- Email: [email protected]
Fabrication of high-resolution conductive lines by combining inkjet printing with soft lithography
Microelectronic Engineering 110 (2013) 219–223
Contents lists available at SciVerse ScienceDirect
Microelectronic Engineering journal homepage: www.elsevier.com/locate/mee
Fabrication of high-resolution conductive lines by combining inkjet printing with soft lithography Jisu Sung, Byung Ju Kang, Je Hoon Oh ⇑ Department of Mechanical Engineering, Hanyang University, 1271 Sa3-dong, Sangrok-gu, Ansan, Gyeonggi-do 426-791, Republic of Korea
a r t i c l e
i n f o
Article history: Available online 15 April 2013 Keywords: Inkjet printing Soft lithography High-resolution printing Surface wettability contrast UV treatment
a b s t r a c t In this study, an effective method to fabricate well-defined conductive lines with high resolution has been proposed by combining inkjet printing with soft lithography. Soft lithography techniques such as nanoimprint lithography for negative SU-8 patterns and micro-contact printing for hydrophobic fluorocarbon layer were used to create surface wettability contrasts so that the spreading of droplets was confined on a hydrophilic region surrounded by hydrophobic regions. Surface wettability contrasts were evaluated by water contact angle measurements, and the maximum contact angle difference between hydrophobic and hydrophilic surfaces was 104°. With the help of such surface wettability contrasts and lift-off process for removing small amount of ink stains, well-defined inkjet-printed lines as narrow as 2.8 lm can be successfully generated. The sintered Ag lines also show good electrical resistivity of 7.6 lX cm, 4.7 times as large as bulk Ag’s resistivity. This combined approach can be used to fabricate high-quality, high-resolution electrodes in printed electronics applications. Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction Recently, a myriad of studies have been focusing on developing solution processes for the implementation of printed electronics because printed electronics has been expected to open new era of flexible and stretchable electronics. Inkjet printing is one of the promising solution processes to replace conventional photolithography in the field of printed electronics due to its simple, low cost and flexible characteristics [1–4]. However, inkjet printing still has several challenges, among which the limitation of resolution is probably the most serious one. Resolution is mostly determined by the initial droplet size and droplet spreading after being placed onto a substrate. The droplet size can be diminished by reducing the nozzle diameter, but this causes inks not to be ejected due to their rheological constraints. The droplet spreading is significantly influenced by surface wettability. Hydrophobic surface is preferred for producing high-resolution features; however, it causes various hydrodynamic instabilities due to high contact angles. Therefore, the resolution of conventional piezo-type inkjet printing still remains in the range of several tens of lm [5,6]. There have been some efforts to fabricate high resolution patterns by using electrohydrodynamic (EHD) jet printing [7,8] or combining inkjet printing with laser ablation [9]. However, each method has its own disadvantages: reliability issue for EHD jet
⇑ Corresponding author. Tel.: +82 31 400 5252; fax: +82 31 400 4705. E-mail address: [email protected] (J.H. Oh). 0167-9317/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.mee.2013.04.004
printing and heat affected damage for laser ablation. Another approach is to create surface wettability patterns that confine the spreading of ink droplets on a hydrophilic region with hydrophobic regions surrounding it. For this approach, one of critical issues is how to make desired surface wettability patterns. Soft lithography techniques [10–12] such as nano-imprint lithography (NIL) and micro-contact printing (lCP) can be considered an acceptable method to fabricate such surface wettability contrasts. In this work, we report on inkjet-printed conductive Ag lines with high resolution which are produced by combining soft lithography. In the soft lithography process, negative SU-8 patterns were fabricated on a glass substrate by placing poly(dimethylsiloxane) (PDMS) stamp onto a spin-coated SU-8 layer, followed by UV treatment to make the imprinted SU-8 surfaces hydrophilic. Hydrophobic fluorocarbon (FC) film was then transferred onto the top surface of SU-8 patterns using lCP, resulting in hydrophilic patterns surrounded by hydrophobic surfaces. After the surface wettability patterns were formed, ink droplets were placed on hydrophilic regions so that those regions can be filled with ink using capillary actions in order to minimize the interfacial energy among solid, air and liquid. Lift-off process of the FC film was performed to obtain well-defined inkjet-printed lines without any ink stains after inkjet printing process. The effects of NIL parameters such as NIL pressure and time on the morphology of imprinted SU-8 patterns were investigated using 3D non-contact surface profiler. Water contact angle (WCA) measurements for both hydrophobic and hydrophilic regions in the surface wettability patterns were conducted to evaluate the surface wettability
220
J. Sung et al. / Microelectronic Engineering 110 (2013) 219–223
contrast. Electrical resistivity of the inkjet-printed lines was also evaluated after sintering process. 2. Experimental details 2.1. Soft lithography process PDMS replicas have been widely used as stamps for the NIL process [13,14]. PDMS stamps are not treated by anti-sticking agents because of its inherent low surface energy [15]. The patterned Si master was spin-coated with a mixture of FC solution (EGC-1720 and FC-40, NovecTM 3M) for anti-sticking between Si master and PDMS mold. The mixture of PDMS base (Sylgad 184, Dow corning) and curing agent (Sylgad 184, Dow corning) (10:1 by volume ratio) was de-gassed in vacuum and then poured onto the patterned Si master as shown in Fig. 1a. The PDMS stamp cured at 60 °C for 4 h was then peeled off from the Si master. Thermoplastic and photosensitive epoxy based SU-8 solution (SU-8 2002, Micro Chem Co.) was selected as a NIL resin. A glass substrate was ultrasonicated in propylene glycol methyl ether acetate (PGMEA) solution for 5 min and then dried on a hot plate at 100 °C for 10 min. SU-8 solution was spin-coated on the glass substrate at 3000 rpm for 30 s to make thin and uniform SU-8 layer with 2 lm thickness and baked on a hot plate at 90 °C for 5 min to fully remove any residual solvents. As shown in Fig. 1b, the SU-8 layer was imprinted using the PDMS stamp under different pressures ranging from 0 to 15 kPa for 1, 2 and 3 min. The temperature of the SU-8 layer was kept at 100 °C to ensure viscous flow and thermal stability during initial NIL stage. The SU-8 pattern was then cooled down to room temperature, and the PDMS stamp was peeled off. The imprinted SU-8 pattern was UV-treated for different times to convert its surface wettability from hydrophobic to hydrophilic (Fig. 1c). For UV treatment, UVO cleaner (AH-1700, AHTech LTS, Korea) which has power of 28 mW/cm2 and two UV lamps with wavelength of 184.0 and 253.7 nm was used in the ambient condition. This UV treatment also leads to increase in the glass transient
temperature of SU-8, above 200 °C, by producing photoacid that acts as a catalyst for cross-linking reaction during post exposure bake, which prevents distortion of structure in the sintering process of inkjet-printed features. Then, the lCP process was conducted using a flat PDMS substrate spin-coated with a mixture of FC solution (EGC-1700 and HFE-7100, NovecTM 3M) at 500 rpm for 20 s (Fig. 1d). The spin-coated FC film on the flat PDMS substrate was carefully contacted with the top surface of imprinted SU-8 patterns for 10 s. As a result, the top surface of SU-8 pattern became hydrophobic while the surfaces inside SU-8 pattern remained hydrophilic. 2.2. Printing process After the soft lithography process, Ag nanoparticle ink (Harima Chemical Co.) was inkjet-printed on the imprinted SU-8 patterns using a piezoelectric single-nozzle printhead with a 30 lm nozzle diameter (MicroFab Co.). The ink contained 63.7 wt.% Ag nanoparticles in tetradecane with average particle diameter of 12 nm. Single droplet was sequentially inkjet-printed at three different positions on the imprinted SU-8 patterns to completely fill the inside of imprinted patterns (Fig. 1e). All inkjet printing processes were performed at room temperature, and all the inkjet-printed patterns were dried for 5 min under ambient condition. For the lift-off process to remove any remaining ink stains, the inkjetprinted patterns on the glass substrate were immersed into the fluorous solvent (HFE-7100, NovecTM 3M), which is a main solvent of the transferred FC film, for 30 s after drying process (Fig. 1f). 2.3. Inkjet-printed pattern evaluation Morphology of the imprinted SU-8 patterns with different NIL pressures and times was observed using a 3D non-contact surface profiler. WCAs were measured on both the UV-treated and FC film-coated regions of the imprinted SU-8 patterns to estimate the surface wettability contrast. Optical microscopic images of
Pressure
PDMS solution
Si master
PDMS stamp
UV/O3
SU-8 structure
Glass substrate
SU-8 layer
UV/O3
PDMS solution
PDMS solution
Si master with negative profile
Glass substrate
Glass substrate
(a) PDMS casting
(b) NIL
(c) UV treatment
PDMS stamp
Inkjet nozzle Droplet placements
FC layer
Filled Ag ink
Transferred FC layer
PDMS stamp Glass substrate
Glass substrate
Glass substrate
(d) μCP
(e) Inkjet printing
(f) Lift-off
Fig. 1. Schematics of fabrication process of a high-resolution Ag line by combining soft lithography with inkjet printing.
221
J. Sung et al. / Microelectronic Engineering 110 (2013) 219–223
the inkjet-printed patterns were obtained to evaluate the inkjetprinted pattern quality and optimize the soft lithography and lift-off processes. Electrical resistivity of the inkjet-printed line patterns was measured using a semiconductor parameter analyzer and probe station after sintering the patterns at 220 °C for 1 h. The microstructure of sintered Ag lines was also examined by scanning electron microscopy (SEM) measurement.
Water contact angle (degree)
(a)
3. Results and discussion 3.1. NIL patterns
60 50 40 30 20 10 0
Fig. 2 shows the surface morphology of the imprinted SU-8 patterns with different NIL pressures and times. Without an externally applied NIL pressure (P < 0.1 kPa from self-weight of PDMS stamp), well-defined negative SU-8 patterns are not able to be obtained because of non-conformal contact, leading to a pile-up profile near both ends of the imprinted SU-8 pattern. When the NIL pressure is increased up to 10 kPa (0.1 kPa 6 P 6 10 kPa), the profiles of imprinted SU-8 patterns agree well with the Si mater’s profile. However, with further increase in the NIL pressure (P > 10 kPa), the imprinted SU-8 patterns show poor quality and quite different surface profiles from the Si master’s one due to the deformation of PDMS stamp. There is little influence of the imprinting time on the surface profile of the imprinted SU-8 patterns since it just takes a few second to complete the NIL process with the SU-8 polymer at 100 °C. 3.2. Surface wettability contrast High surface wettability contrast is required to confine inkjetprinted droplets only on hydrophilic regions surrounded by hydrophobic ones and hence obtain good quality inkjet-printed lines with high resolution. Systematic experiments to increase surface wettability contrast were performed using UV treatment and lCP. Fig. 3a demonstrates the variation in WCAs on the UV-treated SU8 surface as a function of treatment time. A significant decrease in WCA is observed as the treatment time is increased up to 5 min, but there is little change in WCA with further increase in the treatment time. This indicates that longer UV treatment time makes the SU-8 surface more hydrophilic, but the surface wettability remains unchanged after a certain treatment time. After UV treatment, O–H functional groups from phenol group covalently bonded with SU-8 surface and hydrophilic functional groups such C@O and O– C@O are generated on the SU-8 surface. These functional groups
5
10
15
20
Time (min)
(b)
112 °
8°
Patterned area
FC-coated surface
50 µm Fig. 3. (a) Variation of the water contact angle on SU-8 surface with the UV treatment time and (b) optical microscopic image of a SU-8 pattern with surface wettability contrast.
contribute to hydrophilicity of the UV-treated SU-8 surface [16]. Moreover, the SU-8 layer became delaminated from the glass substrate after longer UV treatment, which would be caused by a decrease in adhesion force between cross-linked SU-8 layer and glass substrate. Therefore, all imprinted SU-8 patterns were UV-treated for 5 min in order to increase the surface wettability without any delamination problem. As shown in Fig. 3b, the hydrophobic top surface transferred by lCP has the WCA of 112° whereas the WCA on the UV-treated hydrophilic region is as low as 8°, which means the surface wettability contrast is 104° in terms of the WCA difference. 3.3. Inkjet printing and lift-off process The Ag nanoparticle ink was inkjet-printed on the wettabilitycontrolled SU-8 patterns at three different locations. Fig. 4 shows
NIL pressure (kPa) 0.1 kPa ≤ P ≤ 10 kPa
1.5 1 0.5 0
0
50
100
1.5 1 0.5 0 0
50
100
1.5 1 0.5 0
X distance (μm)
3D profile (NIL for 2 min)
X distance (μm)
P > 10 kPa Pattern height (µm)
Pattern height (µm)
Pattern height (µm)
2D profile
P < 0.1 kPa
Fig. 2. Imprinted SU-8 patterns for various pressures and times.
0
50
X distance (μm)
100
J. Sung et al. / Microelectronic Engineering 110 (2013) 219–223
FC layer thickness (number of μCP) (b) 228 nm (3)
3D profile after μCP
(a) 50 nm (1)
After inkjet printing
the effect of the surface wettability contrast on the inkjet-printed Ag lines. Well-defined Ag lines of 2.8 and 7.8 lm are obtained with the surface wettability contrast (Fig. 4a). The inkjet-printed droplets are repelled from the hydrophobic region, and the imprinted negative pattern regions with hydrophilic surfaces are filled with Ag ink due to the surface wettability contrast. However, small ink stains are observed in the middle of the line in spite of the surface wettability contrast. Since the imprinted line width is very narrow compared with the ejected droplet diameter of 30 lm, the ink placed on the hydrophobic region did not move entirely to the hydrophilic region even when the surface wettability contrast is high, which causes undesired ink stains. On the other hand, when there is no surface wettability contrast, the large amount of inkjet-printed droplets is spread over the imprinted patterns at three printing locations (Fig. 4b). The lift-off process was carried out after inkjet printing in order to remove FC layer, thereby eliminating the undesired ink stains. Fig. 5 represents the effect of the FC film thickness transferred by lCP on the lift-off quality. The FC film thickness is increased by the number of repeated lCP. The ink stain is not removed well for the thin FC film with 50 nm thickness due to retention problem that usually occurs for thin sacrificial layer, FC film in this experiment, in the lift-off process. For thin FC layers, the inkjet-printed line patterns and ink stains on the FC film are connected to each other, creating an ink barrier which prevents FC solvent from penetrating under the ink stains during the lift-off process. As a consequence, the ink stains remain after the lift-off process. In contrast, ink stains placed onto thick FC layer are not connected to the line patterns, hence resulting in the FC layer and ink stains to be readily removed during the lift-off process. As a result, for the 228 nmthick FC film, the well-defined line pattern is fabricated with no ink stain after the lift-off process, indicating that thick FC film is necessary for good lift-off quality. Therefore, it is found that the surface wettability contrast helps to confine ink droplets within a pre-defined narrow region and hence to achieve high-resolution inkjet-printed line with an assistance of lift-off process for removing ink stains.
20 μm
20 μm
After lift-off
222
20 μm
20 μm
Fig. 5. Lift-off results (a) when FC layer thickness is 50 nm and (b) when FC layer thickness is 228 nm.
Fig. 6. Inkjet-printed Ag lines after sintering at 220 °C for 1 h. Insets show the SEM images of grain structures after sintering.
4. Conclusion 3.4. Electrical characterization The inkjet-printed lines were sintered at 220 °C for 1 h to obtain electrical functionality. As shown in Fig. 6, relatively low resistivities less than 10 lX cm are observed for both printed lines with 7.8 lm and 2.8 lm widths, which means that only dried Ag lines are not damaged during the lift-off process. Under the same printing condition, the resistivity of 7.8 lm-wide line is a little higher than that of 2.8 lm-wide line. It would be attributed to different grain growth of each Ag line as depicted in the inset of Fig. 6. The inkjet-printed line with 2.8 lm width has the resistivity of 7.6 lX cm, which is 4.7 times larger than the resistivity of bulk Ag (1.59 lX cm). This indicates that printed Ag lines are conductive enough to be used as electrodes even though their widths are as narrow as a few lm.
7.8
(b) Ag line on SU-8 pattern without wettability contrast
50 μm
50 μm
50 μm
50 μm
2.8
Pattern width (μm)
(a) Ag line on SU-8 pattern with wettability contrast
We have demonstrated a combined approach of soft lithography and inkjet printing to produce inkjet-printed high-resolution Ag lines with variable line widths. This approach is to create surface wettability contrasts that confine the spreading of inkjet-printed droplets on hydrophilic regions surrounded by hydrophobic ones. In order to obtain such surface wettability contrasts, negative SU-8 patterns were first formed on a glass substrate by NIL with PDMS stamps. The SU-8 patterns were UV-treated for 5 min and then a hydrophobic FC layer was transferred onto the top surface of SU-8 patterns by lCP, resulting in the desired surface wettability contrast with WCA difference of 104°: the top surface is hydrophobic and the other surfaces are hydrophilic. Lift-off process was performed to eliminate small amount of undesired ink stains after inkjet printing, and well-defined Ag lines as narrow as 2.8 lm were finally obtained.
Fig. 4. Inkjet-printed Ag lines (a) with and (b) without the surface wettability contrast.
J. Sung et al. / Microelectronic Engineering 110 (2013) 219–223
The resistivity of sintered Ag lines was as low as 7.6 lX cm. This combination of soft lithography and inkjet printing can provide an effective way to fabricate inkjet-printed patterns with high quality and high resolution in the field of printed electronics. Acknowledgements This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Ministry of Education, Science and Technology (NRF-2010-0013807). It was also supported by the SW-embedded Device Development Program (No. 10043828) by the Ministry of Knowledge Economy. References [1] H. Sirringhaus, T. Kawase, R.H. Friend, T. Shimoda, M. Inbasekaran, W. Wu, E.P. Woo, Science 290 (2000) 2123.
223
[2] M. Singh, H.M. Haverinen, P. Dhagat, G.E. Jabbour, Adv. Mater. 22 (2010) 673. [3] B.-J. de Gans, P.C. Duineveld, U.S. Schubert, Adv. Mater. 16 (2004) 203. [4] J. Park, D.J. Lee, S.J. Kim, J.H. Oh, J. Micromech. Microeng. 19 (2009) 095021. [5] B.J. Kang, J.H. Oh, Thin Solid Films 518 (2010) 2890. [6] T.H.J. van Osch, J. Perelaer, Adv. Mater. 20 (2008) 243. [7] D. Byun, V.D. Nguyen, P. Dutta, H.C. Park, Jpn. J. Appl. Phys. 49 (2010) 060 216. [8] S.Y. Back, C.H. Song, S. Yu, H.J. Lee, B.S. Kim, N.Y. Yang, S.H. Jeong, H. Ahn, J. Nanosci. Nanotechnol. 12 (2012) 446. [9] S.H. Lee, D.J. Lee, C.K. Lee, Y.H. Lee, S.T. Wereley, J.H. Oh, Phys. Status Solidi A 209 (2012) 2142. [10] S.Y. Chou, P.R. Krauss, P.J. Renstrom, Appl. Phys. Lett. 67 (1995) 3114. [11] M. Mrksich, G.M. Whitesides, TIBTECH 13 (1995) 228. [12] E. Kim, Y. Xia, G.M. Whitesides, Nature 376 (1995) 581. [13] D.Y. Khang, H. Kang, T.I. Kim, H.H. Lee, Nano Lett. 4 (2004) 633. [14] Y. Zhang, C.T. Lin, S. Yang, Small 6 (2010) 768. [15] M. Bender, U. Plachetka, J. Ran, A. Fuchs, B. Vratzov, H. Kurz, J. Vac. Sci. Technol. B 22 (2004) 3229. [16] C.J. Chang, C.S. Yang, L.H. Lan, P.C. Wang, F.G. Tseng, J. Micromech. Microeng. 20 (2010) 115031.
Contents lists available at SciVerse ScienceDirect
Microelectronic Engineering journal homepage: www.elsevier.com/locate/mee
Fabrication of high-resolution conductive lines by combining inkjet printing with soft lithography Jisu Sung, Byung Ju Kang, Je Hoon Oh ⇑ Department of Mechanical Engineering, Hanyang University, 1271 Sa3-dong, Sangrok-gu, Ansan, Gyeonggi-do 426-791, Republic of Korea
a r t i c l e
i n f o
Article history: Available online 15 April 2013 Keywords: Inkjet printing Soft lithography High-resolution printing Surface wettability contrast UV treatment
a b s t r a c t In this study, an effective method to fabricate well-defined conductive lines with high resolution has been proposed by combining inkjet printing with soft lithography. Soft lithography techniques such as nanoimprint lithography for negative SU-8 patterns and micro-contact printing for hydrophobic fluorocarbon layer were used to create surface wettability contrasts so that the spreading of droplets was confined on a hydrophilic region surrounded by hydrophobic regions. Surface wettability contrasts were evaluated by water contact angle measurements, and the maximum contact angle difference between hydrophobic and hydrophilic surfaces was 104°. With the help of such surface wettability contrasts and lift-off process for removing small amount of ink stains, well-defined inkjet-printed lines as narrow as 2.8 lm can be successfully generated. The sintered Ag lines also show good electrical resistivity of 7.6 lX cm, 4.7 times as large as bulk Ag’s resistivity. This combined approach can be used to fabricate high-quality, high-resolution electrodes in printed electronics applications. Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction Recently, a myriad of studies have been focusing on developing solution processes for the implementation of printed electronics because printed electronics has been expected to open new era of flexible and stretchable electronics. Inkjet printing is one of the promising solution processes to replace conventional photolithography in the field of printed electronics due to its simple, low cost and flexible characteristics [1–4]. However, inkjet printing still has several challenges, among which the limitation of resolution is probably the most serious one. Resolution is mostly determined by the initial droplet size and droplet spreading after being placed onto a substrate. The droplet size can be diminished by reducing the nozzle diameter, but this causes inks not to be ejected due to their rheological constraints. The droplet spreading is significantly influenced by surface wettability. Hydrophobic surface is preferred for producing high-resolution features; however, it causes various hydrodynamic instabilities due to high contact angles. Therefore, the resolution of conventional piezo-type inkjet printing still remains in the range of several tens of lm [5,6]. There have been some efforts to fabricate high resolution patterns by using electrohydrodynamic (EHD) jet printing [7,8] or combining inkjet printing with laser ablation [9]. However, each method has its own disadvantages: reliability issue for EHD jet
⇑ Corresponding author. Tel.: +82 31 400 5252; fax: +82 31 400 4705. E-mail address: [email protected] (J.H. Oh). 0167-9317/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.mee.2013.04.004
printing and heat affected damage for laser ablation. Another approach is to create surface wettability patterns that confine the spreading of ink droplets on a hydrophilic region with hydrophobic regions surrounding it. For this approach, one of critical issues is how to make desired surface wettability patterns. Soft lithography techniques [10–12] such as nano-imprint lithography (NIL) and micro-contact printing (lCP) can be considered an acceptable method to fabricate such surface wettability contrasts. In this work, we report on inkjet-printed conductive Ag lines with high resolution which are produced by combining soft lithography. In the soft lithography process, negative SU-8 patterns were fabricated on a glass substrate by placing poly(dimethylsiloxane) (PDMS) stamp onto a spin-coated SU-8 layer, followed by UV treatment to make the imprinted SU-8 surfaces hydrophilic. Hydrophobic fluorocarbon (FC) film was then transferred onto the top surface of SU-8 patterns using lCP, resulting in hydrophilic patterns surrounded by hydrophobic surfaces. After the surface wettability patterns were formed, ink droplets were placed on hydrophilic regions so that those regions can be filled with ink using capillary actions in order to minimize the interfacial energy among solid, air and liquid. Lift-off process of the FC film was performed to obtain well-defined inkjet-printed lines without any ink stains after inkjet printing process. The effects of NIL parameters such as NIL pressure and time on the morphology of imprinted SU-8 patterns were investigated using 3D non-contact surface profiler. Water contact angle (WCA) measurements for both hydrophobic and hydrophilic regions in the surface wettability patterns were conducted to evaluate the surface wettability
220
J. Sung et al. / Microelectronic Engineering 110 (2013) 219–223
contrast. Electrical resistivity of the inkjet-printed lines was also evaluated after sintering process. 2. Experimental details 2.1. Soft lithography process PDMS replicas have been widely used as stamps for the NIL process [13,14]. PDMS stamps are not treated by anti-sticking agents because of its inherent low surface energy [15]. The patterned Si master was spin-coated with a mixture of FC solution (EGC-1720 and FC-40, NovecTM 3M) for anti-sticking between Si master and PDMS mold. The mixture of PDMS base (Sylgad 184, Dow corning) and curing agent (Sylgad 184, Dow corning) (10:1 by volume ratio) was de-gassed in vacuum and then poured onto the patterned Si master as shown in Fig. 1a. The PDMS stamp cured at 60 °C for 4 h was then peeled off from the Si master. Thermoplastic and photosensitive epoxy based SU-8 solution (SU-8 2002, Micro Chem Co.) was selected as a NIL resin. A glass substrate was ultrasonicated in propylene glycol methyl ether acetate (PGMEA) solution for 5 min and then dried on a hot plate at 100 °C for 10 min. SU-8 solution was spin-coated on the glass substrate at 3000 rpm for 30 s to make thin and uniform SU-8 layer with 2 lm thickness and baked on a hot plate at 90 °C for 5 min to fully remove any residual solvents. As shown in Fig. 1b, the SU-8 layer was imprinted using the PDMS stamp under different pressures ranging from 0 to 15 kPa for 1, 2 and 3 min. The temperature of the SU-8 layer was kept at 100 °C to ensure viscous flow and thermal stability during initial NIL stage. The SU-8 pattern was then cooled down to room temperature, and the PDMS stamp was peeled off. The imprinted SU-8 pattern was UV-treated for different times to convert its surface wettability from hydrophobic to hydrophilic (Fig. 1c). For UV treatment, UVO cleaner (AH-1700, AHTech LTS, Korea) which has power of 28 mW/cm2 and two UV lamps with wavelength of 184.0 and 253.7 nm was used in the ambient condition. This UV treatment also leads to increase in the glass transient
temperature of SU-8, above 200 °C, by producing photoacid that acts as a catalyst for cross-linking reaction during post exposure bake, which prevents distortion of structure in the sintering process of inkjet-printed features. Then, the lCP process was conducted using a flat PDMS substrate spin-coated with a mixture of FC solution (EGC-1700 and HFE-7100, NovecTM 3M) at 500 rpm for 20 s (Fig. 1d). The spin-coated FC film on the flat PDMS substrate was carefully contacted with the top surface of imprinted SU-8 patterns for 10 s. As a result, the top surface of SU-8 pattern became hydrophobic while the surfaces inside SU-8 pattern remained hydrophilic. 2.2. Printing process After the soft lithography process, Ag nanoparticle ink (Harima Chemical Co.) was inkjet-printed on the imprinted SU-8 patterns using a piezoelectric single-nozzle printhead with a 30 lm nozzle diameter (MicroFab Co.). The ink contained 63.7 wt.% Ag nanoparticles in tetradecane with average particle diameter of 12 nm. Single droplet was sequentially inkjet-printed at three different positions on the imprinted SU-8 patterns to completely fill the inside of imprinted patterns (Fig. 1e). All inkjet printing processes were performed at room temperature, and all the inkjet-printed patterns were dried for 5 min under ambient condition. For the lift-off process to remove any remaining ink stains, the inkjetprinted patterns on the glass substrate were immersed into the fluorous solvent (HFE-7100, NovecTM 3M), which is a main solvent of the transferred FC film, for 30 s after drying process (Fig. 1f). 2.3. Inkjet-printed pattern evaluation Morphology of the imprinted SU-8 patterns with different NIL pressures and times was observed using a 3D non-contact surface profiler. WCAs were measured on both the UV-treated and FC film-coated regions of the imprinted SU-8 patterns to estimate the surface wettability contrast. Optical microscopic images of
Pressure
PDMS solution
Si master
PDMS stamp
UV/O3
SU-8 structure
Glass substrate
SU-8 layer
UV/O3
PDMS solution
PDMS solution
Si master with negative profile
Glass substrate
Glass substrate
(a) PDMS casting
(b) NIL
(c) UV treatment
PDMS stamp
Inkjet nozzle Droplet placements
FC layer
Filled Ag ink
Transferred FC layer
PDMS stamp Glass substrate
Glass substrate
Glass substrate
(d) μCP
(e) Inkjet printing
(f) Lift-off
Fig. 1. Schematics of fabrication process of a high-resolution Ag line by combining soft lithography with inkjet printing.
221
J. Sung et al. / Microelectronic Engineering 110 (2013) 219–223
the inkjet-printed patterns were obtained to evaluate the inkjetprinted pattern quality and optimize the soft lithography and lift-off processes. Electrical resistivity of the inkjet-printed line patterns was measured using a semiconductor parameter analyzer and probe station after sintering the patterns at 220 °C for 1 h. The microstructure of sintered Ag lines was also examined by scanning electron microscopy (SEM) measurement.
Water contact angle (degree)
(a)
3. Results and discussion 3.1. NIL patterns
60 50 40 30 20 10 0
Fig. 2 shows the surface morphology of the imprinted SU-8 patterns with different NIL pressures and times. Without an externally applied NIL pressure (P < 0.1 kPa from self-weight of PDMS stamp), well-defined negative SU-8 patterns are not able to be obtained because of non-conformal contact, leading to a pile-up profile near both ends of the imprinted SU-8 pattern. When the NIL pressure is increased up to 10 kPa (0.1 kPa 6 P 6 10 kPa), the profiles of imprinted SU-8 patterns agree well with the Si mater’s profile. However, with further increase in the NIL pressure (P > 10 kPa), the imprinted SU-8 patterns show poor quality and quite different surface profiles from the Si master’s one due to the deformation of PDMS stamp. There is little influence of the imprinting time on the surface profile of the imprinted SU-8 patterns since it just takes a few second to complete the NIL process with the SU-8 polymer at 100 °C. 3.2. Surface wettability contrast High surface wettability contrast is required to confine inkjetprinted droplets only on hydrophilic regions surrounded by hydrophobic ones and hence obtain good quality inkjet-printed lines with high resolution. Systematic experiments to increase surface wettability contrast were performed using UV treatment and lCP. Fig. 3a demonstrates the variation in WCAs on the UV-treated SU8 surface as a function of treatment time. A significant decrease in WCA is observed as the treatment time is increased up to 5 min, but there is little change in WCA with further increase in the treatment time. This indicates that longer UV treatment time makes the SU-8 surface more hydrophilic, but the surface wettability remains unchanged after a certain treatment time. After UV treatment, O–H functional groups from phenol group covalently bonded with SU-8 surface and hydrophilic functional groups such C@O and O– C@O are generated on the SU-8 surface. These functional groups
5
10
15
20
Time (min)
(b)
112 °
8°
Patterned area
FC-coated surface
50 µm Fig. 3. (a) Variation of the water contact angle on SU-8 surface with the UV treatment time and (b) optical microscopic image of a SU-8 pattern with surface wettability contrast.
contribute to hydrophilicity of the UV-treated SU-8 surface [16]. Moreover, the SU-8 layer became delaminated from the glass substrate after longer UV treatment, which would be caused by a decrease in adhesion force between cross-linked SU-8 layer and glass substrate. Therefore, all imprinted SU-8 patterns were UV-treated for 5 min in order to increase the surface wettability without any delamination problem. As shown in Fig. 3b, the hydrophobic top surface transferred by lCP has the WCA of 112° whereas the WCA on the UV-treated hydrophilic region is as low as 8°, which means the surface wettability contrast is 104° in terms of the WCA difference. 3.3. Inkjet printing and lift-off process The Ag nanoparticle ink was inkjet-printed on the wettabilitycontrolled SU-8 patterns at three different locations. Fig. 4 shows
NIL pressure (kPa) 0.1 kPa ≤ P ≤ 10 kPa
1.5 1 0.5 0
0
50
100
1.5 1 0.5 0 0
50
100
1.5 1 0.5 0
X distance (μm)
3D profile (NIL for 2 min)
X distance (μm)
P > 10 kPa Pattern height (µm)
Pattern height (µm)
Pattern height (µm)
2D profile
P < 0.1 kPa
Fig. 2. Imprinted SU-8 patterns for various pressures and times.
0
50
X distance (μm)
100
J. Sung et al. / Microelectronic Engineering 110 (2013) 219–223
FC layer thickness (number of μCP) (b) 228 nm (3)
3D profile after μCP
(a) 50 nm (1)
After inkjet printing
the effect of the surface wettability contrast on the inkjet-printed Ag lines. Well-defined Ag lines of 2.8 and 7.8 lm are obtained with the surface wettability contrast (Fig. 4a). The inkjet-printed droplets are repelled from the hydrophobic region, and the imprinted negative pattern regions with hydrophilic surfaces are filled with Ag ink due to the surface wettability contrast. However, small ink stains are observed in the middle of the line in spite of the surface wettability contrast. Since the imprinted line width is very narrow compared with the ejected droplet diameter of 30 lm, the ink placed on the hydrophobic region did not move entirely to the hydrophilic region even when the surface wettability contrast is high, which causes undesired ink stains. On the other hand, when there is no surface wettability contrast, the large amount of inkjet-printed droplets is spread over the imprinted patterns at three printing locations (Fig. 4b). The lift-off process was carried out after inkjet printing in order to remove FC layer, thereby eliminating the undesired ink stains. Fig. 5 represents the effect of the FC film thickness transferred by lCP on the lift-off quality. The FC film thickness is increased by the number of repeated lCP. The ink stain is not removed well for the thin FC film with 50 nm thickness due to retention problem that usually occurs for thin sacrificial layer, FC film in this experiment, in the lift-off process. For thin FC layers, the inkjet-printed line patterns and ink stains on the FC film are connected to each other, creating an ink barrier which prevents FC solvent from penetrating under the ink stains during the lift-off process. As a consequence, the ink stains remain after the lift-off process. In contrast, ink stains placed onto thick FC layer are not connected to the line patterns, hence resulting in the FC layer and ink stains to be readily removed during the lift-off process. As a result, for the 228 nmthick FC film, the well-defined line pattern is fabricated with no ink stain after the lift-off process, indicating that thick FC film is necessary for good lift-off quality. Therefore, it is found that the surface wettability contrast helps to confine ink droplets within a pre-defined narrow region and hence to achieve high-resolution inkjet-printed line with an assistance of lift-off process for removing ink stains.
20 μm
20 μm
After lift-off
222
20 μm
20 μm
Fig. 5. Lift-off results (a) when FC layer thickness is 50 nm and (b) when FC layer thickness is 228 nm.
Fig. 6. Inkjet-printed Ag lines after sintering at 220 °C for 1 h. Insets show the SEM images of grain structures after sintering.
4. Conclusion 3.4. Electrical characterization The inkjet-printed lines were sintered at 220 °C for 1 h to obtain electrical functionality. As shown in Fig. 6, relatively low resistivities less than 10 lX cm are observed for both printed lines with 7.8 lm and 2.8 lm widths, which means that only dried Ag lines are not damaged during the lift-off process. Under the same printing condition, the resistivity of 7.8 lm-wide line is a little higher than that of 2.8 lm-wide line. It would be attributed to different grain growth of each Ag line as depicted in the inset of Fig. 6. The inkjet-printed line with 2.8 lm width has the resistivity of 7.6 lX cm, which is 4.7 times larger than the resistivity of bulk Ag (1.59 lX cm). This indicates that printed Ag lines are conductive enough to be used as electrodes even though their widths are as narrow as a few lm.
7.8
(b) Ag line on SU-8 pattern without wettability contrast
50 μm
50 μm
50 μm
50 μm
2.8
Pattern width (μm)
(a) Ag line on SU-8 pattern with wettability contrast
We have demonstrated a combined approach of soft lithography and inkjet printing to produce inkjet-printed high-resolution Ag lines with variable line widths. This approach is to create surface wettability contrasts that confine the spreading of inkjet-printed droplets on hydrophilic regions surrounded by hydrophobic ones. In order to obtain such surface wettability contrasts, negative SU-8 patterns were first formed on a glass substrate by NIL with PDMS stamps. The SU-8 patterns were UV-treated for 5 min and then a hydrophobic FC layer was transferred onto the top surface of SU-8 patterns by lCP, resulting in the desired surface wettability contrast with WCA difference of 104°: the top surface is hydrophobic and the other surfaces are hydrophilic. Lift-off process was performed to eliminate small amount of undesired ink stains after inkjet printing, and well-defined Ag lines as narrow as 2.8 lm were finally obtained.
Fig. 4. Inkjet-printed Ag lines (a) with and (b) without the surface wettability contrast.
J. Sung et al. / Microelectronic Engineering 110 (2013) 219–223
The resistivity of sintered Ag lines was as low as 7.6 lX cm. This combination of soft lithography and inkjet printing can provide an effective way to fabricate inkjet-printed patterns with high quality and high resolution in the field of printed electronics. Acknowledgements This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Ministry of Education, Science and Technology (NRF-2010-0013807). It was also supported by the SW-embedded Device Development Program (No. 10043828) by the Ministry of Knowledge Economy. References [1] H. Sirringhaus, T. Kawase, R.H. Friend, T. Shimoda, M. Inbasekaran, W. Wu, E.P. Woo, Science 290 (2000) 2123.
223
[2] M. Singh, H.M. Haverinen, P. Dhagat, G.E. Jabbour, Adv. Mater. 22 (2010) 673. [3] B.-J. de Gans, P.C. Duineveld, U.S. Schubert, Adv. Mater. 16 (2004) 203. [4] J. Park, D.J. Lee, S.J. Kim, J.H. Oh, J. Micromech. Microeng. 19 (2009) 095021. [5] B.J. Kang, J.H. Oh, Thin Solid Films 518 (2010) 2890. [6] T.H.J. van Osch, J. Perelaer, Adv. Mater. 20 (2008) 243. [7] D. Byun, V.D. Nguyen, P. Dutta, H.C. Park, Jpn. J. Appl. Phys. 49 (2010) 060 216. [8] S.Y. Back, C.H. Song, S. Yu, H.J. Lee, B.S. Kim, N.Y. Yang, S.H. Jeong, H. Ahn, J. Nanosci. Nanotechnol. 12 (2012) 446. [9] S.H. Lee, D.J. Lee, C.K. Lee, Y.H. Lee, S.T. Wereley, J.H. Oh, Phys. Status Solidi A 209 (2012) 2142. [10] S.Y. Chou, P.R. Krauss, P.J. Renstrom, Appl. Phys. Lett. 67 (1995) 3114. [11] M. Mrksich, G.M. Whitesides, TIBTECH 13 (1995) 228. [12] E. Kim, Y. Xia, G.M. Whitesides, Nature 376 (1995) 581. [13] D.Y. Khang, H. Kang, T.I. Kim, H.H. Lee, Nano Lett. 4 (2004) 633. [14] Y. Zhang, C.T. Lin, S. Yang, Small 6 (2010) 768. [15] M. Bender, U. Plachetka, J. Ran, A. Fuchs, B. Vratzov, H. Kurz, J. Vac. Sci. Technol. B 22 (2004) 3229. [16] C.J. Chang, C.S. Yang, L.H. Lan, P.C. Wang, F.G. Tseng, J. Micromech. Microeng. 20 (2010) 115031.