The effect of finite pore length on ion structure and charging

Abstract

Nanoporous supercapacitors play an important role in modern energy storage systems, and their modeling is essential to predict and optimize the charging behaviour. Two classes of models have been developed that consist of finite and infinitely long pores. Here, we show that although both types of models predict qualitatively consistent results, there are important differences emerging due to the finite pore length. In particular, we find that the ion density inside a finite pore is not constant, but increases linearly from the pore entrance to the pore end, where the ions form a strongly layered structure. This hinders a direct quantitative comparison between the two models. In addition, we show that although the ion density between the electrodes changes appreciably with the applied potential, this change has a minor effect on charging. Our simulations also reveal a complex charging behaviour, which is adsorption-driven at high voltages, but it is dominated either by co-ion desorption or by adsorption of both types of ions at low voltages, depending on the ion concentration.

Superionic liquids in conducting nanoslits: A variety of phase transitions and ensuing charging behavior

“…[6]. Similarly, in a combined MC and MD study, Breitsprecher et al [54] also observed that the counter-ion adsorption was preceded by co-ion desorption, although they did not report on any phase transition.…”

Ion-ion correlations across and between electrified graphene layers

“…Moving to ILs confined in slit-like pores, several groups 38,39 used simulations to predict oscillations of the capacitance as a function of the pore size. Further work insisted on the importance of pore length 40 and on the necessity to account for the pore size distribution to interpret the experimental data 41 .…”

The effects of slit-pore geometry on capacitive properties: a molecular dynamics study

“…Considering all of the known mechanisms at the same time is not possible, especially in experimental [15][16][17][18] and analytical 11 studies. Because of that, there are different methods invented and used for EDLS simulations such as Molecular dynamics (MD) [19][20][21][22] , Monte-Carlo 23,24 , lattice model 7 , density functional theory (DFT) 13,25 , mixed DFT with MD and post-HartreeFock calculation 6 , and machine-learning [26][27][28] . In addition, modeling and simplifications in simulations gives us some hints about the ongoing microscopic processes in EDLC.…”

“…The geometry of the electrodes is an important factor of supercapacitors that highly affects its performance 22 . There are numerous researches on the different electrodes geometries such as, planar electrodes 20,21 , slit-pores 17,23,29 , combining flat and porous electrodes 30 , cylindrical pores 29,31 , spherical electrodes 31 , carbon nanotubes forest 32 , mathematically flat electrodes vs atomic structured electrodes 33 , atomic rough vs atomic non-rough electrode surfaces 10 and so on.…”

“…They are described as ion-philic and ion-phobic pores for filled and empty pores 34 . Ion-phobic pores are expected to have fast charging and higher energy storage than ion-philic ones 23 . Comparing initially filled and empty pores have been done by Kondrat and Kornyshev 35 with a mean-field model, they have reported that the initially filled pores charging is like diffusive and of the empty pores is a front-like process 8 .…”