Introduction

Transdermal patches have become an essential platform for skin-interfaced electronics in both medical therapy and cosmetic care, offering convenient use, pain-free administration, and compatibility with wearable form factors1. Among noninvasive transdermal delivery strategies, iontophoresis is particularly attractive for flexible patch systems because it enables tunable and localized drug transport through mild electrical current while maintaining a relatively simple electrochemical operating principle2. This physics-based transport mechanism allows iontophoretic devices to be implemented with minimal mechanical complexity, making them well-suited for wearable and flexible electronic patches3,4,5.

Accordingly, recent advances in iontophoretic systems have pursued two long-standing objectives: patch-level simplicity enabling flexible implementation6 and user-adaptive and informative operation for reliable and personalized drug delivery7. To achieve patch-level simplicity, multifunctional components have been adopted to consolidate functions and simplify architecture8,9. Galvanic couples have been employed as both iontophoretic electrodes and built-in power sources, eliminating discrete batteries6,9, while conductive hydrogels have further consolidated functions by serving simultaneously as drug reservoirs and working electrodes, removing separate solid electrodes, adhesives, and reservoir layers10. Together, these strategies promote architectural simplicity, enabling compact, flexible patches compatible with skin-mounted operation. Regarding dosing reliability, substantial variations in skin impedance across individuals and environmental conditions make consistent drug delivery challenging11,12. These variations make simple dose-quantification approaches, such as time-based dosing or fixed-duration operation, clinically unreliable, a limitation that is particularly pronounced in hyperkeratotic disorders such as psoriasis, warts, and tinea pedis13,14. To overcome this limitation, alternative iontophoretic systems have introduced active electronics to regulate delivery. These approaches integrate sensors15,16, microcontrollers17, and external feedback circuits18 to monitor skin impedance or physiological signals and dynamically adjust dosing parameters19. More sophisticated electronic-skin platforms further combine sensing arrays, closed-loop control, wireless communication, and visual displays to enable highly personalized drug delivery20,21,22,23.

In parallel with these engineering challenges, iontophoretic transdermal patches raise growing concerns regarding environmental sustainability because they operate as skin-contact electronic systems that are frequently used and replaced in practice. In iontophoresis, repeated skin contact, occlusive operation, and electrochemical reactions impose strict requirements on hygiene, safety, and performance stability, often limiting device lifetime and promoting replacement rather than long-term reuse24. As a result, the increasing integration of electronic components into iontophoretic patches contributes to plastic and electronic waste, raising concerns about persistent residues and end-of-life management25,26. This trend highlights the need for environmentally accountable iontophoretic systems that maintain therapeutic performance while reducing material persistence across repeated use cycles27. Although eco-degradable materials and transient electronics offer promising routes toward sustainable flexible devices28,29, their integration into feedback-enabled iontophoretic systems remains limited by reliance on conventional electronic components and complex system architectures30.

Taken together, these engineering and environmental considerations expose a persistent three-way trade-off in iontophoretic patch design among patch-level architectural simplicity, user-interactive and informative operation, and eco-degradable system construction. Architectures that prioritize simplicity to achieve compact, patch-level implementation inherently favor uniform, inflexible operation, such as time-based dosing, limiting real-time interaction, and on-patch indication of delivery under user- and condition-dependent skin variability6,11,13,14. Conversely, user-interactive and informative iontophoretic systems rely on electronics-heavy architectures, such as closed-loop control31,32, wireless communication33, and visual displays for dose monitoring34—that increase device footprint, material diversity, and architectural complexity, thereby compromising patch-level simplicity and mechanically compliant, flexible patch formats, and limiting compatibility with eco-degradable materials due to constraints in electrical performance, stability, and system integration16,35,36,37.

In this study, we present an electrochemically synchronized, self-indicating iontophoretic patch with a fully eco-degradable and self-powered system, addressing a long-standing trade-off among patch-level architectural simplicity, user-adaptive and user-informative operation, and system-level sustainability. The patch integrates iontophoretic transdermal transport and electrochromic indication within a single, closed electrochemical platform (ITP–EC patch), in which electrochemical charge transport from a built-in power source is synchronized within one loop. A built-in Mg–MoO3 galvanic cell supplies a single electrochemical current, where ionic current drives iontophoretic delivery through the skin while the corresponding electronic current triggers electrochromic coloration. Because all functional modules operate within this physically closed electro-ionic loop, variations in skin impedance modulate the delivered current, and the resulting cumulative charge is directly rendered as a visible electrochromic color progression. This synchronized electrochemical co-design enables a self-contained, multifunctional patch architecture that converts inherently skin-adaptive iontophoretic transport into user-interactive, electrochromic-informative feedback without external sensors, control electronics, or wiring, while preserving patch-level simplicity compatible with compact and flexible implementation.

During iontophoresis, a citrate buffer at the anode maintains a skin-compatible pH, ensuring stable and safe transdermal operation. For EC actuation, a WO3-based electrochromic system is utilized as an eco-degradable electrochromic material based on previous work38, and advanced through three architectural strategies to electrochemically synchronize with the iontophoretic module: (1) A paste-type WO3 composite layer provides the high charge capacity required for iontophoretic-level currents and exhibits lateral propagation behavior not achievable with conventional thin-film EC materials39,40,41,42. (2) Positioned laterally adjacent to the current collector, the EC gauge displays charge-proportional color propagation, serving as an intuitive, user-facing indicator of patch usage and approximate delivered dose. (3) For an electrolyte of EC actuation, a propylene carbonate (PC)-based electrolyte is employed to suppress unwanted side reactions, ensuring that iontophoretic current is effectively coupled to the electrochromic reaction. Ex vivo porcine skin studies demonstrate that EC gauge progression quantitatively reflects cumulative charge transfer and correlates with transported drug amount. In a psoriasis mouse model, the patch operates safely under skin-compatible conditions and provides a clear visual indication throughout iontophoretic treatment. Importantly, eco-degradability is implemented at the system level rather than as a material or electronic add-on, with all functional components validated to be fully eco-degradable under compost-soil conditions following the ISO 14855 standard, ensuring sustainable end-of-life disposal for skin-mounted transdermal applications.

Results and discussion

Design of the eco-degradable patch for iontophoretic delivery and EC dose tracing

Figure 1a shows the architecture of a fully eco-degradable, self-powered, and electrochemically synchronized ITP–EC patch. The device adopts a vertical stack layout on a flexible poly(butylene adipate-co-terephthalate) (PBAT) substrate. The lower layer contains the galvanic iontophoretic module, and the upper layer contains the EC module. The iontophoretic module consists of anodic and cathodic units positioned on the PBAT substrate. The anodic unit (1.4 × 1.4 × 0.1 cm) is composed of a 50 μm-thick magnesium (Mg) foil in direct contact with a 1 mm-thick citrate-buffered agarose hydrogel loaded with a cationic drug. The cathodic unit (1.4 × 1.4 × 0.1 cm) consists of a 25 μm-thick molybdenum (Mo) foil coated with a 150 μm-thick composite of molybdenum trioxide (MoO3), carbon black (CB), and xanthan gum (XG) binder, interfaced with a 1 mm-thick PBS-based agarose counter hydrogel. Together, these electrodes form an embedded Mg–MoO3 galvanic cell that serves as a shared electrochemical power source for both iontophoretic actuation and electrochromic indication. The EC module (4.0 × 1.0 × 0.1 cm) is fabricated on the PBAT substrate above the iontophoretic module. Paste-type EC gauge consists of WO₃ nanoparticle, UV-treated CB (UV-CB), and XG binder composite laterally deposited on a 150 nm-thick tungsten (W) current collector and a Prussian Blue (PB) counter electrode composed of PB nanoparticle, CB, and XG binder. A 1 mm-thick organogel electrolyte, composed of 1 M NaClO₄ in PC crosslinked with cellulose acetate–pyromellitic dianhydride (CA–PMDA), is sandwiched between the EC gauge and counter electrodes to support Na+ ion transport.

Fig. 1: Architecture and eco-degradation of the self-powered iontophoretic–electrochromic patch (ITP-EC Patch).
Fig. 1: Architecture and eco-degradation of the self-powered iontophoretic–electrochromic patch (ITP-EC Patch).The alternative text for this image may have been generated using AI.
Full size image

a Device structure and working principle of the ITP–EC patch with a built-in Mg–MoO3 battery that drives iontophoretic delivery and the charge-correlated WO3 electrochromic indication of delivered dose in series. Ionic current (yellow arrows) crosses skin to deliver drug, while electronic current (blue arrows) powers the synchronized WO3 gauge, together forming a single closed electro-ionic loop for on-skin, dose-referenced readout. b Photographs of the ITP–EC patch showing the EC module in top view and the galvanic iontophoresis module in bottom view. The top view highlights the patterned WO3 EC layer (overlay red) and counter electrode (overlay blue) with a laminated electrolyte, and the WO3 appears deep blue after coloration. The bottom view shows the anode in red and the cathode in blue on the same flexible substrate. Dashed boxes mark active areas. Scale bar 5 mm. c Eco-degradation of the ITP–EC patch in aerated soil with oat seeds at 25 °C, kept moist by daily tap water. Images at 0, 21, and 42 days show intact EC and electrodes at 0 days, swelling and fragmentation with Mg corrosion by 21 days, and substantial degradation of metallic and EC components with small substrate remnants by 42 days; seed germination is unaffected. Scale bar 1 cm.

The iontophoretic module operates via a spontaneous redox reaction between the Mg anode and MoO3 cathode. At the anode, Mg is oxidized (Mg → Mg2+ + 2e−), releasing electrons into the circuit and generating Mg2+ ions in the hydrogel43. At the cathode, MoO3 is reduced with Na+ from the electrolyte (MoO3 + xNa+ + xe− → NaxMoO3), completing the redox process44,45. This reaction produces an ionic current that flows through the skin from the anode hydrogel to the cathode hydrogel, enabling migration of the cationic drug within an electrochemically synchronized loop. The EC module is operated by the same electrical current from the iontophoretic module, enabling synchronized operation between iontophoretic transport and electrochromic actuation. Electrons generated at the Mg anode flow to the EC gauge, where they are intercalated with Na+ ions from the electrolyte (WO3 + xNa+ + xe− → NaxWO3). causing a blue coloration due to polaron formation within the WO3 lattice37. At the counter electrode, PB is electrochemically oxidized with Na+ deintercalation (NaxFe[Fe(CN)6] → Fe[Fe(CN)6] + xNa+ + xe−), and the released electrons simultaneously flow into the MoO3 cathode of the iontophoretic module. This closed-loop pathway directly couples electrochromic coloration with iontophoretic current, ensuring synchronized electrochemical operation with matching drug delivery and visual indication.

Figure 1b shows the top and bottom views of the ITP-EC patch. With separation by PBAT film substrates, the galvanic electrode-based iontophoretic module is oriented downward to operate ionic flow and iontophoresis toward the skin through the drug-loaded hydrogel and buffer hydrogel layers. The EC module is oriented upward to allow users to conveniently indicate the drug-delivery state during patch operation. The Mg anode and MoO3 cathode of the iontophoretic module are electrically connected to the WO3-based EC layer and PB counter electrode of the EC module, respectively. These electrical connections are established through predefined via-holes in the substrate, which are filled with a W/Beeswax conductive biodegradable paste during assembly46. The EC module exhibits a gauge-like propagation of blue coloration, reflecting localized charge accumulation and lateral propagation driven by the iontophoretic module, thereby providing a charge-correlated, electrochromic indication of iontophoretic drug delivery. The device maintains conformal contact and mechanical flexibility, enabling stable operation and skin-adaptive, user-informative visual feedback under on-skin conditions.

Figure 1c and Supplementary Fig. 1 illustrate the eco-degradation behavior and biocompatibility of the ITP-EC patch in soil where oat seeds are planted. Most of the materials composing the ITP-EC patch can eco-friendly degrade in a natural environment, such as soil or compost, and their byproducts, such as ions, hydroxides, and monomers, have been reported to be biocompatible38,43,47. In the EC layer, W current collector and WO3 EC material degrade via the following reaction47:

$${\rm{W}}+{3{\rm{H}}}_{2}{\rm{O}}\to {{\rm{WO}}}_{3}+{3{\rm{H}}}_{2}$$
(1)
$${{\rm{WO}}}_{3}+{{\rm{H}}}_{2}{\rm{O}}\to {{\rm{H}}}_{2}{{\rm{WO}}}_{4}$$
(2)

In the galvanic electrodes, Mg, MoO3, and Mo current collectors degrade via the following reaction48,49:

$${{\rm{Mo}}+4{\rm{H}}}_{2}{\rm{O}}\to {\rm{Mo}}{{{\rm{O}}}_{4}}^{2-}+{8{\rm{H}}}^{+}+{6{\rm{e}}}^{-}$$
(3)
$${\rm{Mo}}{{\rm{O}}}_{3}+{{\rm{H}}}_{2}{\rm{O}}\to {{\rm{H}}}_{2}{\rm{Mo}}{{\rm{O}}}_{4}$$
(4)
$${\rm{Mg}}+{2{\rm{H}}}_{2}{\rm{O}}\to {\rm{Mg}}{({\rm{OH}})}_{2}+{2{\rm{H}}}_{2}$$
(5)

The PBAT substrate degrades through the cleavage of ester bonds by hydrolysis, and the XG binder, agarose, and CA electrolyte matrix degrade through the cleavage of glycosidic bonds by microorganism44,47,50,51. The eco-degradation mechanism of the remaining PB and CB has not yet been reported; however, they are reported as eco-friendly materials52,53. Tap water was added daily to maintain circulation. XG binder-based EC gauge and galvanic electrodes swelled and gradually decomposed in water, and the Mg foil-based anode also began to corrode. Simultaneously, the decomposition products leached into the soil, and the oat seeds sprouted normally. After 42 days, most of the EC layer and Mg anode were degraded. Remaining materials were subjected to a degradation test according to the ISO 14855 standard, which is an assessment standard for the degradation test in a typical terrestrial compost environment. Specifically, the PBAT substrate gradually developed a yellowish color due to swelling over time and began fragmenting and decomposing after 73 days. The CA-PMDA PC gel formed a white, rigid porous polymer as the PC electrolyte was replaced by water after 1 day, and began fragmenting and decomposing after 45 days. The PBS-based agarose hydrogel softened over time due to hydrolysis of the matrix and started pulverizing after 25 days. The Mo foil developed a darker silvery-colored oxide layer on the surface after 25 days with gradual thinning, internal cracks were formed at 45 days, and irregular corrosion initiated from the edges, which began accelerated degradation. The W foil exhibited a yellowish colored oxide layer on the surface after 66 days.

Galvanic electrode-based self-powered iontophoretic module

Figure 2 focuses on the characterization and optimization of the Mg–MoO3 galvanic electrode-based iontophoretic module, which serves as a power source for both iontophoretic delivery and electrochromic indication by generating ionic current across the skin and electrical current flow to the electrochromic module. The paste-type MoO3 cathode sustains a full-cell operating voltage of ~1.4–1.8 V vs. Mg, sufficient to drive iontophoretic transport and electrochromic actuation6,8,54,55. At the current density of 200 μA cm−2, the cell delivers an areal capacity > 0.65 mAh cm−2 before reaching the 1.4 V cutoff, which is sufficient to sustain both iontophoresis56,57 and EC switching (Supplementary Fig. 2)58.

Fig. 2: pH control, discharge behavior, and iontophoretic performance of the fully eco-degradable, self-powered ITP–EC patch using a citrate buffer.
Fig. 2: pH control, discharge behavior, and iontophoretic performance of the fully eco-degradable, self-powered ITP–EC patch using a citrate buffer.The alternative text for this image may have been generated using AI.
Full size image

a Comparison of anode gel pH changes with citrate buffer and PBS electrolyte after 2 h of discharge at a current density of 200 μA cm−2 (blue: initial pH, red: pH after discharge). b Discharge profiles of Mg-MoO3 iontophoretic patch on porcine skin using pH 5 0.1 M citrate buffer (blue) and PBS (red) electrolyte-based anode gels at a current density of 200 μA cm−2. c Discharge profiles of Mg–MoO3 iontophoretic patch on porcine skin using a pH 5 0.1 M citrate buffer electrolyte-based anode gel at various current densities (black: 50 μA cm−2, red: 100 μA cm−2, blue: 200 μA cm−2). d Cumulative transdermal drug dose delivered through porcine skin by the Mg–MoO3 iontophoretic patch at various current densities (black: 0 μA cm−2, red: 100 μA cm−2, blue: 200 μA cm−2).

At the Mg–MoO3 galvanic cells operation, the Mg anode exhibits an electrochemical potential of 1.65 V versus the standard hydrogen electrode (SHE) in PBS electrolyte59,60. According to the Pourbaix diagram of the Mg–H2O system, the passivation reaction (Mg + 2H2O → Mg(OH)2 + H2) is thermodynamically more favorable than the corrosion reaction (Mg → Mg2+ + 2e−) under these conditions (1.65 V vs. SHE, pH 7.4)61. Mg(OH)2 releases OH−, elevates local pH, which progressively suppresses long-term current output and may also irritate skin with an alkaline transdermal environment. To mitigate pH elevation and improve skin compatibility, a pH 5 citrate buffer was utilized on the Mg side62,63,64,65. Figure 2a shows the pH variation of PBS and pH 5 citrate buffer electrolytes after 2 h of discharge at a current density of 200 μA cm–2 at iontophoretic operation on porcine skin. The pH of PBS increased from 7.4 to 8.86, whereas the pH of citrate buffer was maintained in a near-neutral environment despite a moderate increase due to Mg + 2H+ → Mg2+ + H2, thereby preserving skin-compatible operating conditions. The initially acidic environment in citrate buffer also enhanced Mg reduction potential, raising the initial discharge voltage to ~1.8 V compared with ~1.5 V in PBS; as pH gradually increased during operation, the discharge voltage converged toward ~1.5 V (Fig. 2b).

Figure 2c shows the discharge profiles of the Mg–MoO3 electrode-based iontophoretic module using a pH 5 citrate buffer electrolyte on porcine ear skin under various current densities. At a current density of 50 μA cm−2, the initial voltage of 1.9 V gradually decreased to 1.75 V after 1 h of iontophoresis. When the current density was increased to 200 μA cm−2, the initial voltage of 1.62 V decreased to 1.43 V after 1 h. Higher discharge current densities reduce the initial voltage and accelerate voltage decay. However, the iontophoretic module maintains a potential sufficient to drive the electrochromic reaction across the range of 50–200 μA cm−2. Overall, the system provides stable and continuous DC output at an operating voltage adequate for both iontophoresis and electrochromic actuation, while retaining architectural simplicity and independence from environmental stimuli compared with self-powered iontophoretic patches based on other types of generators66,67,68,69.

Figure 2d illustrates the cumulative transdermal dose of drug delivered through porcine ear skin over time, measured using a Franz diffusion cell under various discharge current densities. Niacinamide, a compound known for skin-brightening effects and treating psoriasis70,71, was employed as the model drug at a concentration of 0.2 M in the Mg anode electrolyte. Passive diffusion produced a delivery rate of 4.73 μg min−1, which increased to 8.82 and 13.97 μg min−1 at 100 and 200 μA cm−2, respectively. A more than twofold enhancement under low-voltage conditions suggests that electromigration substantially contributes to drug delivery beyond passive diffusion due to skin hydration, in agreement with previous reports6,8,54,55. The higher ionic flux at increased current densities enhanced the transdermal transport of ionized niacinamide, and the resulting increase in cumulative charge can be directly visualized through the accelerated electrochromic reaction, establishing a charge-based link between iontophoretic drug delivery and EC visual indication.

Sequential state visualization enabled by EC gauge

The EC module functions as a charge-correlated visual indication by converting the iontophoretic current into a synchronized gauge-like coloration directly linked to drug delivery. Previous lateral electrochromic indicators offer intuitive visual readout39,40,41,42, but their thin-film formats lack the charge capacity required for iontophoretic operation, which demands sustained currents of 100–1000 μA over several hours2,12. To meet this requirement without increasing device form factor, we implemented a paste-type EC layer capable of supporting sufficient iontophoretic reaction charges. Figure 3a shows the morphology of the paste-type EC layer, which was composed of WO3 nanoparticles, UV-CB, and XG binder. The paste-type EC layer exhibited uniform dispersion of WO3 nanoparticles and UV-CB, as the shear-thinning behavior of XG suppressed particle sedimentation and agglomeration during curing. UV-CB also enhanced the dispersibility in H2O solvent-based WO3/XG pastes72. Figure 3b presents the cyclic voltammograms of the paste-type and thin-film EC layers (~100 nm deposited via magnetron sputtering) under the same scan rate. At –1 V (vs. Ag/AgCl), the paste-type EC layer exhibited a current density of –4.73 mA cm−2, approximately one order of magnitude higher than that of the thin-film EC layer. These results indicate the superior charge storage capacity of the paste-type EC layer required to accommodate iontophoretic reaction charges. Supplementary Fig. 3 displays the coloration change of the paste-type EC layer during EC reaction. In its initial bleached state, the layer appears grayish with a peak reflectance of 26.69% at 463 nm. Upon applying –1.6 V, the EC reaction reduces the reflectance across the visible spectrum to below 4%, with the 600–650 nm range dropping to approximately 1.86%, resulting in a deep blue color. The PB counter electrode shows an adequate supply of the charge required for the EC reaction (Supplementary Fig. 4). And by adopting PC, which offers greater electrochemical stability than water as an electrolyte solvent (Supplementary Fig. 5), side reactions such as metal corrosion and solvent decomposition are effectively suppressed, ensuring that the delivered current is primarily consumed by the electrochromic reaction.

Fig. 3: Design optimization of a gauge-like EC module.
Fig. 3: Design optimization of a gauge-like EC module.The alternative text for this image may have been generated using AI.
Full size image

a Optical and SEM images (inset) of paste-type EC layer. b Cyclic voltammograms of paste-type EC layer (blue) and thin-film EC layer (red) in 1 M NaClO4 in propylene carbonate (PC) (scan rate: 50 mV s−1). c Schematic illustration of gauge-like EC layer structure and the mechanism of EC reaction propagation. The upper panel shows the overall architecture, in which the EC layer, composed of WO3 nanoparticles (NPs)/UV-treated carbon (UV-CB)/ xanthan gum (XG), is positioned laterally adjacent to the W current collector. Electrons initially flow along the W current collector, initiating the EC reaction first at the region near the W current collector. The lower panel shows a magnified view of the propagating EC front within the gauge-like EC layer. As the WO3 NPs undergo EC reaction, their electrical conductivity increases, enabling electron transfer to neighboring unreacted WO3 NPs. Simultaneously, Na+ ions intercalate, driven by the electrical field between the EC layer and the counter electrode, supporting lateral propagation of the EC reaction across the layer. d EC reaction propagation images of the gauge-like EC layer under an applied potential of –1.6 V versus the Prussian Blue (PB) counter electrode. Images were captured at 20 min intervals over 1 h. e Amperometric profiles of the gauge-like EC layer (blue) and a layered EC layer (red) measured 1 mm-thick CA-PMDA organogel electrolyte containing 1 M NaClO4 in PC, under an applied potential of −1.6 V versus the Prussian Blue (PB) counter electrode. f Cumulative reaction charge (blue) and corresponding EC propagation distance (red) of the gauge-like EC layer under an applied potential of −1.6 V. g Correlation between cumulative reaction charge and EC propagation distance of the gauge-like EC layer under applied potentials of −1.6 V (red) and −1.4 V (blue).

To enable spatially resolved, charge-based readout, we adopted a gauge-like EC architecture that triggers electrochromic reactions sequentially. Figure 3c illustrates the design and working principle of the EC gauge. In this structure, WO3 near the current collector accepts electrons first and undergoes the EC reaction. WO3 exhibits increased conductivity during EC reaction as electron injection into the conduction band leads to the formation of mobile polarons73. Such an increase in conductivity facilitates electron transport to nearby unreacted WO3 regions, enabling lateral propagation of the EC reaction across the EC layer. Figure 3d and Supplementary Fig 6 present the progression of cumulative reaction charge over time in the two types of EC structures (layered EC layer and EC gauge) under an applied voltage of –1.6 V. The EC gauge shows a clear lateral advancement of the colored region, allowing each charge state to be visually distinguished. This spatially resolved propagation provides a “gauge-filling” visualization, which enables intuitive tracking of cumulative charge state. In contrast, the conventional layered EC structure, in which the EC layer is placed directly on top of the current collector, exhibits uniform blue coloration across the entire surface within 20 min.

Figure 3e presents the current profiles of both EC layers under an applied voltage of –1.6 V. The EC gauge maintained a nearly constant current (~202 μA at 2 min and ~193 μA at 60 min), whereas the layered EC layer exhibited a high initial current (~746 μA at 2 min) that gradually decreased to ~237 μA after 60 min. The high initial current in the layered EC layer is attributed to the EC reaction occurring throughout the entire volume of the EC material, which is in full contact with the current collector and has a relatively thin film thickness. As the reaction progresses, the number of active WO3 sites diminishes, leading to a gradual decline in reaction current. In contrast, the EC gauge enables the reaction to proceed sequentially across a constant cross-sectional area of unreacted WO3 sites, thereby sustaining a nearly constant current over time. This behavior indicates that electrons transported through the conductivity-enhanced WO3 sites efficiently reach the newly reacting regions, supporting continuous lateral propagation of the electrochromic front. Figure 3f illustrates the correlation between the cumulative reaction charge and the propagation distance of the colored region in the EC gauge. Under constant current, the reaction charge increases linearly with time, and the propagation distance follows a matching linear trend. These results confirm that the gauge-like EC layer enables real-time, visually traceable monitoring of cumulative electrochemical charge. Figure 3g presents the propagation distance of the EC reaction as a function of cumulative charge under different applied voltages. Although the EC layers exhibited different currents (~200 μA at –1.6 V and ~120 μA at –1.4 V, Supplementary Fig. 7), the propagation rates with respect to charge remained similar: 1.18 × 10−2 and 1.15 × 10−2 mm/mC, respectively. This consistency demonstrates that the gauge-like EC layer can reliably visualize skin-impedance-induced variations in iontophoretic current through corresponding changes in EC propagation rate.

Validation of EC readout for quantitative transdermal delivery

A central challenge in transdermal iontophoresis is the lack of integrated, real-time monitoring of delivered dose. To establish the ITP-EC patch as a self-reporting therapeutic system, we validated dose quantification via the electrochromic display using ex vivo iontophoretic delivery experiments performed with a Franz diffusion cell under standard conditions (Fig. 4a and b). The ITP–EC patch was mounted on porcine skin, enabling transdermal drug transport into the receptor reservoir, which was sampled every 10 min. The cumulative dose was determined by extracting and quantifying the drug concentration in the reservoir. In parallel, the galvanic current passing between the iontophoretic module and the gauge EC module was continuously monitored by amperemeter to record the discharged charge, while voltammetry confirmed the voltage stability applied to the EC module. This integrated setup allowed direct correlation between electrochromic response, cumulative charge, and transdermal drug flux.

Fig. 4: Ex vivo porcine skin validation of EC readout for quantitative iontophoretic delivery.
Fig. 4: Ex vivo porcine skin validation of EC readout for quantitative iontophoretic delivery.The alternative text for this image may have been generated using AI.
Full size image

a Schematic of the ex vivo Franz-cell setup on porcine skin with the iontophoretic device connected in series to the WO3 EC gauge, with current recorded between the iontophoretic module and the EC gauge, and voltage recorded across the EC gauge, and sampling from the Franz-cell receptor reservoir. b Photograph of the ex vivo Franz-cell setup on porcine skin. c Electrical signals during self-powered, synchronized operation with the iontophoretic module and EC gauge in series: voltage across the WO3 EC module (black) and discharge current between the EC layer and anode (red). d Correlation between EC propagation distance and cumulative transdermal drug dose (n = 4, r = ~0.966).

Figure 4c shows the electrical performances of the integrated ITP-EC system over time during self-powered operation. The operating voltage was stably maintained at ~1.6 V, which is sufficient to drive the EC reaction of the EC gauge. The discharge current exhibited a brief initial decrease due to electrode polarization, but rapidly stabilized to a steady-state value of ~350 μA. This result shows that, under porcine-skin application conditions, the integrated ITP–EC patch exhibits electrically stable operation by supplying a stable potential sufficient to drive the EC gauge reaction and by delivering a steady galvanic current.

Figure 4d and Supplementary Fig. 8 show the cumulative dose released into the Franz cell as a function of the EC gauge propagation distance. At ~3 mm, the cumulative dose was 209 μg (±7 μg), increasing to ~408 μg (±68 μg) at the final distance of ~10 mm. A linear fit yields a slope of ~36 μg mm−1 with a strong correlation between distance and dose (r = 0.966). Because the same electrochemical current drives both iontophoretic delivery and the electrochromic reaction, the EC readout is coupled to the delivered drug dose, enabling visual indication of drug delivery from the display with high accuracy under biophysical conditions representative of porcine skin.

In-vivo therapeutic effect of ITP-EC patch

To demonstrate charge-based, self-indicating iontophoretic delivery under disease-relevant conditions, we applied the niacinamide-loaded ITP–EC patch to an imiquimod-induced psoriasis model until the EC gauge reached saturation. Psoriasis is a chronic inflammatory disease characterized by erythema, scaling, and thickening of the stratum corneum, which reinforces the skin barrier and increases low-frequency impedance, thereby hindering transdermal delivery transport74,75,76. Because these electrical and structural changes evolve with disease progression and recovery, time-based or fixed-duration operation can yield a variable delivered dose, complicating consistent dose tracing. Niacinamide was selected as a model compound because it has been reported to alleviate psoriasis by attenuating NFκB/MAPK-associated inflammation, modulating NAD+/PARP1 metabolism, and supporting ceramide synthesis and keratinocyte differentiation to improve hydration and barrier function71,77,78. Impedance measurements confirmed that imiquimod-induced psoriatic skin exhibited ~1.4 times higher impedance at 100 Hz than normal skin (0.95 vs. 1.3 kΩ), consistent with barrier strengthening (Supplementary Fig. 9).

The ITP-EC patch loaded with 3.3 wt% niacinamide was applied to psoriasis-induced mice for 7 days, and the application time was defined by EC gauge saturation (EC saturation time). Consistent with the disposable design of the patch, a fresh device was applied for each daily treatment session. For mechanistic validation, cumulative charge was recorded during operation (Fig. 5a and Supplementary Fig. 10), whereas a separate patch configuration without electrical recording was used to demonstrate stand-alone, self-indicating operation (Fig. 5b). Control conditions without iontophoresis used passive diffusion for 30 min with two gels, a 3.3 wt% niacinamide gel suitable for cosmetic use (min. dose)79,80, and a 20 wt% gel representing a clinically permitted maximum concentration (max. dose)79,81,82. EC saturation time measured on days 1, 3, 5, and 7 was ~40, ~45, ~43, and ~30 min, respectively, indicating that saturation time decreased as psoriasis improved and impedance declined (Fig. 5c). Similarly, the cumulative charge versus time graph shows the same trend of an increasing charge rate as the days progress (Supplementary Fig. 11). This behavior illustrates that impedance-dependent current variation is directly transduced into charge-correlated EC progression, enabling skin-adaptive and user-informative operation without external sensors or controllers.

Fig. 5: In vivo therapeutic effect of the ITP–EC patch in a psoriasis mouse model.
Fig. 5: In vivo therapeutic effect of the ITP–EC patch in a psoriasis mouse model.The alternative text for this image may have been generated using AI.
Full size image

a Schematic of the in vivo setup with cumulative charge recording during iontophoretic operation and the daily treatment schedule. A fresh single-use patch was applied each day. Created with BioRender.com. b In vivo photographs of the ITP-EC patch on a mouse. c Change in EC saturation time over the treatment period (n = 3). d Representative images of H&E-stained skin sections from the minimum-dose group (left) and ITP–EC group (right). e Epidermal thickness measured after the treatment period for each group (n = 3). Statistical analysis showed a significant difference among the treatment groups (p < 0.05). f Representative images of TUNEL (red)-stained skin sections from the minimum-dose group (left) and ITP–EC group (right).

We performed histological analyses to evaluate treatment-associated tissue responses after repeated application. We assessed the thickness of the epidermis after sacrificing the mice and collecting skin tissue on Day 8 (Fig. 5d, e and Supplementary Fig. 12). Epidermal thickness in the ITP–EC and maximum-dose groups approached values observed in healthy tissue (~21–25 μm), whereas the minimum-dose group remained thicker (~35 μm), consistent with less advanced recovery. Statistical analysis showed a significant difference among the treatment groups (p < 0.05). H&E staining revealed pronounced hyperplasia, thickened epidermis, elongated rete ridges, and dense inflammatory infiltration in the minimum-dose group. In contrast, the ITP–EC group showed epidermal morphology near normal levels with reduced rete ridges and inflammation, comparable to the changes observed in the maximum-dose group. TUNEL assays supported these observations: apoptotic signals were absent in both the minimum-dose and ITP–EC groups but present in the maximum-dose group, indicating tissue stress associated with higher-dose conditions (Fig. 5f and Supplementary Fig. 13). At the experimental endpoint, blood samples were collected for biochemical analysis (Supplementary Fig. 14). The results demonstrated that key hepatic and renal markers remained within normal ranges across all groups, confirming that the ITP-EC implant induces no systemic toxicity. These observations indicate that the ITP–EC patch delivers niacinamide across psoriatic skin under skin-compatible conditions, while the electrochromic gauge provides on-patch, charge-based indication throughout treatment. By synchronizing iontophoretic delivery and electrochromic indication within a single loop, the patch converts recovery-dependent impedance changes into charge-correlated, on-patch feedback, supporting consistent operation while avoiding tissue stress observed under higher-dose conditions.

This study demonstrates a self-powered, fully eco-degradable iontophoretic patch in which iontophoretic drug delivery and electrochromic visual indication are unified through a synchronized electrochemical loop. The same electrochemical current that drives iontophoretic drug delivery simultaneously actuates the EC gauge, enabling on-patch visual indication that quantitatively reflects delivered dose without external sensors, control electronics, or auxiliary power. This electrochemical integration preserves patch-level simplicity using thin, soft, and paste-type components, supporting flexible and conformal implementation. For the iontophoretic module, Mg-MoO3 galvanic electrodes provide a power source with sufficient capacity to simultaneously drive iontophoretic drug delivery and EC actuation under a skin-compatible environment. For the EC module, WO3-based is architected with a paste-type, gauge-like structure and the PC-based electrolyte, which enables to accept iontophoresis charge and intuitive, charge-correlated visual indication synchronized with drug delivery. Ex vivo and in vivo studies show that EC propagation, cumulative charge transfer, and iontophoretic drug dose remain quantitatively correlated under physiologically relevant and disease-altered skin conditions, supporting charge-referenced dose indication without reliance on predefined time- or current-based protocols. System-level eco-degradability is achieved through materials–architecture co-design, with all functional elements validated as transient under compost-soil conditions. Together, these results define an electrochemically unified strategy that resolves the trade-off among simplicity, dose-informative adaptive operation, and environmental sustainability.

The modular and charge-correlated nature of this platform provides a foundation for extending its functionality toward broader practical use and therapeutic scope. Enhancements in skin fixation through adhesive hydrogels by utilizing adhesive additives, such as polydopamine, can improve usability and mechanical stability83,84. Improving shelf life and long-term stability can be achieved by encapsulation of the patch with potential incorporation of dehydration-mitigation strategies such as hygroscopic additives, low-volatility solvent systems85,86,87,88,89. Beyond the demonstrated model compound, the charge-based electrochromic quantification strategy may be adapted to other ionizable therapeutics through empirically established dose–distance calibration libraries, supporting broader transdermal applicability. Moreover, integration with microneedle-based interfaces may further expand delivery capacity to macromolecules and nanoparticles while preserving charge-correlated visual indication. Collectively, these design extensions highlight the potential of the ITP–EC architecture as a scalable platform for sustainable, intuitive, and impedance-responsive transdermal systems.

Methods

Preparation of UV-treated carbon black

UV-CB was prepared as previously reported72. Briefly, carbon black (Super C65, Nanografi) was dried in a vacuum oven overnight, then the dried powder was treated with UV/ozone plasma for 2 h with shaking at 10 min intervals. The 253.7 nm spectra of UV output at UV/ozone generate (UVC-150, Omniscience Co.) was 20 mW cm−2 with a 5 mm distance.

Preparation of Prussian blue nanoparticles

PB nanoparticles were synthesized as previously reported90. Briefly, 15 g sodium citrate tribasic dihydrate (≥99%, Sigma Aldrich) and 3.34 g iron sulfate heptahydrate (≥99%, Sigma Aldrich) were mixed in deionized water (DIW) 200 ml, labeled as solution A. And 15 g sodium citrate tribasic dihydrate and 3.92 g sodium ferrocyanide decahydrate (≥98%, Sigma-Aldrich) were mixed in another DIW 200 ml, labeled as solution B. Both solutions were stirred under N2 conditions. Solution A was added dropwise to solution B, and the mixed solution was stirred at 800 rpm for 6 h. The precipitate was washed with DIW and ethanol 3 times, respectively, and dried in a vacuum oven at 120 °C for 12 h.

Fabrication of the display with EC gauge

A 100 μm-thick PBAT substrate was prepared by drop-casting a 10% w/w PBAT (S-EnPol, Korea)/chloroform (99.5%, Daejung co.) solution onto glass coated with a trichloro(octadecyl)silane (TOS) monolayer, where the glass was immersed in a TOS (>90%, Sigma-Aldrich) 4.228 μM solution in n-hexane (95%, Samchun co.) at RT for 30 min. A W pad was deposited on the TOS-coated glass, and a 280 μm-thick PI template was prepared beside the W pad with 1 mm overlapping. A WO3 paste was prepared by blending WO3 nanoparticles (<100 nm, Sigma Aldrich) 0.5 g, UV-CB 3.2 mg, XG (Xanthan gum from Xanthomonas campestris, Sigma Aldrich) 8.4 mg, and DIW 2.5 ml using a planetary centrifugal mixer (ARM-310, Thinky mixer) with a condition of 2000 rpm for 6 min. An EC gauge was prepared by casting 50 μl of the paste per 1.1 cm2 and drying it at RT overnight.

A PB paste was prepared by blending PB nanoparticles 0.5 g, UV-CB 50 mg, XG 20 mg, and DIW 1.8 ml using a planetary centrifugal mixer at a condition of 2000 rpm for 6 min. A counter electrode was prepared by screen-printing the PB paste via a 280 μm-thick PI mask on a W foil (50 μm thick, 99.95%, Thermo Fisher Scientific), and drying it at RT overnight. Reduction of the PB counter electrode was conducted by applying −0.2 V compared to Ag/AgCl in a 3 M KCl reference electrode (A111, Corrtest Instruments) after immersing in a NaClO4 1 M PC solution with a Pt counter electrode.

A CA-PMDA organogel was prepared as previously reported91. Briefly, a pre-gel solution was prepared by dissolving cellulose acetate (Junsei Co.) 0.6 g, pyromellitic dianhydride (97%, Sigma-Aldrich) 0.66 g in the NaClO4 1 M PC solution 60 ml for 1 day. The triethylamine (≥99%) catalyst was added dropwise and mixed for 1 min, and the mixture was cast onto a glass Petri dish under N2 conditions. After 1 day, the remaining solution in the dish was replaced with a 1 M NaClO4 PC solution.

Electrochemical stability test

Electrochemical stability test of eco-degradable electrolytes was conducted as follows: A W foil and a Pt wire were prepared as working and counter electrodes, respectively. Ag/AgCl in a 3 M KCl reference electrode was utilized, and all the electrodes were immersed in eco-degradable electrolytes. The current was measured using a potentiostat (ZIVE MP2A, ZIVELAB) by applying a voltage range from 2 to −2 V at 10 mV s−1 scan rates.

Electrochemical stability test of eco-degradable metals in PC electrolytes was conducted as follows: Each eco-degradable metal (Mg, Zn, Mo, and W) foil was connected to the working and counter electrode in a NaClO4 1 M PC solution. The current was measured from 0 to −2.5 V at 10 mV s−1 scan rates.

Cyclic voltammetry test

A paste-type EC layer was prepared by casting 50 μl of the WO3 paste per 1.1 cm2 on a W foil and drying it at RT overnight. A thin-film EC layer was prepared by depositing WO3 onto an indium-tin oxide glass (0.7 mm thick, resistivity ≤ 8 Ω/sq, TMALAB co.) using magnetron sputtering a WO3 target (99.95% purity, Krt lab, Korea) with 100 W power and 15 mTorr working pressure under an Ar/O2 mixture gas 27/3 sccm. A PB counter electrode was prepared using the PB paste method as described above. Each electrode was connected to the working electrode, and a Pt wire was connected to the counter electrode. Ag/AgCl in a 3 M KCl reference electrode was utilized, and all the electrodes were immersed in a NaClO4 1 M PC solution. The current of the paste-type EC layer, thin-film EC layer, and PB counter electrode were measured by applying voltage ranges from −1.5 to 1 V, −1 to 1 V, and −1 to 1 V, respectively, at various scan rates.

Reflectance measurement of the EC layer

The paste-type EC layer was prepared and underwent a sufficient EC reaction by applying −1.6 V compared to the PB counter electrode for 30 min. The reflectance at UV–vis spectra (300–800 nm) was measured using a UV–vis spectroscopy (V-770, JASCO).

EC reaction propagation distance analysis from photographic images

Photographic images of the electrochromic (EC) layer were acquired using a digital single-lens reflex camera. The entire EC layer region was cropped from each original image prior to analysis. The percentage of the EC-reacted area was quantified using HSV-based image analysis.

Digital images in red–green–blue (RGB) format were converted to hue–saturation–value (HSV) color space using a standard RGB-to-HSV transformation, where hue ranged from 0 to 179, and saturation and value ranged from 0 to 255. A target color corresponding to the EC-reacted region was defined by reference HSV coordinates (H*, S*, V*) with channel-wise tolerances (ΔH, ΔS, ΔV). Lower and upper bounds for each channel were determined as H*−ΔH to H* + ΔH, S*−ΔS to S* + ΔS, and V*−ΔV to V* + ΔV, and were constrained within the permissible HSV ranges.

Each pixel in the HSV image was evaluated independently. A binary mask (M) was generated by assigning a value of 1 to pixels whose HSV values satisfied all three channel-wise bounds and 0 otherwise. The total number of pixels meeting the threshold criteria (n_tgt) was obtained by summing the positive entries in the mask. The target-area ratio (ρ) was calculated as ρ = 100 × n_tgt/n_tot, where n_tot denotes the total number of pixels in the image. These steps are summarized in Algorithm 1.

Algorithm 1: Target-color area quantification with HSV tolerance

Input: RGB image I in red–green–blue (RGB) color space;

Target color in hue–saturation–value (HSV) space \({c}^{* }=\left({H}^{* },{S}^{* },{V}^{* }\right)\);

tolerances \(\Delta H,\Delta S,\Delta V\);

total pixel count \({n}_{\text{tot}}={|I|}\)

Output: Target-area ratio ρ (in %);

binary mask M where \({M}_{{ij}}=0\) if pixel \(\left(i,j\right)\) matches the target color

range, else 0

Step 1: Preprocess

Convert I from RGB to HSV color space:

   \({I}_{{\rm{HSV}}}\leftarrow \text{ConvertColorSpace}(I,\text{RGB}\to \text{HSV})\)

// ConvertColorSpace: applies standard \(\text{RGB}\)→\(\text{HSV}\)

   Transformation where \(H\,\in \,[0,179],S,\,V\,\in \,[0,255]\)

Step 2: Define tolerance bounds around \({{\boldsymbol{c}}}^{* }\)

Lower bound vector:

   \({\rm{LB}}=[max(0,{H}^{\ast }-\Delta H),max(0,{S}^{\ast }-\Delta S),max(0,{V}^{\ast }-\Delta V)]\)

Upper bound vector:

    \({\rm{UB}}=[max(179,{H}^{\ast }+\Delta H),max(255,{S}^{\ast }+\Delta S),max(255,{V}^{\ast }+\Delta V)]\)

Step 3: Build a binary mask

For each pixel position \(\left(i,j\right)\) in \({I}_{\text{HSV}}\):

Set \({M}_{{ij}}\leftarrow 1\) if

      \({\text{LB}}_{k}\le {I}_{\text{HSV},k}\left(i,j\right)\le {\text{UB}}_{k}\,\text{for}\,\text{all}\,\text{channels}\,k\in \left\{H,S,V\right\}\)

Otherwise, set \({M}_{{ij}}\leftarrow 0\)

Step 4: Compute area ratio

     \({n}_{\text{tgt}}\leftarrow \mathop{\sum }\limits_{i,j}{M}_{{ij}}\)

     \(\rho \leftarrow 100\times {n}_{\text{tgt}}/{n}_{\text{tot}}\)

Return ρ, M

The EC reaction propagation distance was calculated by multiplying the measured area ratio (ρ) by the total EC layer length of 11 mm.

Fabrication of the eco-degradable galvanic iontophoretic module

A MoO3 paste was prepared by blending MoO3 microparticles (6 μm, 99.9%, US Research Nanomaterials) 0.6 g, CB 25 mg, XG 10 mg, and DIW 0.4 ml using a planetary centrifugal mixer with a condition of 2000 rpm for 6 min. A cathode was prepared by screen-printing the MoO3 paste via a 280 μm-thick PI mask on Mo foil (25 μm thick, 99.95%, Thermo Fisher Scientific), and drying it at RT overnight. An anode was prepared by etching the surface of magnesium foil (100 μm, 99.95%, MTI Co.) in a citric acid 3% v/v solution for 5 min, followed by washing with isopropyl alcohol and drying.

A drug solution was prepared by dissolving 0.2 M of niacinamide (≥98%, Sigma Aldrich) in a pH 5, 0.1 M citrate buffer. A drug gel was prepared by dissolving 1.25% v/v agarose powder (higel-Agarose clearTM, E&S Bio Electronics Co.) in the drug solution at 90 °C, followed by casting onto a glass petri dish and gelation at RT. A buffer gel solution was prepared by dissolving 1.25% v/v agarose powder in PBS 1X (Samchun Co.) at 90 °C. A buffer gel was prepared by casting the solution in a glass Petri dish and gelation at RT.

The device components were assembled on a flexible PBAT.

Discharge curve measurement

The discharge curves of the cathode were conducted as follows: The cathode was connected to the working electrode, and a Pt wire was connected to the counter electrode. Surface-etched Mg was connected to the reference electrode, and all the electrodes were immersed in a PBS 1X solution. The voltage was measured under various discharge current densities.

The discharge curves of the iontophoretic module were conducted as follows: 2 mm-thick drug gel and buffer gel were prepared as described above, and were placed on porcine ear skin with a 5 mm interval. The voltage was measured by flowing various discharge current densities.

pH measurement of the drug gel after iontophoresis operation

Two pH conditions of drug gel were prepared as follows: One was prepared by solution of a pH 5, 0.1 M citrate buffer, and the other was a PBS 1X. Both gel solutions were prepared by dissolving 1.25% v/v agarose powder at 90 °C and were cast onto a glass Petri dish. The gels and electrodes were placed on porcine ear skin, and iontophoresis was operated for 2 h at a current density of 200 μA cm−2. After discharge, the gels were pulverized manually and immersed in DIW at 25 times their volume for 1 day. The pH was measured by a pH reader (F20-Std-Kit, Mettler Toledo), then the original pH was calculated.

Assembly of the iontophoretic patch–electrochromic indicator

The iontophoretic patch and the electrochromic module were electrically connected by filling predefined via-holes in a shared PBAT substrate with a biodegradable conductive paste (W powder, beeswax)46. The Mg anode was electrically connected to the current collector of the WO3 layer, sputtered W, and the current collector of the cathode, Mo foil, was electrically connected to the current collector of the EC counter electrode, W foil, through the paste-filled via-holes.

Drug dose measurement via the Franz diffusion cell setup

A Franz diffusion cell apparatus was customized for iontophoretic module operation. The reservoir was filled with a PBS 1X solution with 400 rpm of magnetic stirring, and the porcine ear skin device was loaded. The iontophoretic module was placed on the skin and operated. To measure the delivery performance of the iontophoretic module, 1 cm2 of electrodes was utilized with various discharge current densities. 300 μl of the solution in the reservoir was extracted every 10 min. To measure the performance of ITP-EC, ~2 cm2 of electrodes were utilized. 300 μl of the solution in the reservoir was extracted every 30 min. Niacinamide concentration was calculated by measuring the absorbance at 262 nm using a microplate reader (Epoch 2, BioTek).

Degradation test based on ISO 14855

The degradation test of materials composing ITP-EC was conducted in soil containing a 1:1 mass ratio of commercial compost and tap water at RT. After degradation of the EC layer, PB counter electrode, and Mg anode, further degradation test of remaining parts (PBAT, CA-PMDA organogel, agarose hydrogel, W foil, and Mo foil) was conducted based on ISO 14855 conditions. Each part was immersed in soil containing a 1:1 mass ratio of commercial compost and tap water at 58 °C. Air circulated with small holes, and DIW was added in an amount equal to the weight of the evaporated water every day. The degradation behavior was measured using a photographic image after washing with DIW and drying at RT for 1 h.

Animal study

All animal experiment protocols were performed according to the laboratory animal use protocol approved by the Institutional Animal Care and Use Committee (IACUC) of Korea Institute of Science and Technology (KIST) (Approval ID: KIST-IACUC-2025-060). Balb/c mice (female, 5 weeks old) were provided by Nara biotech (Seoul, Korea).

Imiquimod-induced psoriasis modeling

To induce psoriasis, the dorsal upper backs of mice were shaved, and a dose of 62.5 mg 5% imiquimod cream (Aldara; 3M Pharmaceuticals, UK) was daily applied to the shaved area for 7 consecutive days. Control animals received an equal amount of Vaseline cream as a placebo treatment.

In vivo efficacy test

After PS modeling, the mice were randomly assigned to four groups: non-treated (control), a passive diffusion group treated with 3.3 wt% niacinamide (minimum dose), a passive diffusion group treated with 20 wt% niacinamide (maximum dose), and an ITP–EC group treated with 3.3 wt% niacinamide using the ITP–EC patch. The BALB/c mice were anesthetized with 2% isoflurane gas before each drug patch application. The minimum- and maximum-dose groups received passive delivery for 30 min daily at the same psoriatic skin site for 7 days, whereas the ITP–EC patch was applied until the electrochromic (EC) layer reached saturation. On days 1, 3, 5, and 7, the current during ITP–EC operation was recorded using a source meter (Keithley, 2450 SourceMeter, USA), and the cumulative charge was calculated by integrating the current over time. The serum was isolated from mice at the time of sacrifice for the assessment of systemic toxicity. Biochemical parameters were analyzed by an external diagnostic laboratory (DK Korea Biotech, Seoul, Korea).

Histological analysis

For the histological analysis, the skin samples from each mouse at the end of the experiment were fixed in 10% neutral buffered formalin and embedded in optimal cutting temperature (OCT) compound. The embedded tissues were sectioned at a thickness of 10 μm. For general morphological assessment, the sections were stained with hematoxylin and eosin (H&E) and imaged using an Olympus IX70 microscope at ×400 magnification. To evaluate apoptotic cell death, terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining was performed using the BrdU-Red TUNEL Assay kit and imaged using a confocal microscope, Olympus FV3000.