Recyclable Thin‐Film Soft Electronics for Smart Packaging and E‐Skins

Abstract Despite advances in soft, sticker‐like electronics, few efforts have dealt with the challenge of electronic waste. Here, this is addressed by introducing an eco‐friendly conductive ink for thin‐film circuitry composed of silver flakes and a water‐based polyurethane dispersion. This ink uniquely combines high electrical conductivity (1.6 × 105 S m−1), high resolution digital printability, robust adhesion for microchip integration, mechanical resilience, and recyclability. Recycling is achieved with an ecologically‐friendly processing method to decompose the circuits into constituent elements and recover the conductive ink with a decrease of only 2.4% in conductivity. Moreover, adding liquid metal enables stretchability of up to 200% strain, although this introduces the need for more complex recycling steps. Finally, on‐skin electrophysiological monitoring biostickers along with a recyclable smart package with integrated sensors for monitoring safe storage of perishable foods are demonstrated.

Table S1.Comparison between the contributions of this work and other works in the field, in terms of ink formulation, decomposition process, electrical conductivity, stretchability, usability for digital printing of circuits, and applications.Video S4.Change in LED color in the smart label when the package is stored for more than one hour at temperatures higher than 5 ºC.Video S5.Direct ink writing of soft circuits using the proposed Ag-WPU conductive ink.Table S1.Comparison between the contributions of this work compared to other works in the field, in terms of ink formulation, decomposition process, electrical conductivity, stretchability, usability for digital printing of circuits, and applications.Some recent works have showed recyclable circuits, as shown in Table S1.For instance the work shown in reference [5] proposes a printable conductor with straightforward decomposition in warm water but nevertheless with limited stretchability (80%) and relatively low conductivity (2.1 x 10 4 S/m), as well as no demonstration of direct integration of microchips.Using fully metallic conductors, reference [4] introduces recyclability in rigid circuits made of tin-based conductive tracks that can be melted and reused, while [2] discloses stretchable LM-based stencil printed circuits used for strain sensing.An interesting contribution to the fields of ecological and recyclable electronics is shown in [3] where a PVA-LM mixture in a water medium can be digitally printed as mechanical deformation sensors yet this formulation presents limited stretchability (20%), doesn't prove suitable for microchip integration and the conductive patterns must be activated by an externally applied pressure thus adding another step to the fabrication process.More recently, [1] proposes a variety of highly promising methods for fabricating and recycling highly stretchable fully integrated circuits.Nevertheless, on the one hand the inks used all contain toluene as solvent (which entails environmental concerns) and as well the proposed recycling method requires cumbersome steps such as 'electrowinning' and 'leaching', while also using solvents such as acetone and toluene in the process.Hereupon, having in mind that a printed circuit without microchips would be of limited use, and complex multi-step recycling may prove itself ineffective in terms of cost and scalability, this work represents a striking advance compared to past efforts with recyclable electronics that are soft, stretchable, and printable, none were able to demonstrate a combination of high resolution digital printing, microchip integration, strain tolerance, and simple, efficient and ecological recycling, being this work the first to combine all of these factors.

R/R0
Cycle number

Ink cyclic stability
Ag-WPU Ag-EGaIn-WPU 0,8  The safe zone (green LED) occurs while the packages are not stored above 5 ºC for more than 1 hour.When this 1 hour time limit is passed, the risk zone (red LED) is entered, even if the package is further correctly refrigerated (below 5 ºC). the 1 hour limit is cumulative, as shown in the bottom plot.

Figure S1 .
Figure S1.0402-sized resistors and LEDs (1.0 x 0.5 mm package dimensions and 0.5 mm pad separation) bonded to a set of printed lines with 200 um width.

Figure S2 .
Figure S2.Influence of silver flake and liquid metal concentration in the inks' conductivity at 0% strain.

Figure S3 .
Figure S3.Variation of the viscosity of four Ag-WPU inks as a function of shear rate.The samples were stored at room temperature between 0 and 35 days.

Figure S4 .
Figure S4.Estimated conductivity of Ag-WPU samples upon stretching.The material was assumed to be incompressible.

Figure S5 .
Figure S5.Strain vs resistance curve for Ag-WPU traces printed over a thermoplastic polyurethane (TPU) substrate (nine samples from distinct ink batches).

Figure S6 .
Figure S6.SEM image of Ag-WPU surface Figure S7.SEM image of Ag-WPU surface

Figure S9 .
Figure S9.Printed circuit with integrated SMD components separated in its fundamental elements: conductive ink, TPU substrate, and rigid components.

Figure S10 .
Figure S10.Ag-rich powder after IPA washing.Large PU aggregates can be observed (red highlights).

Figure S11 .
Figure S11.Ag-rich powder after removal of large PU chunks (left) and after being transferred to a glass vial (right).

Figure S12 .
Figure S12.Microscope images of the surface of printed traces from pristine Ag-WPU ink (left) and recycled Ag-WPU ink (right).For the image on the right, a rough surface can be observed, originating from solidified PU residue trapped in the print.

Figure S13 .
Figure S13.Comparison between Strain vs resistance curve for pristine Ag-WPU traces and traces printed from recycled Ag-WPU.

Figure S14 .
Figure S14.Strain-stress curves for pristine TPU film and the same TPU film after undergoing the IPA recycling process.

Figure S15 .
Figure S15.Estimated conductivity of Ag-EGaIn-WPU samples upon stretching.The material was assumed to be incompressible.

Figure S16 .
Figure S16.50 stretching/releasing cycle test of the Ag-WPU and Ag-EGaIn-WPU inks (ε = 10%).The insets on the bottom plots show 5 cycles in the middle of the test.

Figure S17. A :
Figure S17.A: Microscope images of the surface of printed traces from Ag-EGaIn-WPU.A rough surface can be observed, originating from EGaIn droplets fully encapsulated in polyurethane B: DIW printed grids using the Ag-EGaIn-WPU ink with 1.37 mm and 812 um line spacing.A 250 um diameter nozzle was used.

Figure S18 .
Figure S18.SEM image and EDS analysis of Ag-EGaIn-WPU sample cross section.A clear separation between Ag and Ga can be observed, while In is present all over the sample

Figure S19 .
Figure S19.SEM image and EDS analysis of Ag-EGaIn-WPU sample cross section.A clear separation between Ag and Ga can be observed, while In is present all over the sample

Figure S23 .
Figure S23.Dynamic time warping fitting between the signals for two repetitions of a smiling gesture.

Figure S24 .
Figure S24.Dynamic time warping fitting between the signals for two repetitions of a jaw clenching gesture.

Figure S25 .
Figure S25.Dynamic time warping fitting between the signals for two repetitions of an eye blinking gesture.

Figure S26 .
Figure S26.Dynamic time warping fitting between the signals for two repetitions of an eyebrow flashing gesture.

Figure S27 .
Figure S27.Euclidean distance between gesture pairs of the two repetitions, obtained from Dynamic Time Warping of the electrophysiological signals.

Figure S30 .
Figure S30.Example of the printed traces over FR1 substrate to measure ink conductivity.

Figure S32 .
Figure S32.Die C, ASTM D412 dogbone with printed trace and integrated SMD resistor for tensile testing.

Figure S33 .
Figure S33.Example of the printed traces over FR1 substrate using recycled Ag-WPU ink.Rough surface texture arises from the presence of cured polyurethane particles in the recovered Ag-rich powder that become encapsulated in the new print.

Figure S35 .
Figure S35.Setup for thermistor calibration consisting of a hotplate and digital thermometer.

Figure S36 .
Figure S36.Calibration curve for the thermistor-based temperature-monitoring electronic patch.

Figure S37 .
Figure S37.High level block diagram of the developed electrophysiology recording system.

Figure S1 .
Figure S1.0402-sized resistors and LEDs (1.0 x 0.5 mm package dimensions and 0.5 mm pad separation) bonded to a set of printed lines with 200 um width.

Figure S2 .
Figure S2.Influence of silver flake and liquid metal concentration in the inks' conductivity at 0% strain.

Figure S3 .
Figure S3.Variation of the viscosity of four Ag-WPU inks as a function of shear rate.The samples were stored at room temperature between 0 and 35 days.

Figure S4 .
Figure S4.Estimated conductivity of Ag-WPU samples upon stretching.The material was assumed to be incompressible.

Figure S5 .
Figure S5.Strain vs resistance curve for Ag-WPU traces printed over a thermoplastic polyurethane (TPU) substrate (nine samples from distinct ink batches).

Figure S9 .
Figure S9.Printed circuit with integrated SMD components separated in its fundamental elements: conductive ink, TPU substrate, and rigid components.

Figure S10 .
Figure S10.Ag-rich powder after IPA washing.Large PU aggregates can be observed (red highlights).

Figure S11 .
Figure S11.Ag-rich powder after removal of large PU chunks (left) and after being transferred to a glass vial (right).

Figure S12 .
Figure S12.Microscope images of the surface of printed traces from pristine Ag-WPU ink (left) and recycled Ag-WPU ink (right).For the image on the right, a rough surface can be observed, originating from solidified PU residue trapped in the print.

Figure S13 .Figure S14 .
Figure S13.Comparison between Strain vs resistance curve for pristine Ag-WPU traces and traces printed from recycled Ag-WPU.

Figure S15 .
Figure S15.Estimated conductivity of Ag-EGaIn-WPU samples upon stretching.The material was assumed to be incompressible.
Figure S16.50 stretching/releasing cycle test of the Ag-WPU and Ag-EGaIn-WPU inks (ε = 10%).The insets on the bottom plots show 5 cycles in the middle of the test.
Figure S17.A: Microscope images of the surface of printed traces from Ag-EGaIn-WPU.A rough surface can be observed, originating from EGaIn droplets fully encapsulated in polyurethane.B: DIW printed grids using the Ag-EGaIn-WPU ink with 1.37 mm and 812 um line spacing.A 250 um diameter nozzle was used.

Figure S18 .Figure S19 .Figure S20 .Figure S21 .Figure S22 .
Figure S18.SEM image and EDS analysis of Ag-EGaIn-WPU sample cross section.A clear separation between Ag and Ga can be observed, while In is present all over the sample

Figure S23 .
Figure S23.Dynamic time warping fitting between the signals for two repetitions of a smiling gesture.

Figure S24 .Figure S25 .
Figure S24.Dynamic time warping fitting between the signals for two repetitions of a jaw clenching gesture.

Figure S26 .Figure S27 .
Figure S26.Dynamic time warping fitting between the signals for two repetitions of an eyebrow flashing gesture.

Figure S29 .
Figure S29.Steps for synthesis of Ag-EGaIn-WPU conductive ink.An overhead stirrer was used instead of a planetary mixer.

Figure S30 .
Figure S30.Example of the printed traces over FR1 substrate to measure ink conductivity.

Figure S32 .
Figure S32.Die C, ASTM D412 dogbone with printed trace and integrated SMD resistor for tensile testing.

Figure S33 .
Figure S33.Example of the printed traces over FR1 substrate using recycled Ag-WPU ink.Rough surface texture arises from the presence of cured polyurethane particles in the recovered Ag-rich powder that become encapsulated in the new print.