Skin‐Adhesive, ‐Breathable, and ‐Compatible Nanopaper Electronics for Harmonious On‐Skin Electrophysiological Monitoring

On‐skin electronics, which offers an interface for extracting electrophysiological signals from skin, is intensively investigated using electrodes mounted on flexible substrates. Despite numerous efforts toward substrate design to optimize user comfort, substrates with skin‐adhesion, skin‐breathability, skin‐compatibility, mechanical endurance, sterilizability, sustainability, and biodegradability remain desirable candidates for human‐ and environment‐friendly on‐skin electronics. To this end, a wood‐derived cellulose nanofiber paper (denoted nanopaper) with customized porous nanostructures is developed in this study. The customized porous nanopaper enables water‐assisted deformation for skin‐conformability, thereby realizing outstanding skin‐adhesion force, along with high skin‐breathability and compatibility, superior to those of conventional substrates reported for on‐skin electronics. By mounting gold electrodes on the porous nanopaper and adhering them to human skin, the real‐time monitoring of electroencephalogram, electromyogram, and electrocardiogram for diagnosing the human physiological state is successfully achieved. Furthermore, the gold‐electrode‐mounted porous nanopaper affords unique characteristics including durability against skin deformation, reusability, and even sterilizability, owing to its high mechanical endurance, and thermal stabilities. Thus, the as‐prepared porous nanopaper serves a fascinating platform for human‐ and environment‐harmonious on‐skin electronics.

biodegradability, skin-adhesion, -breathability, and -compatibility are considered highly desirable candidates for harmonious onskin electronics with humans and the environment.
To meet these requirements, herein, the use of a woodderived cellulose nanofiber paper substrate is proposed. Cellulose nanofiber paper, denoted nanopaper, has many attractive characteristics, including abundance, sustainability, biodegradability, [17,18] high optical transmittance, [19][20][21][22] high mechanical toughness [19,23] and flexibility, [21,22] high thermal stability, [19,20,22] high organic solvent resistance, [24] and high surface smoothness, [21,22] making it an ideal substrate material for green and flexible electronics. Although these nanopaper substrate characteristics are fascinating, the development of nanopaper-based on-skin electronics is challenging owing to the limited progress inducing skin-breathability and -adhesion properties. For electronic applications, the surface of nanopaper substrates must be sufficiently smooth to mount electrodes and transfer electric signals effectively. [25,26] However, nanopapers with very smooth surfaces have a densely packed structure, which limits their breathability [20,27] and hinders on-skin electronics applications. More importantly, the contact with the skin surface significantly influences the performance of on-skin electronics. [28] Nevertheless, to the best of our knowledge, the skin-adhesion of nanopaper remains unexplored. Therefore, a nanopaper substrate that possesses skin-adhesion and -breathability, while affording electrode mounting for the effective transfer of electrical signals must be designed.
In this study, a skin-adhesive and -breathable nanopaper with customized nanostructures was successfully fabricated. The nanopaper with customized porous nanostructures provides water-assisted skin-conformability and water vapor permeability, which result in high skin-adhesion force and skin-breathability, respectively. The customized nanopaper also has skin-compatibility and allows the mounting of gold (Au) electrodes; this in turn allows the on-skin measurement of electrophysiological signals, including electromyography (EMG), electrocardiography (ECG), and electroencephalography (EEG). The customized nanopaper exhibits readhesive ability for repetitive use even after mounting Au electrodes, owing to its sufficient mechanical endurance. Furthermore, the thermal stability of the customized nanopaper promotes sterilization for safe use. Because of these characteristics, the as-prepared nanopaper serves as an ideal platform for human-and environmentharmonious on-skin electronics.

Customizing Porous Structures for Skin-Adhesive Nanopaper
Cellulose nanofibers with a width of 22 ± 8 nm, [29] which were acquired by the aqueous counter-collision [30] of never-dried softwood bleached kraft pulp, were used for the preparation of the nanopapers. First, nanopaper with a structure similar to that of conventional nanopapers [19,20] was prepared by vacuum filtration of the water suspension of cellulose nanofibers (Figure 1a). The resulting nanopaper was referred to as "dense nanopaper" because it had a densely packed structure ( Figure 1b) with a thickness of 13.7 ± 0.78 µm, a bulk density of 1.36 ± 0.08 g cm -3 , and a porosity of 13.3% ± 4.9%. To test its skin-adhesion capabilities, the dense nanopaper was adhered to sweaty skin Figure 1. Preparation process, porous structure, and skin-adhesion of the nanopapers. a) Preparation processes, photographs, and b,c) field-emission scanning electron microscopy (FE-SEM) images of the dense and porous nanopapers, respectively. d) Pore size distribution, e) shear adhesion force after 1 min of attachment, and f) change in normalized shear adhesion force of nanopapers during attachment/detachment cycles.

www.advmatinterfaces.de
temporarily. However, the dense nanopaper did not remain attached to the skin once after skin deformation; thus, it was found to be unsuitable for on-skin electronics applications.
Hereupon, we attempted to improve the skin-adhesion by customizing the nanostructures within the nanopaper. The densely packed structure of the dense nanopaper was attributed to the agglomeration of cellulose nanofibers caused by water retention before the drying process, [31] as water has a high surface tension (72.14 mN m -1 at 25 °C). [32] To prevent agglomeration, a solvent exchange process was conducted, through which the residual water was replaced by tert-butyl alcohol (t-BuOH), which has lower surface tension (19.96 mN m -1 at 25 °C; [32] Figure 1a). The resulting nanopaper, denoted "porous nanopaper" because of its porous nanostructures (Figure 1c), has a larger thickness (25.3 ± 1.47 µm), lower density (0.68 ± 0.03 g cm -3 ), and higher porosity (56.9% ± 2.1%) than the dense nanopaper. The porous nanopaper was translucent, while the dense nanopaper was transparent, although they have similar total light transmittances of 87.0% and 89.8% at 550 nm, respectively. The difference in appearance was due to the increased light scattering within the porous nanopaper, which also led to a higher diffusion light transmittance of 65.3% at 550 nm compared to that of dense nanopaper (38.7%). Then, the surface areas and pore size distributions of the porous and dense nanopapers were analyzed by Brunauer-Emmett-Teller (BET) and Barrett-Joyner-Halenda (BJH) models. As shown in Figure 1d and Figure S1 in the Supporting Information, the porous nanopaper has a wider pore size distribution (pore size <100 nm) and a higher specific surface area (170.5 m 2 g -1 ) than the dense nanopaper (pore size <20 nm, specific surface area: 3.0 m 2 g -1 ), demonstrating the presence of porous nanostructures within the porous nanopaper. Notably, the porous nanopaper adhered firmly to sweaty human skin. It adhered to the skin for more than 3 h and resisted more than 100 cycles of deformation of the human forehead, indicating its suitability for applications in the field of on-skin electronics.
In addition, the shear adhesion force between the nanopapers and the skin surface, which is important for withstanding skin deformation during on-skin electrophysiological signal measurement, was quantitatively measured ( Figure S2a, Supporting Information). Before the measurement, the nanopapers were attached to the human skin, and the moist conditions of the attached nanopapers and the skin surface were stabilized by pressing with a wet cotton sheet for ≈10 s to eliminate individual and environmental differences. The moisture content of the dense nanopaper, the porous nanopaper, and the skin surface was 26.9% ± 0.8%, 76.4% ± 1.1%, and 86.0% ± 2.7%, respectively ( Figure S3, Supporting Information). The nanopapers were swiftly adhered to the skin after the stabilization treatment. Although the shear adhesion force of the dense nanopaper was 1.27 ± 0.17 N cm -2 after 1 min of attachment on the skin surface (Figure 1e), the dense nanopaper dropped off from the skin after ≈10 min. On the other hand, the shear adhesion force of the porous nanopaper was 2.30 ± 0.60 N cm -2 after 1 min of attachment ( Figure 1e) and remained at 1.30 ± 0.34 N cm -2 after 3 h. The adhesion force values were positively comparable to those of previously reported substrates for on-skin electronics (Table S1, Supporting Information), thereby affording long-term skin-adhesion. In addition, a 90° peel adhesion force was also evaluated to investigate the performance during the intentional detachment of the nanopapers by peeling from the skin ( Figure S2b, Supporting Information). Both the dense and porous nanopaper exhibited low peel adhesion force (0.01 ± 0.005 N cm -2 and 0.02 ± 0.014 N cm -2 for the dense and porous nanopaper, respectively). Thus, the porous nanopaper provided high skin-adhesion capabilities for onskin electronics, and could be gently peeled from human skin, which could prevent skin trauma, such as stripping, irritation, and allergic reactions, which are frequently caused by aggressive adhesives. [33] Additionally, the porous nanopaper allowed readhesion for more than ten iterations of attachment to and detachment from the skin without decreasing its shear adhesion force (Figure 1f), demonstrating significant reusability.

Mechanisms of Skin-Adhesion for Porous Nanopaper
As described above, the nanopaper with the customized porous nanostructures exhibited better skin-adhesion than the dense nanopaper. The putative mechanisms of the superior skinadhesion properties of the porous nanopaper are discussed as follows.
First, the effective water-assisted deformation of the porous nanopaper provided preferable skin-conformability, which can contribute to the improvement in actual contact area with the skin surface. To investigate the skin-conformability, the nanopapers were adhered to a skin replica by pressing with a wet cotton sheet and then observed through an optical microscope. The skin replica had a rough surface with micro-sized wrinkles, whereas the nanopaper had a relatively smooth surface ( Figure S4a-c, Supporting Information). The dense nanopaper did not conform to the rough skin replica surface with microsized wrinkles (Figure 2a). In contrast, the porous nanopaper established conformal contact with the skin surface (Figure 2b,c and Figure S4d,e, Supporting Information). These results indicated that the porous nanopaper undergoes deformation along the micro-sized skin wrinkles, unlike the dense nanopaper. Both porous and dense nanopapers showed sufficient flexibility enough for bending and folding ( Figure S4f,g, Supporting Information), ensuring the contact with skin. However, the porous nanopaper underwent deformation more easily because it has a lower elastic modulus and higher water absorption properties than the dense nanopaper. According to a previous report, the elastic modulus of nanopaper with higher porosity is lower, due to the lower density interfiber bonding between cellulose nanofibers. [23] Moreover, the porous nanopaper can effectively absorb water (such as sweat and water from the skin surface and a wet cotton sheet, respectively; Figure S3, Supporting Information), which further decreases the elastic modulus of the porous nanopaper because the water absorbed in the interfiber interphase region reduces interfiber stress transfer. [34] Thus, despite being thicker (25.3 ± 1.47 µm) than the dense nanopaper (13.7 ± 0.78 µm), the porous nanopaper provided superior water-assisted skin-conformability owing to its low elastic modulus, which led to higher skin-adhesion.
Second, the porous structures would strengthen the liquidinduced attractive capillary interaction between the porous nanopaper and the skin, contributing to the strong skin-adhesion.

www.advmatinterfaces.de
When the nanopapers were attached to the skin, capillary force can be generated owing to the formation of liquid bridges [35] between the nanopapers and the skin. To make the capillary force attractive, the nanopapers should have a high affinity for the liquids on the skin, such as water and sebum. [36] The amphiphilic nature of cellulose molecules [37,38] in the nanopaper contributed to the generation of attractive capillary force. The total attractive capillary force depends on the number of liquid bridges between the nanopapers and the skin. [36,39] Whereas the dense nanopaper would form a small number of liquid bridges, as its packed surface (Figure 1b) tends to aggregate the liquid bridges (Figure 2d), the porous nanopaper, which has isolated cellulose nanofibers (Figure 1c), can form a large number of isolated liquid bridges ( Figure 2e). The porous nanopaper would also prevent the aggregation of liquid bridges while retaining an adequate amount of liquids around the isolated cellulose nanofibers, because its porous structure can absorb excess liquids from sweat or secretion and transfer them effectively to the external environment by the capillary effect. [40] Thus, a large number of isolated liquid bridges form within the structure of the porous nanopaper, thereby providing a strong interaction with skin.
Overall, the porous nanopaper would provide water-assisted deformation for skin-conformability and liquid-induced attractive capillary interactions with the skin, achieving strong skin-adhesion. The porous nanopaper with the water-assisted deformation showed much stronger skin-adhesion (shear adhesion force: 2.30 ± 0.60 N cm -2 ) than that without water-assisted deformation (0.18 ± 0.04 N cm -2 , Figure S5, Supporting Information), further indicating the significance of water-assisted deformation. Once water-assisted deformation occurred within the porous nanopaper, its shear adhesion force was kept at a sufficient level (1.30 ± 0.34 N cm -2 ) even after 3 h, suggesting that the liquid-induced attractive capillary interactions with the skin sufficiently remained for long-term skin-adhesion. On the dried skin surface, abundant accessible hydroxyl groups within the porous nanopaper also contribute to skin-adhesion via hydrogen bonding with proteins in the skin. [41,42]

Skin-Breathability and Skin-Compatibility of Porous Nanopaper
To confirm its applicability to on-skin electronics, the skinbreathability of the nanopapers was evaluated by measuring their water vapor transmission rate (WVTR), as shown in Figure 3a. Figure 3b shows that the weight of water vapor transmitted through a polyethylene terephthalate (PET) film almost remained unchanged during the measurement, indicating that the conventional plastic substrates for on-skin electronics have limited skin-breathability. In contrast, the WVTR of the dense nanopaper was 2103 ± 254 g m -2 d -1 , which was higher than that of the human skin (204 g m -2 d -1 ). [43] The porous nanopaper, which had an even higher WVTR (2912 ± 101 g m -2 d -1 ), owing to its higher porosity, demonstrated ideal skin-breathability for on-skin electronics.
To determine skin-compatibility, skin patch tests were carried out. The nanopapers were affixed on the skin of 20 human subjects for 24 h. The condition of the skin was then examined by an experienced dermatologist and compared with skin before nanopaper attachment. As shown in Table S2 in the Supporting Information, no inflammation was observed in any of the 20 human subjects. Accordingly, both nanopapers exhibited skin-compatibility that is favorable for on-skin measurement, possibly resulting from the high WVTR of the nanopapers and the biocompatible nature of cellulose. Thus, the porous www.advmatinterfaces.de nanopaper exhibited a combination of properties that are essential for applications in on-skin electronics, such as skin-adhesion, skin-breathability, and skin-compatibility.

Applications for Electrophysiological Signal Measurements
To verify the applicability of the nanopaper to on-skin electronics, pattered gold (Au) electrodes were applied to the porous nanopaper to prepare a porous-nanopaper-based electrode (Figure 4a), as Au has shown good skin-compatibility for onskin electronics. [5,44] A thermal evaporation procedure was used to deposit Au on the nanopapers, which is a traditional coating method that can form high-purity coating layers on different substrates. [45] Before on-skin measurements of electrophysiological signals, Au was deposited over the entire surface of the nanopapers (denoted Au-deposited porous nanopaper and Audeposited dense nanopaper) using the same thermal evaporation procedure to estimate the influence of Au deposition on skin-breathability, skin-adhesion, and electrically conductive properties.
To optimize the amount of deposited Au, the sheet resistance was measured against the thickness of the deposited Au layer. As shown in Figure S6a in the Supporting Information, the sheet resistance for both the Au-deposited porous and dense nanopaper significantly decreased with increasing Au thickness from 12 to 15 nm. A limited decline in sheet resistance was achieved by further increasing the Au thickness. Therefore, 15 nm was chosen as the Au thickness for further discussion. The surface structures of the Au-deposited side were partly maintained in the Au-deposited porous and dense nanopaper ( Figure 4b and Figure S6b, Supporting Information), while those of the undeposited side remained almost unchanged ( Figure S6c,d, Supporting Information). This result indicated that the nanopapers have good structural stability against the thermal evaporation procedure for the Au deposition. As a result, the WVTR of the Au-deposited nanopapers was at the same level as that of nanopapers prior to Au deposition; the WVTRs of the Au-deposited porous and dense nanopapers were 3118 ± 192 and 2308 ± 254 g m -2 d -1 , respectively ( Figure S6e, Supporting Information). The higher WVTR after Au deposition possibly resulted from heterogeneous wettability in the Audeposited nanopapers, which would facilitate water penetration from the hydrophobic Au layer to the hydrophilic nanopaper layer. [46,47] The shear adhesion force of the porous nanopaper decreased from 2.30 ± 0.60 to 1.39 ± 0.19 N cm -2 after Au deposition, whereas that of the dense nanopaper decreased from 1.27 ± 0.17 to 0.64 ± 0.03 N cm -2 (Figure 1e and Figure S6f, Supporting Information). Although the adhesion force decreased after Au deposition, only the patterned parts of the nanopaperbased electrodes were deposited with Au ( Figure 4a). Therefore, the partial decrease in the adhesion force would not significantly influence the skin-adhesion of the nanopaper-based electrodes. The electrically conductive properties after Au deposition were then evaluated. The sheet resistance of the Audeposited porous nanopaper was 561 Ω square -1 , and that of the Au-deposited dense nanopaper was 12.4 Ω square -1 . Although the Au-deposited porous nanopaper had a higher sheet resistance, its normalized contact impedance measured at 10 Hz and 0.04 µA (36 kΩ cm 2 ) was lower than that of the Au-deposited dense nanopaper (46 kΩ cm 2 ) immediately after adhering to the human skin, owing to the outstanding skin-conformability of the porous nanopaper. Furthermore, the normalized contact impedance of the Au-deposited dense nanopaper immediately increased beyond the measurement range after 5 min of attachment to the skin (Figure 4c) because of detachment from the skin, which hinders its applicability to electrography signal measurements. In contrast, the Au-deposited porous nanopaper maintained sufficient normalized contact impedance for more than 180 min after adhering to the skin, revealing its potential for long-term on-skin measurement of electrography signals. Thus, the Au-deposited porous nanopaper exhibited sufficiently high normalized electrode/skin contact impedance for effectively transferring electric signals when used in on-skin electronics.
Furthermore, the porous-nanopaper-based electrode with the Au-deposited electrode pattern was deployed for actual on-skin measurements of multitype electrophysiological signals. First, the porous-nanopaper-based electrode was adhered to a forearm for EMG signal acquisition. The amplitude of the EMG signal increased when the forearm contracted and decreased when it relaxed (Figure 4d). This result demonstrates the successful acquisition of EMG signals, which provide information related to nerve and muscle health for muscle status monitoring [48] as well as sport and rehabilitation data analysis. [49] Next, the porous-nanopaper-based electrode was adhered to the chest for ECG acquisition (Figure 4e); the signals, which offer detailed information about the ventricles and atria activity for cardiac disease detection, were successfully acquired. [50] Lastly, the porous-nanopaper-electrode was www.advmatinterfaces.de adhered to the human forehead for EEG acquisition, which is probably the most challenging measurement for on-skin electronics owing to the low signal amplitude. [51] Figure 4f showed the amplitude of the EEG signals captured during eye opening and closing. The characteristic change in alpha wave (α wave) peak was observed during eye closing and opening, [52] which was similar with that acquired from the commercial gel electrode ( Figure S7, Supporting Information). The normalized contact impedance of the Au-deposited porous nanopaper was further decreased by applying moisturizing cream to the human skin before attachment, which allowed low noise EEG signal acquisition even after 180 min of attachment using the wireless instrument according to a previous report. [53] The EEG signals, which provide information related to cognition and memory performance, [54] emotion recognition, [55,56] as well as neurological diseases like epilepsy [57] and Alzheimer's disease, were successfully acquired. [58] These results confirm that the porous-nanopaper-based electrode can adhere to different human parts for multitype electrophysiological signal acquisition which are in a similar amplitude region as in former studies, [59] proving its applicability to on-skin electronics.

Durability against Skin Deformation, Reusability, and Sterilizability
To further demonstrate the benefits of the porous nanopaper, the durability against skin deformation, reusability, and sterilizability of the porous-nanopaper-based electrode were verified.
The porous-nanopaper-based electrode demonstrated outstanding durability against skin deformation. The normalized contact impedance was maintained on a human forehead after 100 deformations of the skin (Figure 5a), whereas the densenanopaper-based electrode peeled off from the skin after one deformation. This phenomenon resulted from the superior adhesion of the porous nanopaper, which provided stable monitoring of electrophysiological signals even with muscle movements during actual use. Moreover, applying moisturizing cream to the human skin before porous-nanopaper-based electrode attachment further decreased the normalized contact impedance from 294 to 66 kΩ cm 2 , without negatively affecting the durability against skin deformation. The durability against skin deformation was further verified by testing the resistance of the porous-nanopaper-based electrode over 1000 times of bending-unbending cycles as shown in Figure 5b. The change in resistance was less than 1.9% and 0.9% when the bending radii was 2 and 4 mm, respectively, indicating that the resistance had little change over multiple bending-unbending cycles. To make the porous-nanopaper-based electrode stretchable, the introduction of kirigami structures to the nanopaper [60] would be effective.
The porous-nanopaper-based electrode also had sufficient mechanical endurance for repeated use after attachment to and detachment from the human skin. Using the porous-nanopaper-based electrode, EMG signals were acquired successfully after ten cycles of attachment and detachment (Figure 5c, see also Figure S8, Supporting Information). Also, the WVTR of the porous-nanopaper-based electrode after detachment was at the same level as the original one at 2473 g cm -2 d -1 ( Figure S9a, Supporting Information), demonstrating persistent skin-breathability.
The durability of the porous-nanopaper-based electrodes against sterilization treatment was confirmed by applying two convenient and effective methods for sterilization: immersion in EtOH for 2 d and heating at 160 °C for 2 h. [12] In both cases, no significant change in the WVTR was observed ( Figure S9a, Supporting Information). As shown in Figure 5d, after immersion in EtOH for 2 d, the normalized contact impedance of the porous-nanopaper-based electrode significantly increased after 5 min, possibly because the skin-conformability of the porous nanopaper decreased as its porosity decreased from Figure 5. Advantages of the porous-nanopaper-based electrode for on-skin electrophysiological signal monitoring. a) Normalized electrode/skin contact impedance of the porous-nanopaper-based electrode versus the number of deformations of the human forehead, demonstrating its durability against skin deformation during actual use. Application of moisturizing cream to the skin surface can further decrease the normalized contact impedance without negatively influencing the durability against skin deformation and skin-adhesion. b) Change in resistance of the porous-nanopaper-based electrode during 1000 bending-unbending cycles with bending radii of 2 and 4 mm. The inset shows the detailed change in resistance within 490-500 bending-unbending cycles. c) Reusability demonstrated by EMG data acquisition after the attachment and detachment of the porous-nanopaperbased electrode to and from the skin, respectively, for one, five, and ten cycles. d) Sterilizability shown by long-term normalized electrode/skin contact impedance of the porous-nanopaper-based electrode on the moisturizing-cream-applied skin before and after sterilization by heating at 160 °C for 2 h or immersion in ethanol for 2 d followed by overnight storage at 23 °C and a relative humidity of 50%.

www.advmatinterfaces.de
56.9% ± 2.1% to 43.7% ± 2.7% after immersion in EtOH. The decrease in the porosity could be due to the partial aggregation of cellulose nanofibers after drying ( Figure S9b, Supporting Information), [32] owing to the higher surface tension of EtOH (21.93 mN m -1 at 25 °C) than that of t-BuOH (19.96 mN m -1 at 25 °C). On the other hand, the porous-nanopaper-based electrode after heating at 160 °C for 2 h demonstrated a similar normalized contact impedance to that of the original porous-nanopaper-based electrode, owing to its maintained porous nanostructures (porosity: 55.6% ± 4.6%, Figure S9c, Supporting Information). These results indicate that the heat treatment is effective for sterilizing the porous nanopaper.
Furthermore, the porous-nanopaper-based electrode was used repeatedly for the EMG signal acquisition after at least ten cycles of attachment and detachment, as shown in Figure 5c. Its normalized contact impedance after 180 min of attachment on skin was successfully measured and remained similar before and after sterilization heat treatment, as shown in Figure 5d. These results corroborate that the Au layer firmly combined on the surface of the porous nanopaper during long-term use.
Owing to the good skin-compatibility of Au [5,44] and nanopapers (Table S2, Supporting Information), the porous-nanopaperbased electrode showed no adverse effect on the human skin during at least ten cycles of attachment and detachment as well as at least 180 min of attachment.
In summary, the porous-nanopaper-based electrode exhibits desirable properties for applications in on-skin electronics, including stable and comfortable monitoring of electrophysiological signals, repeatability, and safe usability. Owing to a combination of excellent skin-adhesion, -breathability, and -compatibility, the porous nanopaper is a promising substrate to realize harmonious on-skin electronics with human and environment (Table S1, Supporting Information).

Conclusion
A porous nanopaper was fabricated and its applicability for onskin electronics was verified. Owing to the presence of the customized porous nanostructure, the porous nanopaper exhibited the excellent skin-adhesion that conventional dense nanopapers hardly achieved. The conformal but reversible adhesion of the porous nanopaper to human skin resulted from its waterassisted skin-conformability and liquid-induced attractive capillary interaction with the skin. Moreover, the porous nanopaper had high skin-breathability and -compatibility, paving a way for on-skin nanopaper electronics. After mounting an Au electrode on the porous nanopaper, multitype electrophysiological signals including those of EMG, ECG, and EEG could be successfully acquired. In addition, because of the high mechanical endurance as well as thermal stability, the porous-nanopaperbased electrode demonstrated durability against skin deformation, reusability, and sterilizability. The porous nanopaper can be prepared from abundant and sustainable bioresources and exhibits biodegradability. Because the porous nanopaper provided many desirable features for on-skin electronics, it could be a promising substrate toward human-and environmentharmonious on-skin electronics.

Experimental Section
Fabrication of Porous and Dense Nanopapers: Cellulose nanofibers of 22 ± 8 nm width, [29] which were prepared according to previously published methods, [30,31] were used for the fabrication of the nanopapers. Briefly, a water suspension of the cellulose nanofibers was obtained by treating a water suspension of never-dried softwood bleached kraft pulp with a high-pressure water-jet system (Star Burst, HJP-25080, Sugino Machine Co., Ltd, Uozu, Japan) equipped with a counter-collision chamber. To prepare porous nanopapers, the water suspension of the cellulose nanofibers (0.1 wt%, 90 mL) was dewatered by vacuum filtration using a hydrophilic polytetrafluoroethylene (PTFE) membrane filter (H020A090C, 0.2 µm pore diameter; Advantec Toyo Roshi Kaisha, Ltd., Tokyo, Japan). Then, 100 mL of t-BuOH (>99.0% purity, Tokyo Chemical Industry Co., Ltd., Tokyo, Japan) was poured in and gently filtered. The resulting wet sheet on the membrane filter was placed on a glass plate, covered with another PTFE membrane filter and a mixed cellulose ester membrane filter (A020A090C, 0.2 µm pore diameter; Advantec Toyo Roshi Kaisha, Ltd., Tokyo, Japan) and finally sandwiched between paper towels (Kimtowel White, NIPPON PAPER CRECIA CO., LTD.). The porous nanopaper was obtained by hot-press drying of the wet sheet (0.35 MPa and 110 °C for 30 min), followed by peeling of the membrane filters. The dense nanopaper was prepared without the t-BuOH treatment.
Evaluation of Porous Structures: The porous structures of the nanopapers were observed by field-emission scanning electron microscopy (FE-SEM, SU-8020, Hitachi High-Tech Science Corp., Tokyo, Japan) at an accelerate voltage of 2 kV. The nitrogen adsorptiondesorption isotherms and pore size distribution curves based on the BET and BJH models were obtained using a surface area and pore size analyzer (NOVA 4200e, Quantachrome Instruments, Kanagawa, Japan). The porosity values of the nanopapers were calculated using the following equation [61] Porosity % 1 100 where ρ b corresponds to the bulk density of the nanopapers and ρ t is the true density of the cellulose nanofibers obtained from softwood kraft pulp, which is 1.57 g m -3 according to a previous study. [62] Evaluation of Skin-Adhesion, -Breathability, and -Compatibility: The shear adhesion force between the human skin and the dense or porous nanopaper was evaluated by lap-shear adhesion measurements according to previously published methods. [63,64] Before analysis, the nanopaper was cut into a rectangle shape (≈1 × 3 cm 2 ). Then, the nanopaper was attached to the back of a male volunteer's hand with an attaching area of 1 × 1 cm 2 , while being pressed with a wet cotton sheet (FC Eye Cleaning Cotton, Hakujuji Corp., Tokyo, Japan) using a finger for ≈10 s to stabilize the moisture conditions of the nanopaper and the skin surface ( Figure S3, Supporting Information). The moisture contents of the dense or porous nanopaper after the stabilization treatment were calculated from the weight difference between the dried and stabilized states. The moisture content on the human skin surface after stabilization was measured by a skin moisture sensor (MoistSense, MORITEX Co., Ltd., Saitama, Japan). After 1 min and 3 h of attachment, the nanopaper adhered on the skin was pulled by a universal testing machine (EZ-SX, Shimadzu Corp., Kyoto, Japan) at an angle of 180° to the skin and a speed of 60 mm min -1 as shown in Figure S2a in the Supporting Information. The shear adhesion force was obtained by dividing the maximum load by the attaching area (1 × 1 cm 2 ). The 90° peel adhesion force was also measured by pulling the adhered nanopaper at an angle of 90° to the skin ( Figure S2b, Supporting Information).
For the optical observation of the skin-conformability of the nanopapers, a template for a skin replica (size: ≈1.5 × 1.5 cm 2 ) was first prepared by applying the mixture of a two-component silicone rubber and catalyst system (Silflo, Flexico Developments Ltd., Potters Bar, UK) to the surface of a volunteer's skin. Then, a water suspension of agar (Hayashi Pure Chemical Ind., Ltd., Osaka, Japan) was hardened www.advmatinterfaces.de on the template to fabricate a skin replica. The dense or porous nanopaper (size: ≈0.5 × 0.5 cm 2 ) was attached to the skin replica (size: ≈1.5 × 1.5 cm 2 ) while pressing with a wet cotton sheet using a finger. 3D optical images of the nanopapers attached to skin replicas were captured using an optical microscope (DM6 M, Leica Microsystems Inc., Wetzlar, Germany). The surface of the skin replica and the nanopapers before and after attachment to and detachment from the skin replica was also observed (VHX-700F, Keyence Corp., Osaka, Japan).
The WVTR values of the nanopapers were evaluated according to the ASTM E96 standard. [8] A glass bottle (internal diameter: Φ18 mm, height: 30 mm) containing 5 g of distilled water was sealed with each nanopaper using polyimide tapes and then placed in an environmental test chamber (SH-642, Espec Corp., Osaka, Japan) at a temperature of 35 °C and a relative humidity of 40%. The mass change of the water was monitored at intervals of 10 min for 3 d using a balance (FX-300i, A&D Co., Ltd., Tokyo, Japan). The WVTR was calculated based on the mass change.
The skin-compatibility of the nanopapers was measured and approved by the Ethics Committee for Human Test of Life Science Laboratories, Ltd., Osaka, Japan (Approval number: 21H-0049). For the evaluation of skin-compatibility, patch test chambers (diameter: 8 mm; Finn chamber Aqua, SmartPractice Japan Corp., Kanagawa, Japan) were used. The nanopapers and a lint cloth (negative control sample) were installed in the chambers and applied to the backs of 20 human subjects for 24 h. After removing the chambers at 1 and 24 h, the skin condition of each subject was assessed by an experienced dermatologist.
Monitoring of Electrophysiological Signals: For electrophysiological signal monitoring, Au electrodes of 15 nm thickness and 1 cm diameter were deposited on the nanopapers by thermal evaporation using the shadow mask technique at a rate of 0.08 nm s -1 . To optimize the thickness of Au layer, the thermal evaporation time was changed to provide Au-deposited nanopapers with different Au thicknesses of 7.5, 10, 12, 15, 18, and 20 nm. The sheet resistance of the Au-deposited nanopapers was then measured by a four-probe method using a surface resistivity meter (Loresta GP T610; Mitsubishi Chemical Analytech Co., Ltd., Kanagawa, Japan). The Au-deposited nanopapers were attached to the skin surface by pressing with a wet cotton sheet. Before attachment, the skin was wiped with alcohol-soaked cotton wool.
The normalized electrode/skin contact impedance of the Au-deposited nanopapers with a contact area of 1 × 1 cm 2 was measured using an assembled system at 0.04 µA and 10 Hz according to the three-electrode method. [65] The reference electrode (conductive paste, TEN20, Weaver and Co., Aurora, USA) was placed on the forehead near the temple, the counter electrode (conductive paste) was located near the mastoid process, and the working electrode (the Au-deposited nanopapers) was placed on the forehead near the brow. Each of the three electrodes had a diameter of 1 cm, and each conductive paste specimen was mounted on an Ag-based disk electrode.
A wireless module [65] featuring a multichannel analog-to-digital converter, Bluetooth low-energy communication system, and lithium-ion battery for sampling at 250 Hz was connected to the porous-nanopaperbased electrode. The electrophysiological signal at one location was obtained from one pair of measuring and reference electrodes. ECG and EMG were obtained from a pair of the porous-nanopaper-based electrodes (distance between the Au electrodes: 2 cm). For ECG acquisition, moisturizing cream (Locobase, Daiichi Sankyo Healthcare, Co. Ltd., Tokyo, Japan) was applied to the skin surface before the attachment of the porous-nanopaper-based electrodes. The EEG was obtained using a porous-nanopaper-based electrode placed on the forehead and a reference electrode placed on the mastoid process. The methods for EMG, ECG, EEG, and impedance measurement were approved by the Osaka University Research Ethics Committee and complied with the research guidelines of Osaka University (Approval number: 31-2-2). Informed consent from all participants was also obtained prior to the experiment.
Real-time monitoring of the electrical resistance of the porousnanopaper-based electrode was performed during repeated bending at different bending radii (2 or 4 mm) using a digital multimeter (34461A; Keysight Technologies Inc., California, USA). A universal testing machine (EZ-SX, Shimadzu Corp., Kyoto, Japan) was used to bend the electrodes and release them back to their initial positions.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.