Fully Printed Flexible Zinc Oxide Patch for Wearable UV Light Sensing

Metal oxide materials are widely adopted in UV sensing due to their advantages of reliability, stability, and high sensitivity. However, the fabrication of metal oxide films on flexible substrates using micro/nano printing techniques still poses challenges. Notably, inks containing oxide nanoparticles often lead to nozzle clogging during the printing process. In addition, methods relying on precursor solutions typically require high post‐processing temperatures that conventional flexible substrates cannot withstand. In this study, a novel precursor ink based on zinc amine hydroxide complexes is proposed, enabling post‐processing at a low temperature of 250 °C. Through a fully printed approach, a wearable UV sensor based on a zinc oxide nanofilm on a flexible substrate is successfully fabricated. The sensor exhibits outstanding sensitivity and rapid response characteristics. By incorporating silver nanoparticles into the ink, the baseline resistance of the sensitive material is adjusted. Furthermore, a smart patch utilizing the NFC communication protocol is developed, allowing for wireless, passive, and wearable UV detection. In this study, a process paradigm for fabricating metal oxide films on flexible substrates using micro/nano printing techniques is introduced, showcasing significant potential in the field of micro/nano sensor fabrication.


Introduction
UV radiation is an electromagnetic wave with a frequency between visible light and X-rays, covering a wavelength range from DOI: 10.1002/aelm.202300469   10 to 400 nm.In our daily lives, UV rays are almost everywhere.Proper exposure to UV radiation can promote the synthesis of vitamin D in the human body, but excessive exposure can cause a series of acute and chronic damage to the skin. [1]Therefore, there is an urgent need for miniaturized, wearable, and lowcost UV detection devices to provide help for UV protection in our daily lives. [2]urrently, UV sensors primarily based on three categories of techniques: photothermal effect, photoelectric effect, and photochromic effect. [3]Among them, the photothermal effect has limited selectivity for UV wavelengths. [4]Based on the photochromic effect, L'Oréal developed a flexible UV sensing patch in collaboration with its sunscreen cream. [5]However, overall, UV sensors based on the internal photoelectric effect have advantages in terms of reliability, stability, and sensitivity. [6]etal oxide is a commonly used photoelectric material for UV sensors due to its high sensitivity, low cost, durability, and nano-structural diversity. [7]3b,8] However, traditional techniques for fabricating metal oxides on sensor substrates are cumbersome and do not guarantee consistency and stability of the devices, making mass production difficult.Microand nano-printing, a new additive manufacturing method, have been implemented in sensor fabrications due to its high throughput, low cost, simplicity, and eco-friendly. [9]However, the metal oxide nano-inks compatible with flexible substrates become a new challenge for the following two aspects: 1) The particles in the ink should be homogeneously dispersed with small size to avoid clogging the printing nozzle; 2) The post-treatment temperature of the ink should be low to accommodate the poor heat resistance of flexible substrates.Unfortunately, direct dispersion of metal oxide nanoparticles often results in poor homogeneity with particle agglomeration, [10] while most sol-gel based metal oxide precursors require high post-processing temperatures. [11]hese factors impede the board application of metal oxides based on the internal photoelectric effect in wearable and portable UV detection.
In this study, we have proposed a novel method for creating a nano-zinc oxide printing ink without particles, utilizing a precursor consisting of an ammine-hydroxo zinc complex.By lowering the annealing temperature to 250 °C, we have been able to form a ZnO nano-film on a polyimide (PI) substrate.Subsequently, we have fabricated a fully printed flexible UV sensor, which exhibited a rapid and linear response to UV light within the intensity range of 50-2000 uWcm −2 .Furthermore, we have incorporated silver nanoparticles into the ZnO ink, which led to the creation of a fully printed ZnO UV flexible sensor with adjustable resistance.In addition, we have designed a flexible patch based on NFC (Near Field Communication) protocol to detect the sensor resistance and interact with a smartphone. [12]By integrating the UV sensor into this patch, we have achieved a wireless, passive wearable system for detecting UV intensity, and demonstrated the reliability and potential of the patch in wearable UV detection.

Results and Discussion
Traditional methods, such as chemical vapor deposition, produce films with high purity and strong substrate adhesion, but their fabrication process involves high temperatures that are incompatible with flexible devices.Transfer printing has been explored to create sensitive films on flexible substrates, yet this approach intricates and demands precise handling, making it unsuitable for large-scale production.On the other hand, methods like vacuum evaporation and sputtering for pattern customization require mask templates, leading to material wastage, and environmental concerns.We have employed micro and nano printing for sensor modification primarily due to its straightforward process, maskless pattern customization, and efficient utilization of raw materials.The concept diagram of this study is shown in Figure 1.Utilizing a fully printed process, we have fabricated a nano-film of zinc oxide, along with fork-shaped electrodes and an NFC communication patch, which enabled wireless and passive wearable UV detection.
To fabricate a zinc oxide nano-film on a flexible substrate, we prepared a precursor solution of zinc ammine-hydroxo complex (Ink 1).Concurrently, to compare the difference in annealing temperature with the traditional sol-gel method precursor solution, we also prepared a sol-gel-based zinc oxide precursor solution (Ink 2), as illustrated in Figure S1.(SupportingInforma-tion) FTIR analysis (Figure S2, Supporting Information) revealed that Ink 1 can form a relatively pure zinc oxide film at 250 °C, whereas the sample using Ink 2 still contained numerous impurities.Only at 500 °C could a pure zinc oxide film be obtained using Ink 2. This disparity can be attributed to the difference in reaction kinetics between the two methods.The decomposition of the zinc ammine-hydroxo complex is a rapid process that does not require a high annealing temperature. [13]Moreover, the preparation route for Ink 1 does not involve organic components that necessitate high-temperature decomposition and removal.
In Figure S3 (Supporting Information), it can be observed that during the entire evaporation process, the changing rate of the side cross-sectional area of the droplets of the unsaturated solution is significantly smaller than that of the saturated solution, which confirmed the effectiveness of reducing the concentration of precursor solution in enhancing its stability.
Figure S4 (Supporting Information) shows that both the asdried and annealed samples exhibit a prominent Zn-O peak at 420 cm −1 , indicating that the initial reaction of precursor decomposition into ZnO has occurred during the drying process at 150 °C. [14]However, the as-dried sample shows peaks at 891 and 1595 cm −1 , which can be attributed to −NH 2 groups, suggesting the presence of partially decomposed zinc ammine hydroxo complex or a small amount of unevaporated ammonia. [15]eaks at 1041 and 1083 cm −1 , corresponding to C─O stretching vibrations, indicate the presence of residual unreacted ethylene glycol. [15]The broad peak ≈3400 cm −1 is attributed partly to residual ethylene glycol and partly to hydroxyl groups of the zinc hydroxide. [16]The peak at 1383 cm −1 indicates the presence of residual NO3 − .However, in the annealed sample's FTIR spectrum, most of the aforementioned impurity peaks, except for the hydroxyl peak from surface adsorption on ZnO, are almost absent.This demonstrates the necessity of the final step of annealing at 250 °C, which effectively removes the impurities.The temperature resistance of polyimide typically ranges from 260 to 350 °C, so annealing at 250 °C does not cause damage to the substrate.However, when considering other flexible substrates, such as PET, it would be prudent to explore the possibility of reducing the necessary annealing temperatures in future studies.Recent advancements in ZnO nanostructure growth, driven by techniques like localized laser annealing [17] or resistive micro/nano heating, [18] offer a promising avenue for addressing this limitation.Notably, the success of solution-deposited and laser-annealed ZnO NP FETs has sparked our inspiration to explore alternative methods of achieving annealing. [19]he nano-film printed for three layers and annealed is shown in Figure 2a. Figure 2b demonstrates that before annealing at 250 °C, some regions of the film have formed spherical nanoparticles with sizes ranging from 100 to 500 nm. Figure 2c shows that after annealing, the diameter of the ZnO nanoparticles is smaller, reaching < 100 nm, and the size distribution is more uniform.Correlating this with the findings from the FTIR testing (Figure S4, Supporting Information), it can be observed that the reaction of precursor decomposition and the formation of ZnO nanoparticles are not fully complete during the drying process.Consequently, the particulate matter in Figure 2b aggregates a significant amount of zinc hydroxide rather than elemental ZnO, leading to larger particle sizes.Figure 2d is a SEM image at a lower magnification, revealing a mesh-like structure of the ZnO nanofilm, suggesting that the film has a porous rather than dense structure, thus having a larger surface area.The smaller nanoparticles and larger surface area are beneficial for the interaction of the thin film material with UV light and environmental oxygen.Figure 2e presents the XRD result of the ZnO film, indicating the presence of the hexagonal wurtzite structure of ZnO. [20]Figure 2f,g, which are TEM images, corroborate the findings from XRD.
Figure S5 (Supporting Information) presents the results of the UV-vis spectroscopy test.It can be observed that the film exhibits high absorption rate and selectivity toward UV light, particularly in the UVA wavelength range.Further calculation reveals a bandgap width of 3.06 eV, which is slightly lower than the typical value for bulk ZnO materials.This difference may be attributed to the defects of ZnO, [21] which caused by various factors such as the fabrication route, annealing temperature. [22]These defects can alter the electronic structure by producing localized resonant states that result in bandgap reduction.
The device with printed silver electrodes and nanoparticle film is shown in Figure 3a.The I-V curve in Figure 3b demonstrates that the dark current of the device is only 0.66 nA in the absence of UV light, leading to a high photocurrent-to-dark current ratio.Both I-V curves with or without the absence of UV light demonstrate straight lines, indicating that the contact between the ZnO sensing layer and the silver electrode is Ohmic contact. [23]For ntype ZnO, it is generally desirable to form Ohmic contact with a metal whose work function is slightly lower than its electron affinity.The electron affinity of ZnO is 4.2 eV, while the work function of Ag is 4.5 eV, slightly higher than the electron affinity of ZnO.However, the printing process and subsequent heat treatment may lead to mutual diffusion and tunneling effect at the interface between Ag and ZnO, resulting in the formation of Ohmic contact. [8,24]Therefore, the device exhibits characteristics similar to a photosensitive resistor.
The calibration results of the sensor are shown in Figure 3c,d, where the light-to-dark current ratio of the device at an intensity of 2000 uWcm −2 is 155, indicating a good signal-to-noise ratio of the device.
Based on the calibration curves in Figure 3c, we calculated the response and recovery time [25] of the sensor to different intensities of UV light, as shown in Figure 4a.The device exhibits slower response speed when exposed to UV light with intensities lower than 100 uWcm −2 .However, as the intensity of UV light increases, the response speed of the device does not continuously increase but gradually reaches a relatively stable range.
The sensitive mechanism of ZnO to UV light primarily relies on the internal photoelectric effect, which can be described by the following reaction equations: Equation ( 1) describes the adsorption of oxygen from the environment in the absence of UV light, where the adsorbed oxygen captures free electrons in the surface layer of ZnO, forming oxygen anions. [26]The consumption of free electrons on the surface gives rise to the formation of a high-resistance depletion layer on the ZnO surface.This effect is notably accentuated in nanofilms due to their expanded surface area, facilitating a higher oxygen adsorption capacity.Consequently, the resultant ZnO material demonstrates low dark current.Furthermore, the layer of oxygen anions triggers the generation of a positively charged space charge layer near the ZnO surface and establishes an electric field directed from the inner layers toward the surface.Equations ( 2) and (3) describe the formation of photo-generated carrier pairs, where electrons, as majority carriers in n-type semiconductor ZnO, increase the device's conductivity.On the other hand, photo-generated holes move along the direction of the electric field near the material surface and combine with oxygen anions, causing desorption of oxygen releasing free electrons in the surface layer of ZnO, which also leads to increased current in the device and contributes to the separation of photogenerated carrier pairs.The extension of carrier lifetimes raised the photocurrent as well.
Equations (1-3) represent first-order chemical reactions, where the reaction rate is proportional to the concentration of reactants. [27]Therefore, the response rate of the sensor is mainly determined by the concentration of the surface adsorbed oxygen ions and the photogenerated carriers.When the UV intensity is very low, the concentration of the generated photocarriers is low, and the increase in UV intensity can produce more carriers, thus speeding up the response speed of the sensor.When the inten-sity of UV light is further increased, the generation of carrier is no longer a rate-limiting step, and the response speed of the device tends to be stable.
The device exhibits the fastest response speed when exposed to 600 uWcm −2 of UV light, with response and recovery times of 1.15 and 4.22 s, respectively, as shown in Figure 4b,c.Figure 4d shows the light response curves under bias voltages of 10 and 5 V.The device under 10 V bias voltage shows a significantly faster response speed, while the recovery speed shows no significant difference.This is because the applied external electric field enhances the extraction efficiency of carriers, thereby reducing the carrier transit time and resulting in a shorter response time.Since the recovery process of the device's current is independent of the applied electric field, the recovery time shows no significant difference under different bias voltages.This study adopted 10 V bias voltage, primarily due to the comparatively larger baseline resistance of the thin film, which correlates with the thickness of the sensitive film.Compared to drip or spin-coating methods, the printed film tends to be thinner.In addition, this reduced thickness enhances the stability of the thin film's performance during stretching or bending, offering a significant advantage for flexible devices.
3a,b,8,28] The proposed device demonstrates significant advantages in in terms of response speed.Combined with morphology characterization analysis, it is possible that the moderate porosity of the thin film allows for moderate oxygen adsorption on the device's surface, facilitating faster adsorption and desorption compared to dense ZnO films and highly porous ZnO films.
Figure 4e,f and Table S2 (Supporting Information) demonstrate the flexibility and stability of the sensor through bending and long-term stability tests, validating its reliability in practical applications.In Figure 4e, both bending states maintained the original response characteristics of the device.However, compared to the response in the flat state, the device's response characteristics showed minimal changes when bent along direction A, while bending along direction B resulted in a 4.3% decrease in photocurrent and a more noticeable slowing of the response speed.This indicates that the internal stress non-uniformity of the sensor becomes more pronounced when bent along an axis perpendicular to the tines' direction, leading to a more significant reduction in the adhesive performance between the sensing layer and the electrode layer.This effect could be attributed to lateral stretching of the tines.Therefore, it is advisable to bend the device along an axis parallel to the tines' direction in practical applications.
To reduce the baseline resistance of the sensor, commercial silver nanoparticle ink was added to the precursor ink.SEM images (Figure S6, Supporting Information) show that the silver nanoparticles are evenly dispersed in ZnO film.The peaks in the XRD spectrum (Figure S7, Supporting Information) confirmed the presence of hexagonal wurtzite structure of ZnO and the silver particles.The diffraction peaks of ZnO remain sharp and narrow, indicating that the addition of silver ink does not affect the crystalline quality of ZnO during post-processing.Figure S6b (Supporting Information) illustrates the presence of two sizes of particles in the silver-doped film.By comparing with Figure 2d,it can be discerned that the particles in the hundred nanometer range correspond to ZnO, while particles in the tens of nanometers range correspond to silver.Therefore, Figure S6b (Supporting Information) showcases the uniform distribution of both these particle sizes within the film.The resistance response of the film to UV light before and after silver doping is shown in Figure S8 (Supporting Information), and it can be observed that the baseline resistance of the silver dropped film is significantly reduced.Figure S9

Conclusion
In summary, our research introduces a novel metal oxide nano ink system capable of producing nanostructured films on flexible polyimide substrates at reduced post-processing temperatures.We have successfully demonstrated the excellent performance of the nanofilm in UV sensing applications.Moreover, we have achieved the fabrication of a wearable smart UV sensor patch through a fully printing-based approach.Most importantly, our research offers a versatile method for fabricating metal oxides on flexible substrates through the utilization of micro-and nano-printing techniques.This approach highlights the significant potential of metal oxides in wearable and portable devices, making them more accessible and practical for everyday use.

Experimental Section
Materials: Zinc nitrate hexahydrate, sodium hydroxide, ammonium hydroxide (25% in water), ethyl alcohol, Zinc acetate dihydrate, and isopropanol were purchased from Shanghai Hushi Co., Ltd.Ethylene glycol and Ethanol amine were purchased from Aladdin Inc.All chemical reagents were used directly without further purification.Ultra-pure water (18 MΩ) was produced by Millipore Direct-Q.
Preparation of Precursor Solutions: Sodium hydroxide (1 g) was dissolved in deionized water to prepare a 2.5 mol L −1 sodium hydroxide aqueous solution.Ammonia solution (25 mL) with a concentration of 25% was taken and diluted with deionized water to obtain a 6.6 mol L −1 dilute ammonia solution.Zinc nitrate hexahydrate solid (4.46g) was added to 15 mL of deionized water and stirred at a speed of 600 rpm until the solid was completely dissolved.While stirring, the prepared sodium hydroxide aqueous solution was added to form zinc hydroxide precipitate.Subsequently, the sample was centrifuged at a speed of 5000 rpm for 5 min, and the precipitate was retained.Deionized water was added to the centrifuge tube and this step was repeated for five times.The repeated centrifugal washing reduced the content of Na + and NO3 − in the precipitate and minimized their influence on the electrical properties of the ZnO final product.
Once the final centrifugation was complete, the precipitate was removed and placed on a hot plate, heated at 60 °C for drying.The dried precipitate was added to the prepared dilute ammonia solution.It was placed on a magnetic stirrer and continuously stirred at a speed of 600 rpm for 6 h to facilitate its dissolution.
During the dissolution of Zn(OH) 2 precipitate in ammonia, it reacts with ammonia to form zinc ammine-hydroxo complex ions.The supernatant obtained after centrifugation served as the precursor solution for ZnO.The equations for the reaction process are as follows [13a] : Zn(OH) 2 (s) +xNH 3 (aq) → Zn(OH) 2 (NH 3 ) x (aq) The undissolved zinc hydroxide (Zn(OH) 2 ) precipitate was weighed after drying in a vacuum drying oven at 60 °C.The amount of undissolved portion was calculated, allowing the determination of the concentration (0.12 mol L −1 ) of Zn 2+ in the precursor saturated solution.Due to the relatively poor stability of the saturated solution, additional ammonia solution was added to adjust the precursor solution to a final concentration of Zn 2+ at 0.06 mol L −1 .
In addition, a sol-gel-based zinc oxide precursor solution was prepared to compare the difference in annealing temperature with the traditional sol-gel method precursor solution.The preparation process is shown in Figure S1b (Supporting Information).
The Fourier-transform infrared spectroscopy (FTIR, IrAffinity, Shimadzu) analyses of the two samples was conducted using the pellet pressing method.For the sample prepared by the sol-gel method, annealing was performed at both 500 and 250 °C.As for the sample prepared from the zinc ammine-hydroxo complex precursor, annealing was conducted at 250 °C.
To confirm the improved stability of the zinc ammine-hydroxo complex solution at reduced concentrations compared to its saturated solution, images of the contact area of droplets from the two solutions were captured using a contact angle meter (SL150E, Kino, USA).The images were collected every 10 s for 150 min.The normalized values of the change in the lateral cross-sectional area of the droplets were calculated from the acquired images using Image J software (NIH, USA).The lateral crosssectional area of the droplets was directly proportional to their volume, allowing for an indirect reflection of the rate of change in droplet volume.
Printing Process of the ZnO Nano-Film: The precursor solution was prepared to form an ink compatible with micro-nano printing.To attain optimal printing outcomes, crucial factors encompass appropriate viscosity, solvent stability, uniform solute dispersion, and low volatility.Moreover, the surface energy between the ink and the substrate needed to be matched. [29]During post-processing, the components of the ink did not decompose into impurities that were difficult to remove.
The resolution of the patterned printing was contingent upon factors such as the size of the printing nozzle, ink viscosity, and the compatibility of surface energies between the ink and the substrate.In this study, a 50 μm nozzle for ink printing using a ultrasonic resonance micro-nano deposition system (Sonoplot Microplotter) was utilized.Typically, it was necessary to ensure that the maximum particle diameter of the ink was less than 1/10 of the nozzle.However, the ink used in this study features a particle-free formulation, eliminating concerns of nozzle clogging.A thermally resistant polyimide (PI) film was chosen as the printing substrate, with a surface energy of 42 mNm −1 . [30]Ethylene glycol (viscosity: 16.1 mPa•S) was used to adjust the viscosity of the ink, while isopropanol (surface energy: 21.7 mNm −1 ) was employed to reduce the surface energy of the ink.The final ink formulation was prepared with a ratio of "precursor solution: ethylene glycol: isopropanol = 5:4:1".The physicochemical properties of the precursor solution and the optimized printing ink are shown in Table S3 (Supporting Information).The ink demonstrated excellent long-term stability, as shown in Figure S10 (Supporting Information).
During the drying process, methods such as direct drip coating or dispensing often result in the formation of undesirable coffee rings.In contrast, high-precision patterned micro-nano printed films (Figure 2e) were typically thinner and exhibited greater uniformity, thereby promoting consistency in device performance during large-scale production.The samples were first dried at 150 °C for 10 min to evaporate the solvents.At this stage, a white semi-transparent film became visible.If multiple layers were printed, the transparency of the film would decrease.After drying, the samples underwent annealing by placing them in a muffle furnace and ramping up the temperature at a rate of 5 °C min −1 .The samples were then kept at 250 °C for 2 h.The FTIR analyses of the samples after drying and annealing were performed respectively.
The nanoscale morphology of the film surface before and after annealing was confirmed using scanning electron microscopy (SEM, SEM-SU-8010, and Hitachi).The crystal structure of the ZnO samples was characterized and verified through X-ray diffraction (XRD, Ultima IV, and Rigaku) testing and High-Resolution transmission electron microscopy (HRTEM, Titan Themis, and FEI) analysis.The optical properties of the printed ZnO film were characterized using a UV-vis spectrophotometer, and the bandgap width was then calculated using Tauc Plot method.
Fully printed ZnO Sensor and its Sensing Properties: Using Sonoplot Microplotter, a conductive silver ink (PeJet-AgINK600, Yixin Kechuang) was first printed on a PI substrate to create a finger-shaped electrode structure (Electrode size: 8 × 8 mm, Interfinger width: 400 μm, spacing: 300 μm).The topological structure and image of the finger-shaped electrode are shown in Figure S11a,b (Supporting Information).SEM images (Figure S11c,d, Supporting Information) demonstrate a smooth and dense surface morphology.Subsequently, three layers of ZnO nanofilms (3800 × 3800 μm) were printed and dried on top of the electrode, followed by annealing.
A UV sensing test system was set up under indoor conditions without ambient UV light.A 365 nm UV Led (Guanghong Tech) was chosen as the light source, with a rated voltage ranging from 3.4 to 3.8 V.It was powered by a DC power supply (2260B-30-36, Keithley), and the output light power was adjusted by varying the voltage.A UV power meter (LH-127, Lianghuicheng) was used to calibrate the UV light intensity.A source meter (B2900B, Agilent) was used to measure the device current and obtain the I-V curve.
The I-V curves of the sensor were measured with and without 400 uW cm −2 UV light illumination, respectively (Figure 3b).Then, the sensor response to different intensities of UV light were tested.In each test, the UV light source was turned on and continuously illuminated the sensor for 30 s.The response current curve of the sensorwas recorded under 10 V voltage excitation.The tested light intensity ranged from 10 to 2000 μW cm −2 .In addition, the response time and recovery time of the sensor were calculated from the response curves.
Under the irradiation of 100 uW cm −2 UV intensity, the response curves of the sensor were measured in three states: flat, horizontally bent and vertically bent with a bending radius of 5 mm.The sensor response curves were measured and normalized.
The sensor was stored in a dust-free environment, and the response curves were measured and normalized on the 0th and 30th day after fabrication, respectively.
NFC Based Smart Patch for Wearable UV Sensing: To reduce the intrinsic impedance of ZnO nanofilm, 16.9 mg of silver ink (PeJet-AgINK600, Yixin Kechuang) was added to 10 mL of ZnO ink, resulting in a mass ratio of 25 wt% of silver to ZnO.The mixture was sonicated for 20 min to ensure uniform mixing.The optimized ink and SEM image of the silverdoped ZnO film were shown in Figure S5 (Supporting Information).Furthermore, the silver-doped ZnO film was characterized using XRD.The impedance response to UV light was tested for both the silver-doped and undoped ZnO films for a comparison.
To enable communication and power transfer between the patch and a smartphone, the NFC protocol was utilized.The resonant frequency of the coil circuit was 13.56 MHz: By simplifying Equation ( 6), LC = 137.8(with the unit of L as μH and the unit of C as pF).Typically, the reasonable range for the inductance was considered to be within 1-4 μH.
The HFSS software was employed for simulating the design of the NFC antenna.The substrate was made of 60 μm thick polyimide, and the coil material was copper foil with a thickness of 18 μm.The single-sided winding was in a rectangular shape ( Figure S12a, Supporting Information).Based on the simulation results, the coil was determined to have seven turns with a line width and spacing of 0.14 mm.The outer dimensions of the coil were 27 × 11 mm.As shown in Figure S12b,(Supporting Information) the inductance at the frequency of 13.56 MHz was measured to be 1.83 μH.According to Equation 3, the total capacitance of the circuit should be 75.3pF.The inductance of the antenna at its self-resonant frequency was 0.50 μH, which allowed to calculate the parasitic capacitance of the antenna as 4 pF.The NFC chip selected was the NTAG213 from NXP Semiconductors, with an equivalent capacitance of 50 pF.Therefore, the required tuning capacitance in the circuit should be 21.3 pF, and a capacitor of 22 pF was chosen for the actual PCB design.
The circuit schematic of the NFC detection patch can be found in Figure S13 (Supporting Information).The patch consisted of the NFC tag minimum system, rectification module, and DC load module.A Schottky diode 1N5819 was used to convert the AC signal generated by the NFC coil into a DC voltage with an effective value of 6 V. Subsequently, the UV sensor was connected in parallel with an LED, and then in series with a resistor.The brightness of the LED, indicated by the current magnitude, provided a visual representation of the resistance value of the UV sensor.
When the resistance value of the sensor droped below a certain threshold, the LED could not illuminate, indicating that the corresponding UV light intensity exceeded the limit.The relationship between the series resistor and the sensor resistance threshold can be found in Figure S14 (Supporting Information).The threshold values for UV light intensity were stored in the NFC chip and could be accessed and modified through a smartphone app ( Figure S15, Supporting Information).

Figure 1 .
Figure 1.Schematic illustration of the fully printed flexible ZnO patch for wearable UV light sensing.

Figure 2 .
Figure 2. Characterization of the ZnO nano-films.a) Photo of the nano-film printed on a PI substrate and annealed at 250 °C.b) SEM image of the ZnO nano film before annealing.c,d) SEM image of the ZnO nano film after annealing.e) XRD of the ZnO nano-film.f,g) TEM image of the ZnO nano film.

Figure 3 .
Figure 3. Calibration of the ZnO sensor to UV light.a) image of the ZnO sensor.b) I-V curve of the sensor with/without UV light conditions.c) Responses of the sensor to different intensities of UV light with the excitation voltage of 10 V. d) Calibration curve of photocurrent of the sensor to UV light.

Figure 4 .
Figure 4. Performance tests of the ZnO sensor.a) Response and recovery time under 10 V voltage excitation to different UV light intensities.b) response time and c) recovery time of the device to 600 uW cm −2 UV light.d) Response of the device to 350 uW cm −2 UV light under 10 and 5 V excitation.e) Response of the sensor under regular or 45°bending states.f) The sensor responses to 100 uW cm −2 UV light on days 0 and 30.
(Supporting Information) demonstrates the response of the silver-doped devices under 2000 μWcm −2 UV irradiation, indicating increased sensitivity compared to pure ZnO films, albeit with a slower response time.As shown in Figure 5, an intelligent NFC tag was fabricated using flexible PCB technology.Integrating the sensor into the NFC tag.The intensity of UV light can be indicated by the on/off state or light intensity of the LED integrated on the patch, which enabled wearable applications.

Figure 5 .
Figure 5. Fully printed smart patch for UV light sensing.a) Image of the smart patch.b) The patch is attached to the wrist and connects to the smartphone via NFC.c) Local close-up of (b).d) The curves of sensor resistance and LED current under different intensities of UV lights.e) The LED states of points A and B marked in (d).