Logic Gates with Ion Transistors

Electronic logic gates are the basic building blocks of every computing and micro controlling system. Logic gates are made of switches, such as diodes and transistors. Ion-selective, ionic switches may emulate electronic switches [1-8]. If we ever want to create artificial bio-chemical circuitry, then we need to move a step further towards ion-logic circuitry. Here we demonstrate ion XOR and OR gates with electrochemical cells, and specifically, with two wet-cell batteries. In parallel to vacuum tubes, the batteries were modified to include a third, permeable and conductive mid electrode (the gate), which was placed between the anode and cathode in order to affect the ion flow through it. The key is to control the cell output with a much smaller biasing power, as demonstrated here. A successful demonstration points to self-powered ion logic gates. Introduction: Electrochemical reactions have been studied since the early eighteenth century [9-10]. Two half-cell reactions are considered. In one, oxidation of the anode takes place. Excess electrons then flow through an external load to the second half-cell, where reduction takes place at the cathode. The circuit is completed by ionic current in the electrolyte. The two-half cells are connected via a permeable membrane, which enables the passage of ions, yet, limits the flow of the bulk electrolyte molecules. We wish to control the ion flow inside electrochemical cells, electrically. Control of a reaction near an electrode (working electrode) is routinely made with an auxiliary electrode and a saturated reference electrode using potentiostats or galvanostats. This approach may affect the surface potential of the working electrode and the control process could become nonlinear. Our approach is different: here, a third permeable electrode (the gate electrode) is placed between the anode and the cathode. Upon biasing of this mid-electrode we form an electrolyte barrier to the flowing ions. Consequently, the external current and voltage of the cell are controlled [11-13]. Our approach is also different than what is accustomed to in the literature. The latter are typically conducted with functionalized membranes [14-15] for the purpose of ion separation. A bipolar Ion Transistor reported in Ref. 1 is a good example – the design was based on ion-selective membranes, and hence was ion specific. In contrast, we aim at controlling both anions and cations by the same electrolyte barrier potential. Simulations: Simulations employed a commercial tool, based on finite elements (COMSOL). We used a very simple Zn-Pt cell: a Zn electrode as the anode and a Pt electrode as the cathode. The model allowed us to deal with a single ion component (Zn) and took into account the reactions at the anode (oxidation of Zn) and on the cathode (formation of hydrogen), yet assumed no reaction at the gate. The diffusion of ions in the cell has considered only excess Zn ions in the electrolyte. The local ion current density was assessed as the negative spatial derivative of the local electrolyte potential (which is proportional to the local electric field) multiplied by the electrolyte conductivity. The effective electrolyte-to-metallic volume ratio in the porous electrode was 1:1. Other simulation parameters were: electrical conductivity of Pt, Zn, porous electrode and Zn concentration in the electrolyte, respectively: 10, 10, 3x10, 0.01 S/m. The upper tip of the Pt cathode was grounded and the upper tip of the Zn anode was kept at (-0.8) V, slightly lower than the standard potential of the Zn anode (E0=0.82 V). This means that the cell's voltage (between Pt cathode and the Zn anode) was +0.8 V. Results are shown in Figs. 1, 2. In the absence of gate reaction, the gate voltage that stops the battery from functioning (the stopping potential) is 0 V when the gate is biased with respect to the grounded cathode. The stopping potential is -2 V when the gate is biased with respect to the grounded anode. The external electrical cell current is negative (meaning flowing towards the anode) and its slope is negative, similar to Fig. 1a. We may conclude that: (1) changes in the electrolyte potential at the gate affect the external current. (2) There are gate bias conditions which can effectively stop the external cell's current from flowing. (a) (b) Fig. 1: (a) External cell's current density as a function of gate voltage. The current density is presented at various times. (b) Electrolyte potential at the mid-gate position as a function the gate voltage. The electrolyte potential is presented at various times. -8 -4 0 4 8 -0.5 -0.25 0 0.25 0.5 C u rr e n t D e n s it y ( A /m 2 ) Gate Voltage, Vg (V) 0 sec 60 sec 180 sec 300 sec 3600 sec -0.3 -0.2 -0.1 0 0.1 0.2 0.3 -0.5 -0.25 0 0.25 0.5 E le c tr o ly te P o te n ti a l (V ) Gate Voltage, Vg (V) 0 sec 60 sec 180 sec 300 sec 3600 sec (a) (b) Fig. 2: Slices of the Electrolyte Potential at (a) Vg=-0.1 V (a) and at (b) Vg=+0.1 V after 60 sec. The rim of the gate electrode is outside the electrolyte and its potential is the biasing potential, Vg. Experiment and Methods: Schematics of the wet-cell battery are shown in Fig. 3a-b. It measured 12 cm x 6 cm x 5 cm. It was made of two compartments: in one, a copper electrode was immersed in 0.1M CuSO4. In the other, a zinc electrode was immersed in 0.1M ZnSO4. The ion-bridge was replaced by a membrane (TS80 polyamide filter made by Sterlitech) on which a thin film of 60:40, Au-Pd film was sputtered using a Hummer V sputtering system under Ar environment (this system is often used for SEM conductive film deposition). The resistance of the film was ca 25 KOhms/cm. The membrane was held tightly between two polymethyl methacrylate (PMMA) plates. One plate was glued to the cell while the other could be removed for gate membrane replacement. A 7-mm hole was drilled through each of the plates in order to let passage to the ions. A copper lead provided contact to the gate membrane. A simple paper filter, placed behind the gate membrane, enabled a better holding of the two plastic plates together.