Chemically Regulated Conical Channel Synapse for Neuromorphic and Sensing Applications

T. M. Kamsma,M. S. Klop,W. Q. Boon,C. Spitoni,B. Rueckauer,R. van Roij

2024-06-05

Abstract:Fluidic iontronics offer a unique capability for emulating the chemical processes found in neurons. We extract multiple distinct chemically regulated synaptic features from a single conical microfluidic channel carrying functionalized surface groups, using finite-element calculations of continuum transport equations. Such channels have long been employed for fluidic sensing and are therefore experimentally well established. By modeling a Langmuir-type surface reaction on the channel wall we couple fast voltage-induced volumetric salt accumulation with a long-term channel surface charge modulation by means of fast charging and slow discharging. These nonlinear charging dynamics are understood through an analytic approximation rooted in first-principles. We show how short-and long-term potentiation and depression, frequency-dependent plasticity, and chemical-electrical signal coincidence detection (acting like a chemical-electrical AND logic gate), akin to the NMDA mechanism for Hebbian learning in biological synapses, can all be emulated with a single channel.

Soft Condensed Matter

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to simulate the chemical processes of biological neurons in artificial systems, especially by designing a chemically - regulated conical channel with multiple synaptic characteristics to achieve this goal. Specifically, the paper explores how to use a single microfluidic conical channel to carry functionalized surface groups, and extract multiple different chemically - regulated synaptic characteristics by calculating the continuous transport equation with the finite - element method. These characteristics include short - term and long - term potentiation (STP and LTP), frequency - dependent plasticity (FDP), and chemo - electrical signal coincidence detection (Hebbian learning similar to the NMDA mechanism), thereby achieving functions similar to biological synapses in artificial systems.

### Main research contents
1. **Chemically - regulated artificial synapses**:
- Designed a conical channel that is widely available experimentally, with functionalized surface groups on the channel wall.
- Solved the Langmuir kinetics and the continuous transport equation by numerical and analytical methods, coupling the nonlinear charging kinetics of the functionalized surface groups with the voltage - induced transient concentration polarization inside the channel.

2. **Simulation of synaptic characteristics**:
- Short - term and long - term potentiation (STP and LTP): Simulate the short - term and long - term potentiation of synapses through rapid salt accumulation and surface charging, and subsequent slow discharge.
- Frequency - dependent plasticity (FDP): Input voltage pulses of different frequencies will lead to different changes in long - term synaptic strength.
- Chemo - electrical signal coincidence detection: By simultaneously applying voltage pulses and reactant ion concentration gradients, achieve an increase or decrease in long - term conductivity, simulating the chemo - electrical signal coincidence detection in the NMDA mechanism.

3. **Combination of theory and experiment**:
- Use the finite - element method (FE) for numerical calculations and provide theoretical support through analytical approximation (AA).
- The results show that this conical channel can effectively simulate multiple functions of biological synapses, providing new possibilities for neuromorphic computing and sensing applications.

### Formula examples
- **Dynamics equation of surface charge density**:
\[
\frac{\partial \sigma}{\partial t}=k\left(\rho_{A}(\Gamma - \sigma)-K\rho_{B}\sigma\right)
\]
where \(\sigma\) is the surface charge density, \(\rho_{A}\) and \(\rho_{B}\) are the concentrations of reactant ions respectively, \(\Gamma\) is the density of surface groups, \(k\) is the reaction rate constant, and \(K\) is the equilibrium constant.

- **Poisson equation**:
\[
\nabla^{2}\Psi = -\frac{e}{\epsilon}\sum_{i}z_{i}\rho_{i}
\]
where \(\Psi\) is the electric potential, \(e\) is the electron charge, \(\epsilon\) is the dielectric constant, \(z_{i}\) and \(\rho_{i}\) are the valence state and concentration of the \(i\)-th ion respectively.

- **Nernst - Planck equation**:
\[
j_{i}=-D\left(\nabla\rho_{i}+\frac{z_{i}\rho_{i}e\nabla\Psi}{k_{B}T}\right)+u\rho_{i}
\]
where \(j_{i}\) is the ion flux, \(D\) is the diffusion coefficient, \(k_{B}\) is the Boltzmann constant, \(T\) is the temperature, and \(u\) is the fluid velocity field.

Through these studies, the paper shows how to achieve multiple chemically - regulated synaptic characteristics in a single conical channel, providing new ideas and methods for the development of neuromorphic computing and sensing technologies.