporting high-throughput measurements.
Although this work focuses on central nervous system applications, the same multimodal approach can be applied to the peripheral nervous system. For example, FSCV has been used to measure real-time norepinephrine release in the splenic neuroimmune circuit [24]. Combining neurotransmitter measurements with biopotential recordings could provide additional information about autonomic regulation and neuroimmune signaling. These applications further highlight the need for scalable, low-noise front ends that can support diverse electrode technologies.
FSCV-based technologies enable both in vitro and in vivo neurochemical measurements [25,26,27,28,29,30,31]. TinyFSCV supports single-channel FSCV with scan rates up to 400 V/s and an input range of ±115 µA using discrete components, at the cost of a large 40 mm × 24 mm footprint [25]. IC implementations reduce area while enhancing noise performance, such as a single-channel cyclic voltammetry chip supporting 0.008–400 V/s scan rates with 25 nARMS of noise with a sampling rate of 50 kS/s in a 1.6 mm × 2.0 mm footprint [27]. A 128-channel array achieves 220 pARMS noise over 20 kHz of bandwidth with a ±5 µA dynamic range [28]. Other designs include an analog front-end (AFE) chip with 39.2 pARMS of noise (5 kHz of bandwidth) [29]. A wireless single-channel system achieves 92 pARMS noise over 2 kHz of bandwidth with a ±430 nA dynamic range [30].
High-density FSCV recordings require a neurochemical front-end that combines high gain with a dynamic range able to accommodate the large background currents at the electrode-electrolyte interface. The current’s magnitude depends on electrode material, geometry, surface treatment, and scan rate. A typical 7 µm carbon fiber electrode produces sub-µA background currents [32,33]. A similarly sized planar gold electrode can produce 0.5–2 µA [34,35,36]. While a coated gold electrode can reach 2+ µA. The background charging current is approximately proportional to the interfacial capacitance and scan rate, and the total input range needs to accommodate both the background charging current and the faradaic currents from neurotransmitter redox reactions. Changes in electrode design and surface modifications can significantly affect the required input range [19,22,37,38,39,40,41]. Selectable gain allows the same front-end to accommodate different electrode configurations while maintaining sensitivity to small redox currents. Different electrodes can therefore be tested while the AFE remains untouched.
Recent CMOS-based devices for in vivo biopotential recording achieve significant advances in channel density, noise performance, and power efficiency [42,43,44,45,46,47,48,49]. A 256-channel integrated circuit (IC) achieves 6.32 µVRMS of noise over a 10 kHz bandwidth with 10-bit ADCs in a compact 1.53 mm2 footprint [42]. The Neuropixels 2.0 extends scalability to 384 simultaneously recordable channels and incorporates 14-bit ADCs, while achieving 7.2 µVRMS of noise over a 10 kHz bandwidth [43]. A 512-channel design achieves 4.8 µVRMS of noise over a 300–6000 Hz bandwidth [44]. A 1024-channel IC, which reduces requirements for data throughput by discarding redundant baseline samples, achieves 7.4 µVRMS of noise (300–6000 Hz) in a 36 × 36 µm2 footprint, consuming only 268 nW of power per channel [45]. At the extreme end of parallel recordings, a microwire array demonstrates ~30,000 in vivo recordings with 6.32 µVRMS noise (300–6000 Hz). However, the design requires a large amount of support circuitry, limiting its use to acute head-fixed environments [46,47,48]. Wireless approaches have also emerged, such as a dual-band transmitter achieving 54 Mb/s with 8.0 µVRMS of noise (10–10,000 Hz bandwidth) in a 3.72 × 3.68 mm2 area [49].
While the devices that have been referenced demonstrate sufficient capability for in vivo applications, they are limited to either biopotential or neurochemical modalities without concurrent recording capabilities. While it is possible to use a single electrode for multiple modalities, the interleaving of measurements can introduce prohibitively long settling periods after switching between modalities [50]. As a result, we focus on systems that present concurrent modalities using separate electrodes for recording. A limited set of integrated devices enables simultaneous non-interleaved multimodal sensing [51,52,53,54,55,56,57]. A two-channel AFE chip supporting both local field potential recording and FSCV achieves 5 µVRMS of noise (1-200 Hz bandwidth) for the biopotential channel and 50.2 pARMS noise (100 kHz bandwidth) with a gain of 100 kω for the neurochemical channel [51]. A 200-channel MEA demonstrates dual-mode operation, with neurochemical sensing (93 pARMS of noise, 10 kHz of bandwidth) and biopotential measurements (4.07 µVRMS of noise, 1–100 kHz of bandwidth), though the neurochemical input range is restricted to ±50 nA [52]. Our prior work is a 512-channel electrophysiology and electrochemical dual-mode CMOS chip with 256 electrophysiology amplifiers, 256 electrochemical amplifiers, and an on-chip microelectrode array of 512 electrodes, designed for in vitro studies of action potential propagation and synaptic neurotransmitter secretion across neuronal networks. This work splits the amplifiers evenly between modalities, achieving 24.9 µVRMS of noise (0.2–10 kHz of bandwidth), and 4.51 pARMS of noise (10.3 kHz of bandwidth) [57]. Despite these advances, the number of devices supporting large-scale, low-noise, simultaneous multimodal recordings is still extremely limited.
This work presents an AFE IC integrating 32 biopotential and 32 neurochemical amplifiers for future simultaneous electrophysiological and neurochemical recording. The neurochemical front-end provides wide dynamic range, low noise, and selectable gain settings to accommodate microelectrodes with varying capacitances and background currents. Each 32-channel array is time-division multiplexed to a dedicated output for scalable integration.

2. A Highly Scalable Neural Amplifier Design

The biopotential and neurochemical amplifier array are fabricated using a 0.35-µm CMOS process. The design balances low-noise integration with high dynamic range capabilities. The array contains 64 neural amplifiers, with 32 allocated to biopotential recording and 32 to neurochemical sensing.

2.1. Neurochemical Amplifier

The amplifier is designed for FSCV, which demands detection of large, rapidly varying displacement currents (microampere scale) caused by the electrode-electrolyte double-layer capacitance at the electrode surface. While the simplest solution to accommodate the large current in the measurement is to reduce the gain, this limits the resolution of small redox peaks caused by electroactive molecules such as dopamine. Prior works have adopted a background subtraction strategy to remove the large displacement current [29], however, this approach is not attractive to existing users who rely on the shape of the background current to evaluate the real-time condition of the electrode, which can deteriorate over time due to various fouling mechanisms. Therefore, the best approach is to maximize the dynamic range while maintaining the resolution. During FSCV measurements using microelectrodes, the targeted transimpedance gain is typically between 0.1–1 µA/V (100 kΩ–1 MΩ).
The device uses 5 V thick-oxide transistors, with an operational amplifier based on a rail-to-rail folded-cascode design to maximize the recording dynamic range (Figure 1a). The rail-to-rail architecture uses complementary differential pairs: an nMOS differential pair (M1–M2) and a pMOS differential pair (M3–M4). The nMOS differential pair folds into a pMOS cascode (M7–M8), while the pMOS differential pair folds into an nMOS cascode (M9–M10).
The neurochemical amplifier is designed using resistive feedback (Figure 1b) [28,58]. The feedback resistor determines the transimpedance gain. Variable gain is implemented with resistors R1–R4 (50 kΩ, 50 kΩ, 100 kΩ, and 800 kΩ) that are controlled by the switches SW1–SW4. Closing a switch removes the corresponding resistor from the feedback loop, reducing the amplifier gain. For example, closing SW3 and SW4 sets the gain to R1 + R2. With all switches open, the gain is equal to R1 + R2 + R3 + R4.
The RC low-pass filter (RLPF, CLPF) is used to set the amplifier bandwidth. Each amplifier occupies 200 µm × 600 µm (0.12 mm2). The total area the 32 neurochemical amplifiers occupy is 3.84 mm2.
During FSCV recordings, it is common to measure large background currents in the range of a few µAs due to the electrode’s double-layer capacitance. The magnitude of the background current is directly related to the material and surface area of the electrode, as well as the scan rate being used. For a typical carbon fiber electrode setup, the background current is below 1 µA [32,33]. However, larger microelectrodes can have background current that is in the range of 10 µA or more [34,35,36]. To accommodate large background currents, the biasing level of the neurochemical amplifier must be chosen carefully. Because the biasing current is determined by an external resistor, the power consumption can be adjusted based on the anticipated background current level. In this experiment, the biasing current (IbiasN) is set to 28.8 µA to provide sufficient headroom over the anticipated background current (~10–15 µA) by the electrodes used in the later section. A total biasing current of 3 × IbiasN (86.4 µA, 432 µW) is used, giving each neurochemical amplifier the ability to handle the broad range of background currents it may face. For a typical CFE producing sub-µA background currents, a ±2 µA dynamic range is sufficient and can be achieved with a lower biasing current, reducing the neurochemical amplifier power consumption to 30 µW.

2.2. Biopotential Amplifier

The biopotential amplifier is designed for compact, low-noise, high-performance signal amplification of neural activities, including both local field potentials (LFP) and action potentials (APs).
The design employs two stages of amplification. The first stage is based on a low-noise operational amplifier (Figure 2a). It adapts a folded-cascode topology, with M2 and M3 as the differential input pair and M8 and M9 as the cascode stage (Figure 2a). The M1 transistor provides feedback regulation in which the current through M2 and M3 matches the biasing current at M4 [59]. This amplifier is represented as OPA1 (Figure 2b). Stable references (Vpos1 and Vpos2) are tied to the non-inverting inputs of both stages and set the DC operating point.
The second stage employs a five-transistor operational amplifier. The inverting inputs of OPA1 and OPA2 are not DC-stable without a resistive path. For DC-stable operation, pseudo-resistors in parallel with C2 and C4 are used [60]. These pseudo-resistors are pMOS transistors with the body and source tied together, and the gate and drain tied together. These provide larger resistance per unit of area than conventional resistors. The total biasing current supplied from M10 and M11 is 19.4 µA, resulting in a power consumption of 97 µW per amplifier. The ratio of the first stage capacitors, C1/C2 (30 pF/200 fF), is multiplied by the ratio of the second stage capacitors, C3/C4 (2 pF/400 fF), to determine the theoretical gain. Unlike the neurochemical amplifier design, the biopotential amplifier is designed to have a fixed gain. The total gain has been set to 750 V/V.
Each biopotential amplifier occupies a 200 µm × 600 µm (0.12 mm2) area, equivalent to the neurochemical amplifiers. In total, the 32 biopotential amplifiers occupy 3.84 mm2. The array minimizes its footprint while maintaining low noise and high performance, enabling high-density amplifier arrays for neural recording systems.

2.3. Multiplexing and CMOS Implementation

The amplifier array output is read out through two separate time-division multiplexing circuits to reduce the total number of outputs and allow greater scalability [28,57,61,62,63,64,65,66]. Time-division multiplexing employs shared lines that enable sequential readouts of amplifier outputs through a single output stage, which significantly reduces wiring complexity which can provide a significant reduction in the footprint size.
The multiplexing circuit is based on a rail-to-rail unity gain amplifier. The multiplexing circuit utilizes a half-shared operational amplifier design in which multiple non-inverting halves of the amplifier (M1N/M1P–M32N/M32P) share a single inverting half of the amplifier (M33N/M33P) (Figure 3) [66]. Each neurochemical and biopotential amplifier output is connected to an input of their respective multiplexing circuit (MUX1–MUX32). Closing each switch (SW1–SW32) connects the non-inverting halves to the inverting half of the amplifier. For instance, when SW1 is closed, the gate of M1N/M1P (MUX1) becomes the positive input, and the gate of the shared inverting transistors (M33N/M33P) is the negative input of the differential pair. By connecting the input (MUX1) with the output of the amplifier (MUXOUT), the pairs form a unity-gain amplifier which outputs the selected input voltage (MUX1). Sequential activation of SW1 to SW32 ensures only one amplifier is being read out, producing a single output (MUXOUT) that multiplexes all the input signals from MUX1 to MUX32. This output connects to external acquisition hardware, requiring the multiplexer to be capable of driving a large capacitive load. A class AB output stage (M38–M43) is employed to handle this requirement. The power consumption of the multiplexing stage is 55.9 µW, and the total power consumption including the class-AB output stage is 479.6 µW. An identical multiplexing circuit is used for both the neurochemical amplifier array and biopotential amplifier array (Figure 4a).
The timing circuit and the multiplexer output interface with a custom data acquisition system. The data acquisition system synchronizes sampling of the analog-to-digital converters (ADCs) with the on-chip timing circuit.
The CMOS chip integrates 32 neurochemical amplifiers, 32 biopotential amplifiers, timing circuits, multiplexers, and bond pads while occupying only a 3 mm × 8.3 mm area, a total area of 24.9 mm2 (Figure 4b). The CMOS chip is designed for use with MEAs, enabling the spatially dense recordings critical for resolving small-scale neuronal activities. The integrated design supports large-scale recordings by minimizing interconnect complexity while maintaining high signal fidelity.

3. Performance and Characteristics

The gain, bandwidth, and noise performances of the neurochemical and biopotential amplifiers are characterized and discussed.

3.1. Characteristics of the Neurochemical Amplifier

All the experiments use a USB-6363 DAQ (National Instruments, Austin, TX, USA) for analog data acquisition. Input signals are generated using either the USB-6363 DAQ or an AFG1062 (Tektronix, Beaverton, OR, USA) function generator. A custom-designed LabVIEW program generates chip control signals. Battery-powered voltage regulators power the chip, minimizing line-frequency noise. A breakout printed circuit board (PCB) is used for straightforward access to each of the chips’ I/O connections.
The neurochemical amplifier has variable feedback resistors that can be switched to determine five gain settings (GS). The 32 neurochemical amplifiers have gains of 92.5 ± 1.1 kΩ, 134.5 ± 1.7 kΩ, 177.1 ± 2.1 kΩ, 691.6 ± 13.0 kΩ, and 854.0 ± 18.8 kΩ (mean ± standard deviation) for GS1–GS5, respectively (Figure 5). The amplifiers exhibit highly linear responses with R2 values of each gain setting exceeding 0.998. The dynamic range of the amplifier ranges from ±15 μA (GS1) to as low as ±2 µA (GS5).
The amplifier bandwidth is measured by injecting sinusoidal currents (10 Hz–100 kHz) (Figure 6). A 1 MΩ resistor at the amplifier input converts an applied sine wave voltage into a known input current. The amplifier bandwidth is 12.6 ± 0.6 kHz and is consistent across all gain settings because of the integrated RC low-pass filter at the amplifier output.
The amplifier array’s input-referred noise, sampled at 20 kHz, for each gain setting are 321.9, 226.9, 195.9, 55.6, and 46.3 pARMS for GS1–5, respectively, with the highest gain setting (GS5) achieving the lowest noise (46.3 pARMS) (Figure 7). This noise represents the noise contribution of the amplifiers themselves; as this device does not have a dedicated electrode array, this value is separate from noise specific to the electrodes being used.
A time-multiplexing scheme sequentially routes the outputs of the neurochemical amplifier array to a single common output. To test the multiplexing circuit, four input signals (sine, triangle, square, sawtooth) are passed into the amplifier array; each individual amplifier’s output has been assigned a distinct color (blue, green, orange, and purple) for visualization; each point represents a single measurement from the associated amplifier (Figure 8).
The data show all 32 neurochemical amplifiers’ outputs multiplexed into a single output. Each individual amplifier is sampled at 20 kHz (50 µs period), evenly divided across the 32 amplifiers for readout (~1.56 µs per amplifier). Sequential markers of matching color indicate an interval of 50 µs that occurs between individual amplifier outputs (Figure 8a). For demultiplexing, the raw MUXOUT output is resampled at a 50 µs interval (Figure 8b). Isolating the individual demultiplexed amplifiers from the multiplexed MUXOUT output yields the sine, triangle, square, and sawtooth input signals (Figure 8c).

3.2. Characteristics of the Biopotential Amplifier

The gain, bandwidth, and noise of the biopotential amplifier array are characterized in this section. The gain and bandwidth are determined using sine-wave voltage inputs. Voltage dividers (10 kΩ/100kΩ) are placed between the function generator’s output and the chip’s inputs to limit the input signal to the µV–mV range. The function generator produces sine-wave inputs ranging from 0.05 Hz to 50 kHz.
The biopotential amplifier array has an average passband gain of 57.10 ± 0.04 dB, a low-frequency cutoff of 0.4 Hz, and a bandwidth of 6.2 kHz (Figure 9). The passband is adequate for typical local field potentials that range from 0.5 Hz to 200 Hz and action potentials from 300 Hz to 5 kHz. The input-referred noise of the array is 6.74 ± 0.92 µVRMS using a sampling rate of 20 kHz (Figure 10).
The multiplexing scheme for the biopotential amplifier array is identical to that of the neurochemical amplifier array. All 32 biopotential amplifier outputs are multiplexed into a single output through time-division multiplexing (Figure 11). The multiplexed output of the individual amplifier signals is separated by time-division (Figure 11a,b) and reassembled digitally (Figure 11c).

4. Silicon-Based Microelectrode Array Probe Fabrication

The silicon-based microelectrode array probe is fabricated from a 4-inch, double-side polished silicon wafer with a thickness of 300 µm. A 200 nm layer of silicon dioxide is deposited on the wafer to electrically isolate the silicon substrate from subsequent metal layers. Using photolithography and lift-off, a 50 nm titanium layer and a 200 nm gold layer are deposited and patterned. A 300 nm layer of silicon dioxide is deposited to form a second insulation layer. Small openings are created to expose the gold electrodes. Probe singulation is achieved using deep reactive-ion etching (60 µm) on the top to define the probe shape, which is followed by ~250 µm of backside etching. Each probe integrates 32 microelectrodes for neurochemical sensing and 32 microelectrodes for biopotential recordings distributed equally into four shanks (Figure 12). For packaging, the probe is wire-bonded to a custom PCB and the bonds are insulated using black epoxy (Resinlab EP965).

5. Catecholamine Recordings Using a Microelectrode Array

To validate the neurochemical amplifiers’ catecholamine measurement capabilities, the 32-microelectrode silicon-based probe is connected to the neural amplifier for in vitro testing. Dopamine is introduced to the probe’s MEA using a flow cell setup, during which FSCV recordings are taken.

5.1. Flow Cell Experiment

An in vitro FSCV microelectrode flow cell (NEC-FLOW-1, Pine Research, Durham, NC, USA) was used to deliver catecholamine solutions to the neurochemical sensing array. The flow cell operates by gravity feed, with the sample flowing toward the electrode array. An Ag|AgCl reference electrode, shared by the biopotential and neurochemical amplifiers, was biased to 2.2 V to maintain the CMOS inputs within the 0–5 V supply range. The FSCV waveform was applied relative to the reference electrode, with a potential range of −0.4 to 0.7 V. PBS was perfused through the flow cell at ~1 mL/min, and 1 µM dopamine (DA) or norepinephrine (NE) solutions were introduced using a switch valve controlled by a syringe pump. In this proof-of-concept implementation, the neurochemical and biopotential amplifiers share a reference electrode.

5.2. Fast-Scan Cyclic Voltammetry Recordings

The neurochemical amplifiers share a common Vpos connection, which is used to apply the FSCV ramp, sweeping the potential from 1.8 to 2.9 V (−0.4 to 0.7 V versus Ag|AgCl held at 2.2 V) at 100 V/s. The ramp repeats every 100 ms with a resting potential of −0.4 V, while PBS is perfused at ~1 mL/min. A five-second baseline recording is collected before catecholamine injection for background subtraction. After injecting 1 µM of dopamine (DA) or norepinephrine (NE), the background-subtracted voltammograms show distinct oxidation and reduction peaks (Figure 13a,d), consistent with prior FSCV measurements [11,13,21]. A 1 µM concentration is selected as a higher concentration test condition for characterization of the front-end. Time-averaged extracellular dopamine concentrations in the rat nucleus accumbens are typically reported around 20–30 nM, with transients reaching several hundred nanomolar [67,68,69]. The DA recording has oxidation and reduction peaks of 151.1 and −159.7 nA, respectively, versus 130.7 and −98.3 nA for NE. The measured DA peak currents match the calibrated response for the tested concentrations (Figure 13c). The similarity between the DA and NE voltamograms is expected given their related chemical structures.
The measured noise depends on the input capacitance of the microelectrode. For the 30 × 100 µm2 gold microelectrode submerged in PBS, the measured noise is ~2.2 nA at the lowest gain setting and ~300 pA at the highest gain setting.
The microelectrode array-based DA/NE recordings validate the neurochemical front-end for catecholamine detection. The recorded cyclic voltammograms are plotted as 2D color maps to visualize DA/NE oxidation and reduction over time (Figure 13b,e). In the color plots, each pixel increment along the x-axis represents a 100 ms interval (FSCV measurements are typically taken at a rate of 10 Hz), each vertical line of pixels represents a single FSCV measurement, in which the color represents the measured current at the potential value occurring during the stimulation, represented along the y-axis. The flow cell setup provides a controlled introduction of DA/NE to the probe over 50 to 150 s, which produces a significant signal with strong oxidation and reduction currents (Figure 13b,d).

6. Biopotential Recordings

The amplifier array’s biopotential recording capabilities are demonstrated using three experimental setups. The first setup (1) directly injects synthesized neural spikes into the amplifier’s input. The second setup (2) places a stimulation electrode and neural probe into a glass beaker of PBS, where neural spikes are injected into the stimulation electrode and recorded through the neural probe. The third setup (3) places the stimulation electrode and neural probe into a phantom brain (agarose gel) and injects the neural spikes for recording. For each of the setups, the synthesized neural spikes are generated using an NI DAQ; the highest DAC resolution is used and the dynamic range is limited to 1V. A voltage divider is set up to reduce the amplitude to simulate in vivo levels. All recordings are processed with a 60 Hz notch filter.
For the direct injection setup (1), the measured data (Figure 14) show voltage inputs with peak-to-peak magnitudes from 340 µV to 40 µV, with each set of neural spikes remaining observable across each magnitude. These measurements validate the biopotential amplifier’s ability to record small amplitudes (10 s of µV).
For the second setup (2), we prepared 30 mL of PBS solution using 10× PBS pH 7.4 (Gibco) stock solution in a 50 mL beaker. A stimulation electrode (a bare platinum/iridium wire of 254 µm diameter, P1 Technologies), reference electrode (Ag|AgCl), and our neural probe are lowered 1 cm into the PBS solution; a 1 cm gap separates each from one another. The stimulation electrode is injected with a series of neural spikes with specific peak-to-peak voltages; the length of the bare stimulation wire results in a low electrode–electrolytic interface impedance. These spikes are recorded through the neural probe’s electrodes, which are connected to the biopotential amplifier. In an in vivo experiment, the recorded signals would be within the range of tens to low hundreds of microvolts. A large peak-to-peak signal of 1.33 mV is successfully recorded, as well as small 40 µV signals in the PBS solution, which demonstrate the device’s ability to capture larger biopotentials without distortion or saturation, in addition to small signals (Figure 15).
In the third experimental setup (3), a 2% (w/w) agarose gel (VWR Chemicals, Radnor, PA, USA) serves as a phantom brain for the experiment. The reference electrode, stimulation electrode and neural probe are placed into the phantom brain 1 cm apart (Figure 16a). A series of neural spikes is applied to the stimulation electrode with peak-to-peak voltages varying between trials. A 130 µV peak-to-peak voltage neural spike series is presented (Figure 16b). A short time span from this series has been chosen to display an isolated neural spike (Figure 16c). For the three testing setups, the microelectrode array-based probe is mounted on a breakout board for connections (Figure 16a). In practice, a dedicated headstage further reduces noise and enables a low-noise recording setup.

7. Discussion and Conclusions

This work presents a CMOS platform integrating 32 neurochemical and 32 biopotential amplifiers for scalable multimodal neural recording. Compared with existing dual-mode front-ends, the proposed architecture provides a neurochemical amplifier with a wide dynamic range and selectable gain to accommodate electrodes with different dimensions and materials (Table 1). The neurochemical amplifier is further compared with recent neurochemical CMOS ICs featuring wide dynamic ranges (Table 2).
Simultaneous neurochemical and biopotential measurements can provide complementary information about neurotransmitter signaling and electrophysiological activity, with potential applications in neuroscience and neuroprosthetics. Future studies will investigate the crosstalk between the neurochemical amplifiers and biopotential amplifiers in conjuction with a microelectrode-array-based probe in in vivo systems.

Author Contributions

Conceptualization, M.A.C., K.A.W., and B.N.K.; methodology, M.A.C., K.A.W., M.D., V.S.S.B., C.S.L., B.S.J., and B.N.K.; software, M.A.C. and B.N.K.; validation, M.A.C., K.A.W., and B.N.K.; formal analysis, M.A.C., K.A.W., and B.N.K.; investigation, M.A.C., K.A.W., M.D., V.S.S.B., C.S.L., B.S.J., and B.N.K.; resources, B.N.K.; data curation, M.A.C., K.A.W., and M.D.; writing—original draft preparation, M.A.C. and B.N.K.; writing—review and editing, M.A.C., K.A.W., V.S.S.B., and B.N.K.; visualization, M.A.C., K.A.W., M.D., V.S.S.B., C.S.L., B.S.J., and B.N.K.; supervision, B.N.K.; project administration, B.N.K.; funding acquisition, B.N.K. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Science Foundation (NSF) under grants #2411566 and #2411567, the United States Air Force Office of Scientific Research (AFOSR) under grant FA9550-21-1-0117, and the National Institute of Drug Abuse (NIDA) under grant 1R01DA062090.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors on request.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
FSCVFast-Scan Cyclic Voltammetry
CMOSComplementary Metal-Oxide-Semiconductor
AFEAnalog Front-End
RMSRoot Mean Square
MEAMicroElectrode Array
ICIntegrated Circuit
nMOSN-type Metal-Oxide-Semiconductor
pMOSP-type Metal-Oxide-Semiconductor
LFPLocal Field Potential
APAction Potential
ADCAnalog-to-Digital Converter
PCBPrinted Circuit Board
GSGain Setting
PBSPhosphate-Buffered Saline
DADopamine
NENorepinephrine

References

  1. Puglisi-Allegra, S.; Ventura, R. Prefrontal/accumbal catecholamine system processes high motivational salience. Front. Behav. Neurosci. 2012, 6, 31. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  2. Felger, J.C.; Treadway, M.T. Inflammation Effects on Motivation and Motor Activity: Role of Dopamine. Neuropsychopharmacology 2017, 42, 216–241. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  3. Stone, E.A.; Lin, Y.; Rosengarten, H.; Kramer, H.K.; Quartermain, D. Emerging Evidence for a Central Epinephrine-Innervated α1-Adrenergic System that Regulates Behavioral Activation and is Impaired in Depression. Neuropsychopharmacology 2003, 28, 1387–1399. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  4. Ma, C.; Dan, Y. The how and why of sleep: Motor theory and catecholamine hypothesis. Neuron 2025, 113, 3315–3332. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  5. Chandler, D.J.; Waterhouse, B.D.; Gao, W.-J. New perspectives on catecholaminergic regulation of executive circuits: Evidence for independent modulation of prefrontal functions by midbrain dopaminergic and noradrenergic neurons. Front. Neural Circuits 2014, 8, 53. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  6. Kim, B.; Kim, H.; Kim, S.; Hwang, Y.-r. A brief review of non-invasive brain imaging technologies and the near-infrared optical bioimaging. Appl. Microsc. 2021, 51, 9. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  7. Li, N.; Jasanoff, A. Local and global consequences of reward-evoked striatal dopamine release. Nature 2020, 580, 239–244. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  8. Lee, T.; Cai, L.X.; Lelyveld, V.S.; Hai, A.; Jasanoff, A. Molecular-Level Functional Magnetic Resonance Imaging of Dopaminergic Signaling. Science 2014, 344, 533–535. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  9. Shapiro, M.G.; Westmeyer, G.G.; Romero, P.A.; Szablowski, J.O.; Küster, B.; Shah, A.; Otey, C.R.; Langer, R.; Arnold, F.H.; Jasanoff, A. Directed evolution of a magnetic resonance imaging contrast agent for noninvasive imaging of dopamine. Nat. Biotechnol. 2010, 28, 264–270. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  10. Venton, B.J.; Cao, Q. Fundamentals of fast-scan cyclic voltammetry for dopamine detection. Analyst 2020, 145, 1158–1168. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  11. Park, J.; Takmakov, P.; Wightman, R.M. In vivo comparison of norepinephrine and dopamine release in rat brain by simultaneous measurements with fast-scan cyclic voltammetry. J. Neurochem. 2011, 119, 932–944. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  12. Zachek, M.K.; Takmakov, P.; Park, J.; Wightman, R.M.; McCarty, G.S. Simultaneous monitoring of dopamine concentration at spatially different brain locations in vivo. Biosens. Bioelectron. 2010, 25, 1179–1185. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  13. Zachek, M.K.; Park, J.; Takmakov, P.; Wightman, R.M.; McCarty, G.S. Microfabricated FSCV-compatible microelectrode array for real-time monitoring of heterogeneous dopamine release. Analyst 2010, 135, 1556–1563. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  14. Amjad, U.; Choi, J.; Gibson, D.J.; Murray, R.; Graybiel, A.M.; Schwerdt, H.N. Synchronous Measurements of Extracellular Action Potentials and Neurochemical Activity with Carbon Fiber Electrodes in Nonhuman Primates. eNeuro 2024, 11, ENEURO.0001-24.2024. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  15. Shnitko, T.A.; Walton, L.R.; Peng, T.-Y.R.; Lee, S.-H.; Chao, T.-H.H.; Verber, M.D.; Wightman, R.M.; Shih, Y.-Y.I. Measurement of electrochemical brain activity with fast-scan cyclic voltammetry during functional magnetic resonance imaging. Nat. Protoc. 2026, 21, 1682–1704. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  16. Shin, M.; Venton, B.J. Fast-Scan Cyclic Voltammetry (FSCV) Reveals Behaviorally Evoked Dopamine Release by Sugar Feeding in the Adult Drosophila Mushroom Body. Angew. Chem. 2022, 134, e202207399. [Google Scholar] [CrossRef] [Scilit]
  17. Xu, X.; Zuo, Y.; Chen, S.; Hatami, A.; Gu, H. Advancements in Brain Research: The In Vivo/In Vitro Electrochemical Detection of Neurochemicals. Biosensors 2024, 14, 125. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  18. Kimble, L.C.; Twiddy, J.S.; Berger, J.M.; Forderhase, A.G.; McCarty, G.S.; Meitzen, J.; Sombers, L.A. Simultaneous, Real-Time Detection of Glutamate and Dopamine in Rat Striatum Using Fast-Scan Cyclic Voltammetry. ACS Sens. 2023, 8, 4091–4100. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  19. Castagnola, E.; Robbins, E.M.; Wu, B.; Pwint, M.Y.; Garg, R.; Cohen-Karni, T.; Cui, X.T. Flexible Glassy Carbon Multielectrode Array for In Vivo Multisite Detection of Tonic and Phasic Dopamine Concentrations. Biosensors 2022, 12, 540. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  20. Nasri, B.; Wu, T.; Alharbi, A.; You, K.D.; Gupta, M.; Sebastian, S.P.; Kiani, R.; Shahrjerdi, D. Hybrid CMOS-Graphene Sensor Array for Subsecond Dopamine Detection. IEEE Trans. Biomed. Circuits Syst. 2017, 11, 1192–1203. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  21. Robinson, D.L.; Venton, B.J.; Heien, M.L.A.V.; Wightman, R.M. Detecting Subsecond Dopamine Release with Fast-Scan Cyclic Voltammetry in Vivo. Clin. Chem. 2003, 49, 1763–1773. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  22. Gatti, I.; Eisawi, K.; Siwakoti, U.; Jadwin, G.; Sigdel, A.; Chiappalone, M.; Kumbhare, D.; Naguib, M.; Castagnola, E. MXene/PEDOT Functional Coatings on Flexible Microelectrode Arrays for Multianalyte In Vivo Neurochemical Sensing and Electrophysiology. Adv. Funct. Mater. 2026, 36, e75402. [Google Scholar] [CrossRef] [Scilit]
  23. Schapira, I.; O’Neill, M.R.; Russo-Savage, L.; Narla, T.; Laprade, K.A.; Stafford, J.M.; Ou, Y. Measuring tryptophan dynamics using fast scan cyclic voltammetry at carbon fiber microelectrodes with improved sensitivity and selectivity. RSC Adv. 2023, 13, 26203–26212. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  24. Mughrabi, I.T.; Gerber, M.; Jayaprakash, N.; Palandira, S.P.; Al-Abed, Y.; Datta-Chaudhuri, T.; Smith, C.; Pavlov, V.A.; Zanos, S. Voltammetry in the spleen assesses real-time immunomodulatory norepinephrine release elicited by autonomic neurostimulation. J. Neuroinflammation 2023, 20, 236. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  25. Adams, S.D.; Doeven, E.H.; Tye, S.J.; Bennet, K.E.; Berk, M.; Kouzani, A.Z. TinyFSCV: FSCV for the Masses. IEEE Trans. Neural Syst. Rehabil. Eng. 2020, 28, 133–142. [Google Scholar] [CrossRef] [PubMed]
  26. Zamani, H.; Chan, S.A.; Smith, C.; Mohseni, P. A Neurochemical Recording Microsystem with Analog Background Current Subtraction and 400V/s FSCV Sensing Using a 1st-Order ΔΣM. In Proceedings of the 2020 IEEE 63rd International Midwest Symposium on Circuits and Systems (MWSCAS), Springfield, MA, USA, 9–12 August 2020; pp. 517–520. [Google Scholar]
  27. Lu, S.Y.; Liao, Y.T. A 19 μW, 50 kS/s, 0.008-400 V/s Cyclic Voltammetry Readout Interface With a Current Feedback Loop and On-Chip Pattern Generation. IEEE Trans. Biomed. Circuits Syst. 2021, 15, 190–198. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  28. White, K.A.; Darroudi, M.; Park, J.; Kim, B.N. A 128-ch Area-Efficient Neurochemical-Sensing Front-End for FSCV Recordings of Dopamine. IEEE Sens. J. 2024, 24, 8788–8797. [Google Scholar] [CrossRef] [Scilit]
  29. Ying, D.; Rosenberg, J.; Singh, N.K.; Hall, D.A. A 26.5 pArms Neurotransmitter Front-End With Class-AB Background Subtraction. IEEE Trans. Biomed. Circuits Syst. 2022, 16, 692–702. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  30. Dorta-Quiñones, C.I.; Wang, X.Y.; Dokania, R.K.; Gailey, A.; Lindau, M.; Apsel, A.B. A Wireless FSCV Monitoring IC With Analog Background Subtraction and UWB Telemetry. IEEE Trans. Biomed. Circuits Syst. 2016, 10, 289–299. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  31. Wang, K.; Xu, W.; Zhang, C.; Yang, Y.; Law, M.K.; Zhou, L.; Chen, J.; Chen, M. A Multi-Channel CMOS Analog Front-End Interface IC with 157.8 dB Current Detection Dynamic Range. Chin. J. Electron. 2024, 33, 393–402. [Google Scholar] [CrossRef] [Scilit]
  32. Walton, L.R.; Verber, M.; Lee, S.-H.; Chao, T.-H.H.; Wightman, R.M.; Shih, Y.-Y.I. Simultaneous fMRI and fast-scan cyclic voltammetry bridges evoked oxygen and neurotransmitter dynamics across spatiotemporal scales. NeuroImage 2021, 244, 118634. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  33. Johnson, J.A.; Hobbs, C.N.; Wightman, R.M. Removal of Differential Capacitive Interferences in Fast-Scan Cyclic Voltammetry. Anal. Chem. 2017, 89, 6166–6174. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  34. Darroudi, M.; White, K.A.; Crocker, M.A.; Kim, B.N. Dopamine Measurement Using Engineered CNT–CQD–Polymer Coatings on Pt Microelectrodes. Sensors 2024, 24, 1893. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  35. Zachek, M.K.; Hermans, A.; Wightman, R.M.; McCarty, G.S. Electrochemical dopamine detection: Comparing gold and carbon fiber microelectrodes using background subtracted fast scan cyclic voltammetry. J. Electroanal. Chem. 2008, 614, 113–120. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  36. Qi, Y.; Ryu, J.; Jang, D.; Schaub, B.; Shin, Y.; Bai, T.; Li, G.; Aronson, J.P.; Leiter, J.C.; Fang, H. Thin-Film Reference Electrodes for Fast-Scan Cyclic Voltammetry. ACS Chem. Neurosci. 2025, 16, 4420–4428. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  37. Ahnood, A.; Chambers, A.; Gelmi, A.; Yong, K.-T.; Kavehei, O. Semiconducting electrodes for neural interfacing: A review. Chem. Soc. Rev. 2023, 52, 1491–1518. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  38. Cogan, S.F. Neural stimulation and recording electrodes. Annu. Rev. Biomed. Eng. 2008, 10, 275–309. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  39. Frank, J.A.; Antonini, M.-J.; Anikeeva, P. Next-generation interfaces for studying neural function. Nat. Biotechnol. 2019, 37, 1013–1023. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  40. Qi, Y.; Jang, D.; Ryu, J.; Bai, T.; Shin, Y.; Gu, W.; Iyer, A.; Li, G.; Ma, H.; Liou, J.-y.; et al. Stabilized carbon coating on microelectrodes for scalable and interoperable neurotransmitter sensing. Nat. Commun. 2025, 16, 3300. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  41. Takmakov, P.; Zachek, M.K.; Keithley, R.B.; Walsh, P.L.; Donley, C.; McCarty, G.S.; Wightman, R.M. Carbon Microelectrodes with a Renewable Surface. Anal. Chem. 2010, 82, 2020–2028. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  42. Park, S.Y.; Na, K.; Vöröslakos, M.; Song, H.; Slager, N.; Oh, S.; Seymour, J.P.; Buzsáki, G.; Yoon, E. A Miniaturized 256-Channel Neural Recording Interface With Area-Efficient Hybrid Integration of Flexible Probes and CMOS Integrated Circuits. IEEE Trans. Biomed. Eng. 2022, 69, 334–346. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  43. Steinmetz, N.A.; Aydin, C.; Lebedeva, A.; Okun, M.; Pachitariu, M.; Bauza, M.; Beau, M.; Bhagat, J.; Böhm, C.; Broux, M.; et al. Neuropixels 2.0: A miniaturized high-density probe for stable, long-term brain recordings. Science 2021, 372, eabf4588. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  44. Papadopoulou, A.; Hermiz, J.; Grace, C.; Denes, P. A Modular 512-Channel Neural Signal Acquisition ASIC for High-Density 4096 Channel Electrophysiology. Sensors 2024, 24, 3986. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  45. Jang, M.; Yu, W.H.; Lee, C.; Hays, M.; Wang, P.; Vitale, N.; Tandon, P.; Yan, P.; Mak, P.I.; Chae, Y.; et al. A 1024-Channel 268 nW/pixel 36x36 μm2/ch Data-Compressive Neural Recording IC for High-Bandwidth Brain-Computer Interfaces. In Proceedings of the 2023 IEEE Symposium on VLSI Technology and Circuits (VLSI Technology and Circuits), Kyoto, Japan, 11–16 June 2023; pp. 1–2. [Google Scholar]
  46. Sahasrabuddhe, K.; Khan, A.A.; Singh, A.P.; Stern, T.M.; Ng, Y.; Tadić, A.; Orel, P.; LaReau, C.; Pouzzner, D.; Nishimura, K.; et al. The Argo: A high channel count recording system for neural recording in vivo. J. Neural Eng. 2021, 18, 015002. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  47. Obaid, A.; Hanna, M.-E.; Wu, Y.-W.; Kollo, M.; Racz, R.; Angle, M.R.; Müller, J.; Brackbill, N.; Wray, W.; Franke, F.; et al. Massively parallel microwire arrays integrated with CMOS chips for neural recording. Sci. Adv. 2020, 6, eaay2789. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  48. Kollo, M.; Racz, R.; Hanna, M.-E.; Obaid, A.; Angle, M.R.; Wray, W.; Kong, Y.; Müller, J.; Hierlemann, A.; Melosh, N.A.; et al. CHIME: CMOS-Hosted in vivo Microelectrodes for Massively Scalable Neuronal Recordings. Front. Neurosci. 2020, 14, 834. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  49. Lyu, L.; Ye, D.; Xu, R.; Mu, G.; Zhao, H.; Xiang, Y.; Tu, Y.; Zhang, Y.; Shi, C.J.R. A Fully-Integrated 64-Channel Wireless Neural Interfacing SoC Achieving 110 dB AFE PSRR and Supporting 54 Mb/s Symbol Rate, Meter-Range Wireless Data Transmission. IEEE Trans. Circuits Syst. II Express Briefs 2020, 67, 831–835. [Google Scholar] [CrossRef] [Scilit]
  50. Avula, A.K.; Goyal, A.; Rusheen, A.E.; Yuen, J.; Dennis, W.O.; Eaker, D.R.; Boesche, J.B.; Blaha, C.D.; Bennet, K.E.; Lee, K.H.; et al. Improved circuitry and post-processing for interleaved fast-scan cyclic voltammetry and electrophysiology measurements. Front. Signal Process. 2023, 3, 1195800. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  51. Chakraborty, I.; Akalkotkar, R.; Krueger, D.; Coffin, T.; Hu, M.; Chen, X.; Gan, X.; Bhat, S.; Zhao, L.; Jia, Y. A Wireless Trimodal Neural Interface Device with Electrical and Electrochemical Recording. In Proceedings of the 2023 IEEE Texas Symposium on Wireless and Microwave Circuits and Systems (WMCS), Waco, TX, USA, 19–20 April 2023; pp. 1–3. [Google Scholar]
  52. Guo, J.; Ng, W.; Yuan, J.; Li, S.; Chan, M. A 200-Channel Area-Power-Efficient Chemical and Electrical Dual-Mode Acquisition IC for the Study of Neurodegenerative Diseases. IEEE Trans. Biomed. Circuits Syst. 2015, 10, 567–578. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  53. Dragas, J.; Viswam, V.; Shadmani, A.; Chen, Y.; Bounik, R.; Stettler, A.; Radivojevic, M.; Geissler, S.; Obien, M.E.J.; Müller, J.; et al. In Vitro Multi-Functional Microelectrode Array Featuring 59 760 Electrodes, 2048 Electrophysiology Channels, Stimulation, Impedance Measurement, and Neurotransmitter Detection Channels. IEEE J. Solid-State Circuits 2017, 52, 1576–1590. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  54. Park, J.S.; Aziz, M.K.; Li, S.; Chi, T.; Grijalva, S.I.; Sung, J.H.; Cho, H.C.; Wang, H. 1024-Pixel CMOS Multimodality Joint Cellular Sensor/Stimulator Array for Real-Time Holistic Cellular Characterization and Cell-Based Drug Screening. IEEE Trans. Biomed. Circuits Syst. 2018, 12, 80–94. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  55. Abbott, J.; Ye, T.; Krenek, K.; Gertner, R.S.; Ban, S.; Kim, Y.; Qin, L.; Wu, W.; Park, H.; Ham, D. A nanoelectrode array for obtaining intracellular recordings from thousands of connected neurons. Nat. Biomed. Eng. 2020, 4, 232–241. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  56. Abbott, J.; Ye, T.; Krenek, K.; Qin, L.; Kim, Y.; Wu, W.; Gertner, R.S.; Park, H.; Ham, D. The Design of a CMOS Nanoelectrode Array With 4096 Current-Clamp/Voltage-Clamp Amplifiers for Intracellular Recording/Stimulation of Mammalian Neurons. IEEE J. Solid-State Circuits 2020, 55, 2567–2582. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  57. Mulberry, G.; White, K.A.; Crocker, M.A.; Kim, B.N. A 512-Ch Dual-Mode Microchip for Simultaneous Measurements of Electrophysiological and Neurochemical Activities. Biosensors 2023, 13, 502. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  58. White, K.A.; Park, J.; Kim, B.N. A 128-ch Neurochemical Microchip with Highly-Scalable Low-Noise Resistive Feedback Amplifiers. In Proceedings of the 2023 IEEE 66th International Midwest Symposium on Circuits and Systems (MWSCAS), Tempe, AZ, USA, 6–9 August 2023; pp. 763–767. [Google Scholar]
  59. Ayers, S.; Gillis, K.D.; Lindau, M.; Minch, B.A. Design of a CMOS Potentiostat Circuit for Electrochemical Detector Arrays. IEEE Trans. Circuits Syst. I Regul. Pap. 2007, 54, 736–744. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  60. Harrison, R.R.; Charles, C. A low-power low-noise CMOS amplifier for neural recording applications. IEEE J. Solid-State Circuits 2003, 38, 958–965. [Google Scholar] [CrossRef] [Scilit]
  61. White, K.A.; Crocker, M.A.; Kim, B.N. High-Speed 1024-Pixel CMOS Electrochemical Imaging Sensor with 40,000 Frames per Second for Dopamine and Hydrogen Peroxide Imaging. Electronics 2025, 14, 3207. [Google Scholar] [CrossRef] [Scilit]
  62. Kim, B.N.; Herbst, A.D.; Kim, S.J.; Minch, B.A.; Lindau, M. Parallel recording of neurotransmitters release from chromaffin cells using a 10×10 CMOS IC potentiostat array with on-chip working electrodes. Biosens. Bioelectron. 2013, 41, 736–744. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  63. White, K.A.; Mulberry, G.; Smith, J.; Lindau, M.; Minch, B.A.; Sugaya, K.; Kim, B.N. Single-Cell Recording of Vesicle Release From Human Neuroblastoma Cells Using 1024-ch Monolithic CMOS Bioelectronics. IEEE Trans. Biomed. Circuits Syst. 2018, 12, 1345–1355. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  64. White, K.A.; Mulberry, G.; Kim, B.N. Rapid 1024-Pixel Electrochemical Imaging at 10,000 Frames Per Second Using Monolithic CMOS Sensor and Multifunctional Data Acquisition System. IEEE Sens. J. 2018, 18, 5507–5514. [Google Scholar] [CrossRef] [Scilit]
  65. White, K.A.; Mulberry, G.; Kim, B.N. Parallel 1024-ch Cyclic Voltammetry on Monolithic CMOS Electrochemical Detector Array. IEEE Sens. J. 2020, 20, 4395–4402. [Google Scholar] [CrossRef] [Scilit]
  66. Mulberry, G.; White, K.A.; Kim, B.N. Analysis of Simple Half-Shared Transimpedance Amplifier for Picoampere Biosensor Measurements. IEEE Trans. Biomed. Circuits Syst. 2019, 13, 387–395. [Google Scholar] [CrossRef] [PubMed]
  67. Garris, P.A.; Ciolkowski, E.L.; Pastore, P.; Wightman, R.M. Efflux of dopamine from the synaptic cleft in the nucleus accumbens of the rat brain. J. Neurosci. 1994, 14, 6084–6093. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  68. Owesson-White, C.A.; Roitman, M.F.; Sombers, L.A.; Belle, A.M.; Keithley, R.B.; Peele, J.L.; Carelli, R.M.; Wightman, R.M. Sources contributing to the average extracellular concentration of dopamine in the nucleus accumbens. J. Neurochem. 2012, 121, 252–262. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  69. Parsons, L.H.; Justice, J.B., Jr. Extracellular Concentration and In Vivo Recovery of Dopamine in the Nucleus Accumbens Using Microdialysis. J. Neurochem. 1992, 58, 212–218. [Google Scholar] [CrossRef] [Scilit] [PubMed]
Biosensors 16 00466 g001
Figure 1. Schematic of the neurochemical amplifier. (a) The operational amplifier design of the neurochemical amplifier adapts a rail-to-rail folded-cascode topology. (b) The neurochemical amplifier employs a series of resistors with parallel switches to allow variable resistance for the feedback loop. The integrated switches select the variable-gain settings.
Figure 1. Schematic of the neurochemical amplifier. (a) The operational amplifier design of the neurochemical amplifier adapts a rail-to-rail folded-cascode topology. (b) The neurochemical amplifier employs a series of resistors with parallel switches to allow variable resistance for the feedback loop. The integrated switches select the variable-gain settings.
Biosensors 16 00466 g001
Biosensors 16 00466 g002
Figure 2. Schematics for the biopotential amplifier. The biopotential amplifier consists of two amplification stages. (a) The first stage of the biopotential amplifier is based on a folded-cascode amplifier; it provides a gain of ~150 V/V. (b) The second stage of the biopotential amplifier provides a gain of ~5 V/V, producing a total theoretical gain of ~750 V/V. Both stages use pseudo-resistors, based on diode-connected transistors, to provide DC stability at the inverting inputs.
Figure 2. Schematics for the biopotential amplifier. The biopotential amplifier consists of two amplification stages. (a) The first stage of the biopotential amplifier is based on a folded-cascode amplifier; it provides a gain of ~150 V/V. (b) The second stage of the biopotential amplifier provides a gain of ~5 V/V, producing a total theoretical gain of ~750 V/V. Both stages use pseudo-resistors, based on diode-connected transistors, to provide DC stability at the inverting inputs.
Biosensors 16 00466 g002
Biosensors 16 00466 g003
Figure 3. The rail-to-rail multiplexing circuit used for the neurochemical and the biopotential amplifiers. The input transistors, M1N/M1P through M32N/M32P and M33N/M33P, form a unity-gain amplifier in which the non-inverting halves (M1N/M1P–M32N/M32P) share the inverting half of the amplifier, M33N/M33P. The switches (SW1–SW32) sequentially provide the non-inverting halves with mutually exclusive access to the inverting half of the amplifier, which is output at MUXOUT.
Figure 3. The rail-to-rail multiplexing circuit used for the neurochemical and the biopotential amplifiers. The input transistors, M1N/M1P through M32N/M32P and M33N/M33P, form a unity-gain amplifier in which the non-inverting halves (M1N/M1P–M32N/M32P) share the inverting half of the amplifier, M33N/M33P. The switches (SW1–SW32) sequentially provide the non-inverting halves with mutually exclusive access to the inverting half of the amplifier, which is output at MUXOUT.
Biosensors 16 00466 g003
Biosensors 16 00466 g004
Figure 4. (a) A system diagram of the neurochemical and biopotential amplifiers. (b) Photomicrograph of the complementary metal-oxide semiconductor (CMOS) chip that includes the neurochemical and biopotential amplifiers, timing circuits, and multiplexers.
Figure 4. (a) A system diagram of the neurochemical and biopotential amplifiers. (b) Photomicrograph of the complementary metal-oxide semiconductor (CMOS) chip that includes the neurochemical and biopotential amplifiers, timing circuits, and multiplexers.
Biosensors 16 00466 g004
Biosensors 16 00466 g005
Figure 5. Gain measurements of the neurochemical amplifiers across the chip’s gain settings (GS). (a) A representative neurochemical amplifier’s gain for all gain settings. (b) A histogram of the measured gain values of all individual amplifiers at each gain setting across the 32 neurochemical amplifiers. From GS1 to GS5, the average (and standard deviation) are respectively: 92.5 ± 1.1 kΩ, 134.5 ± 1.7 kΩ, 177.1 ± 2.1 kΩ, 691.6 ± 13.0 kΩ, and 854.0 ± 18.8 kΩ.
Figure 5. Gain measurements of the neurochemical amplifiers across the chip’s gain settings (GS). (a) A representative neurochemical amplifier’s gain for all gain settings. (b) A histogram of the measured gain values of all individual amplifiers at each gain setting across the 32 neurochemical amplifiers. From GS1 to GS5, the average (and standard deviation) are respectively: 92.5 ± 1.1 kΩ, 134.5 ± 1.7 kΩ, 177.1 ± 2.1 kΩ, 691.6 ± 13.0 kΩ, and 854.0 ± 18.8 kΩ.
Biosensors 16 00466 g005
Biosensors 16 00466 g006
Figure 6. The gain bandwidth measurements of a representative neurochemical amplifier for gain settings (GS) 1 through 5. The neurochemical amplifier bandwidth is 12.6 ± 0.6 kHz (mean ± standard deviation).
Figure 6. The gain bandwidth measurements of a representative neurochemical amplifier for gain settings (GS) 1 through 5. The neurochemical amplifier bandwidth is 12.6 ± 0.6 kHz (mean ± standard deviation).
Biosensors 16 00466 g006
Biosensors 16 00466 g007
Figure 7. The noise characteristics of a representative neurochemical amplifier across each gain setting (GS). The measured input-referred noise (20 kHz sampling rate) for GS1–GS5, respectively, is 321.9 pARMS, 226.9 pARMS, 195.9 pARMS, 55.6 pARMS, and 46.3 pARMS.
Figure 7. The noise characteristics of a representative neurochemical amplifier across each gain setting (GS). The measured input-referred noise (20 kHz sampling rate) for GS1–GS5, respectively, is 321.9 pARMS, 226.9 pARMS, 195.9 pARMS, 55.6 pARMS, and 46.3 pARMS.
Biosensors 16 00466 g007
Biosensors 16 00466 g008
Figure 8. Time-division multiplexing of the neurochemical amplifiers. (a) The grey line represents the raw output of the multiplexer with 32 neurochemical amplifiers, four of which are input with current signals. Blue, green, orange, and purple markers correspond to neurochemical amplifiers input with sine, triangle, square, and sawtooth current signals, respectively. (b) Demonstration of demultiplexing of each neurochemical amplifier’s output by resampling at a 50 µs interval. (c) The demultiplexed signals of the four amplifiers with current signals show clear sine (blue), triangle (green), square (orange), and sawtooth (purple) waves.
Figure 8. Time-division multiplexing of the neurochemical amplifiers. (a) The grey line represents the raw output of the multiplexer with 32 neurochemical amplifiers, four of which are input with current signals. Blue, green, orange, and purple markers correspond to neurochemical amplifiers input with sine, triangle, square, and sawtooth current signals, respectively. (b) Demonstration of demultiplexing of each neurochemical amplifier’s output by resampling at a 50 µs interval. (c) The demultiplexed signals of the four amplifiers with current signals show clear sine (blue), triangle (green), square (orange), and sawtooth (purple) waves.
Biosensors 16 00466 g008
Biosensors 16 00466 g009
Figure 9. The bandwidth measurement of the biopotential amplifiers. Each amplifier is measured from 0.05 Hz to 50 kHz. The bandwidth is measured to be 6.2 ± 0.1 kHz. Error bars indicate standard deviation, with standard deviations less than 0.2% of the maximum value removed for visual clarity.
Figure 9. The bandwidth measurement of the biopotential amplifiers. Each amplifier is measured from 0.05 Hz to 50 kHz. The bandwidth is measured to be 6.2 ± 0.1 kHz. Error bars indicate standard deviation, with standard deviations less than 0.2% of the maximum value removed for visual clarity.
Biosensors 16 00466 g009
Biosensors 16 00466 g010
Figure 10. The noise characteristics of a representative biopotential amplifier. The noise level is measured to be 6.74 ± 0.92 µVRMS at a 20 kHz sampling rate.
Figure 10. The noise characteristics of a representative biopotential amplifier. The noise level is measured to be 6.74 ± 0.92 µVRMS at a 20 kHz sampling rate.
Biosensors 16 00466 g010
Biosensors 16 00466 g011
Figure 11. Time-division multiplexing of the biopotential amplifier. (a) The raw output (shown in grey line) multiplexes 32 individual biopotential amplifiers’ outputs. Blue, green, orange, and purple markers indicate biopotential amplifiers’ inputs with sawtooth, square, sine, and phase-shifted sine voltage signals, respectively. (b) By resampling the raw output at a 50 µs interval, each biopotential amplifier’s output can be demultiplexed. (c) The demultiplexed signals from the four amplifiers show clear sawtooth (blue), square (green), sine (orange), and phase-shifted sine (purple) waves.
Figure 11. Time-division multiplexing of the biopotential amplifier. (a) The raw output (shown in grey line) multiplexes 32 individual biopotential amplifiers’ outputs. Blue, green, orange, and purple markers indicate biopotential amplifiers’ inputs with sawtooth, square, sine, and phase-shifted sine voltage signals, respectively. (b) By resampling the raw output at a 50 µs interval, each biopotential amplifier’s output can be demultiplexed. (c) The demultiplexed signals from the four amplifiers show clear sawtooth (blue), square (green), sine (orange), and phase-shifted sine (purple) waves.
Biosensors 16 00466 g011
Biosensors 16 00466 g012
Figure 12. Silicon-based microelectrode array probe. (a) Microphotograph of the probe with four parallel shanks. Each shank contains eight biopotential and eight neurochemical electrodes, for a total of 64 electrodes across the four shanks. (b) The neurochemical electrodes are 100 µm × 30 µm, and the biopotential electrodes are 15 µm × 15 µm. (c) Microphotograph of the MEA probe wire-bonded to a PCB. (d) Scanning electron microscope image of the MEA probe.
Figure 12. Silicon-based microelectrode array probe. (a) Microphotograph of the probe with four parallel shanks. Each shank contains eight biopotential and eight neurochemical electrodes, for a total of 64 electrodes across the four shanks. (b) The neurochemical electrodes are 100 µm × 30 µm, and the biopotential electrodes are 15 µm × 15 µm. (c) Microphotograph of the MEA probe wire-bonded to a PCB. (d) Scanning electron microscope image of the MEA probe.
Biosensors 16 00466 g012
Biosensors 16 00466 g013
Figure 13. Catecholamine measurements using the microelectrode array probe placed within a flow cell using FSCV. (a) DA recording with an oxidation peak of 151.1 nA and a reduction peak of −159.7 nA. (b) Color plot displaying current amplitude (pixel color) across the scanning potential (y-axis) over time (x-axis) for the DA measurement. (c) Calibration curve for dopamine, plotting oxidation peak versus dopamine concentration. (d) NE recording with an oxidation peak of 130.7 nA and a reduction peak of −98.3 nA. (e) Color plot for the NE measurement.
Figure 13. Catecholamine measurements using the microelectrode array probe placed within a flow cell using FSCV. (a) DA recording with an oxidation peak of 151.1 nA and a reduction peak of −159.7 nA. (b) Color plot displaying current amplitude (pixel color) across the scanning potential (y-axis) over time (x-axis) for the DA measurement. (c) Calibration curve for dopamine, plotting oxidation peak versus dopamine concentration. (d) NE recording with an oxidation peak of 130.7 nA and a reduction peak of −98.3 nA. (e) Color plot for the NE measurement.
Biosensors 16 00466 g013
Biosensors 16 00466 g014
Figure 14. Biopotential recordings with signals applied directly to the amplifier’s input at various amplitudes. Measurements of (a) 340 µV, (b) 130 µV, (c) 75 µV, (d) 40 µV peak-to-peak input signals.
Figure 14. Biopotential recordings with signals applied directly to the amplifier’s input at various amplitudes. Measurements of (a) 340 µV, (b) 130 µV, (c) 75 µV, (d) 40 µV peak-to-peak input signals.
Biosensors 16 00466 g014
Biosensors 16 00466 g015
Figure 15. A biopotential stimulation recording with the stimulation electrode and neural probe within a PBS solution. Measurements of (a) 1.33 mV and (b) 40 µV peak-to-peak neural spikes are shown.
Figure 15. A biopotential stimulation recording with the stimulation electrode and neural probe within a PBS solution. Measurements of (a) 1.33 mV and (b) 40 µV peak-to-peak neural spikes are shown.
Biosensors 16 00466 g015
Biosensors 16 00466 g016
Figure 16. Biopotential recordings using an agarose gel “phantom brain”. (a) Photograph of the agarose gel experimental setup. (b) Measurement of 130 µV peak-to-peak neural spike and (c) a single spike from the series.
Figure 16. Biopotential recordings using an agarose gel “phantom brain”. (a) Photograph of the agarose gel experimental setup. (b) Measurement of 130 µV peak-to-peak neural spike and (c) a single spike from the series.
Biosensors 16 00466 g016
Table 1. State-of-the-art dual-mode (electrochemical/biopotential) sensing front-ends.
Table 1. State-of-the-art dual-mode (electrochemical/biopotential) sensing front-ends.
2015 [52]2017 [53]2018 [54]2020 [55,56]2023 [57]This Work
Technology Node (nm)180180130180350350
Die Size (mm2)5 × 2.6512 × 8.93 × 1.85~10 × ~205 × 53 × 8.3
Operation Voltage (V)1.8-33.63.35
ApplicationIn VitroIn VitroIn VitroIn VitroIn VitroIn Vivo
Electrochemical
Parallel Channels (#)100284409625632
Size Per Channel (µm2)30,00040,000800025,0003070120,000
Sampling Rate (kHz)20201009.44020
Bandwidth (kHz)0.11–10160.74.71012.6
Noise93 pA (FSCV)120–200 pA56 pA~1 pA4.51 pA43–320 pA *
Dynamic Range±50 nA±1 µA--±2–6 µA±2–15 µA *
Gain0.011–2.38 GΩ2.7–3.2 MΩ129 kΩ~700 MΩ0.625–2.08 MΩ92.5–854 kΩ *
Power (µW)12.1178-2491630–432 *
Biopotential
Parallel Channels (#)10020481024409625632
Size Per Channel (µm2)30,000--25,00011,080120,000
Sampling Rate (kHz)2020-9.44020
Bandwidth (kHz)100.3–100.3–64.70.0002–106.2
Noise (µV)4.072.47.12024.96.7
Power (µW)9.116-2252697
* Based on variable gain settings.
Table 2. State-of-the-art large dynamic range neurochemical CMOS ICs.
Table 2. State-of-the-art large dynamic range neurochemical CMOS ICs.
2017 [20]2020 [26]2021 [27]2022 [29]2023 [57]2024 [28]2024 [31]This Work
Technology Node (nm)65180180180350350180350
Supply Voltage (V)1.82.51.2–1.53.33.343.3/1.85
Number of Channels (#)4111256128232
Channel Size (mm2)0.0750.06-0.060.003070.0176-0.12
Total Die Area (mm2)--3.170.256254.186.424.9
Input Current Range (µA)±2.56±1−7/+10±0.375±4.8±5±10 pA–1 mA±2–15 *
Input Referred Noise (pARMS)207108726.54.5122012.8846.3–321.9 *
Bandwidth (kHz)1550510.340–1221012.6
Power Consumption3.1 µW/Ch14.1 µW/Ch1.77 µW/Ch3.7 µW/Ch16 µW/Ch4.4 µW/Ch4.36 mW/Chip30–432 µW/Ch *
* Based on variable gain settings.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Crocker, M.A.; White, K.A.; Darroudi, M.; Bapanapalli, V.S.S.; Lipscomb, C.S.; John, B.S.; Kim, B.N. A Dual-Mode Neural Amplifier Array for Biopotential and FSCV-Based Neurochemical Measurements. Biosensors 2026, 16, 466. https://doi.org/10.3390/bios16090466

AMA Style

Crocker MA, White KA, Darroudi M, Bapanapalli VSS, Lipscomb CS, John BS, Kim BN. A Dual-Mode Neural Amplifier Array for Biopotential and FSCV-Based Neurochemical Measurements. Biosensors. 2026; 16(9):466. https://doi.org/10.3390/bios16090466

Chicago/Turabian Style

Crocker, Matthew A., Kevin A. White, Mahdieh Darroudi, Vishnu Saket S. Bapanapalli, Charles S. Lipscomb, Benjamin S. John, and Brian N. Kim. 2026. "A Dual-Mode Neural Amplifier Array for Biopotential and FSCV-Based Neurochemical Measurements" Biosensors 16, no. 9: 466. https://doi.org/10.3390/bios16090466

APA Style

Crocker, M. A., White, K. A., Darroudi, M., Bapanapalli, V. S. S., Lipscomb, C. S., John, B. S., & Kim, B. N. (2026). A Dual-Mode Neural Amplifier Array for Biopotential and FSCV-Based Neurochemical Measurements. Biosensors, 16(9), 466. https://doi.org/10.3390/bios16090466

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop