🔬 materials science

Universal Framework for Decomposing Ionic Transport into Interpretable Mechanisms

This paper introduces a universal computational framework that decomposes macroscopic ion transport coefficients from molecular dynamics simulations into quantitatively additive, spatiotemporally resolved contributions from specific microscopic mechanisms, thereby enabling the identification of dominant transport modes and design rules across diverse electrolyte materials.

Original authors: KyuJung Jun, Pablo A. Leon, Jurğis Ruža, Juno Nam, Rafael Gómez-Bombarelli

Published 2026-02-19

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: KyuJung Jun, Pablo A. Leon, Jurğis Ruža, Juno Nam, Rafael Gómez-Bombarelli

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to figure out how a crowd of people moves through a busy airport terminal. You can see the overall flow: people are getting from Gate A to Gate B. But how are they doing it?

Are they just walking normally?
Are they hopping on moving walkways?
Are they grabbing a friend and pulling them along?
Are they swapping places with someone else in a chaotic shuffle?

For a long time, scientists studying batteries could only measure the total speed of ions (the "people") moving through battery materials. They knew the battery worked, but they didn't know the specific "moves" that made it fast or slow. It was like knowing a car is traveling at 60 mph but not knowing if it's going fast because the engine is powerful, the tires are good, or the driver is just pressing the gas pedal hard.

This paper introduces a new "super-slow-motion camera" and a smart sorting algorithm called OnsagerDecomposer. It takes the chaotic, high-speed dance of atoms in a battery and breaks it down into clear, understandable stories.

Here is how it works, using simple analogies:

1. The "Virtual Particle" Trick

Imagine every lithium ion in the battery is a person. The new framework says: "Let's pretend this one person is actually five different people wearing different colored hats."

Hat A: Represents when the person is just wiggling in place (vibrating).
Hat B: Represents when the person hops alone.
Hat C: Represents when the person hops while holding hands with a neighbor (a "concerted" hop).
Hat D: Represents when the person is riding on a moving platform (like a polymer chain).
Hat E: Represents when the person swaps places with a neighbor.

The computer tracks the person's movement and assigns every tiny step they take to the correct "hat." If they are just wiggling, that step counts toward Hat A. If they jump, it counts toward Hat B.

The Magic: When you add up all the steps taken by all the "hats," you get the exact total speed of the battery. Nothing is lost or made up. It's a perfect accounting system.

2. The "Zoom Lens" (Time Windows)

The researchers also realized that the answer depends on how fast you look.

Looking at a 0.1-second snapshot: You mostly see people wiggling in place.
Looking at a 1-second snapshot: You start seeing people hopping.
Looking at a 10-second snapshot: You see the whole journey, but you can't tell if they walked or took a bus.

By scanning through different "time windows," the framework finds the sweet spot where you can clearly see the different moves. It tells us: "Ah, the 'hopping' move happens mostly in 1-second bursts, while the 'wiggling' happens in milliseconds."

3. What They Found in Three Different "Airports"

The team tested this on three types of battery materials, and the results were like discovering different travel habits in different cities:

A. The Crystal City (Inorganic Solids)

The Old Theory: Ions just hop from one spot to another, like stepping stones.
The New Discovery: It's not just solo hopping! The most efficient move is a group hop. Imagine a group of friends in a dance line; when one moves, the others move with them in perfect sync.
The Lesson: To make these batteries faster, we need to design materials that encourage these "group dances" (called concerted hops) rather than solo jumps.

B. The Rubber Jungle (Polymer Electrolytes)

The Old Theory: Ions just ride along as the rubbery chains wiggle and stretch (like a surfer on a wave).
The New Discovery: Riding the wave is actually the slowest way to go. The fastest way is for the ion to jump from one rubber chain to another (like a monkey swinging from tree to tree).
The Lesson: In the common polymer PEO, the ions are stuck riding the slow waves. But in a newer polymer called PPM, the ions can swing between chains much more often. This explains why PPM is a better conductor, even if the rubber itself moves slower!

C. The Liquid Pool (Liquid Electrolytes)

The Old Theory: Ions swim through the liquid, dragging their water/solvent friends with them.
The New Discovery: It's all about swapping partners. In some liquids, the ion is like a dancer who quickly swaps partners with the person next to them. This "partner swap" (solvent exchange) is incredibly fast.
The Lesson: A new type of liquid electrolyte (FAN) works amazingly well because the ions can swap partners very easily, creating a "highway" for movement. The old liquids (EC) are slower because the ions hold onto their partners too tightly.

Why This Matters

Before this, scientists were guessing which moves were important. Now, they have a map.

For Engineers: If you want to build a faster battery, don't just guess. Look at the map. If the "group hop" is the winner, design a crystal that encourages groups. If "swapping partners" is the winner, design a liquid that makes swapping easy.
For the Future: This tool turns a blurry video of chaos into a clear instruction manual. It helps us design the next generation of batteries for electric cars and phones by understanding exactly how the energy moves, rather than just hoping it works.

In short, this paper gave us the ability to see the "secret handshake" of atoms, turning a mystery into a blueprint for better energy storage.

1. Problem Statement

Designing next-generation ion conductors for energy storage (e.g., batteries) requires a rigorous understanding of ion transport mechanisms. While macroscopic properties like ionic conductivity and transference numbers are routinely measured, the underlying microscopic processes remain poorly quantified.

The Gap: Experimental techniques (EIS, NMR, QENS) lack simultaneous high spatial and temporal resolution. Conversely, while All-atom Molecular Dynamics (MD) simulations offer high resolution, they have historically been limited to reporting overall transport coefficients or qualitative snapshots.
The Challenge: There is no general, quantitative framework to decompose macroscopic transport into additive contributions from specific microscopic events (e.g., single-ion hops vs. concerted motion, vehicular vs. exchange mechanisms) across different timescales. Existing methods are often system-specific, rely on simplifying assumptions, or fail to guarantee that the sum of mechanistic contributions equals the exact Onsager transport coefficients derived from MD.

2. Methodology: OnsagerDecomposer

The authors introduce OnsagerDecomposer, a general, discrete stochastic framework that analyzes equilibrium MD trajectories to provide a timescale-resolved, additive decomposition of macroscopic transport.

Core Algorithmic Steps:

Discretization & State Definition:
- MD trajectories are subsampled at intervals of $\Delta t$ .
- Microstates: Defined by the coordination environment of an ion (identities of atoms/molecules in the first solvation shell).
  - Crystals: Mapped to Wyckoff sites.
  - Liquids/Polymers: Determined via distance cutoffs or convex-hull methods.
- Macrostates: Coarse-grained labels representing chemically interpretable environments (e.g., Solvent-Separated Ion Pairs [SSIP], Contact Ion Pairs [CIP], Aggregates [AGG]).
Event Classification:
- Transitions between microstates or macrostates within a window $\Delta t$ are mapped to a predefined, complete, and non-overlapping set of event classes.
- Examples:
  - Crystals: Local vibration, single-ion hop, aligned concerted hop, unaligned concerted hop.
  - Polymers: Vehicular motion, intrachain hop, interchain hop, anion exchange.
  - Liquids: Vehicular transport, solvent exchange, anion exchange, simultaneous solvent-anion exchange.
Virtual Particle Construction for Additivity:
- To ensure mathematical rigor, the framework introduces event-specific virtual particles.
- For a given ion $i$ and event type $m$ , a virtual particle undergoes displacement $d_i(t, \Delta t)$ only if the ion undergoes event $m$ during that window.
- This ensures that the sum of displacements of all virtual particles equals the total displacement of the real ion.
Decomposition of Onsager Coefficients:
- The framework calculates displacement correlations ( $C_{AB}$ ) for each event type.
- Using the Green-Kubo/Einstein formalism, it computes event-resolved Onsager transport coefficients ( $L_{AB,m}$ ).
- Key Guarantee: The sum of decomposed coefficients exactly equals the total Onsager coefficient: $L_{AB} = \sum_m L_{AB,m}$ .
- Timescale Scanning: By varying $\Delta t$ , the method identifies characteristic timescales where specific mechanisms emerge, dominate, or coalesce.
Metrics:
- Frequency ( $p_m$ ): Probability of an event occurring.
- Effectiveness: The transport coefficient contribution normalized by frequency ( $L_m / p_m$ ), identifying which mechanisms are most efficient per occurrence.
- Activation Energy: Extracted for specific event types via Arrhenius analysis.

3. Key Contributions

General Framework: A reproducible, system-agnostic tool applicable to inorganic crystals, polymers, and liquids.
Mathematical Rigor: Guarantees exact additivity of mechanistic contributions to macroscopic transport coefficients, resolving previous inconsistencies in decomposition methods.
Timescale Resolution: Uniquely reveals how the dominance and effectiveness of transport mechanisms evolve with observation timescale ( $\Delta t$ ).
Quantitative Validation: Moves beyond qualitative hypotheses to provide quantitative metrics (frequencies, effectiveness, activation energies) for competing transport theories.

4. Key Results

A. Inorganic Crystalline Electrolytes (LATP)

Concerted Motion: The framework quantitatively confirms that concerted hops (multiple ions moving cooperatively) are the dominant transport mechanism in Li-stuffed NASICONs, contributing ~75% of conductivity at 1 ps.
Effectiveness: Aligned concerted hops are >2x more effective per event than single-ion hops.
Activation Energy: The activation energy for aligned concerted hops (0.13 eV) is significantly lower than for single-ion hops (0.29 eV), validating NEB predictions.
Mechanism of "Stuffing": Lithium stuffing (excess Li) creates "activated" local environments (high-energy states due to Li-Li repulsion). Transport in stuffed materials is dominated by transitions involving these activated states, whereas stoichiometric materials rely on rare formation/annihilation of these states.

B. Polymer Electrolytes (PEO vs. PPM)

PEO (Standard): Transport is dominated by vehicular motion and intrachain hops, both coupled to slow polymer segmental dynamics (Rouse dynamics). Interchain hops are rare (<1%) and thus contribute little despite being highly effective.
PPM (Novel Polymer): Despite having slower polymer segmental dynamics (lower $D_{CRU}$ ), PPM exhibits higher Li diffusivity and a 3x higher cation transference number.
Mechanism: PPM facilitates frequent interchain hops (~15-60% of events). These highly effective events decouple Li motion from the slow polymer backbone, compensating for the sluggish matrix.
Design Rule: Increasing the frequency of interchain hops is a more effective strategy for improving conductivity than optimizing vehicular motion.

C. Liquid Electrolytes (EC vs. FAN)

EC (Standard): Transport is a balance between vehicular motion and solvent exchange. Anion exchange is negligible.
FAN (High Conductivity): The high conductivity of FAN-based electrolytes is driven by a highly effective solvent exchange mechanism (ligand-channel transport).
Macrostate Analysis: While FAN has a high population of Aggregates (AGG), these are slow. The high conductivity arises from a small population of SSIP and CIP states that undergo extremely rapid solvent exchange.
Validation: The framework quantitatively validates the "ligand-channel" hypothesis, showing that solvent exchange in FAN is significantly more effective than in EC.

5. Significance and Impact

Unified Understanding: The paper provides a unified, quantitative picture of ion transport across fundamentally different material classes, bridging the gap between atomistic dynamics and macroscopic performance.
Design Guidelines: It offers actionable design rules:
- Crystals: Promote "activated" local environments to enable concerted motion.
- Polymers: Design chemistries that increase the frequency of interchain hops to decouple ions from polymer dynamics.
- Liquids: Weaken Li-solvent interactions just enough to enable rapid solvent exchange without sacrificing salt dissociation.
Closed-Loop Design: The framework enables a "mechanism-aware" computational design loop: diagnose bottlenecks (e.g., rare interchain hops), propose structural changes, simulate, and quantitatively verify the mechanistic shift.
Extensibility: The approach is applicable to other ions (Na, protons, multivalent) and mixed ionic-electronic conductors, accelerating the discovery of next-generation energy storage materials.