Open Access
Add to BibSonomyAbstract
A new phase change material Ge2Sb1.5Bi0.5Te5 (GSBT) with good optothermal effect has been developed as an inorganic photoresist. Masks based on the material can be easily fabricated by home-built laser direct writing (LDW) equipment, and as a result mask patterns have been successfully transferred onto Si substrates by reactive ion etching techniques. Experimental results indicate that maximum etching selectivity of Si to GSBT reaches up to 524:1, which is comparable with the traditional organic photoresists, and the high ratio is also explained theoretically. Because of the merits of the inorganic photoresist, it might prove useful in silicon-based microelectronics
©2012 Optical Society of America
1. Introduction
With the rapid development of the semiconductor and microelectronic industries, the feature size of photolithography is continually reducing by using shorter wavelength lasers [1–3]. For this, it has to develop new organic photoresists corresponding to every new laser wavelength because each photoresist is only sensitive to specific wavelengths, which often needs a very long R&D period and a high investment. For overcoming the obstacles, it is necessary to explore new material systems suitable for a broad wavelength spectrum. Because phase change materials with optothermal response mode have broad spectrum features [4], they have attracted much attention as promising inorganic photoresists recently [5–8]. This kind of new photoresist can be easily deposited onto substrates (both planar and non-planar) in vacuum [9–11], which not only can produce a large area uniform film resist layer, but also is compatible with full vacuum processing technologies in next generation microelectronic technology. In addition, they will greatly simplify production procedures because they need neither a particular light source nor a particular environment and can completely eliminate pre-baking and post-baking steps required for organic photoresists [12].
Ge-Sb-Te (GST), as a typical phase change material, has been widely applied in optical data storage [13–15] and Phase-change Random Access Memory (PRAM) [16–19] based on difference of reflectivity or resistance between amorphous and crystalline states. Recently, some work has been reported on the selective etching property of GST phase material [4,20,21], but it is far from a good inorganic photoresist candidate because of poor surface roughness and blurry patterns after development.
In this paper, a satisfying photoresist material Ge2Sb1.5Bi0.5Te5 (GSBT) is proposed. It can easily be patterned by laser-induced phase change, and can be developed effectively by KOH solution forming a smooth surface and a clear edge. We also investigated selective etching between Si and GSBT by reactive ion etching (RIE), and obtained a high etching selectivity ratio of 524:1. In addition, some patterns transferred successfully onto the Si substrates indicate that the GSBT can be used as a practical photoresist.
2. Experiment
The experimental process is shown in Fig. 1 . GSBT thin films were firstly deposited onto Si substrates at room temperature by RF magnetron sputtering (ULVAC, ACS-4000-C4) under the conditions of a work pressure of 0.1 Pa, a power of 50 W and 25 sccm Ar flow. The as-deposited GSBT film was patterned by a laser direct writing (LDW) system (532 nm laser wavelength, 0.9 NA objective lens) with energy density of 1.31 × 104 J/cm2, then the patterned film was developed in KOH alkaline solution. Using SF6 as the work gas, the developed groove structures were successfully transferred onto the Si substrates by RIE (ETCHLAB 200).
Fig. 1 Schematic showing the experiment steps followed in the study. GSBT resist is patterned by LDW and phase change etching. RIE is applied to transfer formed patterns onto the Si.
The samples were characterized by Hitachi S-4800 field emission scanning electron microscope (FESEM), Dimension 3100 atomic force microscope (AFM) and a stylus profiler (Dektak 150 stylus). The phase changes of the films were studied by Rigaku D/MAX-2000 X-ray diffraction (XRD) and transmission electron microscope (TEM, Tecnai G2 F20 U-TWIN) combined with selected area electron diffraction (SAED) analysis.
3. Results and discussions
3.1 Thermal and laser-induced phase transitions in GSBT
Figure 2(a) is the SEM image of an as-deposited 100 nm GSBT thin film, showing a very smooth surface with ultrafine particles. The mean roughness of the deposited surface was measured by AFM to be 2 nm, which is very suitable for resist thin films. After annealing in a vacuum ambient at 150 °C for 15 min, its surface becomes rough and some small crystal grains appear (Fig. 2(b)), indicating the film has been crystallized. The XRD results in Fig. 2(c) show that the as-deposited film is in an amorphous state, while the emergent peaks after annealing at 150 °C reveal that a phase transformation from the amorphous to a polycrystalline state has occurred in the GSBT film, and these peaks can be indexed as (200), (220), (222) and other planes of a face-centered cubic (FCC) structure [22]. The above results indicate that the as-deposited smooth GSBT film has suitable thermal threshold of phase change, which is very important to carry out thermal mode lithography on GSBT thin films. The UV-Visible absorption spectra of the amorphous state GSBT film in Fig. 2(d) shows the effective absorption of GSBT based on optothermal effect can be suitable for a wide wavelength range, especially for the UV band that is used for lithography in microelectronics.
Fig. 2 The SEM images of as-deposited (a) and annealed (b) GSBT thin films, (c) XRD patterns of as-deposited and annealed GSBT, (d) the UV-Visible absorption spectra of GSBT.
In order to study optothermal effects on the material, a GSBT thin film was deposited on a copper grid coated with an extremely thin carbon diaphragm, and then a stripe was written by LDW with a power density of 1.50 × 104 J/cm2. A TEM image and SAED patterns of the sample are shown in Fig. 3 . The region marked with “A” is the part without laser irradiation, its corresponding diffusive halo of the SAED pattern (Fig. 3(b)) further confirms the as-deposited film is amorphous. The region marked with “B” is the middle area of the stripe where some bigger grains have formed, and the associated regular diffraction spots (shown in Fig. 3(c)) confirm that it is crystalline with a FCC structure the same as that of the thermal phase change, clearly indicating that the essence of laser-induced phase change is a light-induced thermal phase change.
Fig. 3 TEM image and SAED patterns of GSBT resist written by LDW with energy density 1.50 × 104 J/cm2.
3.2 The developing properties of GSBT in KOH solution
The developing properties of GSBT are quite crucial to its capability as a photoresist. Figure 4 shows the development depth (surface height change in the process of development) of the GSBT films in the amorphous and crystalline state varies with development time in the 1.0 wt % KOH solution, respectively. It has been found that the development rate for the crystalline state is about 7.2 Å/min, while the development rate for the amorphous state is about 2.2 Å/min. This demonstrates that the patterned GSBT thin film can be effectively developed by KOH solution.
Fig. 4 Development properties of different phase of GSBT films versus development time in the 1.0 wt % KOH solution.
As an example, Fig. 5(a) shows an AFM image of a groove array on a GSBT film fabricated by LDW before development, the laser-irradiated regions are slightly lower than these regions without laser irradiation by about 3.5 nm (Fig. 5(b)), which might be because the atomic density of the crystalline state is larger than that of the amorphous state [23]. Numerous experiments show that the development is completed after immersing the sample into 1.0 wt % KOH solution for 150 min. The AFM image as well as the corresponding cross section of the developed sample (Fig. 5(c) and Fig. 5(d)) show that the laser-induced regions have been etched away completely and formed uniform trapezoid grooves with an average depth and FWHM of about 74 nm and 250 nm, respectively. The reason for the etched depth not perfectly coinciding with the above development rate may be that the development rate becomes small with the increase of development time because the concentration of solution is reduced. In addition, the developed GSBT surface is very smooth with groove edge roughness of about 3 nm, which is good for pattern transformation.
Fig. 5 (a) AFM image of sample in which selected crystalline regions were induced by LDW. (b) The cross section profile of the laser-induced sample. (c) The appearance of the GSBT sample which has been developed by KOH, and its cross section profile is shown in the (d).
3.3 The etching selectivity between Si and GSBT
Etch selectivity is a key factor to evaluate a photoresist, which is defined as y = VS/VM [24], where VS and VM represent the etching rate of substrates and photoresist, respectively. Figure 6(a) shows the dependence of the etching rates of GSBT film and Si versus SF6 gas flow rate under a given power of 30 W and a pressure of 10 Pa. It is seen that the etching rate of GSBT just increases slightly, although the etching rate of Si increases obviously with an increase in SF6 flow rate from 10 sccm to 40 sccm and decreases when the SF6 flow rate exceeds 40 sccm. Hence, a maximum etching selectivity (~75:1) is obtained at the optimal SF6 flow rate of 40 sccm. These results should be induced by the distinguishing mechanism of Si and GSBT etched by using SF6, the etching of Si with fluorinated compounds is dominated by the following chemical processes:
The etching rate of Si is proportional to the density of fluorine atoms and the reaction probability of fluorine at the surface [25]. Although the increase of SF6 flow rate would initially enhance the etching of Si, the further increasing flow rate would also shorten the gas residence time on the surface of Si and reduce the reaction probability of free fluorine with Si, which results in the decreasing of etching rate of Si when the SF6 flow rate is over 40 sccm. On the other hand, because SF6 doesn’t react with GSBT, the etching of GSBT is governed by physical collision. This etching process is far lower efficiency than the etching of Si, so the etching rate of GSBT is quite low.Fig. 6 The effect of SF6 gas flow rate (a), etching power (b), and etching pressure(c) on the etching rates and etching selectivity of Si to GSBT films. (In order to show the changing trends of all data clearly in the same figure, all the etching rates of GSBT are magnified 20 times).
Figure 6(b) shows that under optimized gas flow of 40 sccm and pressure of 10 Pa, the etching rate of Si increases sharply from 20 W to 50W, but becomes gradually saturated as the power continues to increase, while the etching rate of GSBT rises steadily, and the maximum etching selectivity has reached 201:1 at the etching power of 50 W. This may be because the reaction of fluorine atoms with Si increase with power, but the maximum of the reactive fluorine atoms is limited by the total dose of SF6 flow, thus the etching rate of Si will tend to saturate at higher power. While the etching of GSBT is a physical bombardment process, which would be enhanced by the increasing power, since it makes the mean kinetic energy of the plasma electron higher.
With an increase of etching pressure under the optimized gas flow rate and etching power (40 sccm, 50 W), the etching rates of both GSBT and Si increases first and then decreases (Fig. 6(c)), and the etching selectivity of Si to GSBT reaches a maximum of 524:1 at 14 Pa. These results can be ascribed to the competition of two effects from etching pressure: on the one hand, the increasing pressure enhances the ion bombardment, favoring the etching. On the other hand, the high etching pressure shortens the gas mean free path, so as to lower the energy of fluorine radicals or particles participating in the physical bombardment. This effect is adverse to the etching process. Such competition of these two effects results in the variation tendency of the etching rate observed in Fig. 6(c).
So far, the maximum etching selectivity can reach 524:1, which can be comparable with traditional organic resists and meet the demands of nanofabrication.
3.4 Remove residual GSBT resist
After the pattern transfer, a farther descumming process was needed to clean the residual resist. Here, the residual GSBT resist was cleaned by RIE in the Ar gas ambient, in which the etching rate of GSBT is much larger than that of Si. For example, the etching rate of GSBT is as high as 0.74 nm /s, while the etching rate of Si is only 0.22 nm /s under the conditions of an etching pressure of 4 Pa, a power of 50 W and Ar flow rate of 40 sccm. Finally, patterned Si sample (with the average groove depth of about 350 nm and the FWHM of about 530 nm) was fabricated using the GSBT as the resist layer, which is shown in Fig. 7 . We found that the etched parallel grooves are very uniform with the average line edge roughness of about 10 nm, which can meet the demands of nanofabrication.
Fig. 7 (a) SEM image of Si etched by RIE with the GSBT as photoresist. (b) AFM image of the etched groove. (c) Cross section profile of an etched groove.
4. Conclusion
Ge2Sb1.5Bi0.5Te5 which can be used as an inorganic photoresist over a wide-range of wavelengths has been investigated in detail. Nanostructures have been successfully transferred onto a Si substrate from patterned GSBT resists. The etching selectivity of Si to GSBT is influenced by the gas flow rate, etching power and etching pressure, and the maximum etching selectivity can reach 524:1, which can be comparable with traditional organic resists and meet the demands of nanofabrication.
Acknowledgments
This work is supported by NSFC (10974037, 61006078), NBRPC (2010CB934102), International S&T Cooperation Project (2010DFA51970), and Eu-FP7 (No. 247644).
References and links
1. T. Ito and S. Okazaki, “Pushing the limits of lithography,” Nature 406(6799), 1027–1031 (2000). [CrossRef] [PubMed]
2. M. Trikeriotis, W. J. Bae, E. Schwartz, M. Krysak, N. Lafferty, P. Xie, B. Smith, P. A. Zimmerman, C. K. Ober, and E. P. Giannelis, “Development of an inorganic photoresist for DUV, EUV, and electron beam imaging,” Proc. SPIE 7639, 76390E (2010). [CrossRef]
3. D. V. Myagkov, M. O. Nestoklon, and E. L. Portnoi, “Simple and effective algorithm of inorganic resist As2S3 development simulation,” Proc. SPIE 6732, 67321V (2007). [CrossRef]
4. 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]
5. V. Lyubin, A. Arsh, M. Klebanov, R. Dror, and B. Sfez, “Nonlinear photoresists for maskless photolithography on the basis of Ag-doped As2S3 glassy films,” Appl. Phys. Lett. 92(1), 011118 (2008). [CrossRef]
6. V. Lyubin, M. Klebanov, I. Bar, S. Rosenwaks, N. P. Eisenberg, and M. Manevich, “Novel effects in inorganic As50Se50 photoresists and their application in micro-optics,” J. Vac. Sci. Technol. B 15(4), 823–827 (1997). [CrossRef]
7. V. I. Min’ko, P. E. Shepeliavyi, I. Z. Indutnyy, and O. S. Litvin, “Fabrication of silicon grating structures using interference lithography and chalcogenide inorganic photoresist,” Semicond. Phys., Quantum Electron. Optoelectron. 10(1), 40–44 (2007).
8. K. P. Chiu, K. F. Lai, S. C. Yen, and D. P. Tsai, “Surface plasmon polariton coupling between nano recording marks and their effect on optical read-out signal,” Opt. Rev. 16(3), 326–331 (2009). [CrossRef]
9. B. J. Choi, S. Choi, T. Eom, S. H. Rha, K. M. Kim, and C. S. Hwang, “Phase change memory cell using Ge2Sb2Te5 and softly broken-down TiO2 films for multilevel operation,” Appl. Phys. Lett. 97(13), 132107 (2010). [CrossRef]
10. W. P. Risk, C. T. Rettner, and S. Raoux, “Thermal conductivities and phase transition temperatures of various phase-change materials measured by the 3ω method,” Appl. Phys. Lett. 94(10), 101906 (2009). [CrossRef]
11. 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]
12. H. Jain and M. Vlcek, “Glasses for lithography,” J. Non-Cryst. Solids 354(12-13), 1401–1406 (2008). [CrossRef]
13. A. V. Kolobov, P. Fons, A. I. Frenkel, A. L. Ankudinov, J. Tominaga, and T. Uruga, “Understanding the phase-change mechanism of rewritable optical media,” Nat. Mater. 3(10), 703–708 (2004). [CrossRef] [PubMed]
14. M. L. Lee, K. T. Yong, C. L. Gan, L. H. Ting, S. B. Muhamad Daud, and L. P. Shi, “Crystalline and thermal stability of Sn-doped Ge2Sb2Te5 phase change material,” J. Phys. D Appl. Phys. 41(21), 215402 (2008). [CrossRef]
15. Z. M. Sun, J. Zhou, and R. Ahuja, “Unique melting behavior in phase-change materials for rewritable data storage,” Phys. Rev. Lett. 98(5), 055505 (2007). [CrossRef] [PubMed]
16. K. Nakayama, M. Takata, T. Kasai, A. Kitagawa, and J. Akita, “Pulse number control of electrical resistance for multi-level storage based on phase change,” J. Phys. D Appl. Phys. 40(17), 5061–5065 (2007). [CrossRef]
17. C. Kim, D. M. Kang, T. Y. Lee, K. H. P. Kim, Y. S. Kang, J. Lee, S. W. Nam, K. B. Kim, and Y. Khang, “Direct evidence of phase separation in Ge2Sb2Te5 in phase change memory devices,” Appl. Phys. Lett. 94(19), 193504 (2009). [CrossRef]
18. J. Lee, S. Choi, C. Lee, Y. Kang, and D. Kim, “GeSbTe deposition for the PRAM application,” Appl. Surf. Sci. 253(8), 3969–3976 (2007). [CrossRef]
19. S. J. Park, I. S. Kim, S. K. Kim, S. M. Yoon, B. G. Yu, and S. Y. Choi, “Phase transition characteristics and device performance of Si-doped Ge2Sb2Te5,” Semicond. Sci. Technol. 23(10), 105006 (2008). [CrossRef]
20. 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]
21. J. H. Kim, “Effects of a metal layer on selective etching of a Ge5Sb75Te20 phase-change film,” Semicond. Sci. Technol. 23(10), 105009 (2008). [CrossRef]
22. T. H. Jeong, H. Seo, K. L. Lee, S. M. Choi, S. J. Kim, and S. Y. Kim, “Study of oxygen-doped GeSbTe film and its effect as an interface layer on the recording properties in the blue wavelength,” Jpn. J. Appl. Phys. 40(Part 1, No. 3B), 1609–1612 (2001). [CrossRef]
23. W. K. Njoroge, H. W. Woltgens, and M. Wuttig, “Density changes upon crystallization of Ge2Sb2.04Te4.74 films,” J. Vac. Sci. Technol. A 20(1), 230–232 (2002). [CrossRef]
24. G. M. Feng, B. Liu, Z. T. Song, S. L. Feng, and B. Chen, “Reactive-ion etching of Ge2Sb2Te5 in CF4/Ar plasma for non-volatile phase-change memories,” Microelectron. Eng. 85(8), 1699–1704 (2008). [CrossRef]
25. R. Legtenberg, H. Jansen, M. D. Boer, and M. Elwenspoek, “Anisotrapic reactive ion etching of silicon using SF6/O2/CHF3 gas mixtures,” J. Electrochem. Soc. 142(6), 2020–2028 (1995). [CrossRef]
