3R Electronics: Scalable Fabrication of Resilient, Repairable, and Recyclable Soft‐Matter Electronics

Abstract

E‐waste is rapidly turning into another man‐made disaster. We propose that a paradigm shift toward a more sustainable future can be made through soft‐matter electronics that are resilient, repairable if damaged, and recyclable (3R), provided that they achieve the same level of maturity as industrial electronics. This includes high‐resolution patterning, multi‐layer implementation, microchip integration, and automated fabrication. Herein, a novel architecture of materials and methods for microchip‐integrated condensed soft‐matter 3R electronics is demonstrated. The 3R function is enabled by a biphasic liquid metal‐based composite, a block copolymer with non‐permanent physical crosslinks, and an electrochemical technique for material recycling. In addition, an autonomous laser‐patterning method for scalable circuit patterning with an exceptional resolution of <30 μm in seconds is developed. The phase‐shifting property of the BCPs is utilized for vapor‐assisted “soldering” circuit repairing and recycling. The process is performed entirely at room temperature, thereby opening the door for a wide range of heat‐sensitive and biodegradable polymers for the next generation of green electronics. We demonstrate the implementation and recycling of sophisticated skin‐mounted patches with embedded sensors, electrodes, antennas, and microchips that build a digital fingerprint of the human electrophysiological signals by collecting mechanical, electrical, optical, and thermal data from the epidermis. This article is protected by copyright. All rights reserved

Full text

www.advmat.de
2203266 (1 of 13) ©  The Authors. Advanced Materials published by Wiley-VCH GmbH
3R Electronics: Scalable Fabrication of Resilient, Repairable,
and Recyclable Soft-Matter Electronics
Mahmoud Tavakoli,* Pedro Alhais Lopes, Abdollah Hajalilou, André F. Silva,
Manuel Reis Carneiro, José Carvalheiro, João Marques Pereira, and Aníbal T. de Almeida
M. Tavakoli, P. Alhais Lopes, A. Hajalilou, A. F. Silva, J. Carvalheiro,
J. Marques Pereira, A. T. de Almeida
Soft and Printed Microelectronics Lab
Institute of Systems and Robotics
University of Coimbra
Coimbra -, Portugal
E-mail: mahmoud@isr.uc.pt
M. Reis Carneiro
Soft Machines Lab
Mechanical Engineering
Carnegie Melon University
Pittsburgh, PA , USA
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/./adma..
DOI: 10.1002/adma.202203266
1. Introduction
Today, e-waste production has reached an
alarming level of 7 kg person−1 year−1.[1]
Approximately 20% of the e-waste is sent
for recycling, from which we recover only
a small percentage of precious metals,
particularly gold.[2] The rapid advances
in smart packaging, printed electronics,
sensing stickers, wearable health-moni-
toring patches, and smart e-textiles could
result in the production of billions of
disposable electronics in the next years.
There are already commercial examples of
disposable electronics patches for patient
monitoring, continuous glucose tracking,
sensorized diapers, and underwear for
incontinence management, and electro-
chemical sensor kits for sweat, saliva, and
blood analysis. As technological barriers to
implementing disposable soft-matter elec-
tronics are being rapidly eliminated, large-
scale use of disposable patches is foreseen
in hospitals, elder care units, and profes-
sional athletes.
Despite the social and economic advan-
tages of these systems, this race for a
“smarter” future comes at the cost of expo-
nentially increasing e-waste production. The disposal of bil-
lions of systems that contain highly processed microchips and
scarce metals will lead us to another man-made disaster. Com-
plex biomonitoring systems contain many analog and digital
components that can be reutilized. Therefore, the possibility of
recycling these components, as well as the metals from printed
circuits, should be considered. As we are shifting toward short-
term use of electronics, we need to urgently rethink the mate-
rial architecture to allow recyclability before any fabrication pro-
cedure becomes dominant. The traditional 3Rs politics (reuse,
reduce, and recycle) is not pertinent in the emerging wearable
monitoring applications. Reutilization is not permitted due to
hygiene reasons, and the trend of use is only at the beginning
of an exponential growth.
In the last few years, this problem has attracted the atten-
tion of scientists. Recently, there was a report on a dispos-
able and biodegradable supercapacitor,[3] and there have
been a few studies on the recovery of Ag from used batteries
through electrowinning.[4–8] In the field of transient electronics,
some works focused on biodegradable circuits over starch[9]
E-waste is rapidly turning into another man-made disaster. It is proposed
that a paradigm shift toward a more sustainable future can be made through
soft-matter electronics that are resilient, repairable if damaged, and recy-
clable (3R), provided that they achieve the same level of maturity as industrial
electronics. This includes high-resolution patterning, multilayer implemen-
tation, microchip integration, and automated fabrication. Herein, a novel
architecture of materials and methods for microchip-integrated condensed
soft-matter 3R electronics is demonstrated. The 3R function is enabled by
a biphasic liquid metal-based composite, a block copolymer with nonper-
manent physical crosslinks, and an electrochemical technique for material
recycling. In addition, an autonomous laser-patterning method for scalable
circuit patterning with an exceptional resolution of <30µm in seconds is
developed. The phase-shifting property of the BCPs is utilized for vapor-
assisted “soldering” circuit repairing and recycling. The process is performed
entirely at room temperature, thereby opening the door for a wide range of
heat-sensitive and biodegradable polymers for the next generation of green
electronics. The implementation and recycling of sophisticated skin-mounted
patches with embedded sensors, electrodes, antennas, and microchips that
build a digital fingerprint of the human electrophysiological signals is demon-
strated by collecting mechanical, electrical, optical, and thermal data from the
epidermis.
ReseaRch aRticle
©  The Authors. Advanced Materials published by Wiley-VCH
GmbH. This is an open access article under the terms of the Creative
Commons Attribution-NonCommercial-NoDerivs License, which
permits use and distribution in any medium, provided the original work
is properly cited, the use is non-commercial and no modifications or
adaptations are made.
Adv. Mater. 2022, 34, 

www.advmat.dewww.advancedsciencenews.com
2203266 (2 of 13) ©  The Authors. Advanced Materials published by Wiley-VCH GmbH
and poly(lactic-co-glycolic acid) (PLGA).[10] Further, soluble
poly(vinyl alcohol) (PVA) and polyimine with dynamic covalent
bonds were used as the packaging materials, which allowed the
recovery of liquid metal (LM) from the circuit.[11,12] A review of
biodegradable transient electronics can be found in.[13] While
promising for some applications, transient and biodegradable
electronics are generally composed of nonresilient materials
that function for a short time and in specific environments. For
instance, PVA falls short in applications that demand resilience
to humidity. In general, the degradation profiles of the designed
materials should accommodate the required device lifetimes.[14]
In two recent studies, LM-embedded elastomers with
crosslinked polyurethane (PU)[15] and SIS[16] were used for fab-
ricating stretchable interconnects, from which LM could be
recovered. However, these composites require a mechanical
sintering step to form conductive traces, which hinders the
autonomous fabrication of high-resolution printed circuits. In
addition, a Ga−In−SiO2 paste was introduced,[17] from which
LM could be recovered.
These advances are promising for the move toward recyclable
electronics. However, the previously presented methods lack
the required maturity, including the desired patterning resolu-
tion and the automation level necessary for industrial applica-
tions. Therefore, they fall short compared to the current high
throughput printed circuit board (PCB) manufacturing, in terms
of high-resolution patterning, multilayer implementation, and
scalable and automated fabrication. The reasons for this include
the shortcomings of the existing LM composites, e.g., the
need for mechanical sintering, the smearing behavior, lack of
mechanical integrity, and lack of adhesion to the substrate that
hinders ecient printing of the circuit. Recently, we addressed
some of these problems by demonstrating liquid metal-based
biphasic composites that are sinter-free and nonsmearing.[18–20]
However, the resolution of the circuit printing was limited
because of the mechanical limitations of the extrusion-based
printing. Therefore, it was impossible to develop circuits with
the same resolution as current rigid PCBs that use state-of-
the-art integrated circuits such as surface-mount device (SMD)
microchips. Real progress toward 3R electronics only happens
if we can demonstrate novel fabrication techniques that in one
hand rely on resilient, repairable, and recyclable materials,
and on the other hand can compete with existing techniques
in terms of patterning resolution, multilayer implementation,
microchip integration, and autonomous fabrication.
Here, we present a novel architecture for scalable, autono-
mous, and high-resolution fabrication of 3R electronics
through direct laser patterning of “3R materials,” including
biphasic printable liquid metal composites and “revers-
ible” tough block copolymers (BCPs), which benefit from a
high elasticity and nonpermanent physical crosslinking. The
resulting conductive traces are ultraresilient to mechanical
strain and can withstand >1700% of strain before an electrical
break. We demonstrate the integration of miniature silicon
chips for data acquisition, processing, and wireless communi-
cation through room-temperature chemical soldering by taking
advantage of reversible polymer bonds in block copolymers.
We take advantage of the same reversible bonds for repairing
the damaged circuits and exploit these nonpermanent physical
crosslinks for the rapid and ecient recovery of silicon chips
and other materials, including textile, polymer films, eutectic
gallium indium (EGaIn) liquid metal, silver, and ferrite. We
show how these materials can be employed for fabricating com-
plex electronic devices. We then compare two dierent biphasic
composites based on Ag and Fe and demonstrate techniques
for recycling the metal elements from the biphasic amalgams
for both cases. This includes using magnetic force to sepa-
rate metals in EGaIn–Fe–SIS composite and a combination of
chemical and electrochemical processes, such as leaching and
electrowinning, for recovering individual metals from EGaIn–
Ag–SIS composite. We compare these two composites in terms
of smearing behavior, microstructure integrity, and ease of
recycling.
We also demonstrate a high-resolution laser patterning tech-
nique that significantly improves the circuit patterning speed
by 3–8× (depending on circuit size) and the resolution by ≈10×,
compared to our previous work on digital printing.[19] Usually, a
circuit resolution better than 200µm is desired for modern sil-
icon chips. The low-cost laser processing technique presented
herein allows patterning traces with 30µm width, compared to
300µm in digital printing. Adding to this the single-step vapor-
assisted microchip integration, this work paves an important
path toward scalable fabrication of 3R electronics.
We demonstrate various examples of condensed soft-matter
circuits that embed the necessary state-of-the-art SMD micro-
chips and show examples of skin-mounted patches and e-tex-
tiles that monitor mechanical, electrical, optical, and thermal
changes in the epidermis. In one example, we show a large and
complex multisensor e-skin with several laser-patterned strain
gauges that integrates all the required microchips for acquisi-
tion, amplification, processing, and wireless communication.
This patch continuously obtains a digital fingerprint of the skin
strain profile at multiple positions and direction and utilizes
this for classification of human activities, including breathing,
eating, and various sports activities on the neck and torso. All
these circuits comply with the 3Rs electronics concept. They
are resilient to mechanical strain, repairable if damaged, and
recyclable after disposal. We successfully recovered and reinte-
grated these microchips 10 times.
The entire fabrication process, including the deposition, pat-
terning, and microchip “soldering,” is performed at room tem-
perature. Elimination of temperature from the sintering pro-
cess (as common in printed electronics) and from the soldering
process is an important step toward green electronics as it
enables the fabrication of electronics over a wide range of heat-
sensitive substrates that were not previously possible. At the
same time, this reduces energy consumption. A recent review
of emerging biodegradable elastomers and gels for elastic elec-
tronics can be found in ref. [21].
In summary, to achieve the objectives of 3R electronics, we
tackled three main challenges. First, we introduced a novel
architecture for soft-matter materials such as conductive com-
posites and substrates that satisfy the 3R objectives, do not
require thermal sintering, and embed “reversible” functions
that can be exploited for healing, soldering, and recycling.
Second, we developed autonomous fabrication techniques,
including high-resolution digital patterning and single-step
microchip “soldering,” and third, we developed supporting
technologies for recycling the materials and components. These
Adv. Mater. 2022, 34, 

www.advmat.dewww.advancedsciencenews.com
2203266 (3 of 13) ©  The Authors. Advanced Materials published by Wiley-VCH GmbH
challenges are interrelated and must be addressed simultane-
ously to guarantee a real step toward sustainable 3R electronics.
By addressing these challenges, this work lays the foundation
for the next generation of recyclable soft-matter electronics that
can compete with current PCB manufacturing techniques, thus
paving an important step toward green fabrication and sustain-
able and responsible application.
2. Results
Figure1 shows the concept and building blocks of 3R electronics.
3R electronics relies on a series of 3R materials (Figure1A) that
can be processed to fabricate complex circuits through autono-
mous and room-temperature fabrication techniques (Figure1B
and Video S1, Supporting Information) and applied to a wide
range of applications like wearable monitoring (Figure1C). After
disposal, these devices can be recycled to obtain the costly ele-
ments and microchips (Figure1D). This is especially important
as many of these patches are intended to be used only for a few
hours. We should rely on a material architecture compatible
with the 3R objectives to enable these objectives. Referring to
Figure1A, as a 3R conductor, we use a biphasic LM composite
composed of a block copolymer, EGaIn LM, and conductive
microfillers. This ternary composite is conductive, stretchable,
and sinter-free. The use of EGaIn LM enables a combination
of high conductivity, self-healing property, and high resilience,
which is inherent in its liquid phase. The main objective of the
microfiller is to percolate between the LM droplets, which grants
the composite conductivity right after deposition. Previous
eorts at making a liquid metal embedded elastomer resulted in
printable composites but required a mechanical sintering step
to gain conductivity.[22] Choosing from a wide range of available
conductive fillers allows fine-tuning the viscosity, printability,
and electromechanical properties of the composite. The choice
of filler has a direct and substantial influence on the properties
of the ink, including its recyclability, smearing behavior, homo-
geneity, and ability to retain the liquid metal when subject to
mechanical strain and pressure. We recently demonstrated a ver-
sion of this ink that uses Ag flakes as microparticles.[19,20] This
Adv. Mater. 2022, 34, 
Figure 1. R Electronics. A) Constitutional materials, including biphasic liquid metal-based inks and reversible polymers. B) Autonomous fabrication
techniques including printing/laser patterning i), Pick and place ii), and vapor-assisted soldering and healing iii). Thanks to the reversible polymers,
the microchips penetrate the ink and the substrate, resulting in a seamless integration into the circuit v). C) Example of a laser patterned soft-matter
circuit with eight strain gauges and state of the art SMD microchips for processing, amplification, and communication. D) The concept for recovery of
the elements from Rs electronics, based on dissolving reversible polymeric matrix. E) Utilization of reversible polymers with nonpermanent physical
crosslinks, as the substrate, e-textile adhesive, and binder in ink, contributes to self-soldering, Self-healing, repairing and recycling.

www.advmat.dewww.advancedsciencenews.com
2203266 (4 of 13) ©  The Authors. Advanced Materials published by Wiley-VCH GmbH
can be extended to other conductive particles to obtain dierent
electromechanical properties. This includes ferromagnetic parti-
cles, further discussed in this work.
We use SIS (styrene–isoprene–rubber) BCP as the elasto-
meric binder in the composite and as well for the substrate.
SIS is a thermoplastic elastomer and a triblock copolymer,
including both permanent chemical crosslinks and physical
crosslinks. Longer chains of chemically crosslinked isoprene
sequences include polystyrene blocks at the end of their chains
(Figure1E). These polystyrene blocks collect in small domains
through physical crosslinks. Therefore, SIS BCP benefits from
a reversible phase-shifting property and rapid solubility, which
are exploited in this work for several processes. Unlike many
polymers that only soften in contact with solvents, SIS can be
either softened or fully dissolved in the solvent, depending on
the exposure time and intensity.
This property is exploited in this work for (Figure 1E):
healing of the microcracks that are formed on the ink/substrate
after deposition due to solvent evaporation; repairing of circuits
that suered severe damage or a thorough cut; “soldering” of
SMD components through solvent vapor exposure that causes a
reversible transition from polymer to gel state in the SIS BCP;
and recycling the circuit microchips, metals, and polymer/
textile substrate by decomposition of the circuit to its consti-
tutional elements through dissolving the circuit in the solvent
(C6H5CH3). The use of polymers with nonpermanent bonds
(e.g., physical crosslinks) is the key to achieving the objectives
of 3R. Moreover, SIS BCP provides excellent adhesive proper-
ties, resilience, elasticity, and toughness.
Referring to Figure 1B, automated room temperature fab-
rication of hybrid stretchable circuits starts with the deposi-
tion and patterning of ink through either direct ink writing
(DIW) or laser patterning (Figures 1B-i,iv, and Video S1, Sup-
porting Information). After patterning the circuit, the micro-
chip components are placed over it using a pick and place
machine (Figure1B-ii), and then it is exposed to toluene vapor
(Figure1B-iii). Vapor is generated using a home-made system
composed of a spray nozzle that creates microdroplets of the
solvent at room temperature through intensified sonic energy
using compressed air as the energy source. The resulting atom-
ized droplets of toluene cause a phase transition in the polymer,
from solid to semisolid gel, which happens in both the SIS
substrate and the SIS-containing conductive ink (Figure1B-v).
In this state, microchips penetrate the softened ink and the
substrate (Figure 1B-v). Once the vapor stimuli are removed,
microchips are mechanically locked into the substrate from
5 facets, also assisted by the excellent adhesion properties of the
SIS BCP. Compared to the existing techniques, the soldering
process is convenient, as it does not require selective deposi-
tion of solder paste or conductive adhesives. In fact, the printed
ink and substrate act as the soldering matter, thus eliminating
the need for deposition and sintering of the solder paste. The
entire process is performed at the room temperature, including
the creation of the solvent vapor and the “soldering” process.
The room temperature process is important for the move
toward recyclable electronics. Today, the choice of the substrate
for the fabrication of rigid and flexible PCBs is limited to a few
heat-resistive substrates. For instance, most of today’s flexible
circuits are created over polyimide, which is not stretchable,
soluble, or recyclable. The elimination of the temperature from
the entire fabrication process opens the door to a wider range
of previously impossible substrates due to their limited heat
resistance, including those with reversible functions. Moreover,
a room-temperature process reduces energy consumption.
Figure 1C demonstrates an example of a 3R circuit cre-
ated with this technique that embeds 8 laser-patterned strain
gauges, miniaturized SMD electronic components for signal
acquisition, amplification, and processing, and bluetooth low
energy (BLE) for communication. Referring to Figure1D, both
patch and e-textile circuits can be decomposed to their “ingredi-
ents” thanks to the existing nonpermanent physical crosslinks
in the BCPs (Figure1E).
2.1. Ink Synthesis and Circuit Patterning
As seen in Figure2A, the EGaIn–Ag–SIS composite is pre-
pared by dissolving SIS block copolymers (Sigma-Aldrich) in a
toluene solution, followed by mixing ≈5µm Ag flakes (Silflake
Technic Inc.) and then EGaIn.
The alternative EGaIn–Fe–SIS composite is prepared simi-
larly but replaces the Ag flakes with ferromagnetic particles
(Fe3O4 magnetite 52 µ, 99%, CAS: 1309-38-2). Circuits can be
patterned through direct ink writing (Figure2B,C) or laser pat-
terning (Figure2D–F).
Direct ink writing brings several advantages, including
single-step fabrication, the possibility of creating thick and
highly conductive traces, and eliminating the need for a stencil.
We have successfully printed the ink over textile and various
polymers, including medical-grade wound dressing adhesive
(Tegaderm-Figure 2B-ii). Figure 2C demonstrates an example
of a pressure mapping e-textile that is fabricated through
printing a double-layer mesh of parallel plate capacitive sen-
sors. Figure2D shows an example of a printed circuit applied
over polymer film and transferred to a glove for mapping the
pressure applied to the fingertips and bending of the joints. An
example of surface texture classification is also presented.
One disadvantage of this printing technique is that the cir-
cuit resolution is limited by the size of the nozzle (≈300 µm
linewidth and spacing). While this is sucient for many applica-
tions, increasingly miniaturized SMD microchips require a res-
olution as low as 100µm. To overcome this limitation, we devel-
oped a laser patterning technique that significantly improves
the resolution by ≈10×. We demonstrate high-resolution circuits
with a line width and spacing of 25µm and 52µm, respectively.
Figure S1 (Supporting Information) shows images of the pat-
terned traces and optical profilometry analysis. Laser patterning
is performed using a master oscillator power amplifier (MOPA)
fiber laser with a 1064 nm infrared wavelength. This laser is
very accessible and is widely used by local shops for engraving
metals. First, a PVA sacrificial layer is applied over the glass, fol-
lowed by applying a film of the conductive ink using a thin-film
applicator, which is then patterned using the MOPA laser. Next,
a layer of SIS BCP is applied over the circuit, and the PVA layer
is dissolved in water to peel the circuit. In this way, multiple cir-
cuits can be patterned in a few minutes (Figure2E).
Both printing and laser patterning techniques are shown
in Video S1 (Supporting Information). Optical profilometer
Adv. Mater. 2022, 34, 

www.advmat.dewww.advancedsciencenews.com
2203266 (5 of 13) ©  The Authors. Advanced Materials published by Wiley-VCH GmbH
images of Figure S1 (Supporting Information) show that the
single-layer film thickness is ≈15 µm. The average conduc-
tivity for 5 lines of 100 µm width for a single layer coat was
2.1 × 105 S m–1. Figure S2 (Supporting Information) shows
the detailed steps of the laser patterning process. Compared
to digital printing, laser patterning provides better resolu-
tion and faster fabrication. For instance, the circuit shown in
Figure2F-iii takes ≈320 s via digital printing versus 44s in laser
patterning. In addition, our attempts to fabricate this circuit via
digital printing failed due to the complexity and the resolution.
In the case of hybrid circuits with integrated microchips, these
components are integrated into the circuit using the toluene
vapor exposure technique. As demonstrated in Figure S3 (Sup-
porting Information), the laser processing and vapor-assisted
soldering techniques allow the simultaneous fabrication of mul-
tiple circuits. Figure2E-ii and Figure S4 (Supporting Informa-
tion) show a step-by-step example of fabricating a soft-matter
circuit. Figure 2E-iii and Video S4 (Supporting Information)
show the resulting circuit: a wireless Mechanosensing patch that
integrates eight laser patterned strain gauges and several min-
iaturized SMD components, including BLE for wireless com-
munication, a processor, amplifiers, resistors, and several light
emitting diodes (LEDs) for displaying information. Figure2E-iv
and Figure S3 (Supporting Information) show a wireless tem-
perature monitoring patch. Figure2E-v, Figure S5 (Supporting
Information), and Videos S2 and S3 (Supporting Information)
demonstrate a complex circuit for pulse oximetry and moni-
toring blood oxygen levels. Here, components are installed on
both sides of the patch, allowing LEDs and photo detectors to
interface with the epidermis. Figure2E-iii–vi shows four exam-
ples of skin-interfacing soft-matter patches for mechanical,
thermal, optical, and electrical (biopotentials) sensing. Overall,
Adv. Mater. 2022, 34, 
Figure 2. A) Synthesis of sinter-free stretchable biphasic ink. B) Digital Printing of e-patch and e-textile circuits C) Example of a pressure mapping
e-textile fabricated through digital printing. D) Example of a wearable glove with printed bending and pressure sensors, its application in pressure
sensing, and surface texture classification. E) High-resolution Laser Patterning of the circuits using an accessible IR laser and Single-step vapor-assisted
soldering of miniaturized components: i) Example of the fabrication steps. ii) No solder material or thermal treatment is necessary. Examples of chip-
integrated soft-matter epidermal biostickers for mechano-sensing iii), thermal iv), optical v), and electrical sensing vi). These patches can be used for
monitoring respiration and physical activities iii), body temperature iv), heart rate, and SPO v). The belt presented in E-iv measures ECG, heart rate,
respiration, temperature, and body motions simultaneously.

www.advmat.dewww.advancedsciencenews.com
2203266 (6 of 13) ©  The Authors. Advanced Materials published by Wiley-VCH GmbH
the materials and fabrication techniques presented here pave an
important step toward scalable fabrication of high-resolution,
chip-integrated soft-matter 3R electronics.
In order to create multilayer circuits, we used two dierent
methods. The first technique consist of mounting the com-
ponents on one side of the circuit through vapor-assisted sol-
dering, peeling the circuit after addition of the encapsulating
SIS layer, and mounting the component on the second layer of
the ink through vapor-assisted soldering. Please see SectionS4
of the Supporting Information, and also Figure S21 (Sup-
porting Information) for additional details. This technique is
simple, and fast, and permits epidermal electronics with sen-
sors mounted on skin side, while processing and communica-
tion components mounted on the opposite side of the circuit.
This allows reducing the overall footprint of the sensor. This
is the case for the circuit shown in Figure2E-v, with LEDs and
photo detectors on the skin-interfacing layer, and Bluetooth and
processing chips on the opposite side. The disadvantage of this
technique is that it cannot be extended for additional layers.
The second method permits fabrication of multilayer cir-
cuits through the stacking technique. First single-layer PCBs
are produced via laser patterning, with the connection points
exposed. These layers are then joined together through vapor-
assisted circuit fusion. See details of this process in Figure S22
(Supporting Information), and Section S4 of the Supporting
Information. Figure S23 (Supporting Information) shows an
example of a double layer coil antenna, with integrated LED.
This coil antenna is used for harvesting energy from mobile
phone’s near field communication (NFC) coil, and lighting the
LED without the need for any battery. Inclusion of a second
layer of the coil improves the energy harvesting eciency, com-
pared to a single-layer coil. Video S11 (Supporting Information)
shows the functionality of this double layer circuit.
2.2. Electromechanical Characterization, and Repairing of the
damaged circuits
Printed conductive traces produced by the referred materials
and methods are very resilient and can withstand a record-
breaking maximum strain tolerance of 1700% (Figure3A-i). A
microchip-integrated circuit can withstand strains of >500%
Adv. Mater. 2022, 34, 
Figure 3. A) Printed trace without chip i) and with microchips ii) under strain. B) Electromechanical coupling of the circuits. Electrical resistance versus
strain (i), repetitive cycle test for % strain ii), resistance v/s strain for microchip integrated circuit iii). Cracks are formed usually at >% strain
on the substrate and grow until mechanical failure. C) Vapor-assisted repairing of the circuit which was cut using a sharp knife, and the same circuit
under % strain after healing.

www.advmat.dewww.advancedsciencenews.com
2203266 (7 of 13) ©  The Authors. Advanced Materials published by Wiley-VCH GmbH
(Figure 3A-ii,C-iii), a value >6× compared to previous studies
that demonstrated microchip integrated circuits.[23–25] As seen
in the insets of Figure 3A-ii, the mechanical failure does not
necessarily happen on the chip interface. A premature mechan-
ical failure sometimes occurs at the grippers of the charac-
terization device. Figure 3B-I shows the electromechanical
coupling of 3 printed circuits (no microchip) when subject to
over 1500% mechanical strain. In addition to the extreme resil-
ience, the circuits present a very modest gauge factor and a
quasi-linear behavior until ≈1000% strain. Figure 3B-ii shows
printed traces under 100 repetitive cycles with 30% applied
strain. In the first 40 cycles, the R0 value (resistance at 0%
strain) increases slightly. However, this increase is only from
initial value of ≈2 Ω to a final value of ≈3 Ω and then stabilizes.
Figure3B-iii shows resistance versus strain of a microchip inte-
grated strain and an image of the sample under ≈900% strain,
right before and right after the electrical failure. As expected,
the failure happens at the rigid-soft interface at the edge of the
microchip. The crack initiates at ≈700% strain, which is visible
as an increase in the electrical resistance in Figure 3B-III. It
is noteworthy that although this crack forms and grows under
strain, the ink still maintains its connection to the LED from
two edges of the crack.
Unlike the existing techniques wherein the SMD microchips
are soldered or bonded to the conductive trace only from the
bottom side, in this technique, microchips are surrounded on
all 5 facets and adhere strongly to the substrate and ink due
to the excellent adhesive properties of the SIS. Therefore,
the stress concentration zone, i.e., the interface between the
chip and the substrate, has increased substantially. Thus, this
results in a significant (>6×) improvement over the state-of-
the-art in the maximum strain tolerance of the chip-integrated
circuits.[26,27] Note that the maximum strain tolerance for
chip-integrated circuits largely varies depending on the size,
number, and distribution of these microchips. In most previous
works, characterization is performed with only a single chip.
In addition, 3R circuits are repairable in case of damage.
Referring to Figure3B, after making a thorough cut using a
sharp blade, the circuit was repaired by the same solvent vapor
exposure technique used for microchip interfacing. The repair
is so ecient that the circuit could be subject to large strains.
Figure3C shows an example of a cut circuit that remains func-
tional after the vapor exposure, even when subject to >600%
strain. Exposing the circuit to the solvent vapor results in the
expansion of the polymer volume and the weakening of the
physical crosslinks between the polymer chains. This allows
reconnection and rearrangement of the polymer chains, thus
self-repairing the cut zone.
2.3. Recycling
Referring to Figure4, both e-textile and hybrid patches can be
recycled to their constitutional elements in a few simple steps.
Figure4A and Video S5 (Supporting Information) show the pro-
cess for recovering the components from a hybrid microchip
integrated patch. The circuit is first placed over a filter (mesh
count 100T), then over a paper soaked with toluene, and then
placed in a closed chamber. This generates an intense toluene
vapor environment that dissolves the SIS substrate absorbed by
the underlying paper. In this way, the ink and the components
are separated from the substrate. The vapor-assisted substrate
removal is a simple technique that allows facile recovery of
the microchip components that are usually the costliest ele-
ments in the patch. Afterward, the concentrated ink, which is
a metal-rich chunk with some remaining SIS polymer can be
Adv. Mater. 2022, 34, 
Figure 4. Recycling the e-patch and e-textile circuits. A) e.patch are first decomposed by a vapor-assisted substrate removal, allowing separation of the
microchips and ink as a metal chunk. B) Example of wireless thermal monitoring patch on the skin, after polymer removal. An alternative method for
removing components and the ink is direct dissolving. C) Microchips, battery, and metal chunk removed from the thermal patch. D) e-textile recycling
starts with removing the acetone soluble bond between the e-patch and the textile. After the recovery of the textile, the e-patch is recycled using the
same technique as (A).

www.advmat.dewww.advancedsciencenews.com
2203266 (8 of 13) ©  The Authors. Advanced Materials published by Wiley-VCH GmbH
collected. Alternatively, after removing the substrate, the filter
that contains the ink and microchips can be placed in toluene
for simultaneous recovery of all components. This latter tech-
nique is depicted in Figure4B, which shows the decomposition
of the laser patterned wireless temperature monitoring patch in
Figure2E. Figure 4c depicts the recovered microchips and the
metal-rich chunk.
Figure 4D and Video S6 (Supporting Information) demon-
strate the decomposition steps for the e-textile. A 3R e-textile is
composed of a patch fused into the textile through heat transfer
or solvent-vapor-assisted fusion. The process relies on a revers-
ible and/or soluble thermoplastic such as SIS or thermoplastic
poly urethane (TPU) that fuses seamlessly into textile fibers
during the transfer process. As demonstrated in Figure4C, to
decompose the e-textile, it is first soaked with the solvent of
the adhesive polymer. This dissolves the adhesive layer coated
to the polymer, which permits peeling o the patch from the
textile. Then, the patch can be recycled through the techniques
shown in Figure4A,B.
The recovered metal-rich chunk shown in Figure 4C con-
tains EGaIn, metallic microfiller (Ag or Fe), and the remaining
SIS polymer that could not be fully removed during the steps
shown in Figure4A. Therefore, the metal-rich chunk produced
in this step requires further processing to recover the individual
metals (Figure5). In the case of EGaIn–Fe–SIS ink, the recovery
of liquid metal and the ferromagnetic ferrite filler is straight-
forward through magnetic force (Figure5A and Video S7, Sup-
porting Information). We first mix the metal-rich chunk with
an ecofriendly acid (citric acid). Citric acid removes the Ga2O3
oxide shell around the EGaIn droplets and therefore contrib-
utes to the separation of the EGaIn from the metal chunk. After
removing the EGaIn, we remove the remaining SIS polymer by
mixing the remaining chunk with Toluene. Finally, we collect
the ferrite microparticles using a permanent magnet. As shown
in Video S7 (Supporting Information), the particles maintain
their original ferromagnetic property.
In the case of EGaIn–Ag–SIS (Figure5B), we developed a
mechano-chemical process based on gallium oxide dissolution
in a basic solution, i.e., NaOH, accompanied by mechanical
forcing exerted during mixing and sonication (see Videos S8
and the Experimental Section for details). After recovery of the
LM, we obtain an Ag-rich chunk.
To recover the Ag from this chunk, we performed a leaching
process that included further sonication to disperse particles
and dissolve them in the sulfuric acid (See the Experimental
Section for details). This allowed dissolving of Ag and the
remaining EGaIn coating.
After the leaching, the Ag ions in the solution are trans-
formed back into pure Ag without EGaIn contaminants using
an electrowinning process. Then, the solution was poured into
an electrochemical cell container, and glassy carbon electrodes
were used as the cathode and anode. Pure Ag was recovered
by applying a constant potential of 10V for 4 h. At the cathode
surface, Ag metal aggregates and forms a metal chunk, which
falls to the bottom of the container after gaining some weight.
Video S9 (Supporting Information) shows the process of collec-
tion of the Ag through electrowinning after the leaching step.
The LM recovered from this process can be used directly for
making a new ink. However, the Ag should be first transformed
to flake form, in order to be usable for synthesis of the EGaIn–
Ag–SIS ink.
It is interesting as well to discuss the role of gallium in the
success of the leaching and electrowinning process. In one
experiment we tested the leaching process in equal conditions
for Ag-SIS composite, and an Ag-SIS composite with addi-
tion of EGaIn. The amount of EGaIn/Ag ratio was 0.65/1.65,
which is considerably lower than the biphasic EGaIn–Ag–SIS
composite. This composite was made in order to simulate the
Ag-rich chunk that remains after recovery of EGaIn, and before
the leaching process. Details of the experiment can be seen in
Figure S18 (Supporting Information), and Section S3B of the
Supporting Information. As a result, we found that existence
of the EGaIn in the formulation helps in the leaching process.
This can be as well seen in Videos S10 (Supporting Informa-
tion), which compares the leaching process for both compos-
ites. As can be seen, from the composite that contains EGaIn,
only SIS polymer is remained after 45 min of leaching, whereas
no change in the composite without EGaIn is observed after 3 h
of mixing. To the best of the authors’ knowledge, this is the first
time that combining sulfuric acid leaching and electrowinning
for the extraction valuable silver from the biphasic conductive
composite with high eciency at room temperature has been
discussed.
Figures S19 and S20 (Supporting Information), show the
elemental analysis of the Ag obtained through electrowinning,
showing that it is a pure Ag, with some remaining of sulfur
from the sulfuric acid used in the leaching process.
In summary, the unique combination of materials, i.e.,
reversible block copolymers, EGaIn, and Ag in the composite,
enabled the high eciency of the recycling process.
Although the EGaIn–Fe–SIS has fewer steps to recover
the ink elements, there exists a trade-o between the ease of
recovery and the ink properties. Figure 5C,D demonstrate
optical images of a nonencapsulated EGaIn–Fe–SIS and an
EGaIn–Ag–SIS, respectively, after subjecting them to 10 cycles
of 100% strain. As can be seen, the EGaIn–Ag–SIS remain
nonsmearing while some LM droplets are separated from the
EGaIn–Fe–SIS composite. Therefore, the choice between the
composites is a trade-o. Circuits based on EGaIn–Fe–SIS
are easier to recycle but more complex to fabricate, as they
need an encapsulation layer to remain nonsmearing when
subject to mechanical strain. Scanning electron microscopy
(SEM) from the top surface (Figure 5E), and cross-section
(Figure 5F) of the EGaIn–Fe–SIS composite, and EGaIn–Ag–
SIS (Figure5G,H), shows that the latter is considerably more
homogenous. The liquid phase EGaIn in EGaIn–Ag–SIS is
more entangled with the solid phase Ag, thus resulting in the
liquid metal being retained in the microstructure when sub-
ject to the mechanical strain. However, this comes at the cost
of a more complex element recovery process. Figures S9, S10,
S11, and S12 (Supporting Information), and the related discus-
sion in the supporting information provide additional insights
about the microstructure of the EGaIn–Ag–SIS and EGaIn–Fe–
SIS biphasic composites. Through SEM and energy dispersive
X-ray spectroscopy (EDS) from the top and cross-section of
the printed samples, we demonstrate the microstructure and
material distribution for both composites, and show formation
of AgIn2 intermetallic component (IMC). This IMC permits
Adv. Mater. 2022, 34, 

www.advmat.dewww.advancedsciencenews.com
2203266 (9 of 13) ©  The Authors. Advanced Materials published by Wiley-VCH GmbH
formation of a more homogenous composite that does not
suer from phase separation, and thus results in the desired
nonsmearing behavior, albeit at the cost of making recycling
more dicult.
In addition, it can be seen from Figure S12 (Supporting
Information), that a line printed through extrusion printing
results in a line thickness of ≈100µm, which is 5.5 times higher
than the 15µm thickness in the circuits produced by thin/film
coating and laser patterning (referring to optical profilometer
images in Figure S1 in the Supporting Information). The thick-
ness of the printed trace, had an inverse linear relation with its
electrical resistance, according to the Pouillet’s law of electricity.
While this reduced resistance is generally appreciated in digital
circuits, it also increases the ink consumption. Besides the laser
patterned circuits were able to deliver the intended function in
all laser patterned digital circuits shown in Figure2.
Adv. Mater. 2022, 34, 
Figure 5. Extraction of the metal elements from the recovered metal chunk in the previous step. A) Method for separation of LM and Ferrite from
EGaIn–Fe metal chunk, using magnetic force. B) Method for separation of Ag, and EGaIn, based on the mechano-chemical procedure followed by
dissolving, leaching and Electrowinning. C) Fe–EGaIn–SIS composite after % mechanical strain. The phase separation is expectable due to the
nonhomogeneous microstructure as shown in SEM analysis: E) top surface and F) cross-section. D) Optical image of an Ag–EGaIn–SIS sample post-
strain. SEM microscopy from G) top surface and H) Cross-section

www.advmat.dewww.advancedsciencenews.com
2203266 (10 of 13) ©  The Authors. Advanced Materials published by Wiley-VCH GmbH
Figure 6A and Video S4 (Supporting Information) show
examples of two complex 3R circuits that were produced
through laser patterning and vapor soldering. Both circuits
integrate several laser-patterned strain gauge sensors and all
necessary microchips. The circuit in Figure6A-iii is produced
with recycled ink (recovered liquid metal and new silver flakes)
and microchips that were integrated into the same circuit and
recovered 10 times. Figure6B shows the digital fingerprint of
four strain sensors in 16 dierent scenarios: mild, medium,
and deep respiration (Figure 6B); sequential tapping of the
sensors (Figure 6C); neck stretching exercise in four direc-
tions and swallowing (Figure 6D); and four types of torso exer-
cise (Figure 6E). Additional details from the sensor and the
obtained signals are found in Figures S6–S8 (Supporting Infor-
mation). A distinct digital fingerprint can be observed in each
case, making it even visually classifiable. When combined with
advanced machine learning techniques, these patches enable
monitoring of vital, behavioral, and psychological signals. At the
same time, 3R electronics paves an important step toward sus-
tainable development by enabling a circular economy through
the recovery and reuse of the constitutional elements in these
patches that are intended to be used only for a short time.
3. Conclusion
This work lays the foundations for the autonomous fabrica-
tion of the next generation of 3R electronics. Taking advantage
of sinter-free biphasic composites and innovative fabrication
techniques based on laser patterning and vapor sintering, we
Adv. Mater. 2022, 34, 
Figure 6. A) Soft-matter chip-integrated multinode sensor patch produced by × recycled LM and × recycled microchip. Fingerprint of the gener-
ated signal during mild, normal, and deep respiration (B), tapping (C), four dierent neck movements, and swallowing (D), and four dierent torso
movement (E).

www.advmat.dewww.advancedsciencenews.com
2203266 (11 of 13) ©  The Authors. Advanced Materials published by Wiley-VCH GmbH
demonstrated the fabrication of condensed soft-matter circuits
with integrated microchips that are highly stretchable, resil-
ient, repairable, and recyclable. Compared to digital printing,
the laser sintering technique presented in this work oers 10×
better resolution and considerably faster patterning. We dem-
onstrated that the laser patterning and soldering techniques
could be easily scaled for the rapid fabrication of multiple cir-
cuits. Laser patterning of small/medium circuits (most of the
circuits shown in this work) takes less than 3 min, and sol-
dering was performed in 30–45min.
We believe that the presented fabrication techniques are
simpler than the existing techniques, not only compared to
the existing methods for soft electronics, but also when com-
pared to the rigid PCB manufacturing. Eliminating the need
for thermal sintering, thermal soldering, and solder paste
deposition contribute to the reduction of multiple fabrication
steps. However, the presented techniques are still in their
infancy and require further technological development to reach
the same maturity as the current high throughput PCB tech-
nology. Elimination of the thermal processes from the entire
procedure is a critical advancement in the move toward sus-
tainable electronics, as it opens doors for many heat-sensitive
elastomers that could not withstand sintering or soldering tem-
perature. This includes emerging recyclable and biodegradable
polymers/elastomers.[21,28,29]
These circuits are resilient to mechanical strain and repair-
able if cut. We demonstrated that traces printed using this
material architecture could withstand 1700% of strain, and
microchip-integrated circuits can withstand a record-breaking
maximum strain value of ≈900%. When cut, these circuits are
healed through the vapor exposure technique. The healing is so
ecient that the circuits can be subjected to mechanical strain
post-healing. When disposed of, these circuits can be processed
through simple techniques that allow them to recover their con-
stitutional elements, including textiles, metals, and microchips.
We showed an example of a complex patch with multiple laser
patterned strain gauges that integrate microchips that have
been recycled 10 times. It is noteworthy that the 3R principle is
only partially applied to the microchip-integrated circuits, as the
microchips themselves are not soft, resilient and repairable, but
they can be recycled.
As examples of emerging applications, we demonstrated the
fabrication of complex chip-integrated epidermal patches that
can obtain a fingerprint of optical, thermal, and mechanical sig-
nals from the human skin for human health monitoring and
activity classification.
4. Experimental Section
Synthesis of Biphasic Composites: EGaIn–Ag–SIS composite was
prepared by dissolving SIS (wt% Sigma Aldrich) in toluene (% SIS)
until a clear solution was obtained. For each g of the BCP solution,
. g of Ag flakes (Silflake  Technic Inc.) and  g of EGaIn were
added and mixed using a planetary mixer (rpm).
Similar method to previous composite was followed to prepare
EGaIn–Fe–SIS, replacing Ag flakes with magnetite particles (FeO),
(µm size, %, CAS: --).
Direct Digital Printing: cc Ink cartridges were filled with the
corresponding ink, then printing was performed using a Voltera V-One.
The settings used were based upon manufacturer guidelines (adapted to
the correspondent composite, based on viscosity). To achieve a highly
conductive uniform circuit, usually two to three layers were required.
After printing, the circuits were usually cured for  min− h in an oven
(°C), or overnight at room temperature
Electromechanical Characterization: To characterize the
electromechanical performance of the materials, a “Dogbone” shape was
used with the Die C ASTM D  standard as basis. “dog bones” were
made by laser patterning (VLS ., Universal Laser Systems Inc) mm
thick films. The films were produced by applying a thin-film applicator to
the latex prepolymer (Elastica PVS Latex) with dimensions ( × mm
track with . × .mm pads). For electromechanical testing, an Instron
 with a  N load cell and a data acquisition system composed of
a multimeter (gw nstek gdm-) and -bit DAQ(NI USB ) was
used.
SEM Microscopy: The surface and cross-section morphologies of the
samples were characterized by scanning electron microscopy (SEM)
using a FEI Quanta FEG ESEM equipped with an EDAX Genesis
XM. For the cross-section observation the samples were immersed for
 s in liquid nitrogen. This allowed for a clean fracture of the samples
to be made through mechanical impact.
Optical Microscopy: Microscopic imaging was used for
characterization of the morphology of the deposited composites. It was
done using a Leica SD stereomicroscope, with up to × magnification,
coupled with a MU AmScope .MP camera, and AmScope software
for image capture and measuring.
PVA: g of PVA powder (Selvol , SEKISU) was mixed with mL of
HO. The mixture was stirred at  °C using a hot plate (AGIMATIC-N)
until it became a clear and homogeneous solution.
Conductive Ink: The ink was prepared by dissolving SIS in toluene
( wt% SIS) until a clear solution was obtained. For each g of BCP
solution, . g of Ag flakes (Silflake, Technic) and g of EGaIn were
added and mixed using a planetary mixer (Thinky ARE-) at rpm.
SIS Substrate Solution: The SIS solution was prepared using a : ratio
of SIS (Styrene %, Sigma Aldrich) and toluene.
Patch Fabrication, Laser Patterning: The fabrication started by first
applying a  µm PVA solution over the glass, using a thin-film
applicator (ZUA , ZEHNTNER). This layer served as the sacrificial
layer that permits easy circuit release. Then, two layers of conductive ink
were spread over the PVA using the thin film applicator consecutively.
After curing the ink in the oven, the circuit was patterned to isolate
the circuit traces by laser ablation (pulsed fiber laser with  nm
wavelength) while blowing air to it, in order to expel the created debris.
Because the laser wavelength only aects metals, the PVA layer under the
ablation area remained intact. This turned out to be an eective method
to fabricate high-resolution circuits with spacing as low as µm, thus
enabling the possibility of using smaller IC packages in the applications.
A µm SIS substrate solution was applied over the circuit using a thin
film applicator. After curing, an adhesive Kapton tape was placed on top
of the cured circuit to support it mechanically, avoiding overstraining the
circuit during the next steps. After peeling the circuit from the glass, the
PVA layer was dissolved and removed in water. Using transferable glue,
the circuit was attached to the glass with the Kapton side facing down.
Microchip Interfacing: The circuit was populated with SMD
components using a pick and place machine (eC-placer Eurocircuits)
which were then “soldered” in a toluene vapor chamber for  min.
The vapor was generated using a homemade setup with room-
temperature aerosol generator with the desired profile of intensity. At
these stage components were soldered into the ink, and mechanical
locked, thanks to the adhesive properties of the SIS. After “soldering,”
the circuit was sealed using substrate solution of SIS. The circuit is left
at tilted position in order to remove the excess material by gravity.
Multi Strain Senso: The circuit consisted of a Bluetooth module
(CYBLE-- from Cypress) and two  bits ADC (ADSC from
Texas Instruments) to measure the analogue signals from the strain
sensors. The values from each sensor were captured and sent wirelessly
to a PC. The strain sensors were patterned as part of the conductive
circuit. The strain sensors consisted in a foil patterned design, as the foil
Adv. Mater. 2022, 34, 

www.advmat.dewww.advancedsciencenews.com
2203266 (12 of 13) ©  The Authors. Advanced Materials published by Wiley-VCH GmbH
deforms, the electrical resistance changes due to its dependence to the
conductor geometry.
SpO2 Patch: The circuit measured oxygen saturation. It used light
with two dierent wavelengths pointed to parts of the body with high
blood flow rate and measured the absorbance using photodetectors.
The system comprised several light emitters in the red and infrared
band, controlled by a specialized SpO IC (MAXM from Maxim)
containing two high sensitivity photodiodes. A Bluetooth module
(CYBLE-- from Cypress) was integrated to the patch to
wirelessly exchange information with the host computer.
Temperature Monitoring Circuit: The circuit measures wireless
temperature and humidity with integrated Bluetooth (CYBLE--
from Cypress), battery, temperature, and humidity sensor (Si-A
from Silicon Labs), which sends the data wirelessly to a mobile phone
application.
Recycling e-Patch: The polymer substrate and encapsulation removal
were done by placing the circuit on top of a filter (mesh count T)
and paper towel (soaked with liquid toluene), positioned inside of an
enclosed  L volume chamber. The chamber setup was left in the oven
at °C for  h. During this time the polymer turned into a liquid and
settled down due to gravity, passing through the filter mesh leaving
the SMD components and ink exposed. The filter with the remaining
circuit on top was taken from the chamber, the SMD components were
collected using a tweezer and ink metals were removed with a spatula.
Recycling e-Textile: The polymer substrate and encapsulation removal
were done by placing the circuit on top of a filter (mesh count T)
and paper towel (soaked with liquid toluene), positioned inside of an
enclosed  L volume chamber. The chamber setup was left in the oven
at °C for  h. During this time the polymer turned into a liquid and
settles down due to gravity, passing through the filter mesh leaving
the SMD components and ink exposed. The filter with the remaining
circuit on top was taken from the chamber, the SMD components were
collected using a tweezer and ink metals were removed with a spatula.
Recycling e-Textile: The circuit was separated from the textile by first
impregnating the textile with acetone in order to dissolve the TPU glue
layer and detach it from the textile. The circuit was then removed by hand.
The two TPU sheets containing the conductive ink circuit and SMDs was
immersed inside of a flask filled with acetone. The flask was manually shake
for  min and then the sealing and substrate TPU layers were detached
from each other by hand. Most of the SMD components were released
from the substrate during process, although, to completely separate the
ink and SMD components, additional immersion and shaking was done
until all SMDs and ink were released from the TPU sheets. Components
and ink were recovered from the bottom of the flask using a tweezer.
Recycling EGaIn–Fe–SIS Ink: After removing the metal chunk from
the circuit, using the methods explained in Figure, the LM-Ferrite-SIS
waste was mixed into a Citric acid and isopropanol solution ( RPM
for  min, Thinky Planetary mixer). Most of the LM droplets were
extracted. The remaining metal rich chunk was washed with water,
and put in Toluene and subjected to mixing at  RPM for  min in
order to separate the remaining SIS from other components. Finally,
the magnetic particles were collected by applying an external magnet,
washed with water, and dried in an oven for further use.
Recycling EGaIn–Ag–SIS Ink: After removing the metal chunk from
the circuit, using the methods explained in Figure, the LM-Ag flake-SIS
ink waste was mixed in   NaOH solution at  RPM for  min.
At this stage, most of LM droplets were extracted. After removing the
NaOH, the rest of the waste was put in acetic acid accompanied by
Epsom salt, and subjected to mixing at  RPM for min. Additional
LM droplets were extracted. This step allowed the resulting slurry to
be decomposed. In order to extract additional LM from the waste, the
metal chunk in Isopropanol ( W for min) was sonicated to induce
additional mechanical shear forcing to produce LM nano particles,
similar to previous works on fabrication of EGaIn Nanoparticles. The
mixed solution was dried in an oven and crushed into fine powders,
and dispersed in Toluene ( RPM for  min). At this stage, all the
materials almost decomposed, which was followed by adding HCl. HCL
allowed removal of the gallium oxide shell of the EGaIn Nano particles,
resulting in aggregation into a larger droplet that can be collected. The
rest of the material was an Ag rich chunk that is sent to the leaching and
electrowinning process.
Acid Leaching: To form a suspension containing dissolved particles
from the recovered metal chunk, mL of THF was added into a  mL
flask along with the metal chunk, and mixed (Thinky ARE-, rpm,
min), to dissolve the remaining of the polymer and separate the metal
particles. A sonicator was used to further disperse the particles (min
% power). The mixture was filled with DI water to a total volume of
mL prior to sonication. mL of HHO was added drop by drop to the
suspension while stirring immersed in a water bath. The mixture was
then heated at  °C using a hot plate for an additional  min while
stirring. Then, mL of HSO was added drop by drop and stirred for  h
to completely dissolve the metals.
Electrowinning: The solution was poured on an electrochemical cell
container and glassy carbon electrodes were used as electrodes for
cathode and anode. Recovery of Ag was achieved by applying a constant
potential of V for  h. At the cathode surface the deposition of pure
Ag metal aggregates and forms a metal chunk which normally falls to
the bottom of the container, which was removed using a spatula.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
Acknowledgements
This work was partially supported by the Foundation of Science and
Technology (FCT) of Portugal through the CMU-Portugal project WoW
(Reference Nr: ), and Dermotronics (PTDC/EEIROB//
), financed by the EU structural & investment Funds (FEEI)
through operational program of the center region. Financing also
came from the SMART Display project (reference: POCI---
FEDER-), cofinanced by the European Regional Development
Fund, through Portugal  (PT), and by the Competitiveness and
Internationalization Operational Programme (COMPETE ). The
experiments involving human subject and depicted in Figures  and 
are performed with the full, informed consent of the participant.
Conflict of Interest
The authors declare no conflict of interest.
Data Availability Statement
The data that support the findings of this study are available from the
corresponding author upon reasonable request.
Keywords
biphasic liquid metal, electronic waste, recyclable electronics, soft-
matter electronics, wearable biomonitoring
Received: April , 
Revised: May , 
Published online: June , 
[] Y. Hong, D. Thirion, S. Subramanian, M. Yoo, H.Choi, H. Y.Kim,
J. F.Stoddart, C. T.Yavuz, Proc. Natl. Acad. Sci. USA 2020, 117, .
[] B.Ghosh, M. K.Ghosh, P.Parhi, P. S.Mukherjee, B. K. Mishra, J.
Cleaner Prod. 2015, 94, .
Adv. Mater. 2022, 34, 

www.advmat.dewww.advancedsciencenews.com
2203266 (13 of 13) ©  The Authors. Advanced Materials published by Wiley-VCH GmbH
[] X.Aeby, A.Poulin, G.Siqueira, M. K.Hausmann, G.Nyström, Adv.
Mater. 2021, 33, .
[] Z. Wang, C. Peng, K. Yliniemi, M. Lundström, ACS Sustainable
Chem. Eng. 2020, 8, .
[] U. U. Jadhav, H.Hocheng, J. Achiev. Mater. Manuf. Eng. 2012, 54,
.
[] A.-G. Ghica, E. Vasile, C. Carata, C. I. Covaliu, M.-i. Petrescu,
G.Iacob, M.Buzatu, Rom. J. Mater. 2020, 50, .
[] S. Y.Cho, W. G.Lee, P. P.Sun, Mater. Trans. 2019, 60, .
[] Z. Wang, P. Halli, P. Hannula, F. Liu, B. P. Wilson,
K. Yliniemi, M. Lundström, J. Electrochem. Soc. 2019, 166,
E.
[] J.Miao, H.Liu, Y.Li, X.Zhang, ACS Appl. Mater. Interfaces 2018, 10,
.
[] Y. S. Choi, R. T. Yin, A. Pfenniger, J. Koo, R. Avila, K. B. Lee,
S. W. Chen, G. Lee, G. Li, Y. Qiao, A. Murillo-Berlioz, A. Kiss,
S. Han, S. M. Lee, C. Li, Z. Xie, Y.-Y. Chen, A. Burrell, B. Geist,
H. Jeong, J. Kim, H.-J. Yoon, A. Banks, S.-K. Kang, Z. J. Zhang,
C. R. Haney, A. V. Sahakian, D. Johnson, T. Efimova, Y. Huang,
etal., Nat. Biotechnol. 2021, 39, .
[] C.Shi, Z.Zou, Z.Lei, P. Zhu, W.Zhang, J. Xiao, Sci. Adv. 2020, 6,
eabd.
[] L.Teng, S.Ye, S.Handschuh-Wang, X.Zhou, T. Gan, X.Zhou, Adv.
Funct. Mater. 2019, 29, .
[] W. B.Han, J. H.Lee, J. W.Shin, S. W.Hwang, Adv. Mater. 2020, 32,
.
[] K.Liu, H.Tran, V. R.Feig, Z.Bao, MRS Bull. 2020, 45, .
[] G. Chen, G. Chen, X. Deng, L. Zhu, S. Handschuh-Wang, T. Gan,
B.Wang, Q.Wu, H.Fang, N.Renc, X.Zhou, J. Mater. Chem. A 2021,
9, .
[] R.Tutika, A. B. M. T.Haque, M. D Bartlett, Commun. Mater. 2021,
2, .
[] H. Chang, P.Zhang, R. Guo, Y.Cui, Y.Hou, Z. Sun, W. Rao, ACS
Appl. Mater. Interfaces 2020, 12, .
[] M.Tavakoli, M. H.Malakooti, H.Paisana, Y.Ohm, D. G.Marques,
P. Alhais Lopes, A. P. Piedade, A. T. de Almeida, C. Majidi, Adv.
Mater. 2018, 30, .
[] P. A. Lopes, D. F. Fernandes, A. F. Silva, D. G. Marques,
A. T.deAlmeida, C.Majidi, M.Tavakoli, ACS Appl. Mater. Interfaces
2021, 13, .
[] P. A.Lopes, S.Bruno, A. T.deAlmeida, C.Majidi, M.Tavakoli, Nat.
Commun. 2021, 12, .
[] S.Chen, Z.Wu, C.Chu, Y.Ni, R. E.Neisiany, Z.You, Adv. Sci. 2022,
9, .
[] M. J.Ford, D. K.Patel, C.Pan, S.Bergbreiter, C.Majidi, Adv. Mater.
2020, 32, .
[] K. B. Ozutemiz, J. Wissman, O. B. Ozdoganlar, C. Majidi, Adv.
Mater. Interfaces 2018, 5, .
[] B.Zhang, Q. Dong, C. E. Korman, Z. Li, M. E. Zaghloul, Sci. Rep.
2013, 3, .
[] J.Jiang, B. Bao, M.Li, J. Sun, C.Zhang, Y.Li, F. Li, X.Yao, Y.Song,
Adv. Mater. 2016, 28, .
[] D. G. Marques, P. A. Lopes, A. T. Almeida, M. de Carmel,
M.Tavakoli, Lab Chip 2019, 19, .
[] T. Lu, J. Wissman, C. Majidi, ACS Appl. Mater. Interfaces 2015, 7,
.
[] J. A. Chiong, H. Tran, Y. Lin, Y.Zheng, Z. Bao, Adv. Sci. 2021, 8,
.
[] Y. Deng, R. Dewil, L. Appels, R. Ansart, J. Baeyens, Q. Kang, J.
Environ. Manage. 2021, 278, .
Adv. Mater. 2022, 34, 