![cross.png](/logos/buttonlogos/cross.png)
![cross.png](/logos/buttonlogos/cross.png)
research papers
Synchrotron-radiation-stimulated etching of SiO2 thin films with a tungsten nano-pillar mask
aDepartment of Physics, Shanghai Jiao Tong University, 800 Dong Chuan Road, Shanghai 200240, People's Republic of China, and bDepartment of Vacuum UV Photoscience, Institute for Molecular Science, Myodaiji, Okazaki 444-8585, Japan
*Correspondence e-mail: cswang@sjtu.edu.cn
A nano-pattern of SiO2 on a Si (100) surface has been demonstrated by synchrotron-radiation-stimulated etching with a tungsten nano-pillar mask. The reaction gas was a mixture of SF6 and O2. The mask was fabricated using a focused ion beam with W(CO)6 as the source gas. The width and height of the tungsten nano-pillar were ∼80 nm and 160 nm, respectively. Synchrotron radiation irradiation with flowing SF6 and O2 effectively etches the silicon dioxide, and the etching process followed the surface The etched surface was very flat, and no undercutting occurred during the etching process.
Keywords: synchrotron-radiation-stimulated etching; tungsten nano-pillar mask; SiO2 thin films; surface photochemical reaction; etched pattern.
1. Introduction
Fine patterning techniques in etching have been in great demand in micro-fabrication. Ultra-large-scale integrated circuits have been developed extensively over the past tens of years, and the size of one element has been very much reduced. However, it is difficult to integrate further using conventional processes, and thus a new process has been desired in microelectronics manufacturing. Strong soft X-rays from an undulator in a storage ring are expected to solve this requirement (Ogawa et al., 1994). Synchrotron radiation should be an ideal light source for the study of photochemical reactions owing to various characteristics in the soft X-ray region, such as high intensity, continuity of wavelength and small divergence. Most gas molecules and solids used in semiconductor processes have a large in the soft X-ray region and can be electronically excited at the core or valence levels by synchrotron radiation. Synchrotron-radiation-stimulated etching has attracted considerable attention from the viewpoint of developing new surface micromachining technologies owing to its unique features of high spatial resolution, extremely high material selectivity, anisotropy etching, low damage and clean etching atmosphere (Urisu & Kyuragi, 1987
; Takahashi et al., 1991
). In preceding works synchrotron-radiation-stimulated etching of SiO2 thin films was conducted using a contact cobalt mask, and a micro-pattern of SiO2 thin films was made for the area-selective deposition of self-assembled monolayers (Wang & Urisu, 2003
). The cobalt contact mask on the SiO2 surface was fabricated by sputtering Co thin film on a photolithography resist pattern and using a lift-off technique. In this paper, synchrotron radiation etching of the SiO2/Si surface has been examined using a tungsten nano-pillar mask to manufacture a nano-pattern of SiO2 thin films on the silicon substrate. The tungsten mask was deposited on the SiO2 surface by a focused ion beam with W(CO)6 source gas.
2. Experiment
Fig. 1 shows the schematic process of sample preparation and synchrotron-radiation-stimulated etching. A single-crystal silicon wafer was cut and thoroughly cleaned with acetone, ethanol, deionized water and a solution of concentrated H2SO4 and H2O2. The cleaned wafer was immersed in a dilute HF solution for 2 min to remove its native oxide. The SiO2 thin film was formed on the freshly cleaned Si surface by annealing at 1173 K for 10 h in a dry oxygen atmosphere. The thickness of the SiO2 film measured with an ellipsometer was about 220 ± 10 nm. On the SiO2 surface, a W pillar mask was fabricated using focused ion-beam-induced chemical vapor deposition (FIB-CVD) (SII NanoTechnology) with Ga ions and W(CO)6 source gas. FIB-CVD is a common technique of direct patterning of metallic or insulator films used for repairing electronic devices and photomasks. The method of using FIB-CVD with W(CO)6 source gas to deposit three-dimensional tungsten structures has already been reported (Koops et al., 1988
). The W pillar mask was deposited with a gas pressure, acceleration voltage and ion current of 1 × 10−3 Pa, 30 kV and 0.5 pA, respectively, and the deposited pillar usually contains tungsten, carbon, oxygen and gallium (Ishida et al., 2003
). A portion of the tungsten mask on the SiO2 surface is shown in Fig. 2
. The transverse width and longitudinal height of the tungsten pillar were ∼80 nm and 160 nm, respectively.
![]() | Figure 1 Schematic process of the sample preparation and synchrotron-radiation-stimulated etching. |
![]() | Figure 2 AFM quasi-three-dimensional image of the tungsten nano-pillar. |
Synchrotron radiation etching of SiO2/Si (100) substrate was conducted using a mixture of SF6 and O2 as the reaction gas at beamline 4A2 of the synchrotron radiation facility (UVSOR) at the Institute for Molecular Science of Japan. A large pressure difference between the etching chamber and the beamline was sustained by using a differential vacuum pumping system. The sample was set normal to the incident synchrotron radiation beam. Synchrotron radiation light irradiated the sample surface directly; no window was used. Details of the apparatus for synchrotron radiation etching have been described in an earlier paper (Hirano et al., 1998). The beam diameter on the sample surface was ∼9 mm and the beam current of the storage ring was ∼200 mA.
3. Results and discussion
In the experiment, SiO2 was found to be effectively etched by synchrotron radiation with SF6 + O2 as the reaction gas at room temperature. The etching took place only in the area irradiated with synchrotron radiation and proceeding in the direction of the incident beam, as shown in Fig. 3. In this figure the SiO2 thin film on the silicon substrate was directly irradiated with synchrotron radiation in an atmosphere of SF6 + O2 gas mixture without a mask. No noticeable etching of SiO2 was found when SF6 was not put in the etching chamber, the sample substrate was only exposed to synchrotron radiation at room temperature. It has been reported that an amorphous SiO2 (α-SiO2) film continuously decomposes and evaporates when irradiated with synchrotron radiation at elevated substrate temperatures (Akazawa, 1995
). Why do we call it synchrotron-radiation-stimulated etching? There are two main types of in photo-excited etching at room temperature (Urisu & Kyuragi, 1987
; Kitamura et al., 1994
). One is a gas-phase in which photo-excitation of the surface does not take part in the reaction and the etching reaction occurs spontaneously. The other is a surface in which photo-irradiation of the surface is necessary for the etching reaction. This experimental result suggested that photo-excitation of material surfaces and the following chemical reactions played important roles in synchrotron-radiation-stimulated etching. Under the irradiation of synchrotron radiation, the SiO2 molecule was excited by high-energy photons to a reactive or and SF6 molecules produced F ions. The excited SiO2 reacted with the F ions, and then the resulting compounds decomposed and evaporated. Therefore, the surface excitation was dominant in the above synchrotron-radiation-exited etching SiO2/Si system. Anisotropy etching was achieved owing to a surface This property is practical for fabricating extremely fine patterns with a high aspect ratio.
![]() | Figure 3 CCD image of a portion of the synchrotron-radiation-etched structure without a mask. |
In our previous publication (Wang & Urisu, 2003) it was reported that the etching rate depended on the surface intensity of the synchrotron radiation beam and the of the reaction molecules. It increased with increasing SF6 gas pressure in the range where the attenuation of the beam intensity due to gas absorption was negligible. In the region of higher gas pressure the beam intensity was attenuated by the reaction gas absorption and the etching rate saturated or decreased. In addition, the etching rate increased with decreasing substrate temperature. Thus, the etching rate can be increased by optimizing the reaction gas pressure and by decreasing the substrate temperature. Furthermore, the etching process stopped completely on the SiO2/Si interface if a little O2 was added to the reaction gas. The synchrotron radiation irradiation with flowing SF6 + O2 did not etch the silicon crystal, i.e. high etching selectivity can be obtained between Si and SiO2 layers. Based on these experiments, we carried out the synchrotron-radiation-stimulated etching of SiO2 thin films with a tungsten mask. The synchrotron-radiation-etched pattern with tungsten masks was observed using a scanning electron microscope (SEM) (Jeol, JSM-5510) and an atomic force microscope (AFM) (SII, SPI3800N/SPA400) in dynamic force mode with a 20 µm scanner and a SI-DF40 (spring constant = 42 N m−1) cantilever. Figs. 4
and 5
show the SEM and AFM images of the synchrotron-radiation-etched pattern with tungsten pillar masks, respectively. Patterning SiO2 thin film on the silicon surface in the nano-scale was achieved by synchrotron-radiation-stimulated etching. In the etching experiment the irradiation dose of the synchrotron radiation beam was about 10000 mA min and the gas pressure of SF6 and O2 was about 0.05 torr and 0.002 torr, respectively. From these images it can be concluded that no undercutting had occurred, and the unique features of synchrotron-radiation-stimulated etching such as high spatial resolution, high material selectivity and anisotropy etching are proven. Moreover, the synchrotron-radiation-etched surface looked very flat and uniform, which is important when utilizing this pattern template (Wang & Urisu, 2003
; More et al., 2002
). Similar characteristics were reported in another paper in which the roughness with a line profile on the synchrotron-radiation-etched surface was 0.05 nm. The maximum difference in height in the observed surface was less than 0.2 nm and the sloping angle between two positions in the observed surface was less than 1.2° (Wang & Urisu, 2005
). The hollow appearing around the tungsten pillar was perhaps due to damage of the silicon crystal during fabrication of the tungsten mask. The damaged silicon was likely etched by synchrotron radiation irradiation with flowing SF6 + O2 mixture gas. The exact mechanism of this phenomenon is not yet clear. Further investigations are in progress.
![]() | Figure 4 SEM image of the synchrotron-radiation-etched pattern with a tungsten pillar mask. |
![]() | Figure 5 AFM quasi-three-dimensional image of the synchrotron-radiation-etched pattern with a tungsten pillar mask. |
4. Conclusion
Based on previous synchrotron-radiation-stimulated etching, a SiO2 nano-pattern structure has been fabricated onto a silicon surface using SF6 + O2 as the reaction gas and a tungsten nano-pillar mask. Etching of SiO2 occurred due to a surface excitation reaction which involved the interaction between the excited states of the Si–O bond and F ions or radicals. A high spatial resolution and anisotropy etching was displayed, and the etched surface was very flat. Synchrotron-radiation-etched patterns will be an interesting structure for many kinds of silicon-based devices (Miqin et al., 1998).
Acknowledgements
This work was supported by the National Natural Science Foundation of China (10675083), the Applied Materials Shanghai Research & Development Fund (0416), and a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan.
References
Akazawa, H. (1995). Phys. Rev. B, 52, 12386–12394. CrossRef CAS Web of Science Google Scholar
Hirano, S., Yoshigoe, A., Nagasono, M., Mase, K., Ohara, J., Nonogaki, Y., Takeda, Y. & Urisu, T. (1998). J. Synchrotron Rad. 5, 1363–1368. Web of Science CrossRef CAS IUCr Journals Google Scholar
Ishida, M., Fujita, J., Ichihashi, T., Ochiai, Y., Kaito, T. & Matsui, S. (2003). J. Vac. Sci. Technol. B, 21, 2728–2731. Web of Science CrossRef CAS Google Scholar
Kitamura, O., Goto, T., Terakado, S., Suzuki, S., Sekitani, T. & Tanaka, K. (1994). Appl. Surf. Sci. 79/80, 122–128. CrossRef Google Scholar
Koops, H. W. P., Weiel, R., Kern, D. P. & Baum, T. H. (1988). J. Vac. Sci. Technol. B, 6, 477–481. CrossRef CAS Web of Science Google Scholar
Miqin, Z., Desai, T. & Ferrari, M. (1998). Biomaterials, 19, 953–960. CrossRef PubMed Google Scholar
More, S., Graaf, H., Baune, M., Wang, C. & Urisu, T. (2002). Jpn. J. Appl. Phys. 41, 4390–4394. Web of Science CrossRef CAS Google Scholar
Ogawa, T., Mochiji, K., Ochiai, I. & Yamaoto, S. (1994). J. Appl. Phys. 75, 4680–4685. CrossRef CAS Web of Science Google Scholar
Takahashi, J., Utsumi, Y. & Urisu, T. (1991). J. Appl. Phys. 70, 2958–2962. CrossRef CAS Web of Science Google Scholar
Urisu, T. & Kyuragi, H. (1987). J. Vac. Sci. Technol. B, 5, 1436–1440. CrossRef CAS Web of Science Google Scholar
Wang, C. & Urisu, T. (2003). Jpn. J. Appl. Phys. 42, 4016–4019. Web of Science CrossRef CAS Google Scholar
Wang, C. & Urisu, T. (2005). Appl. Surf. Sci. 242, 276–280. Web of Science CrossRef CAS Google Scholar
© International Union of Crystallography. Prior permission is not required to reproduce short quotations, tables and figures from this article, provided the original authors and source are cited. For more information, click here.