Continuous Patterning of Copper NanowireBased Transparent Conducting Electrodes for Use in Flexible Electronic Applications Zhaoyang Zhong,†,‡,⊥ Hyungjin Lee,§,⊥ Dongwoo Kang,‡ Sin Kwon,‡ Young-Man Choi,∥ Inhyuk Kim,† Kwang-Young Kim,‡ Youngu Lee,*,§ Kyoohee Woo,*,‡,⊥ and Jooho Moon*,† †
Department of Materials Science and Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul, 03722, Republic of Korea Advanced Manufacturing Systems Research Division, Korea Institute of Machinery and Materials (KIMM), 156 Gajeongbuk-Ro, Yuseong-Gu, Daejeon, 34103, Republic of Korea § Department of Energy Systems Engineering, Daegu Gyeongbuk Institute of Science and Technology (DGIST), 333 Techno Jungang-daero, Hyeonpung-Myeon, Dalseong-Gun, Daegu, 42988, Republic of Korea ∥ Department of Mechanical Engineering, Ajou University, 241 Hyowon-ro, Paldal-gu, Suwon-si, Gyeonggi-do 16490, Republic of Korea ‡
S Supporting Information *
ABSTRACT: Simple, low-cost and scalable patterning methods for Cu nanowire (NW)-based flexible transparent conducting electrodes (FTCEs) are essential for the widespread use of Cu NW FTCEs in numerous flexible optoelectronic devices, wearable devices, and electronic skins. In this paper, continuous patterning for Cu NW FTCEs via a combination of selective intense pulsed light (IPL) and roll-to-roll (R2R) wiping process was explored. The development of continuous R2R patterning could be achieved because there was significant difference in adhesion properties between NWs and substrates depending on whether Cu NW coated area was irradiated by IPL or not. Using a custom-built, R2R-based wiping apparatus, it was confirmed that nonirradiated NWs could be clearly removed out without any damage on irradiated NWs strongly adhered to the substrate, resulting in continuous production of low-cost Cu NW FTCE patterns. In addition, the variations in microscale pattern size by varying IPL process parameters/the mask aperture sizes were investigated, and possible factors affecting on developed pattern size were meticulously examined. Finally, the successful implementation of the patterned Cu NW FTCEs into a phosphorescent organic light-emitting diode (PhOLED) and a flexible transparent conductive heater (TCH) were demonstrated, verifying the applicability of the patterned FTCEs. It is believed that our study is the key step toward realizing the practical use of NW FTCEs in various flexible electronic devices. KEYWORDS: copper nanowires, intense pulsed light irradiation, roll-to-roll patterning, phosphorescent organic light-emitting diode, flexible transparent conductive heater u nanowire (NW)-based flexible transparent conducting electrodes (FTCEs) have been the focus of recent research for a wide range of next-generation electronic devices such as flexible optoelectronic devices, wearable devices, and electronic skins.1−5 The reason for this attention is that FTCEs have numerous merits such as low material cost, excellent opto-electrical properties, large-scale solution processability, and outstanding mechanical flexibility. Therefore, many researchers have developed facile synthesis approaches to produce Cu NWs in large quantities,6−8 methodologies to prevent Cu oxidation,9−11 and various postwelding processes to reduce the contact resistance between Cu NWs under ambient atmosphere.12−15 However, despite these significant advance-
C
© 2016 American Chemical Society
ments, the widespread application of Cu NW-based FTCEs is still hindered by critical issues. For example, there are no effective patterning processes for Cu NW-based FTCEs. The conventional photolithography used for patterning is timeconsuming, complicated, and expensive.16,17 Orifice-based direct printing technologies such as inkjet printing cannot be easily adopted because of severe nozzle clogging by NWs with a high hydrodynamic radius.18−20 In addition, a laser direct Received: June 1, 2016 Accepted: July 19, 2016 Published: July 19, 2016 7847
DOI: 10.1021/acsnano.6b03626 ACS Nano 2016, 10, 7847−7854
Article
www.acsnano.org
Article
ACS Nano
Figure 1. (a) Schematic illustration of the patterning method for Cu NW FTCEs using a mask for selective IPL irradiation. SEM images show the (b) surface and (c) cross section of Cu NW-coated areas with and without IPL irradiation. The left side shows the nonirradiated area, and the right side presents the irradiated area. (d) Surface image of the boundary between the nonirradiated area and the irradiated area after removal of NWs (wiped away by wet fabric). The NWs in the nonirradiated area are clearly eliminated.
roll (R2R)-based wiping apparatus. In this process, the size variations of developed patterns as the function of IPL process parameters and the mask aperture sizes (WA) were investigated, and the possible factors related to the broadening or narrowing of pattern size were thoroughly examined with a simulation of light diffraction. Furthermore, both a flexible phosphorescent organic light-emitting diode (PhOLED) and a flexible transparent conductive heater (TCH) were successfully fabricated, demonstrating the potential use of the R2R-patterned Cu NWs. Our continuous patterning approach can accelerate the widespread use of Cu NWs for various low-cost, flexible electronic devices.
writing process has been employed for direct patterning; however, large-area patterning is time-consuming because of the small irradiation spot size of the laser (usually at the mm2 scale).14,15 Therefore, it is necessary to develop a simple and fast patterning method for Cu NW FTCEs to facilitate their practical use in various electronic applications. Song et al. recently proposed a rapid and scalable patterning method for Ag NW-based FTCEs using selective intense pulsed light (IPL) irradiation.21 In their method, IPL-irradiated Ag NWs could not be easily removed because of their strong adhesion developed by fusion between an underlying polymer substrate and the NWs, whereas nonirradiated NWs were easily delaminated from the substrate, enabling the Ag NW FTCEs to be patterned. Similar to their study, it was also observed in our previous study that a strong adhesion between Cu NWs and a polymer substrate can be induced after IPL irradiation.22 After IPL irradiation, Cu NWs are instantly embedded into the substrate at the nanoscale level (we defined this process as “selfnanoembedding”) and were consequently strongly adhered to the substrate; thus, the irradiated NWs were not readily detachable. This observation indicates that the patterning of Cu NW FTCEs can be achieved by taking advantage of significant differences in adhesion properties between NWs and substrates depending on IPL irradiation. In this study, we explored continuous production of patterned Cu NW FTCEs using a combination of selective intense pulsed light (IPL) irradiation and custom-built, roll-to-
RESULTS AND DISCUSSION The process flow for patterning Cu NW-based FTCEs is illustrated in Figure 1a. A Cu NW film with a transmittance (T) of ∼80% (Figure S1, Supporting Information) and a haze of ∼3% was prepared by a Meyer bar coater using Cu NW ink with a concentration of 0.5 wt %. The prepared NW film was shielded by a disposable, thick paper mask carved with a starshaped pattern for selective IPL irradiation. Then, the film covered with the mask was irradiated by IPL at 710 V for 520 μs. After irradiation, the nonirradiated NWs were selectively wiped away by fabric soaked with isopropanol (IPA), which was used as the main solvent for the Cu NW ink. A star-shaped Cu NW conductive pattern was successfully developed through this process, as shown in the photograph of Figure 1. More 7848
DOI: 10.1021/acsnano.6b03626 ACS Nano 2016, 10, 7847−7854
Article
ACS Nano
Figure 2. (a) Schematic diagram of the continuous R2R system for production of patterned Cu NW FTCEs. The dashed box indicates the wiping roll wrapped with wet fabric. The photographs show the selective IPL-irradiated Cu NWs before, during, and after the wiping process. (b) Photographs and RS distribution of 12-star-patterned Cu NW films before and after the R2R-based wiping process. The corresponding stars are marked by numbers (1−12). (c) Line width of the patterned Cu NWs as a function of IPL irradiation voltage and photomask aperture size (WA). NW patterns with line widths similar to the WA could be developed, as marked by arrows.
from the surface of substrate. After wiping by the wet fabric, a SEM image (Figure 1d) showed a clearly defined boundary between nonirradiated and irradiated regions. No NWs were observed in the nonirradiated area, which suggests that the nonirradiated NWs were completely removed. This result indicates that the patterning of Cu NW FTCEs can be achieved. A prototype apparatus for R2R wiping to develop patterns was built to demonstrate that this process can be used for continuous patterning of Cu NW FTCEs, as shown in the schematic diagram of Figure 2a. In this apparatus, a wiping roll wrapped with wet fabric was designed to rotate in the direction opposite to the web movement. While the selectively irradiated Cu NW film was moved forward at a speed of ∼2 m/min, the film surface was wiped by the reversely rotating wet fabric roll. Nonirradiated Cu NWs were clearly removed by this process, and patterned Cu NW films could be continuously fabricated, as shown in the photographs of Figure 2a and Video S1 (Supporting Information). To investigate the change in RS values of the IPL-irradiated conducting area before/after wiping
importantly, it was observed that the sheet resistance (RS) of the area with the IPL-irradiated NWs remained nearly constant (19.6 Ω/sq), even though the irradiated NWs were subjected to the wiping process for the removal of NWs on nonirradiated areas. The microstructures of the surface of selectively irradiated Cu NW films were investigated by field emission scanning electron microscopy (FESEM), as shown in Figure 1b−d. Figure 1b shows that the left side of the Cu NW film was not irradiated by IPL, whereas the right side of the film was irradiated by IPL. A brightness contrast was clearly observed at the boundary (indicated by a light blue dotted line) between the nonirradiated area and the irradiated area. The brightness of the NWs on the nonirradiated area was higher than that of NWs on the irradiated area, which indicates that the NWs on the irradiated area were self-nanoembedded into the plastic substrate. This result is more clearly visible in the crosssectional images of Figure 1c. NWs on the nonirradiated area can be distinguishable from the surface of an underlying substrate. In contrast, on the irradiated area, NWs are embedded into substrate so that they are indistinguishable 7849
DOI: 10.1021/acsnano.6b03626 ACS Nano 2016, 10, 7847−7854
Article
ACS Nano
Figure 3. (a) Geometry for the simulation and (b) simulated light intensity distribution at the observation line based on the air gap distance (L = 0, 5, 10 μm) for aperture sizes (WA) of 15 and 50 μm at a given light intensity (exposure dose) of 1.5. (c) The patterned line width according to exposure dose estimated for aperture sizes (WA) of 15 and 50 μm. It was assumed that the threshold (minimum) intensity for inducing NW embedding on the substrate is 0.7 and L = 10 μm. (d) Schematic diagram for heat conduction, light reflection, and light blocking related to broadening or narrowing of the line pattern developed by selective IPL irradiation on the Cu NW film through the aperture of a photomask.
increased dramatically as NW film was evolved into separated spheres, losing a percolative random-network structure by excessive energy of light, as shown in Figure S3b (Supporting Information). After selective IPL irradiation and the R2R wiping process, the developed line patterns were investigated by FESEM (Figure S4, Supporting Information); the resultant average line widths obtained by measuring five times for every pattern are presented in Figure 2c. After selective IPL irradiation at 710 V for 520 μs, the widths of the line patterns developed with photomask aperture sizes (WA) of 15, 20, 50, and 80 μm were observed to be 18.7, 24.4, 56.9, and 91.3 μm, respectively, which indicates that the developed pattern widths were broader than those of the corresponding WA. It was observed that as the WA was reduced, a higher voltage was necessary to develop the desired patterns. In addition, regardless of the WA, the line width became narrower as the voltage decreased and a line width similar to the corresponding WA could then be achieved, as marked by arrows in Figure 2c. A line width that is narrower than the WA could also be developed by further decreasing the voltage. This broadening or narrowing of the line width is related to light diffraction. Generally, a hard mask cannot be conformably in contact with the substrate, and an air gap of a few microns between the mask and substrate is thus inevitable (Figure 3a). Owing to this air gap, a diffracted light distribution is induced on the nanowire film. To confirm this light diffraction, the diffracted light distribution for an increasing air gap distance (L) of 0−10 μm for WA values of 15 and 50 μm was
by this apparatus, patterned Cu NW FTCEs with 12 stars were fabricated, as shown in Figure 2b. For all star-shaped conducting patterns, uniform RS values were observed, with only slight changes in the average RS and its deviation before (17.8 Ω/sq ±1.8 Ω/sq) and after (17.4 Ω/sq ±1.8 Ω/sq) wiping, indicating that there was little damage on IPL irradiated NWs after R2R wiping process. Moreover, a bending test as a function of the bending radius was performed for the developed NW pattern (Figure S2, Supporting Information), and then it was observed that pattern had excellent mechanical flexibility with no significant change in its RS value at radii ranging from 1.5 mm to 24 mm. It is believed that integration of this R2R wiping process into inline systems incorporating R2R-based slot-die coating and consecutive, fast selective IPL irradiation can realize large-scale, continuous production of Cu NW-based highly flexible and low-cost patterned FTCEs. To investigate the patterning resolution, a line pattern photomask with microscale line widths of 15, 20, 50, and 80 μm was prepared, and a coated Cu NW film was placed in contact with the mask as closely as possible prior to IPL irradiation. For IPL irradiation, the power supply voltage was controlled in the range of 660−710 V with an irradiation time of 520 μs. It should be noted that no significant change was observed in the RS values (16.7−19.6 Ω/sq) of the Cu NW FTCE in this voltage, as shown in Figure S3a (Supporting Information). Below ∼650 V, much higher RS values were observed, which might be due to incomplete welding between NWs caused by low intensity of the light. Over ∼720 V, however, the RS 7850
DOI: 10.1021/acsnano.6b03626 ACS Nano 2016, 10, 7847−7854
Article
ACS Nano
Figure 4. Device structure and performance of a green PhOLED constructed on a patterned Cu NW FTCE. (a) Schematic of a flexible green PhOLED structure using a windmill-shaped Cu NW FTCE as the anode. (b) Normalized emission spectrum for a green PhOLED. The inset shows a photograph of a flexible green PhOLED device. Device characteristics such as (c) current density, (d) luminance, (e) current efficiency, and (f) power efficiency of a green PhOLED.
In addition to the light diffraction, thermal conduction, light reflection, and light blocking may also be responsible for the variations in patterned line width, as shown in Figure 3d. Thermal energy that instantly heats up NWs on an irradiated area can be transferred to NWs located in an adjacent, nonirradiated area. This means that the irradiated NWs lose heat energy to adjacent nonirradiated NWs and, consequently, cannot be embedded into the substrate, which results in the development of narrower line widths. In contrast, nonirradiated NWs that are heated near an irradiated area can be embedded into a substrate, and accordingly, the developed pattern size can become larger. Additionally, because an IPL lamp emits light in all directions, similar to a light bulb, some of the light can be reflected from the stage plate or blocked at the edge of the mask pattern. This light reflection and blocking can affect the light intensity and light diffraction near the pattern edge, which can cause narrowing or broadening of the line width.
numerically simulated using the commercial software optiFDTD (finite-difference time-domain) under simplified illumination conditions at a given light intensity (exposure dose) of 1.5, as shown in Figure 3b. With the assumption that the threshold (minimum) light intensity that can generate NW embedding into the substrate is 0.7, based on the experimental results of Figure 2c, Figure 3c displays the estimated patterned line width as a function of exposure dose. In the case of WA = 15 μm, the developed line width varied from 13.5 to 17.5 μm as the exposure dose increased from 1.5 to 10. When WA was 50 μm, the patterned line width changed from 48.2 to 52.6 μm with the increase of exposure dose from 1.5 to 10. In other words, the blurred intensity distribution results in narrower lines at weaker doses and broader lines at stronger doses. These findings are consistent with our experimental observation in which the patterned line width increased with increasing excitation voltage in the IPL irradiation. 7851
DOI: 10.1021/acsnano.6b03626 ACS Nano 2016, 10, 7847−7854
Article
ACS Nano Table 1. Characteristics of a PhOLED Device Constructed on a Patterned Cu NW FTCE Vturn‑on [V]
Vdriving (at maximum luminance) [V]
max current efficiency [cd/A]
max power efficiency [lm/W]
Lmax [cd/m2]
ELmax [nm]
CIE (500 cd/m2) (x,y)
4.00
14.00
28.11
16.62
1903
516
(0.318, 0.617)
from the smooth Cu NW FTCE and the cathode (Al) in the optimized device structure. It should be noted that these are the highest efficiencies ever achieved in Cu NW FTCE-based OLEDs without light outcoupling structures. The detailed device performance characteristics of the fabricated green PhOLED are summarized in Table 1. A flexible, transparent conductive heater (TCH) was also fabricated as another emerging application to demonstrate the feasibility of our patterned Cu NW FTCE. Figure 5a shows a schematic illustration of the TCH fabrication process based on patterned Cu NW FTCEs. Detailed information on the Cu NW-based heater can be found in the Experimental Methods. A constant DC bias voltage was applied at both ends of the Cu NW-based pattern to induce electrically driven resistive Joule heating. Figure 5b shows the time-dependent average temper-
The patterned Cu NW FTCE was successfully utilized in the fabrication of a green PhOLED. A green PhOLED with a simple bottom-emissive device structure was fabricated on a windmill-shaped Cu NW FTCE with an RS of 20 Ω/sq and a T of 80% that was patterned by selective IPL irradiation, as shown in Figure 4a. All of the layers were successively deposited on top of the patterned Cu NW FTCE at a base pressure of