Open Access
Add to BibSonomy
Abstract
Cu-based micropatterns were fabricated using reductive sintering inside Cu2O nanosphere films induced by green femtosecond laser pulses. The linear and nonlinear absorption coefficients of Cu2O nanosphere films composed of Cu2O nanospheres, polyvinylpyrrolidone, and 2-propanol were 0.062 × 104 /cm and 10−50 cm/GW, respectively. The minimum line width was the same as the laser spot diameter (∼0.7 µm), indicating negligible thermal diffusion for metallization. Cu-based micropatterns with different heights were formed by varying the position of the focal spot along the z-axis by focusing the laser pulses on the films and then scanning. This technique is applied to three-dimensional microfabrication.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Femtosecond laser direct writing is a promising tool for three-dimensional (3D) micro-additive manufacturing. In the initial stage of the development, photosensitive polymers were cross-linked by femtosecond laser-induced two-photon absorption [1,2] to form 3D polymer structures with ∼100 nm resolution. Further, noble metals such as Au and Ag were also precipitated by photoreduction caused by two-photon absorption [3,4]. However, 3D metal microfabrication using multiphoton absorption has been limited to the fabrication of materials composed of noble metals because these metal ions have to be photoreduced and precipitated.
On the Other hand, two-dimensional (2D) Cu and Ni micropatterns have been directly written via laser-induced thermochemical reduction [5–7]. For example, Cu micropatterns were formed by the thermochemical reduction induced by continuous-wave and nanosecond lasers in air. CuO nanoparticle solutions, including CuO nanoparticles, polyvinylpyrrolidone (PVP), and ethylene glycol, were spin-coated on glass or polyimide film substrates. Then, the laser beams were focused and scanned onto the CuO nanoparticle films in air. Finally, non-irradiated CuO nanoparticles were removed in ethanol. Cu micropatterns were formed by the reducing and sintering of CuO nanoparticles in air. Cu micropatterns were also formed using Cu2O nanoparticles solutions [6]. Ni micropatterns were also fabricated by the laser-induced reductive sintering of NiO nanoparticles [7]. A NiO nanoparticle solution was prepared as the raw material by mixing NiO nanoparticles and toluene. Then, a thermochemical reduction was effected by laser-induced linear optical absorption.
We fabricated 3D Cu-based microstructures by nonlinear optical absorption induced by near-infrared femtosecond laser pulses [8,9]. Cu2O nanosphere films, which consist of Cu2O nanospheres of ∼100 nm diameter, PVP, and 2-propanol, were spin-coated on glass and Cu-coated glass substrates. Cu-based microstructures were formed inside the Cu2O nanosphere films by the thermochemical reduction of Cu2O nanospheres to Cu induced by femtosecond laser pulses. In addition, the nonlinear optical absorption properties of Cu2O nanosphere films were investigated at the wavelength of the femtosecond laser pulses (780 nm). However, internal writing of Cu-based microfabrication in the Cu2O nanosphere films has been only achieved on the Cu-coated glass substrates because of the accumulation of heat on bare glass substrates.
In this study, we tried to fabricate Cu micropatterns inside Cu2O nanosphere films via nonlinear optical absorbed thermochemical reduction. Green femtosecond laser pulses, which are expected to provide high transparency at the laser wavelength and intense two-photon absorption, were used for thermochemical reduction. First, the nonlinear optical absorption properties of Cu2O nanosphere films were evaluated at the green femtosecond wavelength of 515 nm. Then, the direct-writing properties were evaluated. Finally, we demonstrated the fabrication of Cu-based micropatterns.
2. Experimental methods
2.1 Direct writing by green femtosecond laser pulse-induced reductive sintering of Cu2O nanospheres
Figure 1 shows a schematic illustration of the direct-writing process of Cu-based micropatterns by using a green femtosecond laser. First, Cu2O nanospheres were synthesized using a polyol method. Then, the Cu2O nanosphere solution consisting of Cu2O nanospheres, 2-propanol, and polyvinylpyrrolidone (PVP) was spin-coated on glass substrates. Subsequently, direct writing using green femtosecond laser pulses was performed in air. Finally, non-sintered Cu2O nanospheres were removed by rinsing the samples in ethanol. The details of each step are described in the following sections.
Fig. 1. Schematic illustration of direct-writing process by the green femtosecond laser pulse-induced reductive sintering of Cu2O nanospheres.
The Cu2O nanosphere solution was prepared by the same process as previously reported [8,9]. First, the Cu2O nanospheres were synthesized by the polyol method. Cu(NO3)2·2.5H2O(72 mM) in ethylene glycol (12 mL) and PVP (Mw∼55000, 288 mM) in ethylene glycol (12 mL) were prepared. Then, the two solutions were simultaneously mixed with ethylene glycol (20 mL) at a rate of ∼90 mL/h in an oil bath at 140°C for 25 min. Then, the suspension solution was centrifuged at 14500 rpm for 30 min to separate Cu2O nanospheres from the solution. Finally, the Cu2O nanospheres were mixed with 2-propanol (3.6 mL) and PVP (Mw∼55000, 0.09 g). The Cu2O nanosphere films which, comprised of Cu2O nanospheres, PVP, and 2-propanol were prepared by spin-coating the solution at 500 rpm for 10 s and dried for solidification by baking at 80°C for 20 min. The films have been confirmed as the raw Cu2O nanospheres without significant changes after baking under the condition [9].
A green femtosecond fiber laser system (Ekspla, FF1000-SHG, wavelength, 515 nm, pulse duration, 100 fs, and repetition rate, 40 MHz) was used for direct writing in air. The laser pulses were focused on the Cu2O nanosphere films by using an objective lens with a numerical aperture of 0.90. The sample substrates coated with the Cu2O nanosphere films were scanned using an xyz-piezo stage.
After direct writing of the micropatterns inside the films, non-sintered Cu2O nanospheres, PVP, and 2-propanol were removed by rinsing the substrates in ethanol. Only micropatterns sintered on the substrates were finally obtained.
2.2 Evaluation of linear and nonlinear absorption
The line width of the fabricated micropatterns was observed by using field emission-scanning electron microscopy (FE-SEM) (Hitachi, SU8230). The linear optical absorption properties were evaluated using an ultraviolet to near-infrared spectrometer (Shimadzu, UV-2600). The nonlinear optical absorption properties were also evaluated using the Z-scan technique [10]. The laser power transmitted through the sample substrate coated with the films of Cu2O nanosphere films was measured using a photodetector by scanning the sample in the z-axis direction using a stage.
Cross-sectional profiles of the micropatterns were measured by atomic force microscopy (AFM, SII, E-sweep). The crystal structures of the micropatterns were examined by X-ray diffraction (XRD) analysis (Rigaku, Miniflex) using Cu-Kα radiation.
3. Results and discussion
3.1 Linear and nonlinear absorption properties of Cu2O nanosphere films
Cu2O nanosphere films consisting of Cu2O nanospheres, PVP, and 2-propanol were prepared. Monodispersed Cu2O nanospheres of ∼100 nm diameter were successfully obtained as previously reported [8,9]. Cu2O nanosphere films were yellow transparent and ∼4 µm thick. The linear optical absorption spectrum of the films is shown in Fig. 2. The linear absorption coefficient at the green femtosecond laser wavelength, 515 nm, was 0.062 × 104 /cm, indicating relatively high transmittance at the wavelength. Moreover, the intense absorption was observed at the half of the laser wavelength, ∼257 nm. These results indicate that multiphoton absorption, specifically two-photon absorption, is expected to be induced at the green femtosecond laser wavelength.
The Z-scan method, which is a well-known evaluation method for nonlinear optical absorption, was used to investigate the two-photon absorption coefficients [10,11]. When the sample substrate was scanned along to z-axis, the transmittance was evaluated by detecting the transmitted laser power. Figure 3 shows the z-scan curves experimentally obtained at the peak intensity of 122 GW/cm2. The Cu2O nanosphere films coated on glass substrate were scanned in z-axis. The scan speed was determined to be 5 mm/s which were appropriate to not cause the permanent changes of the absorption induced by Cu2O nanosphere reduction and sintering. First, the substrate was scanned from the photodetector to the objective lens sides. Subsequently, it was scanned from the objective lens side to the photodetector sides. The calculated curves at the two-photon absorption coefficients of 20 and 25 cm/GW are also shown in Fig. 3(a). It is reported that the nonlinear optical absorption of Cu2O nanoparticles and thin films indicates saturated absorption at low intensity [11–14]. However, the peak intensity, i.e., 122 GW/cm2 in the present work is high, assuming that the saturated absorption is negligible. Therefore, the normalized transmittance T(z) is expressed in Eq. (1) [15].
Fig. 3. (a)Z-scan curves of the experimental (red and blue dots) and theoretical (black lines) data at the peak intensity of 122 GW/cm2. The red and blue dots were obtained by scanning from the detector to lens sides and from lens to detector sides, respectively. (b)Dependence of the peak intensity on the transmittance.
3.2 Direct writing properties
FE-SEM images of the minimum line width on the bare glass and Cu-thin-film-coated glass substrates are shown in Figs. 4(a) and 4(b), respectively. The femtosecond laser pulses were focused on the both substrates by considering the optical path shift inside the Cu2O nanosphere films with high refractive index, assumed as 3.3 + 0.05i [16], first, the laser pulses were focused on the surface of the Cu2O nanosphere films, following that the substrates were shifted by 2 µm from surface to inside the films. Here, the Rayleigh length was estimated as 2.1 µm inside Cu2O nanosphere films, which was smaller than the film thickness. A wider and relatively flat line pattern was formed on the Cu-thin-film-coated glass substrate, though the line pattern on the bare glass substrate was not smooth. The heights of the line patterns on bare and Cu-coated glass substrates were ∼0.3 µm and ∼0.1 µm, respectively, by AFM measurement. These results suggest that the Cu thin film strongly absorbed the green femtosecond laser pulse energy via linear optical absorption, and the absorbed energy heated the Cu2O nanospheres and diffused into the Cu2O nanosphere films, resulting in an increase in the line width. In contrast, a nonplanar line pattern was formed on the bare glass substrates. This result indicates that the line pattern was formed via multiphoton absorption.
Fig. 4. FE-SEM images of line patterns on (a) bare glass and (b) Cu-thin-film-coated glass substrates when the intensity and scan speed were 61 GW/cm2 and 10 µm/s, respectively.
Figure 5 show the relationships between the line width and scan speed at various intensities on (a) bare glass and (b) Cu-thin-film-coated glass substrates, respectively. The line width increased with decreasing scan speed and increasing intensities. The minimum line width on the bare glass and Cu-thin-film-coated glass substrates were ∼0.7 µm and ∼1.1 µm, respectively. The line width on the bare glass substrates was the same as the laser spot diameter of 0.7 µm. A comparison of the line widths of the bare glass and Cu-thin-film-coated glass substrates, showed that wider line patterns were formed on Cu-thin-film-coated glass substrates at lower scan speed and high intensities.
Fig. 5. Relationship between line width and scan speed at (a) bare and (b) Cu-coated glass substrates.
Considering to the large nonlinear optical absorption coefficient of 10–50 cm/GW in Cu2O nanosphere films shown in Fig. 3, the nonlinear optical absorption was perceived as dominant in the initial absorption by the films. However, the linear optical absorption by Cu is intense as explained by the wider line patterns on Cu-thin-film-coated glass substrates in Fig. 5(b). Therefore, we evaluated whether the micropatterns were formed inside the Cu2O nanosphere films using green femtosecond laser pulses.
In comparison of the line patterning using near-infrared femtosecond laser pulses reported previously, [9], the minimum line width written by using green femtosecond laser pulses was same as the focal spot diameter, ∼0.7 µm, although that written by using near-infrared femtosecond laser pulses was twice as the focal spot diameter, ∼2 µm. The results indicate that the efficiency of photothermal conversion used in green femtosecond laser pulses was lower than that used in near infrared femtosecond laser pulses. This is an advantage to write finer patterns on glass substrates.
3.3 Microfabrication inside Cu2O nanosphere films on glass substrates
To evaluate 3D fabrication properties, attempts were made to fabricate micropatterns inside the Cu2O nanosphere films on glass substrates. Figure 6(a) shows a schematic illustration for writing the micropatterns inside the films. Thicker Cu2O nanosphere films were prepared by using the photoresist frames. The film thickness was ∼5 µm which was maximum one with high uniformity. When the surface of the Cu2O nanosphere film was defined at z = 0, the single-layered micropatterns were written at various positions in z-axis from +3.5 to −5.0 µm. The peak intensity and scan speed were 204 GW/cm2 and 10 µm/s, respectively. The raster scan pitch of the micropatterns in xy plane was decided as 0.5 µm to form fully-connected micropatterns. The overlap rate was ∼300%. An optical microscope image of the as-irradiated samples are shown in Fig. 6(b). Figure 6(c) shows the optical microscope image of the samples after removing the Cu2O nanospheres. A comparison of the images in Fig. 6(b) and 6(c) shows that two micropatterns were formed at z = −4.5 and −5.0 µm after the removal shown in Fig. 6(c) though the micropatterns were not observed before the removal. These results suggest that the micropatterns were formed inside the films. The micropatterns written partially outside of the Cu2O nanosphere film, i.e., at z = −4.0, +2.5, 3.0, and 3.5 µm, were unstable. It could be inferred that the Cu2O nanospheres may have been cooled from the surface of the films by natural convection cooling when the micropatterns were formed outside the films. By considering the instability, the dependency of the micropatterns on the focal position was evaluated under the condition that the focal spot was inside the films.
Fig. 6. (a)Schematic illustration of evaluation method for microfabrication inside the Cu2O nanosphere film. Optical microscope images of (b) as-irradiated samples and (c) samples after the removal of Cu2O nanospheres.
To evaluate the dependency of the heights of the micropatterns on the focal position in the film thickness, the frame patterns were also formed as the height standards in the thicker Cu2O nanosphere film as used in Fig. 6. First, the micropatterns were formed at various focal position in the film thickness direction. Single layered square micropatterns of sides 24 µm were formed by raster scanning of the focal spot without piling up in the z-axis direction. The raster scan pitch in xy plane was 0.5 µm. Then, the frames as the standards were formed under the condition that the laser pulses were focused on the surface of the substrates. The peak intensity and scan speed of the micropatterns were 204 GW/cm2 and 10 µm/s, respectively, which were same as those in Fig. 6. The frames were fabricated at the peak intensity of 245 GW/cm2 and the scan speed of 10 µm/s. Figures 7(a)–7(b) show the top views of AFM images of the micropatterns fabricated at z = −2.5 and −3.0 µm, respectively, when the zero in z-axis was defined at the surface of the Cu2O nanosphere film. The difference of the focal positions where the micropatterns were formed inside the film was caused by the slight difference of the thickness of the films because high refractive index of the Cu2O nanosphere films induced the optical path shift. The cross-sectional profiles of the micropatterns are shown in Fig. 7(c). By comparing to the height of the two micropatterns, the height of the micropattern fabricated at z = –2.5 µm, the average of 0.34 µm, was found to be higher than that of the micropattern fabricated at z=−3.0 µm, 0.32 µm. The arithmetic average roughness of the micropatterns fabricated at z = –2.5 µm and z = –3.0 µm, were both 0.01 µm. The heights of the frame was lower than those of the micropatterns, which were caused by the higher density of the sintered micropatterns. These results were consistent with the expectation that the micropatterns were fabricated inside the films by depending on the focal spot. The small dependency of the heights on the focal position suggest that the direct writing inside the Cu2O nanosphere films was not only the initial multiphoton absorption but also following thermal reduction. In the future, higher and complex micropatterns are expected to be formed by preparing thicker Cu2O nanosphere film.
Fig. 7. (a–b) Top views of AFM images of the micropatterns fabricated at z = −2.5 and −3.0 µm, respectively. (c) Cross-sectional profiles of the two micropatterns.
The crystal structures of the micropatterns were evaluated. The larger square micropatterns of sides 1.8 mm was fabricated at peak intensity of 163 GW/cm2 and a scan speed of 100 µm/s. The XRD spectrum of the micropattern (Fig. 8) shows intense peaks corresponding to Cu. These observations suggest that Cu-rich micropatterns including slight Cu2O were fabricated by the reductive sintering of Cu2O nanospheres. The Cu grain size was estimated as ∼65 nm using Scherrer equation. This result suggest that the crystal growth was not induced in photothermochemical reduction of Cu2O nanospheres to Cu. By considering the XRD peak corresponding to Cu2O, Cu2O in the micropatterns was remained raw materials, not reoxidization of reduced Cu.
4. Conclusions
Cu-based micropatterns were directly written inside Cu2O nanosphere films by green femtosecond laser induced reductive sintering.
- (1) The absorption properties of Cu2O nanosphere films were evaluated. The linear and two-photon absorption coefficients of the films at the wavelength of the green femtosecond laser pulses were 0.062 × 104 /cm and 10–50 cm/GW, respectively. These results were consistent with the values corresponding to the wavelength of green nanosecond laser pulses.
- (2) The minimum line widths on the bare glass and Cu-thin-film-coated glass substrates were ∼0.7 and ∼1.1 µm, respectively. For a laser spot diameter of 0.7 µm, the effect of thermal diffusion on the line width is negligible on the bare glass substrates.
- (3) The micropatterns with heights lower than the original thickness of the Cu2O nanosphere films were formed. In addition, the heights of the micropatterns depended on the position of the focal spot in the optical direction, indicating that the micropatterns were fabricated via multiphoton absorption.
- (4) The XRD spectrum exhibited that the Cu-rich micropatterns were fabricated. Thus, Cu-based micropatterns were formed inside Cu2O nanosphere films.
Funding
Japan Society for the Promotion of Science KAKENHI (20H02043, JP16H06064).
Disclosures
The authors declare no conflicts of interest.
References
1. S. Maruo, O. Nakamura, and S. Kawata, “Three-dimensional microfabrication with two-photon-absorbed photopolymerization,” Opt. Lett. 22(2), 132–134 (1997). [CrossRef]
2. S. Kawata, H-B Sun, T. Tanaka, and K. Takada, “Finer features for functional microdevices,” Nature 412(6848), 697–698 (2001). [CrossRef]
3. S. Maruo and T. Saeki, “Femtosecond laser direct writing of metallic microstructures by photoreduction of silver nitrate in a polymer matrix,” Opt. Express 16(2), 1174–1179 (2008). [CrossRef]
4. L. Vurth, P. Baldeck, O. Stephan, and G. Vitrant, “Two-photon induced fabrication of gold microstructures in polystyrene sulfonate thin films using a ruthenium (II) dye as photoinitiator,” Appl. Phys. Lett. 92(17), 171103 (2008). [CrossRef]
5. B. Kang, S. Han, J. Kim, S. Ko, and M. Yang, “One-step fabrication of copper electrode by laser-induced direct local reduction and agglomeration of copper oxide nanoparticle,” J. Phys. Chem. C 115(48), 23664–23670 (2011). [CrossRef]
6. H. Lee and M. Yang, “Effect of solvent and PVP on electrode conductivity in laser-induced reduction process,” Appl. Phys. 119(1), 317–323 (2015). [CrossRef]
7. D. Lee, D. Paeng, H. K. Park, and C. P. Grigoropoulos, “Vacuum-free, maskless patterning of Ni electrodes by laser reductive sintering of NiO nanoparticle ink and its application to transparent conductors,” ACS Nano 8(10), 9807–9814 (2014). [CrossRef]
8. M. Mizoshiri and Y. Kondo, “Direct writing of Cu-based fine micropatterns using femtosecond laser pulse-induced sintering of Cu2O nanospheres,” Jpn. J. Appl. Phys. 58(SD), SDDF05 (2019). [CrossRef]
9. M. Mizoshiri and Y. Kondo, “Direct writing of two- and three-dimensional Cu-based microstructures by femtosecond laser reductive sintering of the Cu2O nanospheres,” Opt. Mater. Express 9(7), 2828–2837 (2019). [CrossRef]
10. M. Sheik-Bahae, A. A. Said, T-H Wei, D. J. Hagan, and E. W. Van Stryland, “Sensitive measurement of optical nonlinearities using a single beam,” IEEE J. Quantum Electron. 26(4), 760–769 (1990). [CrossRef]
11. G. Yang, A. Chen, M. Fu, H. Long, and P. Lu, “Excimer laser deposited CuO and Cu2O films with third-order optical nonlinearities by femtosecond z-scan measurement,” Appl. Phys. A 104(1), 171–175 (2011). [CrossRef]
12. H. Sekhar and D. N. Rao, “Preparation, characterization and nonlinear absorption studies of cuprous oxide nanoclusters, micro-cubes and micro-particles,” J. Nanopart. Res. 14(7), 976 (2012). [CrossRef]
13. B. Karthikeyan, R. Udayabhaskar, T. P. Rose, T. Pandiyarajan, and R. Philip, “Sol-gel prepared Cu2O microspheres: linear and nonlinear optical properties,” RSC Adv. 4(74), 39541–39546 (2014). [CrossRef]
14. J. Serna, E. Rueda, and H. García, “Nonlinear optical properties of bulk cuprous oxide using single beam Z-scan at 790 nm,” Appl. Phys. Lett. 105(19), 191902 (2014). [CrossRef]
15. M. Sheik-Bahae, D. C. Hutchings, D. J. Hagan, and E. W. Van Stryland, “Dispersion of bound electron nonlinear refraction in solids,” IEEE J. Quantum Electron. 27(6), 1296–1309 (1991). [CrossRef]
16. C. Malerba, F. Biccari, C. L. A. Ricardo, M. D’Incau, P. Scardi, and A. Mittiga, “Absorption coefficient of bulk and thin film Cu2O,” Sol. Energy Mater. Sol. Cells 95(10), 2848–2854 (2011). [CrossRef]
