A dynamically equivalent atomistic electrochemical paradigm for the larger-scale experiments.

Electrochemical systems possess a considerable part of modern technologies, such as the operation of rechargeable batteries and the fabrication of electronic components, which are explored both experimentally and computationally. The largest gap between the experimental observations and atomic-level simulations is their orders-of-magnitude scale difference. While the largest computationally affordable scale of the atomic-level computations is ∼ns and ∼nm, the smallest reachable scale in the typical experiments, using very high-precision devices, is ∼s and ∼µm. In order to close this gap and correlate the studies in the two scales, we establish an equivalent simulation setup for the given general experiment, which excludes the microstructure effects (i.e., solid-electrolyte interface), using the coarse-grained framework. The developed equivalent paradigm constitutes the adjusted values for the equivalent length scale (i.e., lEQ), diffusivity (i.e., DEQ), and voltage (i.e., VEQ). The time scale for the formation and relaxation of the concentration gradients in the vicinity of the electrode matches for both smaller scale (i.e., atomistic) equivalent simulations and the larger scale (i.e., continuum) experiments and could be utilized for exploring the cluster-level inter-ionic events that occur during the extended time periods. The developed model could offer insights for forecasting experiment dynamics and estimating the transition period to the steady-state regime of operation.

Dendritic propagation on circular electrodes: The impact of curvature on the packing density.

The dendritic growth in rechargeable batteries is one of the hurdles for the utilization of high energy-density elements, such as alkaline metals, as the electrode. Herein we explore the preventive role of the curved electrode surface in the cylindrical electrode design versus the flat geometry on the stochastic evolution of the dendritic crystals. In this regard we establish a coarse-grained Monte Carlo paradigm in the polar coordinates (r,θ), which runs in a larger scale of time and space (∼µs,∼nm ) than those of interionic collisions (∼fs, Å). Subsequently we track the density and the maximum reach of the microstructures in real time, and we elaborate on the underlying mechanisms for their correlation of the relative dendrite measure with the electrode curvature. Such quantification of the positive impact of the curvature on suppressing dendrites could be utilized as an effective longevity design parameter, particularly for the cases prone to dendritic propagation.

Real-time control of dendritic propagation in rechargeable batteries using adaptive pulse relaxation.

The non-uniform growth of microstructures in dendritic form inside the battery during prolonged charge-discharge cycles causes short-circuit as well as capacity fade. We develop a feedback control framework for the real-time minimization of such microstructures. Due to the accelerating nature of the branched evolution, we focus on the early stages of growth, identify the critical ramified peaks, and compute the effective time for the dissipation of ions from the vicinity of those branching fingers. The control parameter is a function of the maximum interface curvature (i.e., minimum radius) where the rate of runaway is the highest. The minimization of the total charging time is performed for generating the most packed microstructures, which correlate closely with those of considerably higher charging periods, consisting of constant and uniform square waves. The developed framework could be utilized as a smart charging protocol for safe and sustainable operation of rechargeable batteries, where the branching of the microstructures could be correlated with the sudden variation in the current/voltage.

Correction: Elliptic percolation model for predicting the electrical conductivity of graphene-polymer composites.

Correction for 'Elliptic percolation model for predicting the electrical conductivity of graphene-polymer composites' by Asghar Aryanfar et al., Soft Matter, 2021, 17, 2081-2089, DOI: 10.1039/D0SM01950J.

Elliptic percolation model for predicting the electrical conductivity of graphene-polymer composites.

Graphene-based polymers exhibit a conductive microstructure formed by aggregates in a matrix which drastically enhances their transmitting properties. We develop a new numerical framework for predicting the electrical conductivity based on continuum percolation theory in a two dimensional stochastically-generated medium. We analyze the role of the flake shape and its aspect ratio and consequently predict the onset of percolation based on the particle density and the domain scale. Simultaneously, we have performed experiments and have achieved very high electrical conductivity for such composites compared to other film fabrication techniques, which have verified the results of computing the homogenized electrical conductivity. As well, the proximity to and a comparison with other analytical models and other experimental techniques are presented. The numerical model can predict the composite transmitting conductivity in a larger range of particle geometry. Such quantification is exceedingly useful for effective utilization and optimization of graphene filler densities and their spatial distribution during manufacturing.

Finite-pulse waves for efficient suppression of evolving mesoscale dendrites in rechargeable batteries.

The ramified and stochastic evolution of dendritic microstructures has been a major issue on the safety and longevity of rechargeable batteries, particularly for the utilization of high-energy metallic electrodes. We analytically develop criteria for the pulse characteristics leading to the effective halting of the ramified electrodeposits grown during extensive timescales beyond inter-ionic collisions. Our framework is based on the competitive interplay between diffusion and electromigration and tracks the gradient of ionic concentration throughout the entire cycle of pulse-rest as a critical measure for heterogeneous evolution. In particular, the framework incorporates the Brownian motion of the ions and investigates the role of the geometry of the electrodeposition interface. Our experimental observations verify the analytical developments, where the dimension-free developments allows the application to the electrochemical systems of various scales.

Enhanced strength and temperature dependence of mechanical properties of Li at small scales and its implications for Li metal anodes.

Most next-generation Li ion battery chemistries require a functioning lithium metal (Li) anode. However, its application in secondary batteries has been inhibited because of uncontrollable dendrite growth during cycling. Mechanical suppression of dendrite growth through solid polymer electrolytes (SPEs) or through robust separators has shown the most potential for alleviating this problem. Studies of the mechanical behavior of Li at any length scale and temperature are limited because of its extreme reactivity, which renders sample preparation, transfer, microstructure characterization, and mechanical testing extremely challenging. We conduct nanomechanical experiments in an in situ scanning electron microscope and show that micrometer-sized Li attains extremely high strengths of 105 MPa at room temperature and of 35 MPa at 90 °C. We demonstrate that single-crystalline Li exhibits a power-law size effect at the micrometer and submicrometer length scales, with the strengthening exponent of -0.68 at room temperature and of -1.00 at 90 °C. We also report the elastic and shear moduli as a function of crystallographic orientation gleaned from experiments and first-principles calculations, which show a high level of anisotropy up to the melting point, where the elastic and shear moduli vary by a factor of ∼4 between the stiffest and most compliant orientations. The emergence of such high strengths in small-scale Li and sensitivity of this metal's stiffness to crystallographic orientation help explain why the existing methods of dendrite suppression have been mainly unsuccessful and have significant implications for practical design of future-generation batteries.

Lithium batteries: Improving solid-electrolyte interphases via underpotential solvent electropolymerization.

Understanding the mechanism of formation of solid-electrolyte interphases (SEI) is key to the prospects of lithium metal batteries (LMB). Here, we investigate via cyclic voltammetry, impedance spectroscopy and chronoamperometry the role of kinetics in controlling the properties of the SEI generated from the reduction of propylene carbonate (PC, a typical solvent in LMB). Our observations are consistent with the operation of a radical chain PC electropolymerization into polymer units whose complexity increases at lower initiation rates. As proof-of-concept, we show that slow initiation rates via one-electron PC reduction at underpotentials consistently yields compact, electronically insulating, Li+-conducting, PC-impermeable SEI films.

Annealing kinetics of electrodeposited lithium dendrites.

The densifying kinetics of lithium dendrites is characterized with effective activation energy of Ea ≈ 6 - 7 kcal mol(-1) in our experiments and molecular dynamics computations. We show that heating lithium dendrites for 55 °C reduces the representative dendrites length λ¯(T,t) up to 36%. NVT reactive force field simulations on three-dimensional glass phase dendrites produced by our coarse grained Monte Carlo method reveal that for any given initial dendrite morphology, there is a unique stable atomic arrangement for a certain range of temperature, combined with rapid morphological transition (∼10 ps) within quasi-stable states involving concurrent bulk and surface diffusions. Our results are useful for predicting the inherent structural characteristics of lithium dendrites such as dominant coordination number.

Thermal relaxation of lithium dendrites.

The average lengths λ̅ of lithium dendrites produced by charging symmetric Li(0) batteries at various temperatures are matched by Monte Carlo computations dealing both with Li(+) transport in the electrolyte and thermal relaxation of Li(0) electrodeposits. We found that experimental λ̅(T) variations cannot be solely accounted by the temperature dependence of Li(+) mobility in the solvent but require the involvement of competitive Li-atom transport from metastable dendrite tips to smoother domains over ΔE(++)(R) ∼ 20 kJ mol(-1) barriers. A transition state theory analysis of Li-atom diffusion in solids yields a negative entropy of activation for the relaxation process: ΔS(++)(R) ≈ -46 J mol(-1) K(-1) that is consistent with the transformation of amorphous into crystalline Li(0) electrodeposits. Significantly, our ΔE(++)(R) ∼ 20 kJ mol(-1) value compares favorably with the activation barriers recently derived from DFT calculations for self-diffusion on Li(0)(001) and (111) crystal surfaces. Our findings suggest a key role for the mobility of interfacial Li-atoms in determining the morphology of dendrites at temperatures above the onset of surface reconstruction: TSR ≈ 0.65 TMB (TMB = 453 K: the melting point of bulk Li(0)).

Quantifying the dependence of dead lithium losses on the cycling period in lithium metal batteries.

We quantify the effects of the duration of the charge-discharge cycling period on the irreversible loss of anode material in rechargeable lithium metal batteries. We have developed a unique quantification method for the amount of dead lithium crystals (DLCs) produced by sequences of galvanostatic charge-discharge periods of variable duration τ in a coin battery of novel design. We found that the cumulative amount of dead lithium lost after 144 Coulombs circulated through the battery decreases sevenfold as τ shortens from 16 to 2 hours. We ascribe this outcome to the faster electrodissolution of the thinner dendrite necks formed in the later stages of long charging periods. This phenomenon is associated with the increased inaccessibility of the inner voids of the peripheral, late generation dendritic structures to incoming Li(+).

Effects of anodic potential and chloride ion on overall reactivity in electrochemical reactors designed for solar-powered wastewater treatment.

We have investigated electrochemical treatment of real domestic wastewater coupled with simultaneous production of molecular H2 as useful byproduct. The electrolysis cells employ multilayer semiconductor anodes with electroactive bismuth-doped TiO2 functionalities and stainless steel cathodes. DC-powered laboratory-scale electrolysis experiments were performed under static anodic potentials (+2.2 or +3.0 V NHE) using domestic wastewater samples, with added chloride ion in variable concentrations. Greater than 95% reductions in chemical oxygen demand (COD) and ammonium ion were achieved within 6 h. In addition, we experimentally determined a decreasing overall reactivity of reactive chlorine species toward COD with an increasing chloride ion concentration under chlorine radicals (Cl·, Cl2(-)·) generation at +3.0 V NHE. The current efficiency for COD removal was 12% with the lowest specific energy consumption of 96 kWh kgCOD(-1) at the cell voltage of near 4 V in 50 mM chloride. The current efficiency and energy efficiency for H2 generation were calculated to range from 34 to 84% and 14 to 26%, respectively. The hydrogen comprised 35 to 60% by volume of evolved gases. The efficacy of our electrolysis cell was further demonstrated by a 20 L prototype reactor totally powered by a photovoltaic (PV) panel, which was shown to eliminate COD and total coliform bacteria in less than 4 h of treatment.

Dynamics of Lithium Dendrite Growth and Inhibition: Pulse Charging Experiments and Monte Carlo Calculations.

Short-circuiting via dendrites compromises the reliability of Li-metal batteries. Dendrites ensue from instabilities inherent to electrodeposition that should be amenable to dynamic control. Here, we report that by charging a scaled coin-cell prototype with 1 ms pulses followed by 3 ms rest periods the average dendrite length is shortened ∼2.5 times relative to those grown under continuous charging. Monte Carlo simulations dealing with Li(+) diffusion and electromigration reveal that experiments involving 20 ms pulses were ineffective because Li(+) migration in the strong electric fields converging to dendrite tips generates extended depleted layers that cannot be replenished by diffusion during rest periods. Because the application of pulses much shorter than the characteristic time τc ∼ O(∼1 ms) for polarizing electric double layers in our system would approach DC charging, we suggest that dendrite propagation can be inhibited (albeit not suppressed) by pulse charging within appropriate frequency ranges.