Sensitivity increase of 3D printed, self-sensing, carbon fibers structures with conductive filament matrix due to flexural loading

This paper demonstrates that the sensitivity of 3D printed, continuous carbon fiber self-sensing structures can be significantly and irreversibly enhanced through pre-stressing with compressive bending loads, while co-extruding a conductive filament matrix improves their electrical reliability and noise performance.

The Gist

Imagine you have a 3D printer that doesn't just make plastic toys, but can weave strong, invisible threads of carbon fiber into the plastic to make it tough, like a modern-day steel-reinforced concrete. This paper is about turning those strong, 3D-printed beams into their own "nervous system" that can feel when they are being bent or squeezed.

Here is the story of how the researchers made these beams super-sensitive, using simple concepts and analogies.

The Goal: Making a Beam That "Feels"

Usually, if you want to know how much a bridge or a robot arm is bending, you have to glue a separate sensor onto it. The researchers wanted to skip that step. They wanted the carbon fibers inside the beam to act as the sensor themselves.

Carbon fibers are special because when you stretch or squeeze them, their electrical resistance changes (it becomes harder for electricity to flow). This is called being "piezoresistive." However, in their natural, perfect state, these fibers aren't very sensitive to small changes. It's like trying to hear a whisper in a noisy room; the signal is too quiet.

The Secret Trick: "Breaking" the Beam on Purpose

The researchers discovered a counter-intuitive trick to make the beam hear that whisper: they intentionally broke it a little bit.

Think of a bundle of 1,000 tiny guitar strings (the carbon fibers) running inside the plastic.

1. The Setup: When the beam is printed, the plastic cools down faster than the fibers. This creates a "residual stress," kind of like a spring that is already slightly squeezed even before you touch it.
2. The Pre-Stressing: The researchers took the beam and bent it very hard, much harder than it would ever be bent in normal use. This is called "pre-stressing."
3. The Damage: Because of the pre-existing squeeze and the hard bend, some of those tiny internal guitar strings snapped.
4. The Result: Now, imagine you have a bundle of strings where a few are broken. If you bend the beam just a tiny bit, those broken ends rub against each other or lose contact. This causes a massive change in how electricity flows through the bundle.

The Analogy: Imagine a crowded hallway where people are holding hands. If everyone is holding hands tightly, it's hard to break the chain. But if you intentionally let go of a few hands in the middle, a tiny nudge to the crowd will cause a huge ripple effect as the chain breaks apart. The researchers found that by "breaking" the fibers slightly, the beam became incredibly sensitive to even the smallest bends. They achieved sensitivity levels (called "Gauge Factors") over 100, which is much higher than standard sensors.

The Problem: A Noisy Signal

There was a catch. When the fibers broke, the electrical signal became very "noisy." It was like trying to listen to a radio station with static interference. Sometimes the connection would flicker on and off, making the data unreliable. This happened because the plastic (PETG) used to print the beam is an insulator—it doesn't conduct electricity. When a fiber broke, the electricity had nowhere to go, and the signal got lost.

The Solution: The "Safety Net" Filament

To fix the noise, the researchers tried a new printing method. Instead of just printing the carbon fibers, they co-extruded (printed side-by-side) a special conductive filament called "Protopasta" (a plastic mixed with carbon black that conducts electricity).

The Analogy: Think of the carbon fibers as the main highway. When a bridge on the highway collapses (a fiber breaks), traffic stops. The Protopasta acts like a network of side roads and detours. Even if a main fiber breaks, the electricity can still flow through the Protopasta "side roads" to keep the connection alive.

The Result:

• Reliability: The samples printed with Protopasta were much quieter and more reliable. The signal didn't flicker.
• Sensitivity: They kept the high sensitivity created by the broken fibers.
• The Trade-off: The only downside was that the Protopasta clogged the printer nozzle more often, like trying to push thick peanut butter through a straw.

What They Found

1. Compression is Key: The fibers mostly broke when they were being squeezed (compressed), not when they were being pulled (stretched). The sensitivity skyrocketed on the side of the beam that was being squeezed.
2. Permanent Change: Once they bent the beam hard enough to break the fibers, the sensitivity stayed high forever. You couldn't "un-break" the fibers.
3. Noise Reduction: Using the conductive Protopasta filament made the sensor work much better than using regular plastic, proving that having a "safety net" for electricity is crucial for these types of sensors.

In a Nutshell

The researchers took 3D-printed carbon fiber beams, bent them hard enough to snap some of the internal fibers, and found that this damage made the beams incredibly sensitive to touch. To stop the signal from getting noisy, they printed a conductive "safety net" alongside the fibers. The result is a self-sensing structure that is highly sensitive and reliable, created by intentionally introducing a little bit of controlled damage.

Technical Summary

Technical Summary: Sensitivity Increase of 3D Printed, Self-Sensing, Carbon Fiber Structures

Problem Statement
Additive manufacturing (3D printing) has enabled the production of complex parts, but the mechanical strength of standard printed materials often limits their use in engineering-grade applications. While continuous fiber coextrusion (combining pre-impregnated carbon fibers with thermoplastic matrices) addresses strength issues, integrating sensing capabilities remains a challenge. Existing research has demonstrated that continuous carbon fiber bundles exhibit piezoresistive properties, allowing for self-sensing. However, standard carbon fiber strain gauges often possess low sensitivity (gauge factors). Furthermore, previous studies observed that progressive fiber damage under high loads leads to irreversible resistance changes and increased sensitivity, yet this phenomenon has not been systematically utilized to create highly sensitive sensors. Additionally, electrical contact reliability and noise in printed sensors remain significant hurdles.

Methodology
The study utilized an Anisoprint A4 Composer 3D printer to fabricate continuous carbon fiber reinforced beams using a Material Extrusion (MEX) process. The samples consisted of PETG (Polyethylene terephthalate glycol) or Protopasta (conductive PLA) matrices coextruded with 1.5K continuous carbon fibers.

• Sample Design: Beams were printed with carbon fiber reinforced perimeters located in the middle of the cross-section. Post-printing, the samples were sawn to isolate two parallel strain gauges and to expose the fibers for electrical connection via silver paint and screws.
• Experimental Setup: A three-point bending test was conducted using a linear actuator. The samples underwent a specific loading waveform consisting of:
  1. Low-amplitude sine waves (10 N) to measure baseline sensitivity.
  2. Progressive high-amplitude sine waves (up to 70 N) with increasing offsets to induce "pre-stressing."
  3. Waiting periods to allow for viscoelastic relaxation.
• Pre-stressing Protocol: Samples were loaded in an "initial orientation" and then flipped to a "flipped orientation" to test sensitivity in both compression and tension. The goal was to irreversibly alter the gauge factor by inducing controlled fiber failure.
• Theoretical Modeling: Classical beam theory was applied, incorporating residual thermal stress calculations derived from the difference in coefficients of thermal expansion (CTE) between the carbon fibers and the matrix. An electrical model assumed resistance change was proportional to average strain.

Key Contributions

1. Irreversible Sensitivity Enhancement: The paper demonstrates that pre-stressing continuous carbon fiber strain gauges with high compressive bending loads can irreversibly increase their gauge factor (sensitivity) by orders of magnitude.
2. Mechanism Identification: The study attributes the sensitivity increase to local progressive fiber failure (monofilament breakage) caused by the combination of residual thermal stresses from printing and external mechanical loading.
3. Conductive Filament Coextrusion: The research introduces the coextrusion of conductive filament (Protopasta) around the carbon fibers as a method to improve sensor reliability. Micrographs confirmed that the conductive matrix contacts the carbon fiber bundles, providing parallel current paths that bridge broken fiber segments.
4. Comparative Analysis: The study compares PETG and Protopasta matrices, highlighting the superior electrical stability and lower noise of the conductive filament approach.

Results

• Sensitivity Gains: Pre-stressing resulted in gauge factors exceeding 100 (specifically reaching 126 in tension for the shortest samples printed with Protopasta), significantly higher than typical values found in literature for intact fibers.
• Compression vs. Tension: In the initial orientation, sensitivity increased primarily in the strain gauges under compression. In the flipped orientation, sensitivity increased in both compression and tension, though the effect was more pronounced in compression.
• Noise and Reliability: Samples printed with Protopasta exhibited significantly lower electrical noise and more reliable contact compared to those printed with PETG. PETG samples frequently suffered from contact failures or high noise levels, attributed to the inability of the non-conductive matrix to bridge gaps between broken fiber segments.
• Stress Analysis: Calculations indicated that the total compressive stress (loading stress + residual thermal stress) exceeded the compressive yield strength of the carbon fibers (approx. 237 MPa) but remained below their tensile yield strength, supporting the hypothesis that fiber breakage occurred primarily due to compressive failure.
• Microstructural Evidence: Micrographs revealed voids and incomplete resin encapsulation, which allowed the conductive Protopasta to contact the fibers directly. Shearing of fiber bundles was also observed.

Significance and Claims
The paper claims that the combination of pre-stressing and conductive filament coextrusion offers a promising pathway for creating 3D printed structural sensors with high sensitivity. The authors state that this method allows for the creation of sensors that measure their own internal strain, unlike surface-mounted gauges.

However, the authors maintain a modest tone regarding limitations and future applicability:

• Structural Integrity: The method requires inducing fiber fractures to achieve high sensitivity, which reduces the ultimate strength of the structure, potentially making it unsuitable for critical load-bearing applications where maximum strength is required.
• Drift and Noise: While Protopasta reduced noise, viscoelastic drift in the resistance response was still observed, which may affect precision in long-term applications.
• Measurement Limitations: The use of a two-probe setup (rather than four-probe) included contact resistance in the measurements, potentially under-reporting the true gauge factor, though the authors note the measured values were already sufficiently high to validate the concept.
• Modeling Approximations: The theoretical models for residual thermal stress and beam mechanics are acknowledged as simplified approximations due to the complexity of the printing process and material variability.

In conclusion, the work validates that controlled damage (fiber fracture) can be leveraged to create highly sensitive self-sensing structures, provided that the trade-off in structural strength and the need for robust electrical contact management are addressed.