Performance Boosting of Flexible ZnO UV Sensors with Rational

Dec 14, 2015 - Two orders of magnitude speed enhancement in reset time is achieved by ... After a total 3000 cycles bending test, the decay in the res...
1 downloads 0 Views 3MB Size
Subscriber access provided by CMU Libraries - http://library.cmich.edu

Article

Performance Boosting of Flexible ZnO UV Sensors with Rational Designed Absorbing Antireflection Layer and Humectant Encapsulation Heng Zhang, Youfan Hu, Zongpeng Wang, Zheyu Fang, and Lianmao Peng ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.5b09093 • Publication Date (Web): 14 Dec 2015 Downloaded from http://pubs.acs.org on December 27, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Performance Boosting of Flexible ZnO UV Sensors with Rational Designed Absorbing Antireflection Layer and Humectant Encapsulation Heng Zhang, 1 Youfan Hu, 1,* Zongpeng Wang, 2 Zheyu Fang, 2 Lian-Mao Peng1,* 1

Key Laboratory for the Physics and Chemistry of Nanodevices, and Department of Electronics, Peking University, Beijing 100871, China 2

School of Physics, State Key Lab for Mesoscopic Physics, Peking University, Beijing 100871, China

*To whom correspondence should be addressed, Email: [email protected], [email protected]

Abstract Flexible ZnO thin film UV sensors with three orders of magnitude improvement in sensitivity and two orders of magnitude acceleration in speed are realized via light absorption efficiency enhancement and surface encapsulation. Devices are constructed on polyethylene substrate incorporating morphology controlled ZnO nanorod arrays (NRAs) as absorbing antireflection layers. By adjusting the morphology of ZnO NRAs, the light absorptance exceeds 99% through effectively trapping incident photons. As a result, the sensitivity of the UV sensor reaches 109000. Moreover, a mechanism of competitive chemisorption between O2 and H2O at oxygen vacancy sites is proposed to explain the phenomenon of the speed acceleration in moist environment. A new approach of humectant encapsulation is used to make H2O participant rapid processes dominant for speed acceleration. Two orders of magnitude speed enhancement in reset time is achieved by polyethylene glycol encapsulation. After a total 3000 cycles bending test, the decay in the responsivity of UV sensor is within 20%, indicating good mechanical stability. All these results not only demonstrate a simple, effective and scalable approach to fabricate high sensitive and fast response flexible ZnO UV sensors, but also provide meaningful references for performance boosting of photoelectronic devices

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

based on other oxide semiconductors.

Keywords ZnO, nanorod, nanowire, absorbing antireflection, humectant encapsulation, UV sensor

1. INTRODUCTION Recently, nanostructured antireflection (AR) layer is widely introduced in photovoltaic (PV) devices for efficiency enhancement by increasing light absorption.1-6 The construction of the devices can be divided into two groups by different purposes of the AR layer. When the material used for the AR layer is different from the active material in the PV devices,1, 3 the AR layer decreases reflection and enhances transmission into the active area of the device by providing a gradient refraction index profile. When these two materials are the same,4-7 this AR layer decreases reflection and also works as light absorbing medium, and is usually called absorbing AR layer. Comparing to the first construction, the devices in the second group can reduce the material consumption and eliminate the scattering and trapping issues at the interface of two different materials. The latter is important for achieving high performance electronic devices. However, to construct these nanostructures, such as in silicon based PV devices, techniques such as electron-beam lithography, interference lithography, and reactive-ion etching have to be adopted in the top-down fabrication process,8-10 which greatly increases the cost of devices, the complexity of fabrication process, and depresses the performance as a large amount of defects are introduced. So this design is not well-accepted in PV devices. Most recently, solution-grown ZnO nanorod arrays (NRAs) provide a new bottom-up method to construct the moth-eye nanostructure in large areas.11-13 Considering light absorption efficiency is a big issue in the realization of high sensitive photosensors,14 adopting the absorbing AR layer design with easy growth process in ZnO UV sensors is very promising, but related work has not been reported yet. ZnO is an attractive UV sensing material with a wide, direct bandgap of 3.3 eV.15 However, slow response and recovery speed is considered as an intrinsic drawback in ZnO UV sensors as slow oxygen molecular adsorption and desorption are involved in the photoresponse process.16-18 Surface passivation has been used to eliminate the amount of the oxygen adsorption sites to accelerate the

ACS Paragon Plus Environment

Page 2 of 23

Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

speed.19-20 But the cost is a reduced sensitivity. This is because that a hole trapping process on ZnO surfaces related to oxygen adsorption can result in a high photoconductive gain.21 There is more work needs to be done to recheck the underlying mechanism, and find a new way to solve the problem essentially, pushing the technology of ZnO based UV sensors toward practical use. In this work, rational designed absorbing AR layer and humectant encapsulation are cooperated in flexible ZnO UV sensors for greatly enhancing sensitivity and accelerating speed. UV sensors are fabricated on polyethylene (PET) substrates with solution-grown ZnO NRAs constructing the absorbing AR surface structure. By adjusting the morphology of ZnO NRAs, the reflectance on the surface of UV sensors is suppressed to less than 1% at the wavelength of 365 nm, and 99% of the incident UV light can be effectively absorbed in this layer. The sensitivity of the UV sensors can be enhanced by three orders of magnitude after introducing ZnO NRAs, reaching 109000. A mechanism of competitive chemisorption between O2 and H2O at oxygen vacancy sites is proposed to explain the phenomenon of the speed acceleration in moist environment. Based on this, a new method of humectant encapsulation is used to make H2O participant rapid processes dominant, achieving two orders of magnitude acceleration in recovery speed of ZnO UV sensors. Good mechanical stability is revealed by less than 20% responsibility decay after 3000 cycles bending tests. A simple, effective and scalable approach is demonstrated in this work for fabricating high sensitive and fast response flexible ZnO UV sensors. It also provides meaningful references for performance boosting of photoelectronic devices based on other oxide semiconductors.

2. RESULTS AND DISCUSSION The fabrication process for the flexible UV sensor is shown in Figure 1a. At the first step, Ti/Au (5 nm / 25 nm) interdigitated electrodes are patterned on the PET substrate by using a shadow mask. The width and spacing of the electrodes are 2000 µm and 1000 µm, respectively. Then, a 75 nm thick ZnO thin film is deposited on the patterned electrodes using a radio frequency magnetron sputtering deposition system. After finishing this step, a conventional metal-semiconductor-metal UV sensor with planar structure based on ZnO thin

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

film is achieved. In order to construct the absorbing AR layer with subwavelength nanostructures, the planar UV sensor is attached on glass slide, and immersed in an equal molar ratio aqueous solution of zinc nitrate and hexamethylenetetramine (HMTA) to grow ZnO NRAs on the surface of the film. The reaction is kept at 95℃ for 2 h. Figure 1b shows the photograph of an as-fabricated UV sensor covered with ZnO NRAs. The bending of the device reveals its good mechanical flexibility. It should be noted that the whole fabrication process does not require any complex micro-nanofabrication technologies such as photolithography, etching, etc. So the processing costs are reduced to a great extent. To optimize the absorbing AR performance in UV light range, the morphologies of ZnO NRAs are controllably changed by varying the precursor concentration from 20 mM to 90 mM. The top-view and cross-sectional SEM images of the ZnO NRAs grown at different precursor concentrations are shown in Figures 1c and S1. These figures reveal that the nanorods have slightly tapered tips. The average diameters of ZnO nanorods are increased from 30 nm to 156 nm, approximately, as the concentration of the precursor are increased from 20 mM to 90 mM. Finally, the ZnO NRAs change into a quasi-continuous film when the precursor concentration reaches 90 mM. Interestingly, the heights of ZnO NRAs are kept at 1.3 µm and not changed significantly until the precursor concentration is over 70 mM, as shown in Figure S1. More detailed geometric parameters of the ZnO NRAs are shown in Figure S2 in the Supporting Information. Comparing to the general AR layer, the quality of the absorbing AR layer is not only defined by the reflectance, but also the absorptance. To characterize the absorbing AR performance of these ZnO NRAs in UV light range, total reflectance (R) and transmittance (T) spectra of a bare ZnO thin film and ZnO thin films covered with NRAs grown at different precursor concentrations were measured using an ultraviolet–visible spectroscopy (Varian Cary 5000) equipped with an integrating sphere in the wavelength range from 300 to 400 nm. The absorptance (A) is determined by A = 1 − T − R. The results are shown in Figures 2a-c. In these figures, 0 mM represents a bare ZnO thin film without NRAs covering layer. First, the optical interference fringes (Figure 2a, black curve) is eliminated in the presence of NRAs covering layer (Figure 2a, red curve). This is because that the tapered nanorod tips result in a

ACS Paragon Plus Environment

Page 4 of 23

Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

rough interface between ZnO and air, which can eliminate interference effect. Second, from a general view, total reflectance and transmittance are significantly reduced after growing ZnO NRAs. Therefore, the UV light absorptance of ZnO thin film covered with NRAs is higher than the bare ZnO thin film without NRAs, as shown in Figure 2c. In order to present the variation trends of reflectance, transmittance, and absorptance more clearly, the data at the wavelength of 365 nm, which is corresponding to the wavelength of the light source in our UV sensing experiments, is replotted in Figures 2d-f correspondingly (the red lines). Meanwhile, FDTD simulations were performed to further understand the experimental results and reveal the light propagation nature. The geometrical parameters of ZnO NRAs used for FDTD simulation are listed in Figure S2 in Supporting Information. The FDTD simulated data of reflectance, transmittance, and the deduced absorptance of a bare ZnO thin film and ZnO thin films covered with different NRAs are also plotted in Figures 2d-f as the blue lines. The experimental and simulated data shows good agreement, except that the simulated reflectance is slightly higher than experimental one. This is mainly due to the fact that the perfectly smooth top surface and the equal height of nanorods were used in the simulation, while it is not the true case in reality. The simulated electrical field intensity (|E|2) distribution is visualized and shown in Figure 3. All calculated values of the electrical field intensity are normalized to the ones of the excitation source. As presented in Figure 2d, the ZnO thin films with NRAs grown at precursor concentration from 30 mM to 60 mM exhibite optimal general AR property, where the reflectance is decreased to below 1%. Checking the electrical field intensity (|E|2) distribution in Figure 3, it reveals clearly that the electrical field intensity above the air-ZnO interface, which is closely related to the light reflection, dramatically decreases at the presence of NRAs covering. Then, it decreases gradually as the precursor concentration for ZnO NRAs growth increases. While, when the precursor concentration increases further, it starts to increase, and finally goes back to the similar level of a bare ZnO film as the precursor concentration reaches 90 mM. The change of AR properties is induced by the change of a gradient refraction index profile at the air-ZnO interface.14, 22-23 This gradient refraction index profile is introduced by the change of

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

filling factors (the area ratio of ZnO nanorods to the total substrate surface) from top to bottom, which can be tuned by the morphology of the ZnO NRAs. There should be an optimal refraction index profile for the maximum reflection depression. The calculated refraction index profiles in ZnO NRAs with different growth concentration are shown in Figure S3. It can be clearly seen that the refraction index profiles has minimum mutations when the growth concentration is around 40 to 50 mM. They are closest to the quasi-continuous change compared to other concentrations. The actual rough interface between ZnO and air, which is deviated from the simulation assumption as mentioned before, can further smooth the refraction index profile. In our experiments, the ZnO NRAs grown at an extended precursor concentration from 30 mM to 60 mM provide the proper conditions, exhibiting the best general AR property. When the precursor concentration exceeds 70 mM, the top surfaces of densely grown ZnO nanorods start to interconnect to form a quasi-continuous film, giving rise to the reflection on the top surface of ZnO NRAs. About the transmittance, it should be as small as possible for the absorbing AR layer, which is opposite to the general AR layer. As shown in Figure 2e, for the 75 nm thick bare ZnO thin film, the transmittance is around 38%. It is a large number. When ZnO NRAs are grown at 20 mM to make the coverage, the transmittance is only slightly decreased to be around 28%. The simulated electrical field intensity distribution in Figure 3b shows clearly that the NRAs grown at this condition work more like a general AR layer. There are scattering happened between nanorods, but a large amount of light is guided to the underlying ZnO thin film, which can not provide sufficient light absorption. When the precursor concentration increases further, the light scattering between nanorods increases further, and the light is more and more trapped in the upper part of the NRAs. Starting from 40 mM, the transmittance is almost negligible. Taking into account both the effects of reflection and transmission, the ZnO thin film covered with NRAs grown at precursor concentration of 50 mM exhibits the best performance of light absorption at wavelength of 365 nm, as presented in Figure 2f. It is noteworthy that the optimal absorptance is over 99%, which indicated that almost all the incident light is absorbed. It is superior for photosensors.

ACS Paragon Plus Environment

Page 6 of 23

Page 7 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

To check how the rational designed absorbing AR layer affects the performance of UV sensing, the flexible UV sensors were tested. Figure 4a shows I–V curves measured in dark on a UV sensor before and after growing ZnO NRAs on its surface. The growth concentration is 50 mM. The linear I–V curves indicate that good ohmic contacts were formed in both cases, which is preferred for carrier collection in thin film channel. Figure 4b shows the log-scale I-V curves of this ZnO UV sensor measured in dark and under photoluminescence of UV light at the wavelength of 365nm with a power density of 1.5 mW/cm2. The change of photocurrent and dark current of UV sensors when ZnO NRAs were grown at different precursor concentrations are shown in Figure 4c. Both the dark current and photocurrent were measured under a bias voltage of 3V. A UV lamp at 365 nm with a power density of 1.5 mW/cm2 was used as the light source. It can be seen that the UV sensors exhibited larger photocurrent and dark current with ZnO NRAs grown at higher precursor concentrations. The reason for the increase of dark current is that the ZnO nanorods grown at higher concentration merge together at the bottom, leading to an increase of the conductivity in the channel of the UV sensor. However, photocurrent increases quicker than dark current does before the concentration reaches 50 mM. The sensitivity of the device can be defined as  =   



.

△ 

=

As shown in Figure 4d, UV sensors covered with ZnO NRAs grown at

concentration of 50 mM exhibited the highest sensitivity, reaching 109000. It is much higher than the one based on bare ZnO thin film, which is only 29. Notably, the change of sensitivity of the UV sensors agrees well with the change of UV light absorptance shown in Figure 2f, which indicates that the enhancement of UV light absorption with the optimized absorbing AR layer directly improve the sensitivity of UV sensors. Another interesting thing is that although the absorptance only increased from 41% to 99% after introducing the absorbing AR layer, the sensitivity of the UV sensor increased more than three orders of magnitudes for the optimized situation. We propose that some other facts should also play their roles here for the greatly enhanced sensitivity. For a photosensor, beside light absorptance, the collection efficiency of photon generated carriers is also very important. In our device design, before arriving at the electrodes, the transport of carriers experiences

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

two processes in sequence: diffusion in the nanorods from top to the bottom and drift in the bottom thin film by the applied electrical field between two electrodes. When NRAs are first introduced at the precursor concentration of 20 mM, there is no obvious change in dark current. It means that the carrier transport property in the channel does not change too much. However, the photocurrent increases about one order of magnitude with the increase of 28% in light absorptance. This may contribute from the decreased recombination rate of photon generated electrons and holes. Previous investigations show that comparing to thin films, nanowire or nanorod structure can provide larger surface-to-volume ratio and more mid-band-gap surface states.21 It results in decreased hole diffusion length,24 and thus decreased recombination rate of photon generated carriers in NRAs. When the precursor concentration is further increased, the dark current starts to increase. It indicates a better transport property in the bottom thin film channel. The further increased photocurrent comes from both the reduced recombination rate in diffusion and improved drift process. There are still rooms for the optimization. Rechecking Figure 3, considering the optical field distribution, the length of the NRAs and the space occupancy ratio between NRAs and the bottom thin film can be further optimized, or a vertical electrical field can be applied for better carrier collection efficiency. The related work is ongoing. As we discussed above, UV light absorption and sensitivity of the UV sensor are drastically enhanced by growing ZnO NRAs on its surface. However, a slow response and recovery speed of the UV sensor are obtained in our experiments, which is a general problem in ZnO based devices.16-18 Figure 5(a) shows the response process of the as-fabricated UV sensor with NRAs grown at the concentration of 50 mM. It can be seen that the photocurrent of the as-fabricated UV sensor increased gradually when the UV light was turn on, and the current was still unsaturated after 200 s continuous illumination. Moreover, the current did not recover to its initial state in 300 s after the UV light was turned off. Here we use the reset time as a representative parameter to evaluate the speed of the sensor. The reset time (τ) is defined as the time for the photocurrent dropping to 1/e (37%) of its maximum photocurrent. It is 28.19 s in this situation.

ACS Paragon Plus Environment

Page 8 of 23

Page 9 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

To understand the response process of the UV sensor, several other UV response experiments were performed in different conditions. When the UV response experiment was carried out in moist air, the response and recovery speed of the UV sensor increased dramatically compared to that obtained in dry air, as shown in Figure 5b (the violet curve). The obtained reset time is 4.17 s. This phenomenon indicates the possibility that water molecules may accelerate the UV response process of the UV sensor.25-26 To further confirm the influence of H2O, the UV sensor was dried in oven at the temperature of 95℃ for 40 minutes to remove water molecules adsorbed on the sensor. Then, UV response experiment was re-performed in dry air immediately. The result (the green line in figure 5b) shows that the response and the recovery speeds of the UV sensor were greatly decreased after drying, leading to a reset time of 88.94 s. It is generally accepted that oxygen chemisorption at the surface severely limits the response and recovery speed of ZnO UV sensors.17, 19, 21, 27-28 The oxygen chemisorption is greatly promoted at the site of oxygen vacancies with charge transfer from the ZnO surface to the molecular in the form of [ +   →  ;  + 2  →  ]. Thus, the conductivity of ZnO would be reduced due to the consumption of free electrons. When UV light is turned on, electron-hole pairs are generated [hv →   + ℎ], and a part of holes are captured by the adsorbed oxygen ions, leading to the photodesorption of oxygen molecules [ + ℎ →  ]. The unpaired electrons accumulate gradually with time until desorption and re-adsorption of O2 reach equilibrium, resulting in a gradual current rise. When UV light is shut off, the photogenerated holes recombine with electrons quickly, and the oxygen molecules gradually readsorb on ZnO to capture the unpaired electrons, leading to a slow current decay. In the whole process, oxygen molecules have to be exchanged between the ZnO surface and atmosphere to reach equilibrium between desorption and adsorption. The response and recovery speed of the UV sensor is limited by the equilibrium process, which is the slowest step in the whole process. Besides oxygen chemisorption, oxygen vacancies can also work as the active sites for the dissociative chemisorption of water molecules.29-31 The adsorbed H2O may dissociate at the

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

vacancy and hydroxylates a lattice oxygen ( ) neighbored to the vacancy, generating two hydroxyl groups [  +  + 2  → 2  ], resulting in the reduced conductivity of ZnO. Generally, partially hydroxylation happens on ZnO surfaces in atmosphere.32-33 Highly hydroxylated surface of ZnO has also been well demonstrated by intentionally introducing OH group via H2O adsorption.34 So, there is a competition of the chemisorption between O2 and H2O at oxygen vacancy sites. And previous results show that the oxygen vacancies are kinetically more favorable for hydroxyl adsorption than oxygen adsorption.34-35 Upon switching UV light on and off, water molecules are desorbed and adsorbed onto ZnO surface, similar as oxygen molecules. But the rates are very different in these two cases. The hydroxylated surface is hydrophilic, which can condense water molecules on the surface. Then the desorption and adsorption process of H2O can be realized by switching between physisorption and chemisorption on the surface, which can greatly accelerate the speed of the UV sensors. According to the competitive chemisorption mechanism, the observed experimental phenomenon in Figure 5b can be explained as follow. In dry air, there are not adequate amount of H2O molecules available. Therefore, O2 related process dominates the UV light response process, leading to a slow response and recovery speed, as we discussed before. When moist air is introduced, H2O molecules take over the process, resulting in an accelerated speed. At the same time, a little part of the condensed H2O can auto-ionize on the surface of ZnO, [  →   +   ]. The proton (  ) generated by the autoionization can transfer in the physisorbed layer under an applied electric field according to the Grotthuss transport mechanism.36-37 Thus the channel conductance of the UV sensor was slightly increased in moist air, as the purple curve shown in figure 5b. After drying the device at 95℃ for 40 minutes, H2O molecules were all desorbed from ZnO surfaces. We got a speed even slower than the as-fabricated device which may have a small amount of H2O contamination during the fabrication process or from the environment. To further confirm the effect of oxygen vacancy and OH group, the UV sensor was treated by oxygen plasma for 30 s after drying in oven to fill oxygen vacancy and remove the OH group on the ZnO.38 As a result, an even slower response and recovery speed was obtained as the orange curve shown in Figure

ACS Paragon Plus Environment

Page 10 of 23

Page 11 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

5b, which is consistent with our expectation. According to the mechanism discussed above, a new method is proposed to enhance the response and recovery speed of the UV sensors. In our experiments, a dilute polyethylene glycol (PEG) aqueous solution was dropped on the surface of ZnO NRAs. After evaporating the aqueous solvent in air, the PEG was encapsulated on ZnO NRAs. The PEG, as a common humectant,39 can provide a stable moist environment for ZnO. Figure 6a shows the time-dependent UV response of the UV sensor covered with ZnO NRAs after PEG encapsulation. Compared to the as-fabricated UV sensor (Figure 5a), the response and recovery time of the UV sensor was drastically reduced after PEG encapsulation. The photocurrent decay process is enlarged in Figure 6b, from which we obtain the reset time of τ = 0.21 s. It is two orders of magnitude faster in the speed comparing to that of untreated one. The inset (on the left) in Figure 6b shows the SEM image of the ZnO NRAs after PEG coating. It can be seen that the morphology of the ZnO NRAs was not changed obviously. More importantly, spectral response test shows that the UV sensors with PEG encapsulation still exhibit excellent visible-blind property which is not sensitive to the light with wavelengths longer than 400 nm, as shown in the right inset of Figure 6b. In order to confirm the mechanical stability of the flexible UV sensors, bending fatigue tests was performed. As shown in Figure 6c, the sensor was bent 1000 cycles respectively under the strain of 0.64%, 0.77% and 1.24% in sequence. The strain is calculated from the bending geometry as reported elsewhere.40-41 The inset shows the experimental setup for the bending test. It can be clearly seen that, after finishing a total of 3000 bending cycles, the decay of the responsivity is within 20%, which indicates good mechanical stability of the flexible UV sensors.

3. CONCLUSIONS In summary, flexible UV sensors with absorbing AR layers and humectant encapsulation are designed for realizing high sensitivity and fast speed. Devices are fabricated on PET substrate with morphology controlled ZnO NRAs as the rational designed absorbing AR layer.

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Experimental and simulated results show that more than 99% of the incident UV light can be effectively absorbed by adjusting the morphologies of ZnO NRAs, resulting in three orders of magnitude increase in the sensitivity of the UV sensor. The influence of absorbed O2 and H2O molecules on the response speed of the UV sensor is discussed and a competitive chemisorption mechanism between O2 and H2O at oxygen vacancy sites is proposed to explain the phenomenon of speed acceleration in moist air. Ultilizing this mechanism, a new method of PEG encapsulation on ZnO NRAs is presented to make H2O participant rapid processes dominant, achieving two orders of magnitude enhancement in the reset speed. Moreover, bending fatigue tests confirmed the good mechanical stability of the flexible UV sensor. We believe that this study not only offers a simple, effective and scalable approach to boost the performance of ZnO UV sensors without employing complicated fabrication process but also provides meaningful references in enhancing the performance of photoelectronic devices based on other oxide semiconductors.

4. EXPERIMENTAL SECTION 4.1. Fabrication of Flexible UV Sensors. UV sensors were fabricated on polyethylene (PET) substrates. First, the PET substrates were treated by oxygen plasma for 30 s under the power of 50 W, changing the surfaces into hydrophilic. Then the PET substrates were cleaned by a standard cleaning process. Ti/Au (5 nm / 25 nm) interdigitated electrodes were deposited and patterned on the cleaned PET substrate via direct current magnetron plasma sputtering by using a shadow mask. The 5 nm Ti works as an adhesion layer to increase the adhesion between the substrate and Au film. The width and spacing of the electrodes are about 2000 µm and 1000 µm, respectively. Next, a 75-nm-thick ZnO thin film was deposited using radio frequency (rf) magnetron sputter with a rf power of 100 W and a working pressure of 10 mTorr under a flow of Ar gas. Then ZnO NRAs were grown in aqueous solution containing zinc nitrate hexahydrate and hexamethylenetetramine (HMTA). The precursor concentration was adjusted from 20 to 90 mM while the molar ratio of zinc nitrate hexahydrate and HMTA was kept at 1:1. The reaction was kept at 95℃ for 2 h. 4.2. Characterization. The morphological study of ZnO nanostructures has been carried out

ACS Paragon Plus Environment

Page 12 of 23

Page 13 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

with a FEI Quanta 600F field emission SEM. The UV reflectance and transmittance spectra were measured with a Varian Cary 5000 UV-visible spectrometer in the spectral range from 300 to 400 nm. The Keithley 4200-SCS semiconductor characterization system and keithley 2636B sourcemeter were used to measure I-V characteristics and responsivity of the fabricated UV sensors. A UV lamp combined with light filter provided monochromatic light source in the UV sensing experiments. 4.3. FDTD Simulation. FDTD method was used to evaluate the optical properties of the fabricated ZnO nanowires. For convenience and not losing generality, each of the ZnO nanorods was located on point of a hexagonal lattice in the electromagnetic simulations. Periodic boundary conditions were used in x and y directions while PML (perfect match layer) condition was adopted in z direction. Plane wave was injected from –z direction as the excitation source. The values of refractive index (n), and extinction coefficient (k), for FDTD simulation are 2.6 and 0.28, respectively.42 Geometric parameters of ZnO wires were evaluated from SEM images. 4.4. Strain Calculation. The strain applied on the device is calculated from the bending

geometry. The strain of the device ε is related to the thickness t and bending radius of the substrate ρ as ! = "/(2 %).40-41 Acknowledgements This work was supported by National Science Foundation of China (Grant Numbers 61571016 and 61427901), Ministry of Science and Technology of China (Grant Numbers 2011CB933001, 2011CB933002 and 2015CB932403)

Supporting Information Available Cross-sectional SEM images of the ZnO NRAs grown on ZnO thin film at different precursor concentrations;Detailed geometrical parameters of ZnO NRAs used in FDTD simulation; The calculated refraction index profiles in ZnO NRAs with different growth concentration.

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

References (1) Jeong, S.; Garnett, E. C.; Wang, S.; Yu, Z.; Fan, S.; Brongersma, M. L.; McGehee, M. D.; Cui, Y. Hybrid Silicon Nanocone-polymer Solar Cells. Nano Lett. 2012, 12, 2971-2976. (2) Ferry, V. E.; Polman, A.; Atwater, H. A. Modeling Light Trapping in Nanostructured Solar Cells. ACS Nano 2011, 5, 10055-10064. (3) Yu, P.; Chang, C.; Chiu, C.; Yang, C.; Yu, J.; Kuo, H.; Hsu, S.; Chang, Y. Efficiency Enhancement of GaAs Photovoltaics Employing Antireflective Indium Tin Oxide Nanocolumns. Adv. Mater. 2009, 21, 1618-1621. (4) Garnett, E.; Yang, P. Light Trapping in Silicon Nanowire Solar Cells. Nano Lett. 2010, 10, 1082-1087. (5) Oh, J.; Yuan, H. C.; Branz, H. M. An 18.2%-efficient Black-silicon Solar Cell Achieved through Control of Carrier Recombination in Nanostructures. Nat. Nanotechnol. 2012, 7, 743-748. (6) Yuan, H.; Yost, V. E.; Page, M. R.; Stradins, P.; Meier, D. L.; Branz, H. M. Efficient Black Silicon Solar Cell with a Density-graded Nanoporous Surface: Optical Properties, Performance Limitations, and Design Rules. Appl. Phys. Lett. 2009, 95, 123501. (7) Lin, H.; Xiu, F.; Fang, M.; Yip, S.; Cheung, H.; Wang, F.; Han, N.; Chan, K. S.; Wong, C.; Ho, J. C. Rational Design of Inverted Nanopencil Arrays for Cost-Effective, Broadband, and Omnidirectional Light Harvesting. ACS Nano 2014, 8, 3752-3760. (8) Wang, K. X.; Yu, Z.; Liu, V.; Cui, Y.; Fan, S. Absorption Enhancement in Ultrathin Crystalline Silicon Solar Cells with Antireflection and Light-trapping Nanocone Gratings. Nano Lett. 2012, 12, 1616-1619. (9) Han, S. E.; Chen, G. Toward the Lambertian Limit of Light Trapping in Thin Nanostructured Silicon Solar Cells. Nano Lett. 2010, 10, 4692-4696. (10) Kim, J. G.; Choi, H. J.; Park, K. C.; Cohen, R. E.; McKinley, G. H.; Barbastathis, G. Multifunctional Inverted Nanocone Arrays for Non-wetting, Self-cleaning Transparent Surface with High Mechanical Robustness. Small 2014, 10, 2487-2494. (11) Lee, Y. J.; Ruby, D. S.; Peters, D. W.; Mckenzie, B. B.; Hsu, J. W. P. ZnO Nanostructures as Efficient Antireflection Layers in Solar Cells. Nano Lett. 2008, 8, 1501-1505. (12) Chao, Y.; Chen, C.; Lin, C.; Dai, Y.; He, J. Antireflection Effect of ZnO Nanorod Arrays. J. Mater. Chem. 2010, 20, 8134.

ACS Paragon Plus Environment

Page 14 of 23

Page 15 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

(13) Yeh, L.; Lai, K.; Lin, G.; Fu, P.; Chang, H.; Lin, C.; He, J. Giant Efficiency Enhancement of GaAs Solar Cells with Graded Antireflection Layers Based on Syringelike ZnO Nanorod Arrays. Adv. Energy Mater. 2011, 1, 506-510. (14) Tsai, D. S.; Lin, C. A.; Lien, W. C.; Chang, H. C.; Wang, Y. L.; He, J. H. Ultra-high-responsivity Broadband Detection of Si Metal-semiconductor-metal Schottky Photodetectors Improved by ZnO Nanorod Arrays. ACS Nano 2011, 5, 7748-7753. (15) Wang, Z. L. Zinc Oxide Nanostructures: Growth, Properties and Applications. J. Phys.: Condens. Matter 2004, 16, R829-R858. (16) Jin, Y.; Wang, J.; Sun, B.; Blakesley, J. C.; Greenham, N. C. Solution-Processed Ultraviolet Photodetectors Based on Colloidal ZnO Nanoparticles. Nano Lett. 2008, 8, 1649-1653. (17) Kind, H.; Yan, H.; Messer, B.; Law, M.; Yang, P. Nanowire Ultraviolet Photodetectors and Optical Switches. Adv. Mater. 2002, 14, 158. (18) Keem, K.; Kim, H.; Kim, G.; Lee, J. S.; Min, B.; Cho, K.; Sung, M.; Kim, S. Photocurrent in ZnO Nanowires Grown from Au Electrodes. Appl. Phys. Lett. 2004, 84, 4376. (19) Zhou, J.; Gu, Y.; Hu, Y.; Mai, W.; Yeh, P. H.; Bao, G.; Sood, A. K.; Polla, D. L.; Wang, Z. L. Gigantic Enhancement in Response and Reset Time of ZnO UV Nanosensor by Utilizing Schottky Contact and Surface Functionalization. Appl. Phys. Lett. 2009, 94, 191103. (20) Liu, N.; Fang, G.; Zeng, W.; Zhou, H.; Cheng, F.; Zheng, Q.; Yuan, L.; Zou, X.; Zhao, X. Direct Growth of Lateral ZnO Nanorod UV Photodetectors with Schottky Contact by a Single-Step Hydrothermal Reaction. ACS Appl. Mater. Interfaces 2010, 2, 1973-1979. (21) Soci, C.; Zhang, A.; Xiang, B.; Dayeh, S. A.; Aplin, D. P. R.; Park, J.; Bao, X. Y.; Lo, Y. H.; Wang D. ZnO Nanowire UV Photodetectors with High Internal Gain. Nano Lett. 2007, 7, 1003-1009. (22) Southwell, W. H. Pyramid-array Surface-relief Structures Producing Antireflection Index Matching on Optical Surfaces. J. Opt. Soc. Am. A 1991, 8, 549-553. (23) Huang, Y.; Chattopadhyay, S.; Jen, Y.; Peng, C.; Liu, T.; Hsu, Y.; Pan, C.; Lo, H.; Hsu, C.; Chang, Y.; Lee, C.; Chen, K.; Chen, L. Improved Broadband and Quasi-omnidirectional Anti-reflection Properties with Biomimetic Silicon Nanostructures. Nat. Nanotechnol. 2007, 2, 770-774. (24) Soudi, A.; Dhakal, P.; Gu, Y. Diameter Dependence of the Minority Carrier Diffusion Length in Individual ZnO Nanowires. Appl. Phys. Lett. 2010, 96, 253115. (25) Ahn, S. E.; Lee, J. S.; Kim, H.; Kim, S.; Kang, B. H.; Kim, K. H.; Kim, G. T. Photoresponse of Sol-gel-synthesized ZnO Nanorods. Appl. Phys. Lett. 2004, 84, 5022.

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(26) Witkowski, B. S.; Wachnicki, L.; Gieraltowska, S.; Sybilski, P.; Kopalko, K.; Stachowicz, M.; Godlewski, M. UV Detector Based on Zinc Oxide Nanorods Obtained by the Hydrothermal Method. Phys. Status Solidi C 2014, 11, 1447-1451. (27) Sharma, P.; Sreenivas, K.; Rao, K. V. Analysis of Ultraviolet Photoconductivity in ZnO Films Prepared by Unbalanced Magnetron Sputtering. J. Appl. Phys. 2003, 93, 3963. (28) Lu, G.; Linsebigler, A.; Yates, J. T. The Adsorption and Photodesorption of Oxygen on the TiO2 (110) Surface. J. Chem. Phys. 1995, 102, 4657. (29) Meyer, B.; Marx, D.; Dulub, O.; Diebold, U.; Kunat, M.; Langenberg, D.; Woll, C. Partial Dissociation of Water Leads to Stable Superstructures on the Surface of Zinc Oxide. Angew. Chem., Int. Ed. 2004, 43, 6642-6645. (30) Rodriguez, J. A. A Quantum Chemical Study of the Adsorption of Carbon Dioxide and Hydroxyl on Copper and Zinc Oxide Surfaces and Hydroxyl on Platinum Surfaces. Langmuir 1988, 4, 1006-1020. (31) Kunat, M.; Girol, S. G.; Burghaus, U.; Wöll, C. The Interaction of Water with the Oxygen-Terminated, Polar Surface of ZnO. J. Phys. Chem. B 2003, 107, 14350-14356. (32) Gopel, W. Chemisorption and Charge Transfer at Ionic Semiconductor Surfaces: Implications in Designing Gas Sensors. Prog. Surf. Sci. 1985, 20, 9-103. (33) Vohs, J. M. The Surface Science of Metal Oxides. AIChE J. 1998, 44, 502. (34) Sun, R. D.; Nakajima, A.; Fujishima, A.; Watanabe, T.; Hashimoto, K. Photoinduced Surface Wettability Conversion of ZnO and TiO2 Thin Films. J. Phys. Chem. B 2001, 105, 1984-1990. (35) Feng, X.; Feng, L.; Jin, M.; Zhai, J.; Jiang, L.; Zhu, D. Reversible Super-hydrophobicity to Super-hydrophilicity Transition of Aligned ZnO Nanorod Films. J. Am. Chem. Soc. 2003, 126, 62-63. (36) Geissler, P. L.; Dellago, C.; Chandler, D.; Hutter, J.; Parrinello, M. Autoionization in Liquid Water. Science 2001, 291, 2121-2124. (37) Biswas, P.; Kundu, S.; Banerji, P.; Bhunia, S. Super Rapid Response of Humidity Sensor Based on MOCVD Grown ZnO Nanotips Array. Sens. Actuators, B 2013, 178, 331-338. (38) Coppa, B. J.; Davis, R. F.; Nemanich, R. J. Gold Schottky Contacts on Oxygen Plasma-treated, N-type ZnO(0001̄). Appl. Phys. Lett. 2003, 82, 400. (39) Ramsey, R. J. L.; Stephenson, G. R.; Hall, J. C. A Review of the Effects of Humidity, Humectants, and Surfactant Composition on the Absorption and Efficacy of Highly Water-soluble Herbicides. Pestic. Biochem. Physiol. 2005, 82, 162-175.

ACS Paragon Plus Environment

Page 16 of 23

Page 17 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

(40) Lee, G.; Yu, Y.; Cui, X.; Petrone, N.; Lee, C.; Choi, M. S.; Lee, D.; Lee, C.; Yoo, W. J.; Watanable, K.; Taniguchi, T.; Nuckolls, C.; Kim, P.; Hone, J. Flexible and Transparent MoS2 Field-effect Transistors on Hexagonal Boron Nitride-graphene Heterostructures. ACS Nano 2013, 7, 7931-7936. (41) Petrone, N.; Meric, I.; Hone, J.; Shepard, K. L. Graphene Field-effect Transistors with Gigahertz-frequency Power Gain on Flexible Substrates. Nano Lett. 2012, 13, 121-125. (42) Mendoza-Galvan, A.; Trejo-Cruz, C.; Lee, J.; Bhattacharyya, D.; Metson, J.; Evans, P. J.; Pal, U. Effect of Metal-ion Doping on the Optical Properties of Nanocrystalline ZnO Thin Films. J. Appl. Phys. 2006, 99, 014306.

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figures

Figure 1. (a) Schematic illustration of the fabrication process for the flexible ZnO UV sensors covered with NRAs. (b) Photograph of an as-fabricated UV sensor covered with ZnO NRAs. (c1–c8) Top-view SEM images of the ZnO NRAs grown on ZnO thin film at different precursor concentrations from 20 mM to 90 mM.

ACS Paragon Plus Environment

Page 18 of 23

Page 19 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 2. (a) Reflectance spectra, (b) transmittance spectra, and (c) absorptance spectra of a bare ZnO thin film and ZnO thin films covered with ZnO NRAs grown at different precursor concentrations. Experimental and FDTD simulated data of (d) reflectance, (e) transmittance, and (f) absorptance at 365 nm of a bare ZnO thin film and ZnO thin films covered with ZnO NRAs grown at different precursor concentrations. 0 mM represents the bare ZnO thin film without NRA covering.

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. Cross-sectional view of FDTD simulated results of normalized electrical field intensity (|E|2) distribution in (a) a ZnO thin film and (b-i) ZnO thin films covered with NRAs grown at different precursor concentration.

ACS Paragon Plus Environment

Page 20 of 23

Page 21 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 4. (a) Linear I–V curves of a UV sensor before and after growing ZnO NRAs, measured in dark. (b) The log-scale I-V curves of this UV sensor measured in dark and under photoluminescence. The change of (c) photocurrent, dark current and (d) sensitivity of the UV sensors by growing ZnO NRAs on thin film surfaces at different precursor concentrations from 20 mM to 90 mM. 0 mM represents a bare ZnO thin film without NRAs covering.

Figure 5. (a) Photocurrent response of the UV sensor at bias voltage of 3 V in dry air. (b) The change in the response and recovery speed of the UV sensor when it was tested under different conditions.

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 6. (a) Time-dependent response of the UV sensor covered with ZnO NRAs after PEG coating. (b) The photocurrent decay process of the UV sensor coated with PEG. The inset on the left shows the SEM image of the ZnO NRAs coated with PEG. The inset on the right shows the spectral response of the UV sensors measured after PEG coating. (c) Normalized responsivity of the UV sensor with 3000 bending cycles by applying corresponding external strain. The inset photograph shows the experimental setup for bending test.

ACS Paragon Plus Environment

Page 22 of 23

Page 23 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

For Table of Contents Only

ACS Paragon Plus Environment