Ionworks

3D electrochemical battery models

A 3D electrochemical battery model resolves current, concentration, and temperature distributions across the physical geometry of a cell. Available in Ionworks.

A 1D model tells you how a cell performs. A 3D model tells you where in the cell performance breaks down. At some point, the question changes from one to the other.

Live simulation · available in Ionworks

Full 3D DFN over a 1C discharge and relaxation

A browser-rendered time series from a full 3D DFN simulation of a pouch cell, discharged at 1 A for 30 minutes and then relaxed. Step through the profile to see where the thermal response localizes and how the current load drives it.

Loading /models/pouch_dfn_polydata/pouch_dfn_0000.vtk

Applied currentt=0s · I=1.00A
0.00 A1.00 A

What a 3D electrochemical model is

A 3D electrochemical battery model resolves electrode-level current, concentration, and temperature distributions across the physical geometry of a cell, rather than treating the cell as a single point.

Two common sources of confusion: 3D electrochemical is not the same as 3D thermal (which resolves temperature but not electrochemistry), and it is not CFD-coupled simulation (which resolves coolant flow around a pack). Both of those are useful, but they answer different questions.

The dimensionality ladder

SPM / SPMe

Single particle

No spatial resolution across the stack. Fastest. Good at low C-rate and for scoping.

DFN (P2D)

Through-thickness

Through-thickness electrode resolution. The workhorse for 1D cell-level studies.

Pseudo-3D

1D electrochemistry + 3D thermal

1D electrochemistry at each point on the cell, coupled to a 3D thermal and current-collector network. The pragmatic choice when thermal and in-plane current effects drive the decision.

Full 3D

Fully resolved electrochemistry

Spatially resolved electrode electrochemistry across the full cell geometry. The right tool when structured electrodes, tab geometry, or solid-state architectures make the 1D assumption break down.

When 3D matters

Scenario1D (DFN)Pseudo-3DFull 3D
Small-format cell, uniform coolingYes
Large-format pouch, tab-cooledYes
Current-collector Ohmic loss dominatesYes
Hotspot localization for thermal runaway riskYesYes
Structured electrodes, interdigitated geometry, solid-state researchYes
Operating envelope definition and design sweepsYesYesYes (slower)

What 1D cannot tell you

DFN and SPMe resolve electrochemistry through the thickness of the electrode sandwich. The assumption baked into both is lateral uniformity: every point across the cell face sees the same current density, potential, and temperature. The assumption holds for small cells, centered tabs, and modest C-rates. It starts to fail when any of those change.

Current distribution

Current enters through the tabs and spreads across the collectors before passing through the electrode stack. Current density is highest near the tabs and decays toward the far corners. A 1D model averages this away. For a large-format cell at high rate, the local-to-far-corner ratio can drive meaningfully different overpotential, lithium concentration, and aging at each end of the cell.

Tab location and internal resistance

Tab position sets the effective current path through the collectors. Top-top forces current to traverse the full collector length. Top-bottom or diagonal shortens the average path and reduces ohmic losses. Two otherwise identical cells in different tab configurations produce different DCIR profiles under the same operating conditions. A 1D model has no representation of tab geometry and cannot distinguish between configurations.

Thermal hotspots

Local heat generation tracks local current density. Where current concentrates, heat generation is highest, and the region near the tabs runs hotter than the rest of the cell. A lumped thermal model reports the average. A 1D coupled thermal-electrochemical model resolves through-thickness gradients but still assumes lateral uniformity. Only 3D predicts where the hotspot actually forms and how large the gradient to the cell edge really is.

Large-format cell behavior

The effects compound on cells with large electrode area. Lateral gradients in current density create lateral gradients in state of charge: one region can sit at 85% SoC while another is at 70%. Those gradients drive non-uniform aging. Under cold-temperature fast charging, the high-current region near the tabs is also the first place to plate lithium, because local overpotential is highest there. All of these couple, and they only show up in a 3D model.

How Ionworks fits in

How Ionworks supports 3D cell modeling

01

Pseudo-3D and full 3D electrochemical

Both are supported natively. Pseudo-3D couples 1D PyBaMM electrochemistry to a 3D thermal and current-collector network. Full 3D resolves the electrochemistry across the cell geometry directly. Teams pick the level of spatial resolution the decision needs.

02

Same parameters, same environment

Both 3D workflows read the same parameter set you fit for 1D DFN in Train. The equation set (PyBaMM) and the environment your team already works in stay the same. Dimensionality becomes a simulation setting rather than a separate tool.

03

Compare in Predict

Studies let teams run the same protocol through 1D, pseudo-3D, and full 3D side by side and surface where the answers diverge.

04

Sweep in Optimize

Design sweeps over tab placement, stack height, cell geometry, and cooling parameters run inside the same study framework used for 1D optimization.

What Ionworks is not

Ionworks is not a general-purpose 3D multiphysics platform. Teams doing 3D structural analysis, full CFD, or pack-level mechanical integration should keep using COMSOL, Ansys, or Simcenter. The focus here is depth on cell-level electrochemistry, in 1D, pseudo-3D, and full 3D, along with the workflow around it. Teams commonly run both: Ionworks for the cell-level electrochemical model, COMSOL or Simcenter for the pack-level fluid and structural problem.

Example questions teams answer

Tab placement

Does moving the tabs change the cycle life of this pouch cell under the actual duty cycle?

Fast-charge cooling envelope

Is the cooling envelope sufficient at 3C fast charge on this large-format cell?

Hotspot localization

Where is the hottest point in the stack at end-of-discharge, 45 °C ambient?

Frequently asked questions

A pseudo-3D model runs 1D electrochemistry (SPM, SPMe, or DFN) at many points across the cell, coupled to a 3D thermal model and a current-collector network that resolves Ohmic losses and spatial current distribution. It captures most of what matters for large-format cells at a fraction of the cost of a full 3D electrochemical model. Ionworks supports both pseudo-3D and full 3D electrochemical workflows.
When the cell is small-format, cooling is uniform, and the operating window does not create significant in-plane gradients, a 1D DFN or SPMe is enough. 18650 cells at moderate C-rate, coin cells, and first-pass design studies for any format are common examples where 1D is the right tool.
It depends on the model and the mesh. A pseudo-3D pouch-cell run on a cycling protocol is typically minutes to low hours. A full 3D electrochemical run on the same protocol is slower, typically several hours, depending on mesh resolution and protocol length. Teams routinely use full 3D for geometry-sensitive design questions and fall back to pseudo-3D when doing large sweeps.
For cell-level electrochemistry and the workflow around it (data ingest, parameterization, design sweeps, provenance), yes. For pack-level CFD, structural analysis, or general-purpose 3D multiphysics, no. Teams commonly use Ionworks for the cell model and carry the outputs into COMSOL or Simcenter for the pack-level problem.
Yes. The pseudo-3D model uses a 1D electrochemical kernel at each point on the cell, so if that kernel is DFN the parameter set is the same as the 1D DFN used elsewhere in the workflow. The full 3D electrochemical model reads the same parameter set as well. Teams parameterize once in Train and drive 1D, pseudo-3D, and full 3D studies from that one parameter set.
Validation is a combination of terminal-voltage matching against cycling data and spatial validation against measurements that resolve geometry: surface thermocouples, infrared imaging, or instrumented cells. Ionworks supports pulling both kinds of measurements into Measure and tying them back to the parameterized model.

See whether 1D, pseudo-3D, or full 3D changes the answer for your cell

Bring your cell geometry and cycling data. We will run the same protocol through each level of spatial resolution and show you where the answers diverge.