Overview Six Descriptors LFI Scale Environments Quick Setup Live Dashboard GitLab

Lambda-Flow Index (LFI)

ION-Logic introduces the first physics-informed AI framework for quantitative characterization and prediction of ion transport dynamics in complex electrochemical and biological ion-conducting environments — the Lambda-Flow Index (LFI). Built on six orthogonal physico-informational descriptors spanning Nernst-Planck neural transport dynamics, Debye-Hückel coupling efficiency, Butler-Volmer redox kinetics, membrane selectivity, ion concentration fractal topology, and noise-transport inhibition.

GitHub Repository Live Dashboard DOI: 10.5281/zenodo.19702569

93.1%

LFI Prediction Accuracy

42-platform cross-validation

94.8%

Failure Detection Rate

False alert: 3.2%

38 days

Early Warning

Mean lead time

5,148

ITUs

8 years · 42 platforms

The Six ION-Logic Descriptors

NIFP

Neural Ion-Flux Path · 26% Weight · Nernst-Planck Dynamics

Quantifies ion flux topology under Nernst-Planck dynamics. Range: 0.18–3.2 × 10⁻⁹ m²·s⁻¹·V⁻¹·m

DHCT

Debye-Hückel Coupling Tensor · 22% Weight · Electrostatic Theory

Measures departure from ideal Debye-Hückel behavior at finite ionic strength. DHCT > 0.83: COHERENT

RKT

Redox Kinetic Tensor · 20% Weight · Butler-Volmer Kinetics

Central parameter for Butler-Volmer kinetics and exchange current density.

MSC

Membrane Selectivity Coefficient · 16% Weight · Membrane Transport

Stoichiometric balance of ionic information exchange.

ICFD

Ion Concentration Fractal Dimension · 10% Weight · Fractal Electrochemistry

Fractal geometry of ion concentration field. D_f = 1.5–1.71: normal intact

NTII

Noise-Transport Inhibition Index · 6% Weight · Signal Degradation

Noise-driven transport degradation suppression. Mean field: NTII = 0.37

Lambda-Flow Index (LFI) Composite Formula

            LFI = 0.26·NIFP* + 0.22·DHCT* + 0.20·RKT* + 0.16·MSC* + 0.10·ICFD* + 0.06·NTII*
        

LFI_adj = σ(LFI_raw + β_conc + β_therm + β_em)AI correction with learned concentration/thermal/EM bias corrections

Python Interface

            
from ion_logic import LFIParameters, compute_lfi

params = LFIParameters(

    nifp=0.26,

    dhct=0.22,

    rkt=0.20,

    msc=0.16,

    icfd=1.71,

    ntii=0.37

)

result = compute_lfi(params, environment='battery_electrolyte')

LFI Alert Levels

LFI < 0.20
EXCELLENT

0.20–0.38
GOOD

0.38–0.58
MODERATE

0.58–0.78
CRITICAL

> 0.78
COLLAPSE

EXCELLENT: Standard monitoring

GOOD: Seasonal impedance review

MODERATE: Redesign planning

CRITICAL: Emergency recalibration

COLLAPSE: Immediate recovery

Six Electrochemical Environments

95.2%

PEM Fuel Cell / Electrolyzer

8 platforms · Nafion 117/212 · 60–90°C

94.1%

Li-ion Battery Electrolytes

9 platforms · LiPF6/EC-DMC · -20 to +80°C

93.7%

Biological Neural Channels

8 platforms · Na+/K+ ATPase · 35–42°C

92.9%

Seawater Desalination

7 platforms · RO/ED brines · 15–45°C

91.8%

Industrial Electroplating

4 platforms · Cu/Ni/Zn baths · 20–65°C

90.4%

Solid-State Ionic Conductors

6 platforms · LLZO, NASICON · 25–300°C

Quick Setup

            
# Clone repository

git clone https://gitlab.com/gitdeeper11/ION-Logic.git

cd ION-Logic

# Install package

pip install -e .

# Run analysis

python bin/analyze_eis.py --file your_data.csv

# Verify installation

python -c "from ion_logic import __version__; print(__version__)"

Physics-Informed Neural Network + Neural ODE

            
# PINN penalty layer constraints (from paper)

# • Nernst-Planck compliance: ion flux must satisfy electrochemical potential gradient equation

# • Charge electroneutrality: local sum of ionic charges must be zero at equilibrium

# • Thermodynamic consistency: activity coefficients must satisfy extended Debye-Hückel at observed ionic strength

# Python implementation

from ion_logic import IONLogicPredictor

predictor = IONLogicPredictor()

result = predictor.predict(eis_spectrum, current_params)

How to Cite

            
@software{baladi2026ionlogic,

    author = {Samir Baladi},

    title = {ION-Logic: Neural Ion-Kinetic Intelligence for Electrochemical Flow Prediction and Redox Dynamics Control},

    year = {2026},

    version = {1.0.0},

    publisher = {Zenodo},

    doi = {10.5281/zenodo.19702569},

    url = {https://doi.org/10.5281/zenodo.19702569},

    note = {Physics-Informed AI Framework for Ion Transport Dynamics}

}