Transparent photodetectors based on polyoxometalate modified electrospun ZnO homojunction nanowire intersection arrays

Abstract

Transparent photodetectors (PDs) based on binary metal oxides have become a hot topic in the field of photodetection because of their ability to absorb and detect ultraviolet (UV) light. However, PDs constructed with pure binary metal oxides still have the problems of slow response time and high-voltage drive. In addition, due to its natural internal defects, it has the characteristics of an n-type semiconductor and is difficult to achieve p-type conversion. In this work, the top layer of the device is polyoxometalate (POM) modified ZnO NWs which showed the characteristics of p-type semiconductors, its UV absorption was increased by 36.9%, and the bottom layer of the PD is pure ZnO NWs. An interlayer p–n junction is formed between the two layers, which is not only conducive to carrier transmission, but also can complete photodetection without bias. The PD has a transparency of more than 70% in the visible light region. Under the conditions of zero bias and AM 1.5 illumination, the light–dark current difference (ΔI) reaches 23.76 μA, and the rise-fall time is 1.32 s/2.62 s, respectively. After 216 hours of cycling, it still maintains 70% of the original efficiency.

---

1. Introduction

The harm of UV light to the human body is unquestionable. Although the UV light in the sunlight will be absorbed and reflected to a certain extent when the sunlight passes through the ozone layer, the human body is exposed to the sunlight for a long time, and there will still be a series of skin diseases caused by UV light.^1–5 Therefore, how to shield the human body from UV light has received widespread attention, and designing PDs that absorb and detect UV light has gradually become a hot topic in the field of photoelectric detection.^6–9 Among the PDs that absorb and detect UV light, low-latitude devices based on binary metal oxides (such as TiO₂, ZnO, Ga₂O₃, etc.) exhibit a variety of excellent properties such as high responsiveness, rapid response time, and large carrier mobility.^10–19

As a typical wide band gap n-type semiconductor, zinc oxide (ZnO) has been applied in the field of UV photodetection due to its suitable energy band (E_g = 3.37 eV) and good optical properties.²⁰ PDs based on pure ZnO low-latitude nanomaterials show good performance, but still have a long response time and the need for a high-voltage drive, which affect the continued promotion and application of ZnO-based PDs.^21–23 In addition, as a wide band gap semiconductor, wurtzite-type ZnO has internal defects, making it an n-type semiconductor. It is difficult for this type of n-type semiconductor to achieve p-type conversion. Attempts have been made to construct p–n heterojunctions with other common p-type materials and ZnO, but the effect is not ideal due to the energy level mismatch.^18,24–29 Researchers also tried to dope or modify ZnO with some materials to achieve p–n modification.^30–40 However, the materials used are expensive and their difficulty to be widely applied is still an important problem.

POMs are a class of metal oxide clusters with nano-scale size, which have semiconductor-like characteristics, and they also have characteristics not found in general metal oxide semiconductors.^41–50 First of all, POMs have an adjustable energy band, which can be adjusted well after being compounded with semiconductor materials. Secondly, POMs have good light absorption, especially strong absorption in the UV region, and after the semiconductor materials are combined, the absorbance of the composite material can be improved. Finally, after POMs are compounded with semiconductor materials, they can act as shallow electron traps, separating photogenerated excitons under illumination, and enhancing electron transmission. In view of the above characteristics, POMs have been applied to dye-sensitized solar cells (DSSCs) and perovskite solar cells (PSCs), proving that they have good application prospects in the field of optoelectronics.^51–56

According to the characteristics of POMs, this work uses POMs to modify ZnO NWs for PDs, and completes the device preparation by electrospinning. POMs tend to agglomerate, so electrospinning is used to solve this problem.^7,31,57 The three kinds of Keggin-type POMs selected in this work are Na₈[HPW₉O₃₄] (PW₉), Na₇[PW₁₁O₃₉] (PW₁₁) and H₃[PW₁₂O₄₀] (PW₁₂). They all have the function of adjusting the energy band.^58–60 The E_g values of ZnO NWs modified by POMs have been changed. The UV-visible absorption spectra proved that after being combined with POMs, the UV absorption of the pure ZnO NWs had increased by 32.7% (PW₉), 36.9% (PW₁₁), and 20.1% (PW₁₂), respectively. The top layer of PD is ZnO NWs modified with POMs. The Mott–Schottky curve proves that ZnO NWs modified with POMs exhibit the characteristics of p-type semiconductors. Through infrared (IR) spectroscopy, X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS), it is proved that POMs and ZnO NWs are more than simple physical doping. The bottom layer of the PD is formed by pure ZnO NWs, and a p–n homojunction is formed between the two layers of spinning to better complete the carrier transmission and make the PD work under unbiased conditions. The UV-visible transmission spectrum proves that the PD maintains a transparency of more than 70% in the visible light region. Under the working conditions of zero bias voltage and AM 1.5 illumination, the light–dark current difference (ΔI) reaches 23.76 μA, and the rise-fall time of PD is 1.32 s/2.62 s. After 216 hours of cycling, the PD still maintains 70% of the original efficiency.

2. Experimental

2.1 Preparation of materials

PW₉, PW₁₁ and PW₁₂ were synthesized according to the literature reported previously, and the synthesis method was described in the ESI.† The materials required for the synthesis, including phosphoric acid (A.R.), sodium tungstate (A.R.), binary hydrogen phosphate (A.R.), glacial acetic acid (A.R.), concentrated nitric acid (A.R.) and acetone (A.R.), were purchased from Aladdin and have not been further purified.

2.2 The precursor preparation method of pure ZnO NWs and POMs modified ZnO NWs

0.9 g of polyvinylpyrrolidone (PVP, M_n = 1 300 000) was dissolved in 15 g of ethanol and stirred at room temperature for 6 to 8 hours until the solution was completely clear and transparent. At the same time, 0.8 g of Zn(NO₃)₂·6H₂O was dissolved in 2 g of deionized water and stirred for 15 min at room temperature to make it completely dissolved. Then the prepared zinc nitrate solution was added to the prepared PVP solution drop by drop, and stirred at room temperature for 8 h until the solution is completely mixed. For the precursor modified by POMs, the prepared POMs were added into a solution of zinc nitrate, and the mass doping ratios of PW₉ were 1 : 10 and 1 : 20, the doping ratios of PW₁₁ were 1 : 10, 1 : 20 and 1 : 50, and the doping ratio of PW₁₂ was 1 : 15.

2.3 Preparation of a PD and its working principle

The simulation schematic diagram of spinning is shown in Fig. 1a. By applying a positive high voltage of +10 kV to the needle, the spinning solution is extruded from the spinneret through the syringe pump to form droplets. Under the action of high voltage electricity, the droplet deformation occurs, and if the voltage is low, it will appear tapered. When the voltage further increases until it exceeds a certain critical value, under the excitation of the high voltage, the droplets form a jet. Through violent vibration in the air, the jet is stretched and refined. At this time, metal electrodes are placed at both ends of the collector. As shown in the right picture in Fig. 1a, the jet is fixed on the surface of the current collector in an orderly manner to achieve the goal of unidirectional rotation. The collector is FTO conductive glass, and the distance between the collector and the injection needle is 10 cm.

Fig. 1 (a) Schematic diagram of the simulation of the basic principle of electrospinning. (b) Schematic diagram of the simulation of the photodetector (not drawn to scale). (c) Schematic diagram of the reaction mechanism and energy band of the PW₁₁@ZnO NWs/pure ZnO NWs p–n homojunction under 0 bias.

Fig. 1b shows a schematic diagram of the entire device (not drawn to scale). The entire device has a vertical structure. FTO conductive glass was used as a collecting device. The bottom layer was spun with an undoped ZnO precursor and the top layer was doped with POMs. The precursor fluid controlled the direction of spinning of the two layers by changing the direction of the collector. The pure zinc nitrate precursor solution was used for bottom spinning, and the POM dissolved precursor solution was used for top spinning. After the spinning was complete, annealing was completed quickly at 550 °C. The photo of the obtained device is shown in Fig. S4 (ESI†). The light transmittance is shown in Fig. S5 (ESI†).

The simulation schematic diagram of the device for photodetection is shown in Fig. 1c. As seen in the figure, the valence band of PW₁₁@ZnO NWs is 0.454 eV higher than that of pure ZnO NWs, and our PW₁₁@ZnO/ZnO homojunction NWs have been cross-arranged, which formed a p–n homojunction structure. In this case, due to the generation of electron–hole pairs under the light, the electrons will move to the pure ZnO NW side, and the holes will move to the PW₁₁@ZnO side, thereby helping in the formation of a charge transfer state, and conducive to the separation of light-generated carriers in NWs.

3. Results and discussion

Fig. 2a–c show the comparison of IR spectra of three pure POMs and POMs@ZnO NWs. Fig. 2a shows the IR spectra of pure ZnO NWs, pure PW₁₁ and PW₁₁@NWs. The characteristic peaks of the IR spectrum of PW₁₁ are at 1099 cm⁻¹ (P–O), 954 cm⁻¹ (W=O), 897 cm⁻¹ and 857 cm⁻¹ (W–O_b–W), 812 cm⁻¹ and 738 cm⁻¹ (W–O_c–W), which belong to the stretching vibration peaks of the PW₁₁ anion. This is consistent with the previously reported single-vacancy Keggin-type POM compounds.⁵⁸ The characteristic peaks of PW₁₁ can be observed in the IR spectrum of PW₁₁@ZnO NWs, and the shift of these peak positions can be observed, indicating that there is an interaction between PW₁₁ and ZnO.^58,61,62Fig. 2b shows the IR spectra of pure PW₉ and PW₉@ZnO NWs. The characteristic peaks of the IR spectrum of PW₉ are at 1059 cm⁻¹ (P–O), 932 cm⁻¹ (W–O_t), 880 cm⁻¹ (W–O_b), 811 cm⁻¹ and 729 cm⁻¹ (W–O_c–W), which belong to the vibration peak of the PW₉ anion, which is consistent with the previously reported multi-vacancy Keggin-type POM compound.⁵⁹ The characteristic peaks of PW₉ can be observed in the IR spectrum of PW₉@ZnO NWs, and the shift of these peak positions can be observed, indicating that there is an interaction between PW₉ and ZnO. Fig. 2c shows the IR spectra of pure PW₁₂ and PW₁₂@ZnO NWs. Among them, the characteristic peaks of the IR spectrum of PW₁₂ are at 1074 cm⁻¹, 973 cm⁻¹, 902 cm⁻¹ and 773 cm⁻¹. These strong peaks are better than the stretching of the PO₄ unit and W–O–W bond in PW₁₂. The vibration is caused by the vibration peak belonging to the PW₁₂ anion, which is consistent with the previously reported saturated Keggin-type POM compound.⁶⁰ The characteristic peaks of PW₁₂ can be observed in the IR spectrum of PW₁₂@ZnO NWs, and the shift of these peak positions can be observed, indicating that there is an interaction between PW₁₂ and ZnO.

Fig. 2 (a) IR spectra of pure ZnO NWs, pure PW₁₁ and PW₁₁@ZnO NWs. (b) IR spectra of pure PW₉ and PW₉@ZnO NWs. (c) IR spectra of pure PW₁₂ and PW₁₂@ZnO NWs. (d) XRD full spectrum comparison chart of pure ZnO NWs and POMs@ZnO NWs. (e) XRD full-spectrum wurtzite crystal images of pure ZnO NWs and PW₁₁@ZnO NWs. (f) Comparison of (100), (002) and (101) peaks in the XRD spectra of pure ZnO NWs and PW₁₁@ZnO NWs.

The structure of the prepared NWs can be further analyzed by XRD. Fig. 2d shows the XRD full spectrum of pure ZnO NWs and POMs@ZnO NWs. It can be seen from the figure that the ZnO in POMs@ZnO NWs still maintains its wurtzite crystal form. In Fig. 2e, it can be seen that the distance (d) between POMs@ZnO NWs and pure ZnO NWs remains unchanged under the full spectrum measurement. Fig. 2f shows a fine analysis of the diffraction peaks from 30° to 38°. The figure shows that the deviation of POMs@ZnO compared with pure ZnO is very small (≤0.17°). In contrast, the 2θ angle of POMs@ZnO is shifted to the left. Among them, the distance d can be calculated using Bragg's law

2d sin θ = nλ (1)

where n is the order of diffraction, usually n = 1, and λ is the X-ray wavelength (Cu Kα in the instrument used in this article, λ = 1.5406 Å). In this work, the sintered ZnO has a wurtzite structure, and it can be proved by XRD full spectrum analysis that its lattice constant can be calculated using the following formulawhere the lattice constants of pure ZnO were calculated to be a = 3.244 Å and c = 5.196 Å. The lattice constants of PW₁₁@ZnO are a = 3.250 Å and c = 5.214 Å; both lattice constants are slightly increased compared with those of pure ZnO. The shift of the diffraction angle to the left indicates that in the ZnO modified by PW₁₁, the spacing of the (100), (002) and (101) crystal planes are all increased, which indicates that the crystal lattice has expanded. PW₁₁ has a complex anionic structure. With the introduction of PW₁₁, the crystal lattice of ZnO has undergone a certain deformation.

Fig. 3a shows a field emission scanning electron microscopy (SEM) image of a single layer of pure ZnO NWs. In this work, by controlling the electrode arrangement of the electrospinning receiver (Fig. 1c), the NW array ejected by electrospinning can be controlled to be in a unidirectional state. Fig. 3b shows the SEM image of the array of PW₁₁@ZnO NWs and pure ZnO NWs after laminated spinning, and the two layers are arranged in a cross. Fig. 3c–f shows the distribution of each element in this region, which proves that PW₁₁ and ZnO NWs are well combined together.

Fig. 3 (a) SEM image of a single layer of pure ZnO NWs. (b) SEM image of the stack of PW₁₁@ZnO NWs and pure ZnO NWs. The elemental mapping images of (c) O, (d) Zn, (e) P and (f) W of the laminated NWs.

We characterized the exact chemical states of Zn and O in pure ZnO NWs, pure PW₁₁ and PW₁₁@ZnO NWs by XPS.⁶³ Through a small difference in the binding energy of certain atoms in different chemical environments, and after peak fitting, the bonding state between them and other atoms is shown. As shown in Fig. 4a, the binding energies of Zn 2p in pure ZnO NWs are 1020.6 eV and 1043.8 eV, respectively. In contrast, the binding energies of Zn 2p in PW₁₁@ZnO NWs changed to 1021.1 eV and 1044 eV, respectively, as shown in Fig. 4d. Fig. 4b shows that the O 1s of pure ZnO has two peaks at 530.9 eV and 529.7 eV, named O2 and O1, respectively. In contrast, the peak position of O 1s in PW₁₁@ZnO NWs has changed, and the peaks at binding energies of 531.8 eV and 530.2 eV correspond to O2 and O1, respectively, as shown in Fig. 4e. The O 1 peak is composed of the metal bond Zn–O, and the O2 peak is the surface hydroxyl group. XPS also shows that the content of the surface hydroxyl group increases significantly after modification of POMs.^64–66 We calculated the lattice oxygen atoms (O_lattice) and vacant oxygen atoms (O_O–H) according to the area of each part under the O 1s curve, as shown in Table S2 (ESI†). As shown in Fig. 4c, the binding energies of W 4f in pure PW₁₁ are 35.4 eV and 37.4 eV, respectively. In contrast, the binding energies of W 4f in PW₁₁@ZnO NWs changed to 35.2 eV and 37.1 eV, respectively, as shown in Fig. 4f.⁵⁸ IR, XRD and XPS tests proved that ZnO maintains the original structure and properties, and the two are not merely a simple physical combination.

Fig. 4 XPS measurement spectrum of (a) Zn 2p and (b) O 1s of pure ZnO NWs. XPS measurement spectrum of (c) W 4f of the pure PW₁₁. XPS measurement spectrum of (d) Zn 2p, (e) O 1s and (f) W 4f of PW₁₁@ZnO NWs.

Fig. 5a–c show the thermogravimetric (TG) curves of three pure POMs. Fig. 5a shows the TGA curve of PW₉. In the range of 30–800 °C, PW₉ has three weight loss steps. In the first step, the mass loss occurs at 30–223.4 °C and is 10.78%, which corresponds to the loss of crystal water. The mass loss in the second step is 223.4–456.2 °C and is 1.09%, and the corresponding loss is bound water. In the third step, the mass loss is between 456.2 and 655.4 °C, and is 0.72%. The corresponding loss is structural water. There is no significant weight loss after 655.4 °C, and the total weight loss is 12.59% within the range of 30–800 °C. The TGA curve of PW₁₁ is shown in Fig. 5b. In the range of 30 to 800 °C, PW₁₁ has three weight loss steps. In the first step, the mass loss occurs at 30–140.9 °C, and is 6.30%. The corresponding loss is crystal water. The mass loss in the second step is 140.9–429.9 °C and is 1.72%. The corresponding loss is bound water. In the third step, the mass loss is 429.9–619.6 °C and is 7.77%. The corresponding loss is structural water. There is no significant weight loss after 619.6 °C, and the total weight loss is 16.22% within the range of 30–800 °C. Fig. 5c shows the TGA curve of PW₁₂. In the range of 30 to 800 °C, PW₁₂ has three weight loss steps. In the first step, the mass loss occurs at 30–218.9 °C, and is 4.50%, which corresponds to the loss of crystal water. The mass loss in the second step is 218.9–520.4 °C, and is 2.09%. The corresponding loss is bound water. In the third step, the mass loss is between 520.4 and 627.9 °C, and is 0.72%, which corresponds to the loss of structural water. There is no significant weight loss after 627.9 °C, and the total weight loss is 7.42% within the range of 30–800 °C. The annealing temperature of POMs@ZnO NWs is 550 °C. According to the above TGA curve, the three kinds of POMs will not decompose at this temperature.

Fig. 5 Thermogravimetric curves of (a) PW₉, (b) PW₁₁ and (c) PW₁₂. (d) UV-vis spectra of pure ZnO NWs, PW₉@ZnO NWs, PW₁₁@ZnO NWs and PW₁₂@ZnO NWs. (e) Mott–Schottky curves of pure ZnO NWs and (f) PW₁₁@ZnO NWs.

Fig. 5d shows the comparison of UV-visDRS (UV-visDRS) between pure zinc oxide NWs and POMs@ZnONWs. It can be seen from Fig. 5d that the UV absorption of POMs@ZnO NWs has been significantly improved, and PW₁₁@ZnO NWs have the most significant improvement effect among several materials. Besides, it can be seen from the figure that the light absorbance of PW₉ and PW₁₂@ZnO NWs has also been improved. The absorption peak of POMs@ZnO mainly falls in the ultraviolet region, so the wavelength of the light absorbed by the PD to generate a photoelectric response is also in the ultraviolet region. In this case, the ultraviolet light absorption of the composite material is increased, which improves the light utilization rate of the PD.

Fig. 5e, f and Fig. S13, S14 (ESI†) are the Mott–Schottky curves of pure ZnO NWs and POMs@ZnO NWs.⁶⁷ The data is obtained through electrochemical impedance testing, and then the Mott–Schottky curve is obtained by calculation. For pure ZnO NWs, the slope of the Mott–Schottky curve is positive and monotonically increasing, corresponding to n-type semiconductors. For POMs@ZnO NWs, the slope of the Mott–Schottky curve is negative and monotonously decreasing, which corresponds to p-type semiconductors.

Fig. 6a, b, d and e show the cyclic voltammetry curves (CV curve) of pure ZnO NWs and POMs@ZnO NWs. The conduction band positions of the four types of NWs are determined by the first reduction peak in the negative direction of the CV curve that is less than 0.⁶⁸

Fig. 6 Cyclic voltammetry curves of (a) pure ZnO NWs and (b) PW₉@ZnO NWs. (c) E_g calculation diagram of pure ZnO NWs and POMs@ZnO NWs. (d) Cyclic voltammetry curves of PW₁₁@ZnO NWs and (e) PW₁₂@ZnO NWs. (f) Energy level band gap diagrams of pure ZnO and POMs@ZnO.

Fig. 6c shows the E_g calculation diagrams of four types of NWs, calculated using the UV-vis DRS test data, and the calculation formula is as follows

(αhν)^1/n = A(hν − E_g) (3)

hν = hc/λ (4)

where α is the light absorption index, h is the Planck constant, ν is the frequency, E_g is the semiconductor forbidden bandwidth, A is a constant, and n is a constant. Among them, the direct bandgap semiconductor n = 1/2, indirect bandgap semiconductor n = 2, c is the speed of light, and λ is the wavelength of light. According to the calculations, the E_g value of pure ZnO NWs is 3.37 eV, the E_g value of PW₉@ZnO NWs is 3.18 eV, the E_g value of PW₁₁@ZnO NWs is 3.16 eV and the E_g value of PW₁₂@ZnO NWs is 3.18 eV. Through the above calculations, energy level band gap of four NWs is shown in Fig. 6f. In summary, through the modification of POMs, the band gap of ZnO nanowires is narrowed and the conduction band is improved. Since PW₁₁ has a more suitable HOMO and LUMO, PW₁₁@ZnO has the most suitable energy level and band gap among the three kinds of POMs@ZnO.

As shown in Fig. 7a, the mass ratios of PW₁₁ and Zn (NO₃)₂ doped in the precursor were 1 : 10, 1 : 20 and 1 : 50. The figure shows that as the doping ratio increases, the photocurrent of the PD increases significantly. However, continuing to increase the content of PW₁₁ in the precursor, the PW₁₁ in the precursor no longer dissolves, indicating that the mass doping ratio is about 1 : 10 for the solubility of PW₁₁ under these conditions. After PW₁₁ is combined with ZnO, it can act as shallow electron traps, separating photogenerated excitons under illumination, and enhancing electron transmission. Theoretically speaking, the improvement of the device performance by an increase of the POM content generally shows an increasing trend first and then decreases. After reaching the extreme value, if POMs continue to be added to the precursor, and POMs will agglomerate on the material and affect the performance of the device. However, in this work, since the POMs in the precursor can no longer be dissolved in the precursor solution before reaching the extreme value, in this experiment, the maximum solubility in PW₁₁ is the doping ratio for the optimal device performance.

Fig. 7 (a) The I–t curve of the photodetector when the mass doping ratios of PW₁₁ are 1 : 10, 1 : 20 and 1 : 50 (spinning time 1 min). (b) The I–t curve of the photodetector at the electrospinning times of 30 s, 60 s and 90 s (the mass doping ratio of PW₁₁ is 1 : 10). (c) Comparison of the optimal photodetector performance between PW₉@ZnO and PW₁₁@ZnO. (d) The stability I–t curve of the optimal photodetector modified by PW₁₁. (e) The time curve of the highest current of the photodetector modified by PW₁₁. (f) The rise/decay time of the optimal photodetector modified by PW₁₁.

As shown in Fig. 7b, under the optimal doping ratio of PW₁₁, the performance of the device was tested by controlling the spinning time. It can be seen from the figure that as the spinning time increases, the dark current decreases. This phenomenon can be attributed to the increase in the thickness of the NW layer and the increase in the internal resistance as the spinning time increases. The dark current of the device is reduced. In addition, with the extension of the electrospinning time, the photocurrent showed an increasing trend first and then decreased. This phenomenon can be attributed to the increase in the number of NWs with the extension of the electrospinning time. Under light conditions, the increase of photogenerated carriers increases the photocurrent. But similar to the dark current, as the internal resistance increases further, the photocurrent begins to decrease. Therefore, the performance of the device is best when the rotation time is 60 s. Fig. 7c shows the performance comparison chart of the optimal device modified by PW₉ (quality doping ratio is 1 : 10, spinning time 60 s) and the optimal device modified by PW₁₁. It can be seen from the figure that the performance of the PW₁₁ modified device is better than that of PW₉. As shown in Fig. 6f, we tested the energy level band gaps of pure ZnO NWs and POMs@ZnO NWs, respectively. Among them, PW₁₁@ZnO NWs have energy levels and band gaps that are more matched with pure ZnO NWs, so they exhibit the best photodetection performance.

Fig. 7d and e show the stability test of the PD modified by PW₁₁, and the test interval is 48 h. It can be seen from the figure that as time increases, after 216 h, the device maintains 70% performance. The main reason for this decrease in stability can be attributed to the poor chemical stability of the POMs selected in this work in the air as traditional semiconductor materials, and will slowly decompose in the air, which will cause the performance of the device to decrease over time. Fig. 7f shows part of the I–t curve of the optimal device modified by PW₁₁. The time required for the current to increase from 10% of the peak value to 90% (and vice versa) is defined as the rise time and the decay time, respectively. It can be seen from the figure that the rise/decay time is 1.32 s/2.62 s. Under light conditions, the average photocurrent is stable at 28.81 μA, and under no light conditions, the dark current is stable at 5.05 μA, the light–dark current difference (ΔI) reaches 23.76 μA, and the on–off ratio is 5.7. Since the chemical stability of POMs in the air is not as good as that of traditional semiconductor materials, it is easy to generate a certain dark current and reduce the on/off ratio.

The test data of PW₉ and PW₁₂ modified PDs are shown in Fig. S15 and S16 (ESI†). The doping ratios of PW₉ are 1 : 10 and 1 : 20. When the doping ratio is 1 : 50, its photoelectric response is very weak. Moreover, after continuing to increase the doping ratio of PW₉, the precursor solution becomes too viscous to electrospin. Therefore, the optimal ratio is also 1 : 10. Since PW₁₂ is a saturated Keggin-type POMs it has fewer active sites than PW₉ and PW₁₁, so the performance of the device is not greatly improved, and its light response speed has also become very slow. The photoelectric response of the device is very weak when PW₁₂ is doped into the precursor at ratios of 1 : 50 and 1 : 20. When PW₁₂ was mixed into the precursor at a ratio of 1 : 15, floccules were formed in the precursor solution at the beginning, and the floccules were dissolved after continuous stirring. After continuing to increase the doping ratio, the floccules appearing cannot dissolve after continuous stirring. Therefore, the optimal doping ratio is 1 : 15.

As shown in Fig. 8a, this work tested the I–t curves of the photodetectors prepared by PW₁₁@ZnO NWs at different light intensities, and the light intensities were 60, 80, 100, 120 and 140 mW cm⁻². The responsivity of the device at different light intensities is calculated using the following formula:

R = I_P − I_d/P_light × S (5)

where I_P means the photocurrent, I_d is the dark current, P_light is the incident light intensity in the experiment, and S is the active area.40 The relationship between the responsivity and photocurrent and light intensity is shown in Fig. 8b. Fig. 8c shows the relationship between the detection rate and light intensity. The detection rate D* is calculated according to the following formula:where R means the responsivity, S is the active area, q is the electron charge, I_d is the dark current. Fig. 8d shows the responsivity at different wavelengths. It can be seen that the device has good responsivity in the ultraviolet region, while in the visible region, the photodetector has almost no response.

Fig. 8 (a) The I–t curve at different light intensities. (b) The responsivity and photocurrent at different light intensities. (c) Detection rate (D*) at different light intensities. (d) The responsivity at different wavelengths.

4. Conclusions

In summary, we successfully modified the ZnO NWs with Keggin-type POMs, and made the modified ZnO NWs exhibit p-type semiconductor characteristics. Next, a series of characterization methods are used to prove the existence of POMs in the whole structure. Additionally, by controlling the electrodes of the collecting device during electrospinning, we constructed two vertical NW arrays and formed p–n homojunctions in the two layers to achieve the purpose of photodetection. The made PD has good transparency in the visible light region. Under unbiased voltage and standard AM1.5 illumination, the light–dark current difference (ΔI) can reach 23.76 μA, and its rise/decay time can reach 1.32 s/2.62 s. It is worth noting that the on/off ratio of the PD is only 5.7. Compared with other ZnO-based PDs, its good responsivity and unbiased working conditions can better adapt to possible needs. Furthermore, its preparation method is simple, which has more advantages in practical applications that may be put into use in the future. In addition, we provide new ideas for the study of photodetectors modified by polyoxometalates. There are many types of POMs that have semiconductor-like properties and diverse anionic structures. The application in the field of photodetection should not stop there.

Conflicts of interest

The authors declare no competing interests.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (No. 21871041), the Natural Science Foundation of Jilin Province (No. 20180101298JC), the Technology Foundation for Selected Overseas Chinese Scholars of Personnel Ministry of China, the Science and Technology Research Project of the Education Department of Jilin Province, the Science and Technology Activities Project Preferential Funding for Selected Overseas Chinese Scholars of Jilin Province Human Resources and Social Bureau.

References

1. J. Han, M. He, M. Yang, Q. Han, F. Wang and F. Zhong, et al., Light-modulated vertical heterojunction phototransistors with distinct logical photocurrents, Light: Sci. Appl., 2020, 9, 167.
2. C. Li, H. Wang, F. Wang, T. Li, M. Xu and H. Wang, et al., Ultrafast and broadband photodetectors based on a perovskite/organic bulk heterojunction for large-dynamic-range imaging, Light: Sci. Appl., 2020, 9, 31.
3. D. Wang, A. E. L. Allcca, T. F. Chung, A. V. Kildishev, Y. P. Chen and A. Boltasseva, et al., Enhancing the graphene photocurrent using surface plasmons and a p-n junction, Light: Sci. Appl., 2020, 9, 126.
4. A. Aiello, Y. Wu, Z. Mi and P. Bhattacharya, Deep ultraviolet monolayer GaN/AlN disk-in-nanowire array photodiode on silicon, Appl. Phys. Lett., 2020, 116, 061104.
5. J. Zhou, Y. Gu, Y. Hu, W. Mai, P.-H. Yeh and G. Bao, et al., Gigantic enhancement in response and reset time of ZnO UV nanosensor by utilizing Schottky contact and surface functionalization, Appl. Phys. Lett., 2009, 94, 191103.
6. B. H. Kind, H. Yan, B. Messer, M. Law and P. Yang, Nanowire Ultraviolet Photodetectors and Optical Switches, Adv. Mater., 2002, 14, 158–160.
7. N. Nasiri, R. Bo, F. Wang, L. Fu and A. Tricoli, Ultraporous Electron-Depleted ZnO Nanoparticle Networks for Highly Sensitive Portable Visible-Blind UV Photodetectors, Adv. Mater., 2015, 27, 4336–4343.
8. W. Tian, T. Zhai, C. Zhang, S. L. Li, X. Wang and F. Liu, et al., Low-cost fully transparent ultraviolet photodetectors based on electrospun ZnO-SnO2 heterojunction nanofibers, Adv. Mater., 2013, 25, 4625–4630.
9. P. Wan, M. Jiang, T. Xu, Y. Liu and C. Kan, High-mobility induced high-performance self-powered ultraviolet photodetector based on single ZnO microwire/PEDOT:PSS heterojunction via slight ga-doping, J. Mater. Sci. Technol., 2021, 93, 33–40.
10. X. Liu, L. Gu, Q. Zhang, J. Wu, Y. Long and Z. Fan, All-printable band-edge modulated ZnO nanowire photodetectors with ultra-high detectivity, Nat. Commun., 2014, 5, 4007.
11. Z. Li, Z. Li, Z. Shi and X. Fang, Facet-Dependent, Fast Response, and Broadband Photodetector Based on Highly Stable All-Inorganic CsCu₂I₃ Single Crystal with 1D Electronic Structure., Adv. Funct. Mater., 2020, 30, 2002634.
12. A. K. Rana, J. T. Park, J. Kim and C.-P. Wong, See-through metal oxide frameworks for transparent photovoltaics and broadband photodetectors, Nano Energy, 2019, 64, 103952.
13. J. Dai, C. Xu, X. Xu, J. Guo, J. Li and G. Zhu, et al., Single ZnO microrod ultraviolet photodetector with high photocurrent gain, ACS Appl. Mater. Interfaces, 2013, 5, 9344–9348.
14. H. Wang, H. Chen, L. Li, Y. Wang, L. Su and W. Bian, et al., High Responsivity and High Rejection Ratio of Self-Powered Solar-Blind Ultraviolet Photodetector Based on PEDOT:PSS/beta-Ga2O3 Organic/Inorganic p-n Junction, J. Phys. Chem. Lett., 2019, 10, 6850–6856.
15. L. Chen, C. Yang and C. Yan, High-performance UV detectors based on 2D CVD bismuth oxybromide single-crystal nanosheets, J. Mater. Sci. Technol., 2020, 48, 100–104.
16. Y. Wang, L. Li, H. Wang, L. Su, H. Chen and W. Bian, et al., An ultrahigh responsivity self-powered solar-blind photodetector based on a centimeter-sized beta-Ga₂O₃/polyaniline heterojunction, Nanoscale, 2020, 12, 1406–1413.
17. B. Zhao, F. Wang, H. Chen, L. Zheng, L. Su and D. Zhao, et al., An Ultrahigh Responsivity (9.7 mA W^-1) Self-Powered Solar-Blind Photodetector Based on Individual ZnO–Ga₂O₃ Heterostructures., Adv. Funct. Mater., 2017, 27, 1700264.
18. Z. Li, M. K. Joshi, J. Chen, Z. Zhang, Z. Li and X. Fang, Mechanically Compatible UV Photodetectors Based on Electrospun Free-Standing Y³⁺-Doped TiO₂ Nanofibrous Membranes with Enhanced Flexibility, Adv. Funct. Mater., 2020, 30, 2005291.
19. K. Hu, F. Teng, L. Zheng, P. Yu, Z. Zhang and H. Chen, et al., Binary response Se/ZnO p-n heterojunction UV photodetector with high on/off ratio and fast speed, Laser Photonics Rev., 2017, 11, 1600257.
20. S. Dhar, T. Majumder and S. P. Mondal, Graphene Quantum Dot-Sensitized ZnO Nanorod/Polymer Schottky Junction UV Detector with Superior External Quantum Efficiency, Detectivity, and Responsivity, ACS Appl. Mater. Interfaces, 2016, 8, 31822–31831.
21. F. Teng, L. Zheng, K. Hu, H. Chen, Y. Li and Z. Zhang, et al., A surface oxide thin layer of copper nanowires enhanced the UV selective response of a ZnO film photodetector, J. Mater. Chem. C, 2016, 4, 8416–8421.
22. Y. Tu, S. Chen, X. Li, J. Gorbaciova, W. P. Gillin and S. Krause, et al., Control of oxygen vacancies in ZnO nanorods by annealing and their influence on ZnO/PEDOT:PSS diode behaviour, J. Mater. Chem. C, 2018, 6, 1815–1821.
23. C. Soci, A. Zhang, B. Xiang, S. A. Dayeh, D. P. R. Aplin and J. Park, et al., ZnO Nanowire UV Photodetectors with High Internal Gain, Nano Lett., 2007, 7, 1003–1009.
24. M. Zheng, F. Yang, J. Guo, L. Zhao, X. Jiang and G. Gu, et al., Cd(OH)₂@ZnO nanowires thin-film transistor and UV photodetector with a floating ionic gate tuned by a triboelectric nanogenerator., Nano Energy, 2020, 73, 104808.
25. A. K. Rana, M. Kumar, D. K. Ban, C. P. Wong, J. Yi and J. Kim, Enhancement in Performance of Transparent p-NiO/n-ZnO Heterojunction Ultrafast Self-Powered Photodetector via Pyro-Phototronic Effect., Adv. Electron. Mater., 2019, 5, 1900438.
26. A. Bera and D. Basak, Photoluminescence and photoconductivity of ZnS-coated ZnO nanowires, ACS Appl. Mater. Interfaces, 2010, 2, 408–412.
27. Z. Zhang, Y. Ning and X. Fang, From nanofibers to ordered ZnO/NiO heterojunction arrays for self-powered and transparent UV photodetectors, J. Mater. Chem. C, 2019, 7, 223–229.
28. Y. Xiang, Q. Zhou, Z. Li, Z. Cui, X. Liu and Y. Liang, et al., A Z-scheme heterojunction of ZnO/CDots/C₃N₄ for strengthened photoresponsive bacteria-killing and acceleration of wound healing., J. Mater. Sci. Technol., 2020, 57, 1–11.
29. B. Deka Boruah, S. Naidu Majji, S. Nandi and A. Misra, Doping controlled pyro-phototronic effect in self-powered zinc oxide photodetector for enhancement of photoresponse, Nanoscale, 2018, 10, 3451–3459.
30. G. D. Yuan, W. J. Zhang, J. S. Jie, X. Fan, J. X. Tang and I. Shafiq, et al., Tunable n-Type Conductivity and Transport Properties of Ga-doped ZnO Nanowire Arrays, Adv. Mater., 2008, 20, 168–173.
31. Y. Ning, Z. Zhang and F. Teng, X. Fang. Novel Transparent and Self-Powered UV Photodetector Based on Crossed ZnO Nanofiber Array Homojunction, Small, 2018, 14, 1703754.
32. C. L. Hsu, Y. D. Gao, Y. S. Chen and T. J. Hsueh, Vertical p-type Cu-doped ZnO/n-type ZnO homojunction nanowire-based ultraviolet photodetector by the furnace system with hotwire assistance, ACS Appl. Mater. Interfaces, 2014, 6, 4277–4285.
33. X. Zhu, Q. Xie, H. Tian, M. Zhang, Z. Gou and S. He, et al., High photoresponse sensitivity of lithium-doped ZnO (LZO) thin films for weak ultraviolet signal photodetector, J. Alloys Compd., 2019, 805, 309–317.
34. W. Ko, S. Lee, N. Myoung and J. Hong, Solution processed vertically stacked ZnO sheet-like nanorod p–n homojunctions and their application as UV photodetectors, J. Mater. Chem. C, 2016, 4, 142–149.
35. J. Lu, C. Xu, J. Dai, J. Li, Y. Wang and Y. Lin, et al., Improved UV photoresponse of ZnO nanorod arrays by resonant coupling with surface plasmons of Al nanoparticles, Nanoscale, 2015, 7, 3396–3403.
36. G. C. Park, S. M. Hwang, J. H. Lim and J. Joo, Growth behavior and electrical performance of Ga-doped ZnO nanorod/p-Si heterojunction diodes prepared using a hydrothermal method, Nanoscale, 2014, 6, 1840–1847.
37. W. Dai, X. Pan, S. Chen, C. Chen, W. Chen and H. Zhang, et al., ZnO homojunction UV photodetector based on solution-grown Sb-doped p-type ZnO nanorods and pure n-type ZnO nanorods, RSC Adv., 2015, 5, 6311–6314.
38. O. Lupan, V. Cretu, V. Postica, M. Ahmadi, B. R. Cuenya and L. Chow, et al., Silver-doped zinc oxide single nanowire multifunctional nanosensor with a significant enhancement in response, Sens. Actuators, B, 2016, 223, 893–903.
39. H. Zhou, L. Yang, P. Gui, C. R. Grice, Z. Song and H. Wang, et al., Ga-doped ZnO nanorod scaffold for high-performance, hole-transport-layer-free, self-powered CH₃NH₃PbI₃ perovskite photodetectors, Sol. Energy Mater. Sol. Cells, 2019, 193, 246–252.
40. T. H. Flemban, V. Singaravelu, A. A. Sasikala Devi and I. S. Roqan, Homogeneous vertical ZnO nanorod arrays with high conductivity on an in situ Gd nanolayer, RSC Adv., 2015, 5, 94670–94678.
41. Y. R. Wang, Q. Huang, C. T. He, Y. Chen, J. Liu and F. C. Shen, et al., Oriented electron transmission in polyoxometalate-metalloporphyrin organic framework for highly selective electroreduction of CO2, Nat. Commun., 2018, 9, 4466.
42. J. Lin, N. Li, S. Yang, M. Jia, J. Liu and X. M. Li, et al., Self-Assembly of Giant Mo240 Hollow Opening Dodecahedra, J. Am. Chem. Soc., 2020, 142, 13982–13988.
43. L. Chen, W. L. Chen, X. L. Wang, Y. G. Li, Z. M. Su and E. B. Wang, Polyoxometalates in dye-sensitized solar cells, Chem. Soc. Rev., 2019, 48, 260–284.
44. S. Zhang, N. Liu, H. Wang, Q. Lu, W. Shi and X. Wang, Sub-Nanometer Nanobelts Based on Titanium Dioxide/Zirconium Dioxide-Polyoxometalate Heterostructures, Adv. Mater., 2021, 33, e2100576.
45. M. Lu, M. Zhang, C. G. Liu, J. Liu, L. J. Shang and M. Wang, et al., Stable Dioxin-Linked Metallophthalocyanine Covalent Organic Frameworks (COFs) as Photo-Coupled Electrocatalysts for CO2 Reduction, Angew. Chem., Int. Ed., 2021, 60, 4864–4871.
46. L. Z. Dong, L. Zhang, J. Liu, Q. Huang, M. Lu and W. X. Ji, et al., Stable Heterometallic Cluster-Based Organic Framework Catalysts for Artificial Photosynthesis, Angew. Chem., Int. Ed., 2020, 59, 2659–2663.
47. Q. Liu, H. Yu, Q. Zhang and D. Wang, X. Wang. Temperature-Responsive Self-Assembly of Single Polyoxometalates Clusters Driven by Hydrogen Bonds., Adv. Funct. Mater., 2021, 31, 2103561.
48. G. Yang, X. Xie, M. Cheng, X. Gao, X. Lin and K. Li, et al., H₄SiW₁₂O₄₀-catalyzed cyclization of epoxides/aldehydes and sulfonyl hydrazides: An efficient synthesis of 3,4-disubstituted 1H-pyrazoles., Chin. Chem. Lett., 2021 DOI:10.1016/j.cclet.2021.08.037.
49. G. Yang, K. Li, X. Lin, Y. Li, C. Cui and S. Li, et al., Regio- and Stereoselective Synthesis of (Z)-3-Ylidenephthalides via H₃PMo₁₂O₄₀-Catalyzed Cyclization of 2-Acylbenzoic Acids with Benzylic Alcohols, Chin. J. Chem., 2021, 39, 3017–3022.
50. G.-P. Yang, X. Zhang, Y. Liu, D. Zhang, K. Li and C. Hu, Self-assembly of Keggin-type U(VI)-containing tungstophosphates with a sandwich structure: an efficient catalyst for the synthesis of sulfonyl pyrazoles, Inorg. Chem. Front., 2021, 8, 4650–4656.
51. T. Wang, T. Ji, W. Chen, X. Li, W. Guan and Y. Geng, et al., Polyoxometalate film simultaneously converts multiple low-value all-weather environmental energy to electricity, Nano Energy, 2020, 68, 104349.
52. T. Wang, M. Xu, C. Ma, Y. Gu, W. Chen and Y. Li, et al., Strategic Design of a Bifunctional NiFeCoW@NC Hybrid to Replace the Noble Platinum for Dye-Sensitized Solar Cells and Hydrogen Evolution Reactions, ACS Appl. Mater. Interfaces, 2021, 13, 25010–25023.
53. T. Wang, M. Xu, F. Li, Y. Li and W. Chen, Multimetal-based nitrogen doped carbon nanotubes bifunctional electrocatalysts for triiodide reduction and water-splitting synthesized from polyoxometalate- intercalated layered double hydroxide pyrolysis strategy, Appl. Catal., B, 2021, 280, 119421.
54. L. Chen, W. Chen, H. Tan, J. Li, X. Sang and E. Wang, The improved efficiency of quantum-dot-sensitized solar cells with a wide spectrum and pure inorganic donor–acceptor type polyoxometalate as a collaborative cosensitizer, J. Mater. Chem. A, 2016, 4, 4125–4133.
55. X. Xu, L. Zhang, T. Wang, Y. Li, T. Ji and W. Chen, et al., The dual effect of “inorganic fullerene” {Mo132} doped with SnO2 for efficient perovskite-based photodetectors, Mater. Chem. Front., 2021, 5, 6931–6940.
56. X. Xu, M. Xie, K. Xu and Y. Zhao, g-C₃N₄@PMo₁₂ composite material double adjustment improves the performance of perovskite-based photovoltaic devices, Sol. Energy, 2020, 209, 363–370.
57. S. Yang, H. Ma, Y. Luo and J. Gong, Photocurrent Response of Surface-Functionalized Metal Oxides with Well-Matched Energy Levels: From Nothing to Something, ChemPhysChem, 2012, 13, 2289–2292.
58. H. Wang, L. Fang, Y. Yang, R. Hu and Y. Wang, Immobilization Na₇PW₁₁O₃₉ on quanternary ammonium functionalized chloromethylated polystyrene by electrostatic interactions: An efficient recyclable catalyst for alcohol oxidation, Appl. Catal., A, 2016, 520, 35–43.
59. W. H. Fang, W. D. Wang and G. Y. Yang, A novel poly(polyoxometalate) built by {Cu₉}/{Cu₅} clusters and {PW₉}/{PW₁₀}/{PW₁₁} lacunary fragments, Dalton Trans., 2015, 44, 12546–12549.
60. H. Shi, T. Zhao, J. Wang, Y. Wang, Z. Chen and B. Liu, et al., Fabrication of g-C₃N₄/PW₁₂/TiO₂ composite with significantly enhanced photocatalytic performance under visible light., J. Alloys Compd., 2021, 860, 157924.
61. L. D'Souza, M. Noeske, R. M. Richards and U. Kortz, Polyoxotungstate stabilized palladium, gold, and silver nanoclusters: a study of cluster stability, catalysis, and effects of the stabilizing anions, J. Colloid Interface Sci., 2013, 394, 157–165.
62. Q. Zhang, T. Wang, Z. Sun, L. Xi and L. Xu, Performance improvement of photoelectrochemical NO₂ gas sensing at room temperature by BiVO₄-polyoxometalate nanocomposite photoanode, Sens. Actuators, B, 2018, 272, 289–295.
63. W. Liu, F. Zhu, B. Ge, L. Sun, Y. Liu and W. Shi, MOF derived ZnO/C@(Ni,Co)Se₂ core–shell nanostructure on carbon cloth for high-performance supercapacitors, Chem. Eng. J., 2022, 427, 130788.
64. H. B. Lee, R. T. Ginting, S. T. Tan, C. H. Tan, A. Alshanableh and H. F. Oleiwi, et al., Controlled Defects of Fluorine-incorporated ZnO Nanorods for Photovoltaic Enhancement, Sci. Rep., 2016, 6, 32645.
65. H. B. Lee, N. Kumar, M. M. Ovhal, Y. J. Kim, Y. M. Song and J. W. Kang, Dopant-Free, Amorphous–Crystalline Heterophase SnO₂ Electron Transport Bilayer Enables >20% Efficiency in Triple-Cation Perovskite Solar Cells, Adv. Funct. Mater., 2020, 30, 2001559.
66. X. Wang, Q. Di, X. Wang, H. Zhao, B. Liang and J. Yang, Effect of oxygen vacancies on photoluminescence and electrical properties of (200) oriented fluorine-doped SnO₂ films., Mater. Sci. Eng., B, 2019, 250, 114433.
67. A. Allagui, H. Alawadhi, M. Alkaaby, M. Gaidi, K. Mostafa and Y. Abdulaziz, Mott-Schottky analysis of flower-like ZnO microstructures with constant phase element behavior, Phys. Status Solidi A, 2016, 213, 139–145.
68. F. R. Li, T. Wang, Y. J. Li, X. Y. Xu, C. H. Ma and W. L. Chen, et al., Heteropoly Blue/Protonation-Defective Graphitic Carbon Nitride Heterojunction for the Photo-Driven Nitrogen Reduction Reaction, Inorg. Chem., 2021, 60, 5829–5839.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1qm01246k

---