Open Access
Add to BibSonomyAbstract
Metallic glass film of Pr60Al10Ni10Cu20 is proposed to be used as a resist of phase-change lithography (PCL). PCL is a mask-less lithography technology by using laser-direct-writing to create the intended nanopatterns. Thermal distribution in the PrAlNiCu film after exposure is calculated by finite element method (FEM). Thin films are exposed by continuous-wave laser and selective etched by nitric-acid solution, and the patterns are discerned by optical and atomic force microscope. The etching rate of as-deposited PrAlNiCu is thus nearly five times of the crystalline film. These results indicate that PrAlNiCu metallic glass film is a promising resist for phase-change lithography.
© 2016 Optical Society of America
1. Introduction
Currently, in order to follow the prediction of Moore's Law [1,2], lithography is facing unprecedented challenges, including resolution improving, efficiency ameliorating [3,4]. Due to the presence of the optical diffraction limit [5–8], conventional optical lithography techniques should use a shorter wavelength laser light source or a larger numerical aperture of optical lens to improve the lithographic resolution [9]. Phase change lithography (PCL), which based on the etching rate difference between amorphous state and crystalline state of phase change material in specific etching solution [10,11], has a higher resolution as a kind of special thermal lithography. Since there is no cumulative effect of photons [11], the accuracy could be controlled at atomic scale, achieving the smaller resolution than the diffraction limit of the laser spot [12].
A large number of phase change materials were used for PCL, such as chalcogenide GeSbTe [13–15]. Metallic glasses of MgCuY were first proposed by our group [16]. However, the etching selectivity of GeSbTe films is unsatisfactory, exhibiting a low etching ratio only about 2. In comparison, a better etching selectivity with a value of 3 was achieved using MgCuY films. However, due to the high thermal conductivity, a much higher laser power is required to crystallize the thin films locally, and the size of the crystallized region is difficult to control. Pr60Al10Ni10Cu20 (PrAlNiCu) exhibits a low crystallization temperature relatively [17–19] and a significant difference between the crystalline and amorphous states, which brings a considerable etching selectivity. Moreover, the thermal conductivity of PrAlNiCu is moderate, so that the resolution of lithography could be better controlled than other metallic glasses such as MgCuY. At the same time, the crystallized region at the upper and bottom surface of thin films could be shown a low difference. These characteristics greatly develop the potential of PrAlNiCu film directly used as a photoresist, and it is easy and simple to be etched by the acid solution. Therefore, it is necessary to investigate the lithography properties of PrAlNiCu metallic glass films.
This paper explores the etching ratio of PrAlNiCu films in amorphous and crystalline states etched by aqueous solution of nitric-acid. It also gives a reasonable explanation, and compares the thermal distribution in the films of PrAlNiCu and MgCuY based on FEM.
2. Simulation
To investigate the advantage of PrAlNiCu as resist compared to the other phase change materials, FEM was carried out to simulate the thermal distribution inside the PrAlNiCu and MgCuY thin films. Two models were both built with the same dimensions of 100 μm × 100 μm × 10 μm. The 0.3-μm-thickness films were glued on the top of two models respectively. To be more intuitive, only one fourth piece of the model was loaded and simulated owing to the symmetry [Fig. 1]. The laser beam was loaded on the surface, which the coordinate is (0, 0, 0). The temperature outside the entire model was set at 20 °C, and the free convection throughout the entire surface. Although the thermal conductivity and heat capacity are function of temperature, and they show the difference between amorphous state and crystalline state. In order to simplify the simulation and only explore the trends of temperature change, the thermal conductivity and heat capacity in the simulation were set as constant. All the parameters of material are listed in the Table 1.
Fig. 1 3D model of PrAlNiCu or MgCuY thin film on the substrate for FEM simulation. The film was 300-nm-thickness and glass substrate was 10-μm-thickness.
Table 1. Material parameters [18,20–24] applied in the simulation.
3. Laser direct writing exposure and wet etching experiment
A 300-nm-thickness PrAlNiCu thin film was deposited on a glass substrate by Magnetron sputtering (JZCK-640, JUZHI Co., China), with 0.3 Pa Argon pressure and 60 W DC power. The sample is fixed on an X-Y stepping translation stage, whose displacement is controlled by stepping orders. Relative movement with the stationary continuous-wave laser beam could be generated, thereby producing the desired crystallization nanopatterns. The principle of laser direct writing exposure is as follows: computer drives the semiconductor laser (Cube-100C, Coherent Co) emitting a 661-nm laser, towards to the auto-focus system (SGSP-OBL-3, Sigma KOKI Co) and 0.4-NA objective lens. After the two components focusing, laser hit the sample precisely. The patterns of the sample surface can be observed by optical microscope (VHX-1000e, KEYENCE Co., Japan). Figure 2 shows the detailed schematic diagram of exposure system. There are two laser output ways: continuous-wave laser and pulse laser, which create patterns of lines or dots, respectively. Laser power, pulse width and movement speed of the translation stage are variable.
Fig. 2 Schematic diagram of exposure system. The wavelength of laser beam is 661 nm, beam expander collimates and expands laser beam, auto-focus system makes the thin film focused accurately, the numerical aperture of the objective lens is 0.4, sample moves with X-Y stepping translation stage based on the data we imported in the computer to form the crystalline patterns.
Amorphous PrAlNiCu thin films were annealed at 350 °C for 20 minutes in a vacuum oven (5 × 10−4 Pa). X-Ray diffraction instrument (PANalytical Company, Netherlands) was carried out to determine the film structure of both the as-deposited and annealed PrAlNiCu thin films. XRD scans are depicted in Fig. 3: before annealing, the diffraction pattern displayed a smooth curve while produced three diffraction peaks after annealing, which corresponded to the phases of AlPr at 26°, Cu6Pr at 47°and Al4Cu9 at 54°. Both the amorphous and crystalline PrAlNiCu thin films were wet etched in the 0.5-wt% aqueous solution of nitric-acid for seconds and immersed in the pure deionized water to remove residual solution. Step profiler (P-16 + , KLA-Tencor Co) was used to measure the height from substrate to surface of etched thin films. Etching time was changed to investigate the relationship between the etching quantity and etching time. The exposed and etched thin films were observed by optical microscope and atomic force microscope (AFM).
Fig. 3 XRD curves for the as-deposited and annealed PrAlNiCu thin films. Red curve stands for the annealed thin film while black curve stands for the as-deposited thin film. The annealing condition was at 350 °C for 30 min.
4. Results and discussion
Figure 4(a) shows the simulated temperature distribution at the upper surface of PrAlNiCu and MgCuY thin films. Lasers with the same power but different pulse widths were applied to increase the central (0, 0, 0.1) temperature of both two samples up to 300 °C. The laser power was set at 20 mW; the exposure width of PrAlNiCu was 150 ns while MgCuY was 80 ms. The black line represents the temperature distribution of upper surface of MgCuY, the temperature droped very slowly along the X-direction, while the temperature of upper surface of PrAlNiCu droped rapidly as the red line indicated.
Fig. 4 Temperature distribution simulation curves along X-direction of PrAlNiCu or MgCuY. (a) Temperature curves for the upper surface of the PrAlNiCu or MgCuY thin films along X-direction. The laser power was 20 mW for the both samples, pulse width of MgCuY was 80 ms and PrAlNiCu was 150 ns. The center temperature was roughly the same at 300 °C. (b) Temperature curves for the upper surface of PrAlNiCu in six different laser powers and pulse widths. The center temperature was controlled at 275 °C.
It can be intuitively realized that PrAlNiCu shows much lower thermal conductivity than MgCuY, therefore under the same exposure power, shorter exposure width is required to attain the same temperature, and the temperature droped more rapidly along X-direction. Since the crystalline temperature of PrAlNiCu is about 180 °C [18], as for MgCuY film, that is about 200 °C [25]. Based on the curves in Fig. 4(a), the crystallized region of PrAlNiCu can be controlled within a radius of 0.5 μm; however, MgCuY far exceeded 4 μm. In accordance with this principle, the phase transition region of PrAlNiCu can be controlled more precisely. This indicates that PrAlNiCu-based resist is capable of achieving higher resolution.
To investigate the rate of temperature drop from center to the edge of PrAlNiCu thin films, simulated temperature distribution with different laser parameters were drawn in Fig. 4(b). The PrAlNiCu thin films were under six different laser-irradiation conditions: 10 mW-5 μs, 15 mW-320 ns, 20 mW-100 ns, 30 mW-29 ns, 40 mW-13.6 ns, and 50 mW-8 ns. These six curves have the same temperature of about 275 °C at the center (X = 0), but the rate of temperature drop was different. The lower laser power and longer exposure time, the slower temperature droped. In other words, higher laser power and shorter exposure time can produce a higher drop rate and a smaller crystallized region. Rely on this principle, crystallized region can be more accurately controlled by modulating laser power and pulse width.
Although the thermal conductivity of PrAlNiCu is lower than MgCuY, it is sufficient to cause a 300-nm-thickness PrAlNiCu film to achieve roughly the same crystallized regions at both upper and bottom surface. In Fig. 5, the chromatic regions signify areas in which the temperatures are greater than crystalline temperature of PrAlNiCu. From A to E, the crystallized regions presented a steep angle, and exhibited a trend of approaching to right angle with the increase of the laser pulse width.
Fig. 5 Temperature contour images on X-Z cross section of PrAlNiCu. The laser power was 10 mW and the pulse widths were 200 ns, 250 ns, 300 ns, 400 ns, 500 ns From A to E. The crystalline temperature of PrAlNiCu was approximately 180 °C, regions in which temperatures are greater than crystalline temperature were set as chromatic and the rest regions were black.
The thermal diffusion rate is quickly enough in the vertical direction so that the crystallized regions at both upper and bottom surface of PrAlNiCu film reach a substantially uniform level instantly. In fact, if the thickness of film came to micron or greater level, the temperature would drop rapidly just like the X-direction of PrAlNiCu thin film in Fig. 4. However, as for the thickness of PrAlNiCu films at nanometre length scale, the thermal diffusion in the vertical direction (Z-direction) could run through the film, allowing the crystallized region to form an approximate cylindrical body, exactly meet the condition to be used as photoresist.
Step profiler was carried out to investigate the surface morphology of as-deposited, exposed and etched thin films. The stepping distance was set at approximately 160 μm. Figure 6(a) shows a relatively flat surface with 2-nm-roughness. A 60-mW continuous-wave laser irradiated the surface to crystallize the exposed region and create line patterns. Figure 6(b) shows that the exposed region produced about 20-nm convexity. This indicates the density of PrAlNiCu thin films decreases with crystallization and the volume increased thereby forming a ridge. After 5 s wet etching in 0.5-wt% aqueous solution of nitric-acid, owing to the etching selectivity ratio between the amorphous and crystalline states, the convexity increased substantially and reached about 70nm, as Fig. 6(c) shows.
Fig. 6 Surface step diagram of PrAlNiCu thin films. (a) Step diagram of as-deposited PrAlNiCu thin film before exposure. (b) Step diagram of PrAlNiCu thin film after exposure and before wet etching. (c) Step diagram of PrAlNiCu thin film after wet etching. (d) Surface topography of PrAlNiCu thin film after wet etching. The laser was continuous-wave laser and the laser power was set at 60 mW, the etching solution was 0.5-wt% aqueous solution of nitric-acid and the etching time was 5 s.
As the crystallization of MgCuY films form concave but the wet etching form convex, this property makes the selective etching ratio relatively small. In comparison, PrAlNiCu could form convex crystallized regions and the wet etching increased the height of convex, greatly increased the selective etching ratio.
AFM was used to further research the PrAlNiCu thin films etching performance. In Fig. 6(d), since the etching rate of amorphous states is greater than crystalline, the crystallized line patterns will become ridges after etching. The etched regions were smooth and the raised ridges were very obvious. This explained the good etching performance of PrAlNiCu thin films.
To fabricate sequentially crystallization line patterns, PrAlNiCu amorphous thin film was irradiated by gradient laser power. The laser powers were 60 mW (a), 70 mW (b), 80 mW (c) and 90 mW (d) respectively. Optical microscope was used to observe and measure the width of lines.
With the increase of the laser power, the width of the line concomitant increased. As the Fig. 7 shows, four different power lasers created four crystallized lines with the width of 2.46 μm, 3.31 μm, 3.91 μm and 4.42 μm correspondingly. Obviously, the width growth rate decreased gradually. However, what is certain is that the transverse dimension of lithography can be adjusted by varying the laser power, and the resolution depends on the minimum precision of laser power.
Fig. 7 Optical micrographs of laser-induced crystallized lines on the PrAlNiCu thin film. Four lines from left to right (a-d), the laser power was 60 mW, 70 mW, 80 mW and 90 mW respectively.
Since the etching rate of amorphous sample is very high, the corresponding etching time was set at 2.5 s, 5 s, 7.5 s, 10 s, 12.5 s, 15 s, 17.5 s, 20 s. While the etching rate of crystalline film is slower relatively, the etching time was set from 5 s to 50 s with 5 s step. As depicted in Fig. 8, the red line and black line stand for the etching quantity of crystalline and amorphous films over time respectively. After curve fitting, for crystalline films, the function was:
For amorphous films, the function was:The slope ratio of these two curves was about 1:5, which represent the etching ratio between crystalline and amorphous thin film. After 5 s etching, the etching quantity of crystalline film was about 10 nm, while the amorphous counterpart was about 50 nm. There was about 40 nm difference, roughly in line with the results of previous experiments shown in Figs. 6(a)-6(c).Fig. 8 Etching rate curves for amorphous and crystalline PrAlNiCu thin films. The etching solution was 0.5-wt% aqueous solution of nitric-acid, etching time of amorphous sample was set from 2.5 s to 20 s with 2.5 s step, etching time of crystalline sample was set from 5 s to 50 s with 5 s step. The etching rate of amorphous sample was about 10 nm/ s while crystalline sample was about 2 nm/ s.
The PrAlNiCu thin films exhibited better corrosion resistance after annealing against nitric-acid solution. Nitric-acid can chemically react with most metallic materials because of its strong oxidizing. Therefore, the amorphous PrAlNiCu can react quickly in nitric-acid solution. After annealing, however, the XRD pattern showed that the crystalline PrAlNiCu film generated new phases of AlPr, Cu6Pr and Al4Cu9, which are not active in the nitric-acid solution. Moreover, the corrosion resistance of copper and aluminum alloy performs outstanding [26,27]. This leads to a greater etching selective ratio between crystalline and amorphous thin films. For monolayer films, such an etching ratio is very impressive.
Moreover, in the range of 200-nm-thickness, there is a liner relationship between the etching quantity and etching time. The deviation is only a few nanometers around. The results demonstrate the ability to precisely control the etching depth of thin films by controlling the etching time in the Z-dimension.
5. Conclusion
PrAlNiCu metallic glass film has been proposed for the resist of phase change lithography. The horizontal temperature control for laser direct writing in PrAlNiCu film is easier than that in MgCuY film by simulation and comparison, because its thermal conductivity is much lower than latter. But in the vertical direction, the thermal could pass through the film to form a crystallization cylinder. 0.5-wt% aqueous solution of nitric-acid is approved as the suitable etching solution. The achieved etching selective ratio is about 5:1 in amorphous and crystalline states. It is more than twice that in GeSbTe. Moreover, the etching quantity expresses a linear relationship with the etching time, providing the possibility to accurately control the etching depth. However, the obtained experimental etching ratio is not the best one because the optimized factors such as etching temperature, solution concentration and material composition, etc will obtain a preferable result. Our results prove that PrAlNiCu metallic glass film is indeed a very promising resist for lithography due to the excellent etching selectivity and outstanding properties. We hope that PrAlNiCu will play an important role in the future lithography.
References and links
1. G. Marsh, “Moore’s law at extremes,” Mater. Today 6(5), 28–33 (2003). [CrossRef]
2. S. Hamilton, “Intel research expands Moore’s law,” IEEE Computer 36(1), 31–40 (2003). [CrossRef]
3. Z. Luo, Y. Jiang, B. D. Myers, D. Isheim, J. Wu, J. F. Zimmerman, Z. Wang, Q. Li, Y. Wang, X. Chen, V. P. Dravid, D. N. Seidman, and B. Tian, “Atomic gold-enabled three-dimensional lithography for silicon mesostructures,” Science 348(6242), 1451–1455 (2015). [CrossRef] [PubMed]
4. Y. Huang, Q. Zhao, L. Kamyab, A. Rostami, F. Capolino, and O. Boyraz, “Sub-micron silicon nitride waveguide fabrication using conventional optical lithography,” Opt. Express 23(5), 6780–6786 (2015). [CrossRef] [PubMed]
5. X. Zhang and Z. Liu, “Superlenses to overcome the diffraction limit,” Nat. Mater. 7(6), 435–441 (2008). [CrossRef] [PubMed]
6. S. Fan, F. Qi, T. Notake, K. Nawata, Y. Takida, T. Matsukawa, and H. Minamide, “Diffraction-limited real-time terahertz imaging by optical frequency up-conversion in a DAST crystal,” Opt. Express 23(6), 7611–7618 (2015). [CrossRef] [PubMed]
7. A. Grbic and G. V. Eleftheriades, “Overcoming the diffraction limit with a planar left-handed transmission-line lens,” Phys. Rev. Lett. 92(11), 117403 (2004). [CrossRef] [PubMed]
8. M. Mivelle, P. Viktorovitch, F. I. Baida, A. El Eter, Z. Xie, T. P. Vo, E. Atie, G. W. Burr, D. Nedeljkovic, J. Y. Rauch, S. Callard, and T. Grosjean, “Light funneling from a photonic crystal laser cavity to a nano-antenna: overcoming the diffraction limit in optical energy transfer down to the nanoscale,” Opt. Express 22(12), 15075–15087 (2014). [CrossRef] [PubMed]
9. T. Ito and S. Okazaki, “Pushing the limits of lithography,” Nature 406(6799), 1027–1031 (2000). [CrossRef] [PubMed]
10. R. Drevinskas, M. Gecevičius, M. Beresna, Y. Bellouard, and P. G. Kazansky, “Tailored surface birefringence by femtosecond laser assisted wet etching,” Opt. Express 23(2), 1428–1437 (2015). [CrossRef] [PubMed]
11. Y. Lin, M. H. Hong, T. C. Chong, C. S. Lim, G. X. Chen, L. S. Tan, Z. B. Wang, and L. P. Shi, “Ultrafast-laser induced parallel phase-change nanolithography,” Appl. Phys. Lett. 89, 0411081 (2006).
12. T. Shintani, Y. Anzai, H. Minemura, H. Miyamoto, and J. Ushiyama, “Nanosize fabrication using etching of phase-change recording films,” Appl. Phys. Lett. 85(4), 639–641 (2004). [CrossRef]
13. X. Yu and J. Robertson, “Modeling of switching mechanism in GeSbTe chalcogenide superlattices,” Sci. Rep. 5, 12612 (2015). [CrossRef] [PubMed]
14. C. H. Chu, C. Da Shiue, H. W. Cheng, M. L. Tseng, H.-P. Chiang, M. Mansuripur, and D. P. Tsai, “Laser-induced phase transitions of Ge2Sb2Te5 thin films used in optical and electronic data storage and in thermal lithography,” Opt. Express 18(17), 18383–18393 (2010). [CrossRef] [PubMed]
15. M. L. Tseng, B. H. Chen, C. H. Chu, C. M. Chang, W. C. Lin, N. N. Chu, M. Mansuripur, A. Q. Liu, and D. P. Tsai, “Fabrication of phase-change chalcogenide Ge2Sb2Te5 patterns by laser-induced forward transfer,” Opt. Express 19(18), 16975–16984 (2011). [CrossRef] [PubMed]
16. B. J. Zeng, J. Z. Huang, R. W. Ni, N. N. Yu, W. Wei, Y. Z. Hu, Z. Li, and X. S. Miao, “Metallic resist for phase-change lithography,” Sci. Rep. 4, 5300 (2014). [CrossRef] [PubMed]
17. Q. G. Meng, S. G. Zhang, J. G. Li, and X. F. Bian, “Dilatometric measurement and glass-forming ability in Pr-based bulk metallic glasses,” Scr. Metall. 55(6), 517–520 (2006). [CrossRef]
18. Z. F. Zhao, P. Wen, R. J. Wang, D. Q. Zhao, M. X. Pan, and W. H. Wang, “Formation and properties of Pr-Based bulk metallic glasses,” J. Mater. Res. 21(02), 369–374 (2006). [CrossRef]
19. Z. F. Zhao, Z. Zhang, P. Wen, M. X. Pan, D. Q. Zhao, W. H. Wang, and W. L. Wang, “A highly glass-forming alloy with low glass transition temperature,” Appl. Phys. Lett. 82(26), 4699–4701 (2003). [CrossRef]
20. N. H. Pryds, M. Eldrup, M. Ohnuma, A. S. Pedersen, J. Hattel, and S. Linderoth, “Preparation and properties of Mg-Cu-Y-Al bulk amorphous alloys,” Mater. Trans. 41(11), 1435–1442 (2000). [CrossRef]
21. X. X. Xia, W. H. Wang, and A. L. Greer, “Plastic zone at crack tip: a nanolab for formation and study of metallic glassy nanostructures,” J. Mater. Res. 24(09), 2986–2992 (2009). [CrossRef]
22. Q. G. Meng, S. G. Zhang, J. G. Li, and X. F. Bian, “Strong liquid behavior of Pr55Ni25Al20 bulk metallic glass,” J. Alloys Compd. 431(1-2), 191–196 (2007). [CrossRef]
23. X. M. Long, X. S. Miao, J. J. Sun, X. M. Cheng, H. Tong, Y. Li, D. H. Yang, J. D. Huang, and C. Liu, “Dynamic switching characteristic dependence on sidewall angle for phase change memory,” Solid-State Electron. 67(1), 1–5 (2012). [CrossRef]
24. K. E. Petersen, “Dynamic micromechanics on silicon: techniques and devices, IEEE. Trans. Electron. Dev 25(10), 1241–1250 (1978). [CrossRef]
25. M. Mezbahul-Islam and M. Medraj, “Phase equilibrium in Mg-Cu-Y,” Sci. Rep. 3, 3033 (2013). [CrossRef] [PubMed]
26. F. Su and K. Yao, “Facile fabrication of superhydrophobic surface with excellent mechanical abrasion and corrosion resistance on copper substrate by a novel method,” ACS Appl. Mater. Interfaces 6(11), 8762–8770 (2014). [CrossRef] [PubMed]
27. T. B. Wei, F. Y. Yan, and J. Tian, “Characterization and wear- and corrosion-resistance of microarc oxidation ceramic coatings on aluminum alloy,” J. Alloys Compd. 389(1-2), 169–176 (2005). [CrossRef]
