*3.1. Sintering of Printed Silver*

after the heat treatment of the ZnO pattern.

about 50 nm of the unsintered particles (Figure 3a).

about 50 nm of the unsintered particles (Figure 3a).

The structure of the device is depicted in Figure 2, which shows three main components of the devices deposited on Si/SiO<sup>2</sup> substrate. While the common electrodes and the micro-heater were printed on the first layer, the zinc precursor was printed at last. The zinc precursor appears as a white rectangle pattern over the silver electrodes. Thus, the photodetector utilizes the metal-semiconductor contacts with the two-terminals structure. Two low-magnification FE–SEM images of samples before and after heat treatment was presented in Figure A2 (Appendix A) to provide a better understanding of the effect of heat treatment to the film. *3.1. Sintering of Printed Silver*  The structure of the device is depicted in Figure 2, which shows three main components of the devices deposited on Si/SiO2 substrate. While the common electrodes and the micro-heater were printed on the first layer, the zinc precursor was printed at last. The zinc precursor appears as a white rectangle pattern over the silver electrodes. Thus, the photodetector utilizes the metal-semiconductor contacts with the two-terminals structure. Two low-magnification FE–SEM images of samples before and after heat treatment was presented in Figure A2 (Appendix A) to provide a better understanding of the effect of heat treatment to the film. *3.1. Sintering of Printed Silver*  The structure the device is depicted Figure which shows three main components of the devices deposited on Si/SiO2 substrate. While the common electrodes and the micro-heater were printed on the first layer, the zinc precursor was printed at last. The zinc precursor appears as a white rectangle pattern over the silver electrodes. Thus, the photodetector utilizes the metal-semiconductor contacts with the two-terminals structure. Two low-magnification images of samples before and after heat treatment was presented in Figure A2 (Appendix A) to provide a better understanding of the effect of treatment to the film.

*Micromachines* **2020**, *11*, 490 4 of 11

*Micromachines* **2020**, *11*, 490 4 of 11

The effect of Joule heating on the nanostructure of printed silver could be observed in Figure 3, which shows the sintering of silver nanoparticles. Printed silver film composes of separated particles after the evaporation of the solvent. The sintering effect is significantly depending on the distance from the heater. At the electrode section (Figure 3b), which is distinct from the radiation source, the particle size of approximately 80 to 100 nm could be observed, which is an increase from the size of The effect of Joule heating on the nanostructure of printed silver could be observed in Figure 3, which shows the sintering of silver nanoparticles. Printed silver film composes of separated particles after the evaporation of the solvent. The sintering effect is significantly depending on the distance from the heater. At the electrode section (Figure 3b), which is distinct from the radiation source, the particle size of approximately 80 to 100 nm could be observed, which is an increase from the size of about 50 nm of the unsintered particles (Figure 3a). The effect of heating on the nanostructure of printed silver could be in Figure which shows the sintering of silver nanoparticles. Printed silver film composes of separated after the of sintering effect is significantly depending on the from the heater. At the electrode section (Figure is distinct from the radiation source, the particle size of approximately 80 100 could an increase from the size

**Figure 3.** Field emission–scanning electron microscope (FE–SEM) images show the sintering of printed silver nanoparticles ink by the resistive heating. (**a**) The printed silver nanoparticles without annealing. (**b**) The silver nanoparticles at electrode section after annealing. **Figure 3.** Field microscope images show the sintering of printed silver by the resistive heating. **a**) The silver nanoparticles without annealing. (**b**The silver nanoparticles at after annealing. **Figure 3.** Field emission–scanning electron microscope (FE–SEM) images show the sintering of printed silver nanoparticles ink by the resistive heating. (**a**) The printed silver nanoparticles without annealing. (**b**) The silver nanoparticles at electrode section after annealing.

Heating radiation from resistive effect is the main source that induces the sintering of the printed silver pattern. In Figure 4a, the heater witnesses a notable change of film morphology, such as particle agglomeration up to the size of 200 nm. As the previous discussion has pointed out, the remote pattern exhibits minor sintering effect, while at the center of heat source, major agglomeration could be observed. Later analysis of temperature distribution shows that this is indeed a result of gradually Heating radiation from resistive effect is the main source that induces the sintering of the printed silver In Figure 4a, the heater witnesses a notable change of film such particle agglomeration up to the size of 200 nm. As the previous discussion has pointed out, the remote pattern minor sintering effect, while at the center of heat source, major agglomeration could be observed. Later analysis of temperature distribution shows that this is indeed a result of gradually reduce of temperature in the substrate. Although the sintering of silver could improve the properties of the conductive pattern, a severe agglomeration of silver nanoparticles could lead to the Heating radiation from resistive effect is the main source that induces the sintering of the printed silver pattern. In Figure 4a, the heater witnesses a notable change of film morphology, such as particle agglomeration up to the size of 200 nm. As the previous discussion has pointed out, the remote pattern exhibits minor sintering effect, while at the center of heat source, major agglomeration could be observed. Later analysis of temperature distribution shows that this is indeed a result of gradually

reduce of temperature in the substrate. Although the sintering of silver could improve the electrical properties of the conductive pattern, a severe agglomeration of silver nanoparticles could lead to the reduce of temperature in the substrate. Although the sintering of silver could improve the electrical properties of the conductive pattern, a severe agglomeration of silver nanoparticles could lead to the interruption of the conductive track and open the circuit. In order to reach a sufficient temperature for the later heat treatment of the sensing material, multi-layers printing of silver was carried on. The heat treatment was conducted after all the layers were printed to reduce dislocation of printed layers due to handling during heat treatment. Figure 4b shows a notable improvement in film morphology when four-layer printing was employed, which does not exhibit severe agglomeration and any gap in film after heat treatment. *Micromachines* **2020**, *11*, 490 5 of 11 interruption of the conductive track and open the circuit. In order to reach a sufficient temperature for the later heat treatment of the sensing material, multi-layers printing of silver was carried on. The heat treatment was conducted after all the layers were printed to reduce dislocation of printed layers due to handling during heat treatment. Figure 4b shows a notable improvement in film morphology when four-layer printing was employed, which does not exhibit severe agglomeration and any gap *Micromachines* **2020**, *11*, 490 5 of 11 interruption of the conductive track and open the circuit. In order to reach a sufficient temperature for the later heat treatment of the sensing material, multi-layers printing of silver was carried on. The heat treatment was conducted after all the layers were printed to reduce dislocation of printed layers due to handling during heat treatment. Figure 4b shows a notable improvement in film morphology

**Figure 4.** FE–SEM images show the sintering of micro-heater with different numbers of printed silver layers: (**a**) single-layer printing and (**b**) four-layer printing. **Figure 4.** FE–SEM images show the sintering of micro-heater with different numbers of printed silver layers: (**a**) single-layer printing and (**b**) four-layer printing. **Figure 4.** FE–SEM images show the sintering of micro-heater with different numbers of printed silver

#### *3.2. Temperature Survey of Joule Heating 3.2. Temperature Survey of Joule Heating* layers: (**a**) single-layer printing and (**b**) four-layer printing.

emissivity of SiO2 of 0.9 [24].

in film after heat treatment.

The temperature of the device during the Joule heating was investigated using the thermal camera and the results are presented in Figure 5. Thermal photo was taken from the back of the device to avoid the complexity of different emissivity of materials. The temperature was calibrated using the software provided by the thermal camera's manufacturer (NS9205 Viewer, Avio) with applied emissivity of SiO2 of 0.9 [24]. The temperature of the device during the Joule heating was investigated using the thermal camera and the results are presented in Figure 5. Thermal photo was taken from the back of the device to avoid the complexity of different emissivity of materials. The temperature was calibrated using the software provided by the thermal camera's manufacturer (NS9205 Viewer, Avio) with applied emissivity of SiO<sup>2</sup> of 0.9 [24]. *3.2. Temperature Survey of Joule Heating*  The temperature of the device during the Joule heating was investigated using the thermal camera and the results are presented in Figure 5. Thermal photo was taken from the back of the device to avoid the complexity of different emissivity of materials. The temperature was calibrated using the software provided by the thermal camera's manufacturer (NS9205 Viewer, Avio) with applied

(**a**) 4-W electrical power. (**b**) 5-W electrical power. The unit of temperature scale bar is degree Celsius. After calibration using emissivity of 0.9 of SiO2, an average temperature of 184 °C could be **Figure 5.** Temperature survey of the device with different heating power by the thermal camera. (**a**) 4-W electrical power. (**b**) 5-W electrical power. The unit of temperature scale bar is degree Celsius. **Figure 5.** Temperature survey of the device with different heating power by the thermal camera. (**a**) 4-W electrical power. (**b**) 5-W electrical power. The unit of temperature scale bar is degree Celsius.

measured at the resistive heater, while that at the zinc precursor film was 171 °C when input electrical power was set at 4 W. There was a significant increase of temperature when raising the input power. When the input power was set at 5 W, the recorded average temperature is 267 °C at the heater and 239 °C at the precursor pattern. The temperature of the resistive heater depends on the input power, and their correlation might possibly be expressed by the relation [25]: *P = a(T -T0) +b(T -T0)2 +c(T4 -T04)*, (1) After calibration using emissivity of 0.9 of SiO2, an average temperature of 184 °C could be measured at the resistive heater, while that at the zinc precursor film was 171 °C when input electrical power was set at 4 W. There was a significant increase of temperature when raising the input power. When the input power was set at 5 W, the recorded average temperature is 267 °C at the heater and 239 °C at the precursor pattern. The temperature of the resistive heater depends on the input power, and their correlation might possibly be expressed by the relation [25]: After calibration using emissivity of 0.9 of SiO2, an average temperature of 184 ◦C could be measured at the resistive heater, while that at the zinc precursor film was 171 ◦C when input electrical power was set at 4 W. There was a significant increase of temperature when raising the input power. When the input power was set at 5 W, the recorded average temperature is 267 ◦C at the heater and 239 ◦C at the precursor pattern. The temperature of the resistive heater depends on the input power, and their correlation might possibly be expressed by the relation [25]:

$$P = a(T - T\_0) + b(T - T\_0)^2 + c(T^4 - T\_0^4),\tag{1}$$

*a*, *b,* and *c* are fitting parameters. where *P*, *T,* and *T*<sup>0</sup> are input power, heater temperature and ambient temperature, respectively, and *a*, *b,* and *c* are fitting parameters.

where *P*, *T,* and *T0* are input power, heater temperature and ambient temperature, respectively, and

### *3.3. Generation of ZnO by Joule Heating Micromachines* **2020**, *11*, 490 6 of 11

Thermogravimetric analysis (TGA) was utilized to investigate the calcination of the precursor and formation of ZnO by high-temperature treatment, which could be cataloged into two stages: the vaporization stage and the decomposition of zinc precursor stage [26]. Figure 6 shows the result of thermal analysis of zinc acetate dihydrate in ambient air. When temperature raising from 60 to 100 ◦C, water vaporization occurs, which could be correlated to the sharp decline of the salt weight of 15% in TGA result. There is a slight decay from 150 to 200 ◦C, which denotes the starting of decomposition process. From 200 ◦C, there is a considerable reduction of weight as the reaction is promoted by temperature. The weight is stable at approximately 23% after 370 ◦C, indicating that the thermo-decomposition has completed and most of zinc salt has been transformed to ZnO. Although this result recommends that temperature of above 370 ◦C is necessary to thoroughly decompose the zinc precursor, it also suggests that lower temperature still could partially form ZnO. Starting with zinc acetate dihydrate, the reaction is finished with most of the products of the process are volatile, such as water, acetone ((CH3)2CO), acetic acid (CH2COOH), and carbon dioxide (CO2) [27]. *3.3. Generation of ZnO by Joule Heating*  Thermogravimetric analysis (TGA) was utilized to investigate the calcination of the precursor and formation of ZnO by high-temperature treatment, which could be cataloged into two stages: the vaporization stage and the decomposition of zinc precursor stage [26]. Figure 6 shows the result of thermal analysis of zinc acetate dihydrate in ambient air. When temperature raising from 60 to 100 °C, water vaporization occurs, which could be correlated to the sharp decline of the salt weight of 15% in TGA result. There is a slight decay from 150 to 200 °C, which denotes the starting of decomposition process. From 200 °C, there is a considerable reduction of weight as the reaction is promoted by temperature. The weight is stable at approximately 23% after 370 °C, indicating that the thermodecomposition has completed and most of zinc salt has been transformed to ZnO. Although this result recommends that temperature of above 370 °C is necessary to thoroughly decompose the zinc precursor, it also suggests that lower temperature still could partially form ZnO. Starting with zinc acetate dihydrate, the reaction is finished with most of the products of the process are volatile, such as water, acetone ((CH3)2CO), acetic acid (CH2COOH), and carbon dioxide (CO2) [27].

**Figure 6. T**hermogravimetric analysis (TGA) survey of zinc precursor in the air to study the formation of ZnO. **Figure 6.** Thermogravimetric analysis (TGA) survey of zinc precursor in the air to study the formation of ZnO.

Elemental and morphological studies of the heat-treated film could provide more evidence for the mechanism of the generation of ZnO. Because the decomposition reaction of zinc acetate dihydrate shows the loss of oxygen element via product vaporization, an investigation of Zn:O atomic ratio would be meaningful for tracking the formation of ZnO. Using EDS analysis, which is shown in Table A1 (Appendix A), it was found that the as-printed film exhibits an average Zn:O atomic ratio of 0.378. In the case of using 4-W Joule heating, the atomic ratio of 0.588 could be observed. This Zn:O atomic ration further increases to 0.605 when 5-W heating power was applied. The rise of Zn:O atomic ratio is an important evidence of thermal decomposition reaction and the formation of ZnO by Joule heating. It could be seen that the Zn:O atomic ratio is not significantly different between 4- and 5-W heating power. Furthermore, there is a noticeable variation in this atomic ratio investigation through the pattern, which could be contributed to the grading of Elemental and morphological studies of the heat-treated film could provide more evidence for the mechanism of the generation of ZnO. Because the decomposition reaction of zinc acetate dihydrate shows the loss of oxygen element via product vaporization, an investigation of Zn:O atomic ratio would be meaningful for tracking the formation of ZnO. Using EDS analysis, which is shown in Table A1 (Appendix A), it was found that the as-printed film exhibits an average Zn:O atomic ratio of 0.378. In the case of using 4-W Joule heating, the atomic ratio of 0.588 could be observed. This Zn:O atomic ration further increases to 0.605 when 5-W heating power was applied. The rise of Zn:O atomic ratio is an important evidence of thermal decomposition reaction and the formation of ZnO by Joule heating. It could be seen that the Zn:O atomic ratio is not significantly different between 4- and 5-W heating power. Furthermore, there is a noticeable variation in this atomic ratio investigation through the pattern, which could be contributed to the grading of temperature through the surface.

temperature through the surface. Furthermore, morphological study depicts a remarkable change in film structure after Joule heating process (Figure 7). Figure 7a demonstrates the zinc precursor film before any heat treatment was applied. The film appears to be full of fractures, which could be the result of the solvent evaporation and the condensation of salt. There are remarkable changes in film morphology after the Joule heating treatment, such as the vanishing of notable fractures and the appearance of wrinkles on the surface of the film (Figure 7b,c). However, there is no significant difference between these two treated films. These film structures are commonly observed in the sol-gel derived film as our previous report [3]. These wrinkles could be originated from the internal stress of film during rapid solvent Furthermore, morphological study depicts a remarkable change in film structure after Joule heating process (Figure 7). Figure 7a demonstrates the zinc precursor film before any heat treatment was applied. The film appears to be full of fractures, which could be the result of the solvent evaporation and the condensation of salt. There are remarkable changes in film morphology after the Joule heating treatment, such as the vanishing of notable fractures and the appearance of wrinkles on the surface of the film (Figure 7b,c). However, there is no significant difference between these two treated films. These film structures are commonly observed in the sol-gel derived film as our previous report [3]. These wrinkles could be originated from the internal stress of film during rapid solvent withdrawal,

caused by the difference in thermal expansion of the gelation and underlying layer [28]. In addition, the transition from viscous to viscoelastic of the zinc precursor ink also contributes to the formation of these wrinkles [29]. withdrawal, caused by the difference in thermal expansion of the gelation and underlying layer [28]. In addition, the transition from viscous to viscoelastic of the zinc precursor ink also contributes to the formation of these wrinkles [29].

*Micromachines* **2020**, *11*, 490 7 of 11

**Figure 7. E**nergy-dispersive X-ray spectroscopy (EDS) analysis of zinc precursor film with different treatment conditions: (**a**) 0 W, (**b**) 4 W, (**c**) 5 W. Insets are FE–SEM images of film morphology according to each condition. **Figure 7.** Energy-dispersive X-ray spectroscopy (EDS) analysis of zinc precursor film with different treatment conditions: (**a**) 0 W, (**b**) 4 W, (**c**) 5 W. Insets are FE–SEM images of film morphology according to each condition.

### *3.4. UV Light Sensing Performance of the Sensor*

*3.4. UV Light Sensing Performance of the Sensor*  The sensing of UV light was demonstrated by the fabricated device and the result was presented in Figure 8. A bias voltage of 5 V was applied to the two terminals of device, which also serves as electrode by making metal contacts with the semiconductor film. The electrical current was recorded while the UV light was turned on/off (Figure 8a). It could be noted that the sample without applying heat treatment did not shows a notable response to UV light. On the other hand, sample with the treatment exhibits remarkable response to the short-wavelength light. When UV was turn on, the current raised significantly until it reached the equilibrium. On the other hand, when UV was turned off, the current quickly decay to the initial value. It is also worth noting that there is a remarkable difference of the samples with different heating conditions, such as the sample treated with 5-W power has photocurrent with a maximum value of 1.6 × 10<sup>−</sup>7 A, which is about ten times higher than that of the sample treated with 4-W power. The sensing of UV light was demonstrated by the fabricated device and the result was presented in Figure 8. A bias voltage of 5 V was applied to the two terminals of device, which also serves as electrode by making metal contacts with the semiconductor film. The electrical current was recorded while the UV light was turned on/off (Figure 8a). It could be noted that the sample without applying heat treatment did not shows a notable response to UV light. On the other hand, sample with the treatment exhibits remarkable response to the short-wavelength light. When UV was turn on, the current raised significantly until it reached the equilibrium. On the other hand, when UV was turned off, the current quickly decay to the initial value. It is also worth noting that there is a remarkable difference of the samples with different heating conditions, such as the sample treated with 5-W power has photocurrent with a maximum value of 1.6 <sup>×</sup> <sup>10</sup>−<sup>7</sup> A, which is about ten times higher than that of the sample treated with 4-W power.

The responsivity could be calculated using following expression: [30]

$$R = \frac{I\_{ph}}{PS} \tag{2}$$

where *Jph* is the photocurrent, *P* is the light intensity, and *S* is the effective area, which could be determined by the area of the ZnO pattern between two electrodes, such as 0.2 mm × 0.6 mm. Further analysis shows the responsivity values at 5.42 mW/cm<sup>2</sup> light intensity of the 5-W sample and the 4-W sample is 0.029 and 0.0027 A/W, respectively. Although the responsivity is lower than that has been reported in other inkjet-printed ZnO-based UV photodetector [31], it could be due to the obtained temperature is lower than the point where the reaction is totally finished. In Figure 8b, the I–V characteristics of prepared devices were presented, which show a linear current–voltage relationship

for both of the devices with treatment. This behavior indeed shows the Ohmic contact nature of the silver–ZnO interface. *Micromachines* **2020**, *11*, 490 8 of 11

**Figure 8.** Photo-sensing properties of the fabricated device. (**a**) Sequential illumination of the sensor by UV light and obtained photocurrent. (**b**) *I–V* curve of the devices under UV illumination. **Figure 8.** Photo-sensing properties of the fabricated device. (**a**) Sequential illumination of the sensor by UV light and obtained photocurrent. (**b**) *I–V* curve of the devices under UV illumination.

### The responsivity could be calculated using following expression: [30] **4. Discussion**

, (2) where *Jph* is the photocurrent, *P* is the light intensity, and *S* is the effective area, which could be determined by the area of the ZnO pattern between two electrodes, such as 0.2 mm × 0.6 mm. Further analysis shows the responsivity values at 5.42 mW/cm2 light intensity of the 5-W sample and the 4- W sample is 0.029 and 0.0027 A/W, respectively. Although the responsivity is lower than that has The photosensitivity of the device is strongly influenced by the heat treatment condition. As the discussion in previous section has pointed out, temperature of higher than 150 ◦C is critical to facilitate the reaction to form ZnO from the zinc salt. It is worth noting that photosensitivity is a common characteristic of a semiconductor. ZnO is a wide bandgap semiconductor (*Eg* = 3.35 eV at room temperature [32]). Therefore, it only be sensitive with larger photon energy of the incident light, such as the UV light used in this work which has photon energy of 3.40 eV according to 365 nm central wavelength of the LED.

been reported in other inkjet-printed ZnO-based UV photodetector [31], it could be due to the obtained temperature is lower than the point where the reaction is totally finished. In Figure 8b, the I–V characteristics of prepared devices were presented, which show a linear current–voltage relationship for both of the devices with treatment. This behavior indeed shows the Ohmic contact nature of the silver–ZnO interface. **4. Discussion**  The sample without heat treatment could not form ZnO by the thermal decomposition process. Therefore, the prepared film failed to work as a photodetector. Meanwhile, the sample processed with 4-W power demonstrates a remarkable response to UV illumination. It is because the temperature generated by the Joule heating process has facilitated the formation of ZnO semiconductor. As a wide bandgap semiconductor, the interaction of ZnO crystalline with high energy photons excites the generation of the electron-hole pair which increases the carrier concentration in the lattice [33]. Therefore, film resistance reduces and causes a surge in current running through the device.

The photosensitivity of the device is strongly influenced by the heat treatment condition. As the discussion in previous section has pointed out, temperature of higher than 150 °C is critical to facilitate the reaction to form ZnO from the zinc salt. It is worth noting that photosensitivity is a common characteristic of a semiconductor. ZnO is a wide bandgap semiconductor (*Eg* = 3.35 eV at room temperature [32]). Therefore, it only be sensitive with larger photon energy of the incident light, such as the UV light used in this work which has photon energy of 3.40 eV according to 365 nm central wavelength of the LED. The sample without heat treatment could not form ZnO by the thermal decomposition process. Therefore, the prepared film failed to work as a photodetector. Meanwhile, the sample processed with 4-W power demonstrates a remarkable response to UV illumination. It is because the Furthermore, the sample processed with 5-W power possesses a better performance in terms of responsivity to UV light. This improvement indicates that the magnitude of input power for Joule heating of semiconductor film could significantly influence the film properties. The root of this enhancement could be originated from the fact that higher temperatures could promote more formation of ZnO as previous discussion, therefore the interaction with photon is improved and more electron-hole pairs could be generated when exposing to the UV illumination. In addition, higher temperatures could also improve the contact between ZnO nanoparticles, therefore reduce the band bending at the interfaces and promote the transportation of electron through those grain boundaries [30].

temperature generated by the Joule heating process has facilitated the formation of ZnO semiconductor. As a wide bandgap semiconductor, the interaction of ZnO crystalline with high energy photons excites the generation of the electron-hole pair which increases the carrier concentration in the lattice [33]. Therefore, film resistance reduces and causes a surge in current running through the device. Furthermore, the sample processed with 5-W power possesses a better performance in terms of Current work employed silicon wafer due to its excellent heat resistance, because of the temperature during Joule heating could excess 250 ◦C, which most of polymer will not be able to withstand. One possible alternative solution is using high temperature polymer such as polyimide. However, due to the low thermal conductivity of polyimide comparing that of silicon wafer [34,35], the structure of device must be changed to reduce the distance from heater to the printed zinc salt in order to obtain a sufficient calcination.

responsivity to UV light. This improvement indicates that the magnitude of input power for Joule

#### heating of semiconductor film could significantly influence the film properties. The root of this **5. Conclusions**

enhancement could be originated from the fact that higher temperatures could promote more formation of ZnO as previous discussion, therefore the interaction with photon is improved and more electron-hole pairs could be generated when exposing to the UV illumination. In addition, higher temperatures could also improve the contact between ZnO nanoparticles, therefore reduce the band In this work, on-substrate heating synthesis of ZnO thin film by the printed silver resistive heater is proposed. Both conductive patterns and the precursor pattern were printed on only one inkjet printer with an exchange of cartridge for each material. Electrical current running through the

**5. Conclusions** 

order to obtain a sufficient calcination.

boundaries [30].

silver conductive pattern generates heat, which is utilized to facilitate the thermal decomposition of printed zinc precursor film to form ZnO. The magnitude of supplied power for heating process has significant influence on the film formation via the achieved temperature. Therefore, it is found that higher power could produce a better property of printed ZnO semiconductor film. Despite the humble performance of printed photodetector, this work demonstrates a promising approach to additively manufacture electronic devices, which reduces the number of equipment involved and the amount of energy consumed. Hence, integrated 3D printing might also be possible. printer with an exchange of cartridge for each material. Electrical current running through the silver conductive pattern generates heat, which is utilized to facilitate the thermal decomposition of printed zinc precursor film to form ZnO. The magnitude of supplied power for heating process has significant influence on the film formation via the achieved temperature. Therefore, it is found that higher power could produce a better property of printed ZnO semiconductor film. Despite the humble performance of printed photodetector, this work demonstrates a promising approach to additively manufacture electronic devices, which reduces the number of equipment involved and the amount of energy consumed. Hence, integrated 3D printing might also be possible.

is proposed. Both conductive patterns and the precursor pattern were printed on only one inkjet

*Micromachines* **2020**, *11*, 490 9 of 11

bending at the interfaces and promote the transportation of electron through those grain

Current work employed silicon wafer due to its excellent heat resistance, because of the temperature during Joule heating could excess 250 °C, which most of polymer will not be able to withstand. One possible alternative solution is using high temperature polymer such as polyimide. However, due to the low thermal conductivity of polyimide comparing that of silicon wafer [34,35], the structure of device must be changed to reduce the distance from heater to the printed zinc salt in

**Author Contributions:** Conceptualization, V.-T.T. and H.D.; methodology, V.-T.T. and Y.W.; formal analysis, V.Y.T.; investigation, V.-Y.T. and Y.W.; writing—original draft preparation, V.-T.T.; writing—review and editing, H.D.; visualization, V.-T.T.; supervision, H.D.; project administration, H.D. All authors have read and agreed to the published version of the manuscript. **Author Contributions:** Conceptualization, V.-T.T. and H.D.; methodology, V.-T.T. and Y.W.; formal analysis, V.Y.T.; investigation, V.-Y.T. and Y.W.; writing—original draft preparation, V.-T.T.; writing—review and editing, H.D.; visualization, V.-T.T.; supervision, H.D.; project administration, H.D. All authors have read and

**Funding:** This research was funded by Ministry of Education Academic Research Fund, Singapore. agreed to the published version of the manuscript

**Acknowledgments:** We appreciate the equipment support from Singapore Centre for 3D Printing (SC3DP). **Funding:** This research was funded by Ministry of Education Academic Research Fund, Singapore.

**Conflicts of Interest:** The authors declare no conflict of interest. **Acknowledgments:** We appreciate the equipment support from Singapore Centre for 3D Printing (SC3DP).

### **Appendix A Conflicts of Interest:** The authors declare no conflict of interest.

**References** 

435.

60.

doi:10.1039/C6NR08220C.

*Films* **2013**, *544*, 509–514.

doi:10.1039/C9NR05776E.

**Table A1.** EDS Investigation of Zn:O atomic ration at random positions on printed pattern. **Appendix A** 


**Figure A1.** Waveform of signal applying to cartridge nozzles to create jetting of ink: (**a**) silver ink and (**b**) zinc salt ink. **Figure A1.** Waveform of signal applying to cartridge nozzles to create jetting of ink: (**a**) silver ink and *Micromachines* (**b**) zinc salt ink. **2020**, *11*, 490 10 of 11

**Figure A2.** Low-magnification FE–SEM images of two samples before (**a**) and after heat treatment (**b**). **Figure A2.** Low-magnification FE–SEM images of two samples before (**a**) and after heat treatment (**b**).

3. Tran, V.-T.; Wei, Y.; Yang, H.; Zhan, Z.; Du, H. All-inkjet-printed flexible ZnO micro photodetector for a

4. Yus, J.; Gonzalez, Z.; Sanchez-Herencia, A.J.; Sangiorgi, A.; Sangiorgi, N.; Gardini, D.; Sanson, A.; Galassi, C.; Caballero, A.; Morales, J.; et al. Semiconductor water-based inks: Miniaturized NiO pseudocapacitor

5. Sanchez, J.G.; Balderrama, V.S.; Garduno, S.I.; Osorio, E.; Viterisi, A.; Estrada, M.; Ferre-Borrull, J.; Pallares, J.; Marsal, L.F. Impact of inkjet printed ZnO electron transport layer on the characteristics of polymer solar

6. Ganesan, S.; Gollu, S.R.; Alam khan, J.; Kushwaha, A.; Gupta, D. Inkjet printing of zinc oxide and P3HT:ICBA in ambient conditions for inverted bulk heterojunction solar cells. *Opt. Mater.* **2019**, *94*, 430–

7. Hasan, K.u.; Nur, O.; Willander, M. Screen printed ZnO ultraviolet photoconductive sensor on pencil

8. Liang, Y.N.; Lok, B.K.; Wang, L.; Feng, C.; Lu, A.C.W.; Mei, T.; Hu, X. Effects of the morphology of inkjet printed zinc oxide (ZnO) on thin film transistor performance and seeded ZnO nanorod growth. *Thin Solid* 

9. Jiang, L.; Li, J.; Huang, K.; Li, S.; Wang, Q.; Sun, Z.; Mei, T.; Wang, J.; Zhang, L.; Wang, N.; et al. Low-Temperature and Solution-Processable Zinc Oxide Transistors for Transparent Electronics. *ACS Omega* 

10. Friedmann, D.; Lee, A.F.; Wilson, K.; Jalili, R.; Caruso, R.A. Printing approaches to inorganic semiconductor

11. Gagnon, J.C.; Presley, M.; Le, N.Q.; Montalbano, T.J.; Storck, S. A pathway to compound semiconductor

12. Chae, D.; Kim, J.; Shin, J.; Lee, W.H.; Ko, S. A low-temperature and short-annealing process for metal oxide thin film transistors using deep ultraviolet light for roll-to-roll processing. *Curr. Appl. Phys.* **2019**, *19*, 954–

13. Jeon, J.B.; Kim, B.J.; Bang, G.J.; Kim, M.-C.; Lee, D.G.; Lee, J.M.; Lee, M.; Han, H.S.; Boschloo, G.; Lee, S.; et al. Photo-annealed amorphous titanium oxide for perovskite solar cells. *Nanoscale* **2019**, *11*, 19488–19496,

14. Palneedi, H.; Park, J.H.; Maurya, D.; Peddigari, M.; Hwang, G.-T.; Annapureddy, V.; Kim, J.-W.; Choi, J.-J.; Hahn, B.-D.; Priya, S.; et al. Laser Irradiation of Metal Oxide Films and Nanostructures: Applications and

15. Sharma, S.; Pande, S.S.; Swaminathan, P. Top-down synthesis of zinc oxide based inks for inkjet printing.

16. Rao, A.; Long, H.; Harley-Trochimczyk, A.; Pham, T.; Zettl, A.; Carraro, C.; Maboudian, R. In Situ Localized Growth of Ordered Metal Oxide Hollow Sphere Array on Microheater Platform for Sensitive, Ultra-Fast

Gas Sensing. *ACS Appl. Mater. Interfaces* **2017**, *9*, 2634–2641, doi:10.1021/acsami.6b12677.

photocatalyst fabrication. *J. Mater. Chem. A* **2019**, *7*, 10858–10878, doi:10.1039/C9TA00888H.

drawn circuitry over paper. *Appl. Phys. Lett.* **2012**, *100*, 211104, doi:10.1063/1.4720179.

flexible electronics. *J. Mater. Chem. C* **2019**, *7*, 8771–8795, doi:10.1039/C9TC01630A.

wearable UV monitoring device. *Nanotechnology* **2017**, *28*, 095204.

electrodes by inkjet printing. *J. Eur. Ceram. Soc.* **2019**, *39*, 2908–2914.

cells. *RSC Adv.* **2018**, *8*, 13094–13102, doi:10.1039/C8RA01481G.

**2017**, *2*, 8990–8996, doi:10.1021/acsomega.7b01420.

additive manufacturing. *MRS Commun.* **2019**, *9*, 1001–1007.

*RSC Adv.* **2017**, *7*, 39411–39419, doi:10.1039/C7RA07150G.

Advances. *Adv. Mater.* **2018**, *30*, 1705148, doi:10.1002/adma.201705148.
