- Email: [email protected]
Direct patterning of silver electrodes with 2.4 μm channel length by piezoelectric inkjet printing
Journal of Colloid and Interface Science 487 (2017) 68–72
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science journal homepage: www.elsevier.com/locate/jcis
Regular Article
Direct patterning of silver electrodes with 2.4 lm channel length by piezoelectric inkjet printing Honglong Ning, Ruiqiang Tao, Zhiqiang Fang, Wei Cai, Jianqiu Chen, Yicong Zhou, Zhennan Zhu, Zeke Zheng, Rihui Yao ⇑, Miao Xu, Lei Wang, Linfeng Lan, Junbiao Peng ⇑ Institute of Polymer Optoelectronic Materials & Devices, State Key Laboratory of Luminescent Materials & Devices, South China University of Technology, PR China
g r a p h i c a l a b s t r a c t
Droplets
5
Channel length: 2.4 μm 41
43
47
50
55
5859
64
66 67
Droplet diameter: 50 μm 4
43
47
3
55
47
2
58
64
58
64
66
58
40
45
50 55 60 Drop space (um)
a r t i c l e
65
i n f o
Article history: Received 27 July 2016 Revised 24 September 2016 Accepted 8 October 2016 Available online 11 October 2016 Keywords: Short channel length Inkjet printing Direct patterning Untreated substrate Thin film transistors
a b s t r a c t The control of channel length is of great significance in the fabrication of thin film transistors (TFTs) with high-speed operation. However, achieving short channel on untreated glass by traditional piezoelectric inkjet printing is problematic due to the impacting and rebounding behaviors of droplet impinging on solid surface. Here a novel method was proposed to obtain short channel length on untreated glass by taking advantage of the difference in the retraction velocities on both sides of an ink droplet. In addition, droplets contact mechanism was first introduced in our work to explain the formation of short channel in the printing process. Through printing droplets array with optimized drop space and adjusting appropriate printing parameters, a 2.4 lm of channel length for TFT, to the best of our knowledge, which is the shortest channel on substrate without pre-patterning, was achieved using piezoelectric inkjet printing. This study sheds light on the fabrication of short channel TFT for large size and high-resolution displays using inkjet printing technology. Ó 2016 Elsevier Inc. All rights reserved.
1. Introduction Over the past decades, printed electronics have attracted significant attention due to their low-cost, eco-friendly, non-contact, direct-writing, fast, and large scale processing [1–3]. A variety of printed electronics with desirable performance were fabricated through using new printable materials [4,5] and advanced ⇑ Corresponding authors. E-mail addresses: [email protected] (R. Yao), [email protected] (J. Peng). http://dx.doi.org/10.1016/j.jcis.2016.10.016 0021-9797/Ó 2016 Elsevier Inc. All rights reserved.
nozzle-based methods [6–10], such as inkjet printing, aerosol-jet printing, and electron-hydrodynamic jet printing. As the increasing demand on the miniaturization of printed electronics, highresolution electronics printed by traditional drop-on-demand (DOD) piezoelectric inkjet printer becomes a big challenge [6,11,12]. To overcome this problem, many efforts have been devoted to obtain small channel length for electronics, for example, surface treatment of substrates [6,13], lithography [14] or laser ablation [15].
H. Ning et al. / Journal of Colloid and Interface Science 487 (2017) 68–72
However, little attention has been paid to control channel length by manipulating the spreading and relaxation processes of ink droplet impinging on glass [16–21]. The impinging process for a single droplet is well reported, but little know about the process for multiple droplets. As multiple droplets impact on glass, it’s inevitable to induce droplets contact phenomenon. The impinging process of later released droplet can be significantly influenced by the wetted substrate caused by former droplet if the position of these droplets is close enough. Here we proposed a method based on new pattern design rules to obtain electrodes with short channel length on glass using inkjet printing technology. Through printing droplets array with optimized drop space and adjusting appropriate printing parameters, the channel length between printed source and drain electrodes can be easily obtained without any additional treatments of glass. Thus the real advantages of the direct writing inkjet printing technology using in the fabrication of electronics can be achieved. Besides, droplets contact mechanism was first introduced in our work to explain the formation of short channel in the printing process.
2. Experimental Silver nanoparticles dispersed in triethylene glycol monoethyl ether, with solid contents of 45 wt% and average particle diameter of about 50 nm, were used in this experiment (DGP-45LT-15C, Advanced Nano Products). The ink was ejected from a piezoelectric drop on demand inkjet printing system (DMP-2831, FUJIFILM Diamtix, USA) with the DMC-11610 (10 pL, Nozzle diameter 21 lm) cartridge onto clean glass substrate under ambient conditions (22 °C, 70% relative humidity). Most of the patterns were printed at temperature of 28 °C, and droplets firing rate of 2 m/s. The patterns as printed were 180 s UV cured for the fast solidification using a 600 W UV curing equipment (IntelliRay, Uvitron, USA). The diameter of droplets printed onto glass substrate in our experiment is approximately 50 lm. Optimized drop space is 35 lm for depositing continuous films. Two 100 lm width lines with 35 lm drop space and 173 lm lines pitch were designed to present the smallest channel length that can be achieved using ordinary printing method. The final channel length is about 37.2 lm, which is not enough for the demand of advanced electronics. However, when further decreasing the line pitch by only 1 lm, the channel disappears, which is also verified as following. Droplet array with different drop space in lateral and vertical directions, was achieved by designing the array with unit of 10 lm side length square, which is actually one single droplet released by our printer through the automatic calculation of the belonging DMP software. We printed patterns consist of two to five droplets with different drop space to explore the contact mechanism of droplets. The obtained rules were introduced to print droplet array, in order to find the optimized drop space in lateral and vertical directions, for the forming of desired short channel. Then the substrate temperature was adjusted to fill the defects of
69
interspace besides the channel, and meanwhile the channel length was further reduced to only 2.4 lm. 3. Results and discussion 3.1. The model for single droplet impinges on solid surface For glass substrate, the droplet impact process can be classified into five time-sequential phases: kinematic, spreading, retracting, wetting, and equilibrium [18,21]. The spreading and equilibrium states, which can significantly influence the printing process, are shown in Fig. 1(a) and (b), respectively. The diameter of the maximum spreading state (Dmax) follows equations: 1
Dmax / We4
ð1Þ
qv 20 D0 ð2Þ r where v0 is the droplet firing rate, and r represents the surface ten-
We ¼
sion of the droplet. The small surface tension of the ink makes the Dmax much higher than the diameter of the equilibrium state (De). 3.2. Limited channel length printed using conventional pattern design method As shown in Fig. 2(a), pattern of two lines with line width of 100 lm and drop space of 35 lm was designed to obtain the shortest channel by the optimization of lines pitch. When lines pitch decreased from 173 lm to 172 lm, the channel disappeared completely, which means the shortest channel length of about 37.2 lm was achieved. Droplets were released line by line with certain travel route, which is presented in Fig. 2(b), to form patterns. We suppose that the interval time of droplets releasing is smaller than both of the spreading time and the retracting time of a single droplet impinges substrate referred to the recent report [18], which also suggested that the spreading time is smaller than the retracting time. Based on these facts, the intermediate state of printed droplets can be expressed in Fig. 2(c) for lines pitch above 173 lm. The residual area was formed during the retracting process of the released droplets of the former printed line, and will slow down the retraction velocity of the contacted side of droplet. For large lines pitch, the first droplet of the latter printed line spreads its maximum diameter to the residual area, but not contact with the last droplet of the former printed line. Thus a small migration distance of the latter printed line was induced, but the channel still exists. However, when the lines pitch was decreased below 172 lm, not only the residual area, but also the last droplet of the former printed line can be contacted by the first droplet of the latter printed line, as shown in Fig. 2(d). The merging process of droplets will produce a large driving force for the migration of droplet, which finally make the channel disappear. Accordingly,
Fig. 1. Schematic illustration of the process of single droplet impinges on glass substrate. (a) The spreading state of a droplet, Dmax denotes the maximum diameter of the droplet. (b) The equilibrium state of a droplet, De is the diameter of the droplet.
70
H. Ning et al. / Journal of Colloid and Interface Science 487 (2017) 68–72
Fig. 2. (a) 100 lm width lines with lines pitch of 173 lm and 172 lm, respectively. (b) Schematic of the travel route and released droplets calculated by software for the formation of lines with a drop space of 35 lm. (c) Schematic of droplets behavior with a large line pitch for the formation of a short channel. (d) Schematic of droplets behavior with a small line pitch that reveals the limitation of the shortest channel formed by using conventional pattern design method.
we consider that the droplets contact phenomenon is the main reason of the sudden disappearance of the channel by decreasing the lines pitch for only 1 lm. Actually, in our other experiment, the smallest lines pitch for the achieving of the shortest channel length are not complete lineally increased as the increasing of lines width (Fig. S1). This phenomenon implies that the channel was not only influenced by the droplets of the edge of lines. It’s convincing that further exploration of droplets contact mechanism may have possibilities to push the limits of the shortest channel length direct patterned on glass by traditional piezoelectric inkjet printing. 3.3. Droplets contact rules of printed droplets based on DOD inkjet printing The printed patterns shown in Fig. 3(a) reveal the presence of several micron width interspace defects with drop space of 45 lm, droplet firing rate of 2 m/s and substrate temperature of 28 °C. The defects, which distribute asymmetric and random, was firstly thought to be resulted from the inadequate location precision of Dimatix 2800 printer and the instability of the ink. In order to further explain this phenomenon, we printed several lines of different line width using the same printing parameters as shown in Fig. 3(b). The obvious defects circled were found to be consistent with the defects of the patterns printed in Fig. 3(a). The essential physical idea was explored by printing patterns with two to five droplets with different drop space. As shown in Fig. 4(a), for patterns of two droplets, the merged droplets separated until the drop space was increased to 58 lm; for patterns of three droplets, the first two droplets also separated at a drop space of 58 lm, but the drop space should be increased to 64 lm for the separation of all droplets and decreased to 46 lm for the merging of all droplets. Although the situation of patterns of four droplets and five droplets is more complex, it follows the same drop contact regularities.
Fig. 3. (a and b) Printed patterns with drop space of 45 lm, droplets firing rate of 2 m/s, and substrate temperature of 28 °C.
Fig. 4(b) shows that the droplets merging phenomenon happens more difficult with the increasing of drop space. For example, if we want to make sure that two droplets merging phenomenon happen, more droplets need to be printed when required drop space is increased. Meanwhile, when drop space is fixed, the least number of droplets for the happening of droplets contact phenomenon with fixed merged droplets number can be determined. Fig. 4(c) indicates that as the decreasing of drop space, the number of droplets merged with the first printed droplet increased. Fig. 4(d) indicates that as the increasing of drop space, the number of individual droplets before the merging of droplets increased. Besides the general tendency described above, the other processes also follow the similar rules. When the first droplet impinges on glass, the spreading and retraction processes is uniform, but the induced residual area will make the retraction velocity of the closer side of the second droplet slower. The difference in
H. Ning et al. / Journal of Colloid and Interface Science 487 (2017) 68–72
71
Fig. 4. (a) Contact phenomenon of two droplets, three droplets, four droplets and five droplets, (b) the least number of droplets for the forming of larger droplet merged by two droplets, three droplets, four droplets and five droplets, (c) the number of the first printed droplet and the droplets merged with it, and (d) the number of individual droplets before the merging of droplets, with different drop space.
the retraction velocities on both sides of the second droplet will make its residual area different from the first droplet, which make the retraction process of the third droplet differs from the second droplet. The wetting effect of the former deposited droplets accumulate drop by drop, thus the happening of droplets merging phenomenon starts from the last printed droplets as the decreasing of drop space. This accumulation effect can be released by the merging of droplets, thus droplet deposited after merged droplets separates again. The lateral droplets contact experiment carried out exhibits the same tendency for drop contact. We repeated the printing of the
patterns in Fig. 4 for ten times, both of them present the same regularity. 3.4. Direct patterning of the shortest channel length of 2.4 lm The real printing precision of inkjet printing technology was ignored in the past, because of the inadequate understanding of the significance of the droplets wetting and the accumulation effect. The droplets distance shown in Fig. 4, which is below 10 lm, reveals the potential of achieving much shorter channel length using the droplets contact pattern design method.
Table 1 The optimization of drop space for the obtaining of patterns with the shortest channel (X represents the drop space of the lateral direction; Y represents the drop space of the vertical direction).
72
H. Ning et al. / Journal of Colloid and Interface Science 487 (2017) 68–72
Fig. 5. Optimized patterns for the printing of the shortest channel. (a) With drop space of 45 lm, droplets firing rate of 2 m/s and substrate temperature of 28 °C. (b) With defect concentrated area removed. (c) With the substrate temperature increased to 45 °C.
We printed droplet arrays with different lateral and vertical drop space, for the optimization of patterns with the shortest channel length. The patterns shown in Table 1 reveal the general tendency of the formation of interspace. Most of the interspace have a width below 10 lm. The resulted structures of printed droplet changed sensitively with the increasing of drop space by only 1 lm, accordingly providing possibilities for this method to obtain desired patterns. The different droplets contact phenomenon for the lateral and the vertical directions owing to the different interval time of printed droplets for these two directions, which can be explained by the fixed travel route of nozzles as described in Fig. 2(b). Pattern with only one channel is needed for the fabrication of source and drain electrodes of TFTs. Fig. 5(a) shows one of the optimized drop space, based on the results in Table 1. The drop space for both directions was 45 lm, the firing rate was 2 m/s, and the substrate temperature was 28 °C. Besides the channel, interspace also existed in Fig. 5(a). In order to eliminate the unwanted interspace, we reduced the side length of the designed pattern from 500 lm to 400 lm, and the optimized pattern is shown in Fig. 5(b). The substrate temperature was increased from 28 °C to 45 °C for further eliminating of the unwanted interspace. More details about the influences of substrate temperature to the short channel structure are shown in Fig. S2. The surface energy of substrate was increased with the increasing of substrate temperature. Thus the Dmax and De of droplets also slightly increased. As shown in Fig. 5(c), the interspace was filled and the channel length was reduced to only 2.4 lm, which is the shortest channel length ever reported achieved by direct patterning of inkjet printing. Moreover, the channel becomes smoother with the increased substrate temperature. We considered this is the real limitation of short channel printing by traditional piezoelectric inkjet printer with 10 pL cartridge (Nozzle diameter 21 lm). The short channel structure can be easily repeated using the method we proposed, and the results are shown in Fig. S3. 4. Conclusion In summary, a novel approach was proposed to prepare short channel structure silver electrodes on untreated solid substrates by making use of the difference in the retraction velocities on both sides of an ink droplet during printing process. Consequently, a 2.4 lm channel length was obtained based on this novel approach by traditional piezoelectric inkjet printing. Moreover, droplets contact mechanism was first introduced in our work to explain the formation of short channel in the printing process. The method for preparing short channel structure is repeatable and extremely
simple, thus provides referential significance for new design rules of printed patterns. Acknowledgements This work was supported by National Program on Key Basic Research Project (973) (No. 2015CB655004), National Key R&D Program (Nos. 2016YFB0401504 and 2016YFF0203603), Guangdong Natural Science Foundation (No. 2016A030313459), Science and Technology Project of Guangdong Province (Nos. 2014B090915004, 2016B090907001, 2014A040401014, 2016B090906002, 2015A010101323, 2014B090916002, 2015A010101323, 2015B090915001 and 2015B090914003), Educational Commission of Guangdong Province (Nos. 2014KZDXM010 and 2015KTSCX003), the Fundamental Research Funds for the Central Universities (Nos. 2015ZP024 and 2015ZZ063), State Key Laboratory of Luminescence and Applications (SKLA-2016-11). Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2016.10.016. References [1] W.J. Hyun, E.B. Secor, G.A. Rojas, M.C. Hersam, L.F. Francis, C.D. Frisbie, Adv. Mater. 27 (2015) 7058. [2] Z. Zhang, X. Zhang, Z. Xin, M. Deng, Y. Wen, Y. Song, Adv. Mater. 25 (2013) 6714. [3] D.J. Finn, M. Lotya, J.N. Coleman, ACS Appl. Mater. Interfaces 7 (2015) 9254. [4] Y. Kim, S. Jang, J.H. Oh, Appl. Phys. Lett. 106 (2015) 14103. [5] J. Park, J. Hwang, J. Phys. D Appl. Phys. 47 (2014) 405102. [6] Y. Kwon, Y. Lee, K. Lee, Y. Choi, Y. Choa, J. Electron. Mater. 44 (2015) 2608. [7] X. Wu, Z. Chen, T. Zhou, S. Shao, M. Xie, M. Song, Z. Cui, RSC Adv. 5 (2015) 20924. [8] K. Hong, S.H. Kim, A. Mahajan, C.D. Frisbie, ACS Appl. Mater. Interfaces 6 (2014) 18704. [9] R. Aga, C. Jordan, R.S. Aga, C.M. Bartsch, E.M. Heckman, IEEE Electr. Device Lett. 35 (2014) 1124. [10] L. Kuai, J. Wang, T. Ming, C. Fang, Z. Sun, B. Geng, J. Wang, Sci. Rep.-UK 5 (2015) 9923. [11] Y. Li, L. Lan, P. Xiao, S. Sun, Z. Lin, W. Song, E. Song, P. Gao, W. Wu, J. Peng, ACS Appl. Mater. Interfaces (2016). [12] M. Mashayekhi, A. Conde, T.N. Ng, P. Mei, E. Ramon, J. Display Technol. (2015). [13] T. Wei, F. Linrun, Z. Jiaging, C. Qingyu, C. Sujie, G. Xiaojun, J. Mater. Chem. C 2 (2014) 1995. [14] H. Wang, C. Cheng, L. Zhang, H. Liu, Y. Zhao, Y. Guo, W. Hu, G. Yu, Y. Liu, Adv. Mater. 26 (2014) 4683. [15] I. Theodorakos, F. Zacharatos, R. Geremia, D. Karnakis, I. Zergioti, Appl. Surf. Sci. 336 (2015) 157. [16] J. de Ruiter, D. van den Ende, F. Mugele, Phys. Fluids 27 (2015) 12105. [17] H. Fujimoto, S. Watanabe, T. Okamoto, T. Hama, H. Takuda, Exp. Therm. Fluid Sci. 60 (2015) 66. [18] H.K. Huh, S. Jung, K.W. Seo, S.J. Lee, Microfluid Nanofluid 18 (2015) 1221. [19] S. Ganesan, S. Rajasekaran, L. Tobiska, Int. J. Heat Mass Transfer 78 (2014) 670. [20] M. Taghilou, M.H. Rahimian, Int. J. Therm. Sci. 86 (2014) 1. [21] J.B. Lee, S.H. Lee, Langmuir 27 (2011) 6565.
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science journal homepage: www.elsevier.com/locate/jcis
Regular Article
Direct patterning of silver electrodes with 2.4 lm channel length by piezoelectric inkjet printing Honglong Ning, Ruiqiang Tao, Zhiqiang Fang, Wei Cai, Jianqiu Chen, Yicong Zhou, Zhennan Zhu, Zeke Zheng, Rihui Yao ⇑, Miao Xu, Lei Wang, Linfeng Lan, Junbiao Peng ⇑ Institute of Polymer Optoelectronic Materials & Devices, State Key Laboratory of Luminescent Materials & Devices, South China University of Technology, PR China
g r a p h i c a l a b s t r a c t
Droplets
5
Channel length: 2.4 μm 41
43
47
50
55
5859
64
66 67
Droplet diameter: 50 μm 4
43
47
3
55
47
2
58
64
58
64
66
58
40
45
50 55 60 Drop space (um)
a r t i c l e
65
i n f o
Article history: Received 27 July 2016 Revised 24 September 2016 Accepted 8 October 2016 Available online 11 October 2016 Keywords: Short channel length Inkjet printing Direct patterning Untreated substrate Thin film transistors
a b s t r a c t The control of channel length is of great significance in the fabrication of thin film transistors (TFTs) with high-speed operation. However, achieving short channel on untreated glass by traditional piezoelectric inkjet printing is problematic due to the impacting and rebounding behaviors of droplet impinging on solid surface. Here a novel method was proposed to obtain short channel length on untreated glass by taking advantage of the difference in the retraction velocities on both sides of an ink droplet. In addition, droplets contact mechanism was first introduced in our work to explain the formation of short channel in the printing process. Through printing droplets array with optimized drop space and adjusting appropriate printing parameters, a 2.4 lm of channel length for TFT, to the best of our knowledge, which is the shortest channel on substrate without pre-patterning, was achieved using piezoelectric inkjet printing. This study sheds light on the fabrication of short channel TFT for large size and high-resolution displays using inkjet printing technology. Ó 2016 Elsevier Inc. All rights reserved.
1. Introduction Over the past decades, printed electronics have attracted significant attention due to their low-cost, eco-friendly, non-contact, direct-writing, fast, and large scale processing [1–3]. A variety of printed electronics with desirable performance were fabricated through using new printable materials [4,5] and advanced ⇑ Corresponding authors. E-mail addresses: [email protected] (R. Yao), [email protected] (J. Peng). http://dx.doi.org/10.1016/j.jcis.2016.10.016 0021-9797/Ó 2016 Elsevier Inc. All rights reserved.
nozzle-based methods [6–10], such as inkjet printing, aerosol-jet printing, and electron-hydrodynamic jet printing. As the increasing demand on the miniaturization of printed electronics, highresolution electronics printed by traditional drop-on-demand (DOD) piezoelectric inkjet printer becomes a big challenge [6,11,12]. To overcome this problem, many efforts have been devoted to obtain small channel length for electronics, for example, surface treatment of substrates [6,13], lithography [14] or laser ablation [15].
H. Ning et al. / Journal of Colloid and Interface Science 487 (2017) 68–72
However, little attention has been paid to control channel length by manipulating the spreading and relaxation processes of ink droplet impinging on glass [16–21]. The impinging process for a single droplet is well reported, but little know about the process for multiple droplets. As multiple droplets impact on glass, it’s inevitable to induce droplets contact phenomenon. The impinging process of later released droplet can be significantly influenced by the wetted substrate caused by former droplet if the position of these droplets is close enough. Here we proposed a method based on new pattern design rules to obtain electrodes with short channel length on glass using inkjet printing technology. Through printing droplets array with optimized drop space and adjusting appropriate printing parameters, the channel length between printed source and drain electrodes can be easily obtained without any additional treatments of glass. Thus the real advantages of the direct writing inkjet printing technology using in the fabrication of electronics can be achieved. Besides, droplets contact mechanism was first introduced in our work to explain the formation of short channel in the printing process.
2. Experimental Silver nanoparticles dispersed in triethylene glycol monoethyl ether, with solid contents of 45 wt% and average particle diameter of about 50 nm, were used in this experiment (DGP-45LT-15C, Advanced Nano Products). The ink was ejected from a piezoelectric drop on demand inkjet printing system (DMP-2831, FUJIFILM Diamtix, USA) with the DMC-11610 (10 pL, Nozzle diameter 21 lm) cartridge onto clean glass substrate under ambient conditions (22 °C, 70% relative humidity). Most of the patterns were printed at temperature of 28 °C, and droplets firing rate of 2 m/s. The patterns as printed were 180 s UV cured for the fast solidification using a 600 W UV curing equipment (IntelliRay, Uvitron, USA). The diameter of droplets printed onto glass substrate in our experiment is approximately 50 lm. Optimized drop space is 35 lm for depositing continuous films. Two 100 lm width lines with 35 lm drop space and 173 lm lines pitch were designed to present the smallest channel length that can be achieved using ordinary printing method. The final channel length is about 37.2 lm, which is not enough for the demand of advanced electronics. However, when further decreasing the line pitch by only 1 lm, the channel disappears, which is also verified as following. Droplet array with different drop space in lateral and vertical directions, was achieved by designing the array with unit of 10 lm side length square, which is actually one single droplet released by our printer through the automatic calculation of the belonging DMP software. We printed patterns consist of two to five droplets with different drop space to explore the contact mechanism of droplets. The obtained rules were introduced to print droplet array, in order to find the optimized drop space in lateral and vertical directions, for the forming of desired short channel. Then the substrate temperature was adjusted to fill the defects of
69
interspace besides the channel, and meanwhile the channel length was further reduced to only 2.4 lm. 3. Results and discussion 3.1. The model for single droplet impinges on solid surface For glass substrate, the droplet impact process can be classified into five time-sequential phases: kinematic, spreading, retracting, wetting, and equilibrium [18,21]. The spreading and equilibrium states, which can significantly influence the printing process, are shown in Fig. 1(a) and (b), respectively. The diameter of the maximum spreading state (Dmax) follows equations: 1
Dmax / We4
ð1Þ
qv 20 D0 ð2Þ r where v0 is the droplet firing rate, and r represents the surface ten-
We ¼
sion of the droplet. The small surface tension of the ink makes the Dmax much higher than the diameter of the equilibrium state (De). 3.2. Limited channel length printed using conventional pattern design method As shown in Fig. 2(a), pattern of two lines with line width of 100 lm and drop space of 35 lm was designed to obtain the shortest channel by the optimization of lines pitch. When lines pitch decreased from 173 lm to 172 lm, the channel disappeared completely, which means the shortest channel length of about 37.2 lm was achieved. Droplets were released line by line with certain travel route, which is presented in Fig. 2(b), to form patterns. We suppose that the interval time of droplets releasing is smaller than both of the spreading time and the retracting time of a single droplet impinges substrate referred to the recent report [18], which also suggested that the spreading time is smaller than the retracting time. Based on these facts, the intermediate state of printed droplets can be expressed in Fig. 2(c) for lines pitch above 173 lm. The residual area was formed during the retracting process of the released droplets of the former printed line, and will slow down the retraction velocity of the contacted side of droplet. For large lines pitch, the first droplet of the latter printed line spreads its maximum diameter to the residual area, but not contact with the last droplet of the former printed line. Thus a small migration distance of the latter printed line was induced, but the channel still exists. However, when the lines pitch was decreased below 172 lm, not only the residual area, but also the last droplet of the former printed line can be contacted by the first droplet of the latter printed line, as shown in Fig. 2(d). The merging process of droplets will produce a large driving force for the migration of droplet, which finally make the channel disappear. Accordingly,
Fig. 1. Schematic illustration of the process of single droplet impinges on glass substrate. (a) The spreading state of a droplet, Dmax denotes the maximum diameter of the droplet. (b) The equilibrium state of a droplet, De is the diameter of the droplet.
70
H. Ning et al. / Journal of Colloid and Interface Science 487 (2017) 68–72
Fig. 2. (a) 100 lm width lines with lines pitch of 173 lm and 172 lm, respectively. (b) Schematic of the travel route and released droplets calculated by software for the formation of lines with a drop space of 35 lm. (c) Schematic of droplets behavior with a large line pitch for the formation of a short channel. (d) Schematic of droplets behavior with a small line pitch that reveals the limitation of the shortest channel formed by using conventional pattern design method.
we consider that the droplets contact phenomenon is the main reason of the sudden disappearance of the channel by decreasing the lines pitch for only 1 lm. Actually, in our other experiment, the smallest lines pitch for the achieving of the shortest channel length are not complete lineally increased as the increasing of lines width (Fig. S1). This phenomenon implies that the channel was not only influenced by the droplets of the edge of lines. It’s convincing that further exploration of droplets contact mechanism may have possibilities to push the limits of the shortest channel length direct patterned on glass by traditional piezoelectric inkjet printing. 3.3. Droplets contact rules of printed droplets based on DOD inkjet printing The printed patterns shown in Fig. 3(a) reveal the presence of several micron width interspace defects with drop space of 45 lm, droplet firing rate of 2 m/s and substrate temperature of 28 °C. The defects, which distribute asymmetric and random, was firstly thought to be resulted from the inadequate location precision of Dimatix 2800 printer and the instability of the ink. In order to further explain this phenomenon, we printed several lines of different line width using the same printing parameters as shown in Fig. 3(b). The obvious defects circled were found to be consistent with the defects of the patterns printed in Fig. 3(a). The essential physical idea was explored by printing patterns with two to five droplets with different drop space. As shown in Fig. 4(a), for patterns of two droplets, the merged droplets separated until the drop space was increased to 58 lm; for patterns of three droplets, the first two droplets also separated at a drop space of 58 lm, but the drop space should be increased to 64 lm for the separation of all droplets and decreased to 46 lm for the merging of all droplets. Although the situation of patterns of four droplets and five droplets is more complex, it follows the same drop contact regularities.
Fig. 3. (a and b) Printed patterns with drop space of 45 lm, droplets firing rate of 2 m/s, and substrate temperature of 28 °C.
Fig. 4(b) shows that the droplets merging phenomenon happens more difficult with the increasing of drop space. For example, if we want to make sure that two droplets merging phenomenon happen, more droplets need to be printed when required drop space is increased. Meanwhile, when drop space is fixed, the least number of droplets for the happening of droplets contact phenomenon with fixed merged droplets number can be determined. Fig. 4(c) indicates that as the decreasing of drop space, the number of droplets merged with the first printed droplet increased. Fig. 4(d) indicates that as the increasing of drop space, the number of individual droplets before the merging of droplets increased. Besides the general tendency described above, the other processes also follow the similar rules. When the first droplet impinges on glass, the spreading and retraction processes is uniform, but the induced residual area will make the retraction velocity of the closer side of the second droplet slower. The difference in
H. Ning et al. / Journal of Colloid and Interface Science 487 (2017) 68–72
71
Fig. 4. (a) Contact phenomenon of two droplets, three droplets, four droplets and five droplets, (b) the least number of droplets for the forming of larger droplet merged by two droplets, three droplets, four droplets and five droplets, (c) the number of the first printed droplet and the droplets merged with it, and (d) the number of individual droplets before the merging of droplets, with different drop space.
the retraction velocities on both sides of the second droplet will make its residual area different from the first droplet, which make the retraction process of the third droplet differs from the second droplet. The wetting effect of the former deposited droplets accumulate drop by drop, thus the happening of droplets merging phenomenon starts from the last printed droplets as the decreasing of drop space. This accumulation effect can be released by the merging of droplets, thus droplet deposited after merged droplets separates again. The lateral droplets contact experiment carried out exhibits the same tendency for drop contact. We repeated the printing of the
patterns in Fig. 4 for ten times, both of them present the same regularity. 3.4. Direct patterning of the shortest channel length of 2.4 lm The real printing precision of inkjet printing technology was ignored in the past, because of the inadequate understanding of the significance of the droplets wetting and the accumulation effect. The droplets distance shown in Fig. 4, which is below 10 lm, reveals the potential of achieving much shorter channel length using the droplets contact pattern design method.
Table 1 The optimization of drop space for the obtaining of patterns with the shortest channel (X represents the drop space of the lateral direction; Y represents the drop space of the vertical direction).
72
H. Ning et al. / Journal of Colloid and Interface Science 487 (2017) 68–72
Fig. 5. Optimized patterns for the printing of the shortest channel. (a) With drop space of 45 lm, droplets firing rate of 2 m/s and substrate temperature of 28 °C. (b) With defect concentrated area removed. (c) With the substrate temperature increased to 45 °C.
We printed droplet arrays with different lateral and vertical drop space, for the optimization of patterns with the shortest channel length. The patterns shown in Table 1 reveal the general tendency of the formation of interspace. Most of the interspace have a width below 10 lm. The resulted structures of printed droplet changed sensitively with the increasing of drop space by only 1 lm, accordingly providing possibilities for this method to obtain desired patterns. The different droplets contact phenomenon for the lateral and the vertical directions owing to the different interval time of printed droplets for these two directions, which can be explained by the fixed travel route of nozzles as described in Fig. 2(b). Pattern with only one channel is needed for the fabrication of source and drain electrodes of TFTs. Fig. 5(a) shows one of the optimized drop space, based on the results in Table 1. The drop space for both directions was 45 lm, the firing rate was 2 m/s, and the substrate temperature was 28 °C. Besides the channel, interspace also existed in Fig. 5(a). In order to eliminate the unwanted interspace, we reduced the side length of the designed pattern from 500 lm to 400 lm, and the optimized pattern is shown in Fig. 5(b). The substrate temperature was increased from 28 °C to 45 °C for further eliminating of the unwanted interspace. More details about the influences of substrate temperature to the short channel structure are shown in Fig. S2. The surface energy of substrate was increased with the increasing of substrate temperature. Thus the Dmax and De of droplets also slightly increased. As shown in Fig. 5(c), the interspace was filled and the channel length was reduced to only 2.4 lm, which is the shortest channel length ever reported achieved by direct patterning of inkjet printing. Moreover, the channel becomes smoother with the increased substrate temperature. We considered this is the real limitation of short channel printing by traditional piezoelectric inkjet printer with 10 pL cartridge (Nozzle diameter 21 lm). The short channel structure can be easily repeated using the method we proposed, and the results are shown in Fig. S3. 4. Conclusion In summary, a novel approach was proposed to prepare short channel structure silver electrodes on untreated solid substrates by making use of the difference in the retraction velocities on both sides of an ink droplet during printing process. Consequently, a 2.4 lm channel length was obtained based on this novel approach by traditional piezoelectric inkjet printing. Moreover, droplets contact mechanism was first introduced in our work to explain the formation of short channel in the printing process. The method for preparing short channel structure is repeatable and extremely
simple, thus provides referential significance for new design rules of printed patterns. Acknowledgements This work was supported by National Program on Key Basic Research Project (973) (No. 2015CB655004), National Key R&D Program (Nos. 2016YFB0401504 and 2016YFF0203603), Guangdong Natural Science Foundation (No. 2016A030313459), Science and Technology Project of Guangdong Province (Nos. 2014B090915004, 2016B090907001, 2014A040401014, 2016B090906002, 2015A010101323, 2014B090916002, 2015A010101323, 2015B090915001 and 2015B090914003), Educational Commission of Guangdong Province (Nos. 2014KZDXM010 and 2015KTSCX003), the Fundamental Research Funds for the Central Universities (Nos. 2015ZP024 and 2015ZZ063), State Key Laboratory of Luminescence and Applications (SKLA-2016-11). Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2016.10.016. References [1] W.J. Hyun, E.B. Secor, G.A. Rojas, M.C. Hersam, L.F. Francis, C.D. Frisbie, Adv. Mater. 27 (2015) 7058. [2] Z. Zhang, X. Zhang, Z. Xin, M. Deng, Y. Wen, Y. Song, Adv. Mater. 25 (2013) 6714. [3] D.J. Finn, M. Lotya, J.N. Coleman, ACS Appl. Mater. Interfaces 7 (2015) 9254. [4] Y. Kim, S. Jang, J.H. Oh, Appl. Phys. Lett. 106 (2015) 14103. [5] J. Park, J. Hwang, J. Phys. D Appl. Phys. 47 (2014) 405102. [6] Y. Kwon, Y. Lee, K. Lee, Y. Choi, Y. Choa, J. Electron. Mater. 44 (2015) 2608. [7] X. Wu, Z. Chen, T. Zhou, S. Shao, M. Xie, M. Song, Z. Cui, RSC Adv. 5 (2015) 20924. [8] K. Hong, S.H. Kim, A. Mahajan, C.D. Frisbie, ACS Appl. Mater. Interfaces 6 (2014) 18704. [9] R. Aga, C. Jordan, R.S. Aga, C.M. Bartsch, E.M. Heckman, IEEE Electr. Device Lett. 35 (2014) 1124. [10] L. Kuai, J. Wang, T. Ming, C. Fang, Z. Sun, B. Geng, J. Wang, Sci. Rep.-UK 5 (2015) 9923. [11] Y. Li, L. Lan, P. Xiao, S. Sun, Z. Lin, W. Song, E. Song, P. Gao, W. Wu, J. Peng, ACS Appl. Mater. Interfaces (2016). [12] M. Mashayekhi, A. Conde, T.N. Ng, P. Mei, E. Ramon, J. Display Technol. (2015). [13] T. Wei, F. Linrun, Z. Jiaging, C. Qingyu, C. Sujie, G. Xiaojun, J. Mater. Chem. C 2 (2014) 1995. [14] H. Wang, C. Cheng, L. Zhang, H. Liu, Y. Zhao, Y. Guo, W. Hu, G. Yu, Y. Liu, Adv. Mater. 26 (2014) 4683. [15] I. Theodorakos, F. Zacharatos, R. Geremia, D. Karnakis, I. Zergioti, Appl. Surf. Sci. 336 (2015) 157. [16] J. de Ruiter, D. van den Ende, F. Mugele, Phys. Fluids 27 (2015) 12105. [17] H. Fujimoto, S. Watanabe, T. Okamoto, T. Hama, H. Takuda, Exp. Therm. Fluid Sci. 60 (2015) 66. [18] H.K. Huh, S. Jung, K.W. Seo, S.J. Lee, Microfluid Nanofluid 18 (2015) 1221. [19] S. Ganesan, S. Rajasekaran, L. Tobiska, Int. J. Heat Mass Transfer 78 (2014) 670. [20] M. Taghilou, M.H. Rahimian, Int. J. Therm. Sci. 86 (2014) 1. [21] J.B. Lee, S.H. Lee, Langmuir 27 (2011) 6565.