⚡ electrical engineering

Direct-Write Transfer Printing of Low-Loss Copper Interconnects onto Arbitrary Substrates

This paper presents a direct-write transfer printing technique that decouples copper deposition from the target substrate via meniscus-confined electrodeposition, enabling the creation of low-loss, near-bulk conductivity interconnects with smooth surfaces on arbitrary and textured substrates for high-frequency wearable and bioelectronic applications.

Original authors: Yuxin Jiao, Andrew Gleadall, Timothy D. Hall, Majid Minary-Jolandan

Published 2026-08-14

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Yuxin Jiao, Andrew Gleadall, Timothy D. Hall, Majid Minary-Jolandan

Original paper licensed under CC BY 4.0 (https://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

The Invisible Highway and the Sticky Problem

Imagine your smartwatch or the clothes you wear are trying to talk to the world. To do this, they need to send and receive invisible radio waves, like whispers traveling at the speed of light. But here's the catch: when these whispers get really fast (at frequencies used for 5G and beyond), the metal wires inside them start to act weird. Instead of flowing smoothly through the whole wire, the electricity gets squeezed into a tiny, thin layer right on the surface. This is called the "skin effect." If that surface is bumpy or the metal is a bit "spongy" with tiny holes, the signal gets lost, like a whisper getting drowned out by static.

To fix this, engineers usually have two choices, and both have big flaws. One way is to spray metal onto a surface in a super-clean, vacuum chamber. This makes smooth, perfect metal, but it's like trying to paint a fuzzy teddy bear with a spray gun; the metal can't reach into the fur, and the high heat needed might melt the fabric. The other way is to print metal using special "ink" made of tiny metal particles. This works on any surface, even fuzzy fabrics, but the ink leaves behind a rough, bumpy surface full of tiny gaps, which ruins the signal. Scientists have been stuck trying to choose between a smooth surface that can't touch soft materials, or a rough surface that works on soft materials but kills the signal.

The "Sticky Tape" Solution

In this paper, researchers Yuxin Jiao and her team at The University of Texas at Dallas, Loughborough University, and Faraday Technology Inc. found a clever way to have their cake and eat it too. They realized they didn't need to grow the metal on the final object (like the fabric or the plastic). Instead, they grew the metal on a temporary, reusable "staging ground" and then stuck it onto the final object, like applying a high-tech sticker.

Here is how they did it: They used a tiny nozzle to write copper onto a smooth, stainless-steel sheet. They didn't just spray it; they used a trick called "meniscus-confined electrodeposition." Imagine the nozzle holding a tiny, invisible drop of salty water (electrolyte) right at its tip. As the nozzle moves, this drop acts like a tiny, moving battery that pulls copper out of the water and builds it up on the steel sheet. Because the drop is so small and controlled, they can draw very precise lines, even making them loop back on themselves to form complete circles or spirals without any gaps.

Once the copper is built up on the steel, it forms a thin, free-standing foil. The team then takes a piece of sticky tape (or a special fabric) and presses it onto the copper. When they peel the tape away, the copper sticks to the tape and lifts off the steel sheet cleanly, leaving the steel ready to be used again. This means they can make the copper on a perfect, smooth steel sheet and then transfer it onto a bumpy woven shirt, a stretchy plastic band, or a piece of paper, without ever having to heat them up or put them in a vacuum.

What They Found

The results were impressive. The copper they transferred was incredibly smooth on the side that touched the final object, with a roughness of only 40–55 nanometers (that's about 1,000 times thinner than a human hair). It was also very dense, with a resistivity (a measure of how hard it is for electricity to flow) of 2.3–2.6 µΩ·cm, which is very close to the best solid copper you can buy.

When they tested how well these copper lines carried high-frequency signals (from 5 to 30 GHz), they found that the signal loss was 1.4 to 7.1 times lower than what you get with standard printed copper inks. In fact, their copper was so good that its performance was within just 1.2 times that of solid bulk copper. They also showed that these thin copper films were strong enough to be handled like free-standing foils without breaking, even though they were only about 2 to 4 micrometers thick (roughly the width of a few bacteria).

The team also discovered that the way they moved the nozzle mattered a lot. If they just used standard computer instructions meant for 3D printers (which assume you are squeezing out thick plastic), the copper would get stuck or break at the seams where the line loops back. By using a custom, step-by-step control system, they could tell the nozzle exactly when to pause and how fast to move, allowing them to draw complex shapes like closed loops and filled squares without any defects.

Why It Matters

This work suggests a new path for making electronics that can be worn or bent. By separating the "growing" of the metal from the "wearing" of the metal, they solved the problem of making smooth, high-speed wires on rough, soft, or porous materials like textiles. The paper shows that this method works on many different surfaces, from plastic films to woven fabrics, and that the copper lines can be as thin as 550 µm. While the researchers note that more work is needed to make these circuits into full, working devices, they have proven that this "transfer printing" technique is a viable way to create low-loss, high-frequency copper connections on almost any surface, opening the door for smarter, more flexible electronics in the future.

Technical Summary: Direct-Write Transfer Printing of Low-Loss Copper Interconnects

Problem Statement
Conformal and wearable electronic systems are increasingly requiring metal interconnects capable of operating at gigahertz-to-millimeter-wave frequencies for wireless communication and power transfer. At these frequencies, the "skin effect" confines current to a nanometer-scale surface layer (approximately 380–470 nm in copper at 20–30 GHz). Consequently, conductor loss is governed by two critical properties: the electrical resistivity and the surface roughness of the metal. Existing metallization routes fail to simultaneously optimize both properties on arbitrary or textured substrates:

Vapor-phase deposition (e.g., sputtering, evaporation) produces smooth, near-bulk resistivity films but requires vacuum, elevated temperatures, and line-of-sight access, making it incompatible with porous, fibrous, or textured substrates like textiles.
Particle-based inks (e.g., inkjet, screen printing) can pattern on arbitrary substrates under ambient conditions but result in porous, rough films with high resistivity. While post-processing can improve one property, it often degrades the other (e.g., smoothing increases resistivity, or densification leaves roughness exceeding the skin depth).

Methodology
The authors introduce a direct-write transfer printing process that decouples metal growth from the target substrate. The methodology consists of three primary stages:

Meniscus-Confined Electrodeposition (MCED): Copper is deposited via MCED onto a reusable, polished stainless-steel cathode. A nozzle containing an electrolyte and an anode moves across the steel surface, confining the redox reaction to the wetted footprint beneath the nozzle tip. This allows for direct writing of copper into arbitrary planar geometries without masks or vacuum.
Deterministic Toolpath Control: To address challenges in creating closed paths and avoiding seam defects, the authors utilize the FullControl library. Unlike conventional slicers that assume volumetric extrusion, this approach treats manufacturing as an explicitly ordered sequence of kinematic and process-state changes. It allows for deterministic control over nozzle position, velocity, and dwell time. This enables the design of paths that account for electrochemical factors, such as preferential deposition on existing copper and mass-transport limitations at seam intersections, preventing film stress accumulation and delamination during growth.
Transfer Printing: Once the copper film reaches a sufficient thickness (typically >2 µm), a transfer tape (or a polymer solution for specific substrates) is applied to the film. Uniform pressure is applied, and the substrate is peeled away. The process exploits the differential adhesion between the copper-tape and copper-steel interfaces to lift the film intact from the cathode and place it onto the receiver substrate.

Key Results
The study demonstrates the successful transfer of copper patterns onto a diverse range of substrates, including polymer films (PVDF-TrFE, Kapton), adhesive tapes, woven fabrics, and nonwoven electrospun mats.

Geometric Versatility: The process successfully transferred complex geometries, including serpentine traces, spiral coils, closed circular loops, and filled bonding pads. Feature linewidths were preserved from 1.5 mm down to 550 µm.
Mechanical Integrity: Free-standing copper films (1–4 µm thick) demonstrated sufficient mechanical robustness for handling and mounting. Tensile testing showed fracture stresses of approximately 224 ± 77.3 MPa at strains below 0.52%, confirming the films can survive the transfer process without fracturing.
Surface Morphology and Resistivity: The transfer process results in a "substrate-templated" working surface. The side of the copper that was in contact with the polished steel during deposition becomes the functional surface after transfer.
- Roughness: This functional surface exhibits a smooth, spatially uniform roughness ( $S_q$ ) of 40–55 nm, which is well below the skin depth at 20–30 GHz.
- Resistivity: The films exhibit near-bulk resistivity of 2.3–2.6 µΩ·cm (approximately 1.3–1.5 times the bulk value of 1.72 µΩ·cm) for films thicker than ~2 µm.
High-Frequency Performance: Across the 5–30 GHz band, the transferred films achieved a surface resistance within 1.2 times that of bulk copper. Compared to printed copper interconnects, the conductor loss was 1.4 to 7.1 times lower. An Ashby plot analysis positions this method in the optimal region of property space, achieving both low resistivity and low roughness in a single deposition step followed by transfer, whereas other methods typically require multi-step post-processing to achieve only one of these properties.

Significance
The paper claims that this approach reframes metallization as a two-step problem—deposition and placement—rather than a single process constrained by the target surface's properties. By separating metal growth from the receiving surface, the method provides a mask-free, vacuum-free, and thermal-budget-free route to high-frequency copper interconnects on arbitrary and textured substrates.

The authors state that this work opens a pathway toward high-frequency smart textiles and conformal bioelectronics, enabling the integration of gigahertz-compatible interconnects into materials (such as woven and nonwoven textiles) that are incompatible with traditional vapor-phase deposition and difficult to wet uniformly with particle-based inks. The process maintains the geometric freedom of additive manufacturing while delivering electrical performance approaching that of dense, electrodeposited foil.

The Invisible Highway and the Sticky Problem

The "Sticky Tape" Solution

What They Found

Why It Matters

Technical Summary: Direct-Write Transfer Printing of Low-Loss Copper Interconnects

More like this