🔬 materials science

Full-Field Damage Monitoring in Architected Lattices Using In situ Electrical Impedance Tomography

This paper demonstrates the first in situ implementation of electrical impedance tomography within tunable architected lattices to achieve real-time, full-field monitoring of damage evolution and fracture localization in 3D-printed multifunctional composites, establishing a scalable sensing modality for autonomous intelligent materials and digital twin applications.

Original authors: Akash Deep, Andrea Samore, Alistair McEwan, Andrew McBride, Shanmugam Kumar

Published 2026-02-18

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Akash Deep, Andrea Samore, Alistair McEwan, Andrew McBride, Shanmugam Kumar

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 have a complex, 3D-printed sponge made of a special, stretchy plastic mixed with tiny, invisible carbon threads. This sponge is designed to be strong, but like any material, it can crack or break under pressure.

Usually, to know if something is breaking, you have to stick tiny sensors (like strain gauges) on specific spots. It's like trying to figure out where a leak is in a giant, complex pipe system by only checking a few valves. If the leak happens far away from your valves, you won't know until the whole system fails.

This paper introduces a smarter way: turning the entire object into a giant, self-sensing "electrical map."

Here is the simple breakdown of what the researchers did, using some everyday analogies:

1. The "Smart Sponge" (The Material)

The team 3D-printed a lattice structure (think of it like a complex honeycomb or a branching tree skeleton) using a special ink. This ink wasn't just plastic; it was mixed with Carbon Nanotubes (CNTs).

The Analogy: Imagine the plastic is the road, and the carbon nanotubes are the cars driving on it. As long as the road is smooth, the cars (electricity) flow easily. If the road cracks, the cars get stuck, and the flow stops.

2. The "Electrical X-Ray" (EIT)

Instead of sticking sensors on specific spots, they placed 16 tiny electrical contacts around the edge of the sponge. They then sent electrical currents through the object from different angles and measured the voltage coming out the other side.

The Analogy: Think of this like a CT scan for a broken bone, but instead of X-rays, they use electricity. By sending currents through the object from many different directions (like shining a flashlight from every angle in a dark room), a computer can mathematically "guess" what's happening inside. If a part of the road (the lattice) cracks, the electricity has a harder time getting through that specific spot. The computer uses this data to draw a live, color-coded map of the damage.

3. Designing the "Sensing Skin"

The researchers didn't just print any shape; they used a special mathematical design (inspired by wallpaper patterns and tree branches) to make the lattice.

The Analogy: They engineered the "roads" inside the sponge so that electricity naturally flows through the most critical areas. It's like designing a city's traffic grid so that if a bridge collapses, the traffic jam is immediately obvious to the control center. They found that by tweaking the shape of the lattice, they could make the "electrical X-ray" much sharper and more sensitive to tiny cracks.

4. Catching the "Pre-Failure" (The Results)

When they stretched the sponge until it broke, two things happened:

The Old Way: A simple resistance meter told them, "Hey, the whole thing is getting harder to push electricity through." But it couldn't tell them where the crack was.
The New Way (EIT): The computer generated a live video map. It showed a tiny red spot appearing before the material actually snapped. It pinpointed exactly which "branch" of the lattice was about to break, even if that branch was far away from the sensors on the edge.

Why This Matters

This is a huge leap forward because it turns a passive building material into an active, intelligent one.

Before: You had to guess when a bridge or a plane part was getting tired.
Now: The material itself can "feel" its own damage and tell you exactly where it is happening in real-time.

The Big Picture:
Imagine a future where your car's chassis, a building's support beams, or even a medical implant can "talk" to you. They could say, "I'm fine, but there's a tiny crack forming in the top-left corner, let's fix it before it gets worse." This research is a major step toward creating those "self-aware" materials that can monitor their own health, preventing disasters before they happen.

1. Problem Statement

Architected lattice materials and metamaterials offer tailored mechanical properties and multifunctionality (e.g., sensing, actuation) through their geometric design rather than just composition. However, monitoring the internal state of these complex 3D structures presents significant challenges:

Limitations of Conventional Sensing: Traditional piezoresistive sensing relies on discrete point measurements or global end-to-end resistance. While effective for simple structures, these methods cannot provide full-field spatial resolution. They fail to detect damage occurring far from electrodes or to track the specific spatial evolution of fracture propagation in real time.
Complexity of Current Models: Existing multiphysics models that attempt to reconstruct resistance fields are computationally intensive, require extensive constitutive calibration, and often suffer from additive manufacturing-induced errors that limit predictive accuracy.
Need for Real-Time Monitoring: There is a critical gap in robust, non-invasive techniques capable of mapping internal conductivity changes and damage evolution (including pre-fracture events) within complex, 3D-printed lattice composites during mechanical loading.

2. Methodology

The authors developed a novel framework integrating Additive Manufacturing (AM), conductive nanocomposites, and Electrical Impedance Tomography (EIT).

Material Fabrication:
- Composite: Multi-walled carbon nanotubes (MWCNTs) were dispersed into a photocurable resin (PlasGray) to create a conductive nanocomposite. Scanning Electron Microscopy (SEM) confirmed uniform CNT dispersion.
- Geometry: 2D auxetic lattices were designed using a Voronoi-based branch–trunk–branch motif inspired by 2D wallpaper symmetries ( $p4$ group). A computational framework allowed systematic tuning of branch lengths, junction angles, and relative density.
- Printing: Specimens were fabricated using Digital Light Processing (DLP) 3D printing with optimized parameters for layer thickness and exposure to ensure structural integrity.
Experimental Setup:
- Electrode Configuration: Sixteen electrodes were distributed along the lattice periphery using conductive silver paint and epoxy.
- Loading: Quasi-static uniaxial tensile tests were performed.
- Sensing Schemes: Two current injection protocols were tested:
  1. Adjacent Injection: Current between neighboring electrodes.
  2. Across Injection: Current between geometrically opposite electrodes.
- Data Acquisition: A Spectro-EIT system measured boundary voltages at 1 kHz. Data was processed in real-time using the EIDORS MATLAB toolbox.
Reconstruction Algorithm:
- The study employed a Maximum a Posteriori (MAP) reconstruction framework to solve the inverse problem.
- It estimated conductivity changes ( $\Delta\sigma$ ) between sequential loading steps using a Jacobian matrix ( $H$ ) derived from Finite Element Method (FEM) forward simulations.
- Regularization (Laplace prior) was used to handle the ill-posed nature of the inverse problem and noise.

3. Key Contributions

First In Situ EIT in Lattices: This is the first demonstration of applying EIT in situ within a tunable architected lattice framework to monitor damage evolution in 3D-printed composites.
Geometry as a Sensing Component: The study proves that lattice geometry can be deliberately engineered to enhance EIT sensitivity. By tuning the Voronoi motifs, the structure itself becomes an active component of the imaging system, improving spatial uniformity and detection of remote damage.
Dual-Scale Sensing: The work integrates full-field EIT imaging (localizing specific fracture sites) with global four-point probe (4PP) measurements (tracking macroscopic stress-strain), providing a comprehensive view of material failure.
Calibration-Free Imaging: Unlike numerical modeling approaches that require complex constitutive calibration, EIT provides a measurement-driven, non-destructive method to visualize strain localization and damage without prior material models.

4. Key Results

Sensitivity Analysis: A Jacobian-based sensitivity analysis identified a specific lattice design (Lattice A) with the highest mean and minimum sensitivity. This design was selected for subsequent experiments, demonstrating that geometric tuning directly impacts EIT detectability.
Damage Localization:
- Pristine Lattices: EIT successfully reconstructed conductivity maps that resolved sequential ligament fractures with high temporal resolution. Localized conductivity loss coincided precisely with fracture sites, even those remote from the electrodes.
- Pre-Cracked Lattices: The system detected damage initiation and propagation from a pre-existing central crack, proving its ability to monitor internal defects away from the boundary.
Quantitative Metrics:
- Gauge Factor: The lattice exhibited an effective gauge factor ( $GF_e$ ) of approximately 0.64–0.72 (summed EIT resistance), which is lower than the bulk composite ( $GF \approx 4.2$ ) but represents a robust, spatially averaged response less susceptible to local artifacts.
- Fracture Signatures: Discrete ligament fractures manifested as abrupt jumps in global resistance and sharp, localized drops in reconstructed conductivity.
Comparison with Conventional Methods:
- While 4PP and standard piezoresistive measurements captured the onset of failure via global resistance spikes, they provided no spatial information regarding where the failure occurred.
- EIT provided a visual "tomogram" of the damage, distinguishing between isolated ligament breaks and progressive network collapse.

5. Significance and Impact

Intelligent Materials: This work establishes a pathway toward autonomous, self-sensing materials. By embedding sensing capabilities directly into the lattice architecture, these materials can report their own structural health in real time.
Digital Twin Frameworks: The experimentally validated, spatially resolved data provides a critical foundation for Digital Twin applications, allowing for the validation of predictive models and real-time structural health monitoring (SHM) in aerospace, biomedical, and energy systems.
Scalability: The approach is scalable and compatible with standard AM processes, offering a new paradigm for designing multifunctional metamaterials that are not only load-bearing but also intrinsically intelligent.
Future Outlook: The study highlights that optimizing electrode placement, reconstruction algorithms, and lattice geometry can further improve spatial resolution, paving the way for next-generation materials capable of detecting early-stage, pre-fracture damage before catastrophic failure occurs.