The role of polyelectrolyte brushes in tunable synaptic devices

Abstract

With the ever-increasing digitization of society, the development of materials with low-power memory storage -similar to synapses- is becoming more relevant. The field of iontronic artificial synapses has gained traction, in particular with polymers as the memory-active material which allows for additional bio-compatibility, flexibility and tunability. Polyelectrolyte brushes are an example of stimulus-responsive materials that can be used in iontronic devices. However, the complexity of current neuromorphic devices does not allow us to isolate and understand the role of polyelectrolyte brushes in their synaptic response. In this paper, we show that polyelectrolyte brushes are capable of synaptic behavior in the most simple of electrochemical cell designs. Furthermore, by combining theory and experimental work, we shed light on the role of brushes in this synaptic behavior and their dynamic stimuli-responsiveness to polarity changes for different salt concentrations. The obtained trends and interpretations of the nonlinear potential-current response, paired-pulse experiments, and accumulative learning lay the foundation for designing and developing polymer brush-based neuromorphic devices.

Figures (4)

• Figure 1: Polyelectrolyte brushes on gold. A schematic representation of our material; from the molecular structure of our PSPMA brushes (left) to a microscopic side-view of these brushes (middle), which is mounted in an electrochemical cell as a coated gold WE (right).
• Figure 2: Nonlinear responses of PSPMA brushes. a) Visual representation of an osmotic (left) and salted (right) brush. b) Equivalent circuit of a PSPMA brush in a KCl electrolyte solution. c) Chronoamperometry in osmotic and salted conditions, with polarities $\pm$ 0.2 V. d) Potential (input) and current (output) trace of a spike-number dependent plasticity (SNDP) experiment. Forgetting curves of an osmotic (e) and salted (f) brush. Pulse numbers are varied between 1-100 pulses. g) Stability of the PSPMA brushes across $>$25.000 pulses. h) Potential (input) and current (output) traces to assess the non-linear ion rectification behavior and potential-current coupling. i) Peak current as a function of the potential amplitude, as derived from (h).
• Figure 3: Paired-Pulse Characteristics. a) Time trace of a paired-pulse experiment with $\Delta$t = 200 ms. Paired-pulse experiments using pulses of U = -0.2 V (b) or U = +0.2 V (c) with PSPMA brushes in the osmotic (0.01 M) and salted (0.5 M) regime. d) Mechanistic picture of the trends in (b). e) Ion distributions along the z-direction using MD simulations of a representative charged brush (see Supporting Information). f) Mechanistic picture of the trend in (c), specifically the osmotic regime. Here, region I indicates the area close to the WE that is rich in K+ counter-ions due to the collapse of PSPMA chains, while region II comprises the area that will be depleted of K+ counter-ions upon applying a positive bias.
• Figure 4: Accumulative learning behavior. a) Time trace of a SRDP experiment with f = 50 Hz. b) SNDP tests performed with U = +0.2 V in 0.01 M and 0.5 M KCl. Pulse numbers were varied between 1-100 pulses, while f = 100 Hz. c) SRDP tests performed with U = +0.2 V in 0.01 M and 0.5 M. Pulse frequencies were varied between 1-100 Hz, with #pulses = 10. d) Discharging of salted PSPMA brushes (0.5 M) during the pulse train. Color coded from 1 Hz (red) to 100 Hz (black). e) Mechanistic picture of the trend in (c), specifically the salted regime.