Abstract: We report the initial results from a 4X reduction interferometric lithography technique using extreme ultraviolet (EUV) radiation from a new undulator on the Aladdin storage ring at the Synchrotron Radiation Center of the University of Wisconsin-Madison. We have extended traditional interferometric lithography by using 2nd diffraction orders instead of 1st orders. This change considerably simplifies mask fabrication by reducing the requirements for mask resolution. Interferometric fringes reduced by 4X (from 70 nm half-period grating to 17.5 nm) have been recorded in a 50 nm thick hydrogen silsesquioxane photoresist using 13.4 nm wavelength EUV radiation.
1. Burn J. Lin, “Sober view on extreme ultraviolet lithography,” J. Microlith., Microfab., Microsyst. 5, 033005 (2006), and the references there in.
2. E. H. Anderson, K. Komatsu, and H. I. Smith, “Achromatic holographic lithography in the deep ultraviolet," J. Vac. Sci. Technol. B 6, 216 (1988)
3. A. Yen, E. H. Anderson, R. A. Ghanbari, M. L. Schattenburg, and H. I. Smith, “Achromatic holographic configuration for 100-nm-period lithography,” Appl. Opt. 31, 4540 (1992).
4. M. Wei, D.T. Attwood, T.K. Gustafson, and E.H. Anderson, “Patterning a 50-nm Period Grating using Soft Xray Spatial Frequency Multiplication”, J. Vac. Sci. and Tech., 12/6, 3648-3652, (1994).
5. M. L. Schattenburg, C. Chen, P. N. Everett, J. Ferrera, P. Konkola, and H. I. Smith, "Sub-100 nm metrology using interferometrically produced fiducials," J. Vac. Sci. Technol. B 17, 2692-2697 (1999).
6. H. H. Solak, D. He, W. Li, S. Singh-Gasson, F. Cerrina, B. H. Sohn, X. M. Yang, and P. Nealey, "Exposure of 38 nm period grating patterns with extreme ultraviolet interferometric lithography," Appl. Phys. Lett. 75, 2328 (1999).
7. Y.-C. Cheng, A. Isoyan, J. Wallace, M. Khan, and F. Cerrina, “Extreme Ultraviolet Holographic Lithography: Initial Results,” Appl. Phys. Lett. 90, 023116 (2007).
8. A. Isoyan, Y.-C. Cheng, F. Jiang, J. Wallace, F. Cerrina, and S. Bollepalli, “Progress in extreme ultraviolet interferometric and holographic lithography,” J. Vac. Sci. Technol. B 25, 2145 (2007).
9. H. H. Solak, C. David, J. Gobrecht, V. Golovkina, F. Cerrina, S. O. Kim, and P. F. Nealey, "Sub-50 nm period patterns with EUV interference lithography," Microelectron. Eng. 67-68, 56 (2003).
10. K. Eidmann, M. Kuhne, P. Muller, and G. D. Tsakiris, “Characterization of pinhole transmission gratings,” J. X-Ray Sci. Technol. 2, 259–273 (1990)
11. H. W. Schnopper, L. P. Van Speybroeck, J. P. Delvaille, A. Epstein, E. Kallne, R. Z. Bachrach, J. Dijkstra, and L. Lantward, “Diffraction grating transmission efficiencies for XUV and soft x rays,”Appl. Opt. 16, 1088 (1977).
12. M. Born and E. Wolf, “Principles of Optics,” Macmillan, New York, (1964), 7th Ed., Chap. 8, p. 412.
#94978 - $15.00 USD Received 14 Apr 2008; revised 21 May 2008; accepted 23 May 2008; published 4 Jun 2008
(C) 2008 OSA 9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 9106
13. L. E. Ruggles, M. E. Cuneo, J. L. Porter, D. F. Wenger, and W. W. Simpson, “Measurement of the efficiency of gold transmission gratings in the 100 to 5000 eV photon energy range.,” Rev. Sci. Instrum. 72, 1218 (2001).
14. Artak Isoyan, Yang-Chun Cheng, Fan Jiang, John Wallace, Mikhail Efremov, Paul Nealey, and Franco Cerrina, ‘Progress in extreme ultraviolet interferometric lithography at the University of Wisconsin,” Proc. SPIE 6921, 6921R (2008).
15. John Wallacea, Yang-Chun Cheng, Artak Isoyan, Quinn Leonard, Mike Fisher, Mike Green, Joseph Bisognano, Paul Nealey, Franco Cerrina, "A novel EUV exposure station for nanotechnology studies," Nucl. Instrum. Methods Phys. Res. A 582, 254–257, (2007).
16. M. Goldstein, Sematech (private communication).
1. Introduction
Interferometric lithography is usually used in the 1st order, generating interference fringes displaying a period equal to half the period of the original mask gratings. Although beneficial, a 2X reduction still poses fabrication challenges for nanopatterning. A larger reduction factor (e.g., 4X instead of 2X) would be highly desirable. This change is not trivial, because of the properties of the imaging interferometric system, as discussed below. Careful optimization of the gratings design, fabrication process, and illumination is necessary to achieve high resolution fringes from large period gratings.
2. EUV Interferometric Lithography
Extreme ultraviolet lithography (EUVL) research involves a broad range of topics, including tools, the source, projection optics, resists, and masks. Exposure systems in the EUV region around 13.4 nm are needed for the development of imaging materials, and advanced photoresists are needed for upcoming lithography nodes at 20 nm and below. Typically, high-resolution patterning at or less than 32 nm period lines and spaces requires optical systems that are complex and expensive. Today, only a few such EUV exposure tools exist [1]. Over the course of the last decade, we have developed a new lithographic technique, EUV interference lithography (EUV-IL), to support research and development efforts in many areas that use periodic patterns for research and production [2–6]. As we have shown, a natural extension of EUV-IL is EUV holographic lithography (EUV-HL), in which a computer-generated hologram patterned on a suitable carrier is used to form the image of an arbitrary pattern on a given substrate [7]. The EUV-IL setup is simple, requiring only a bright and small emittance EUV source (ideally an undulator on a synchrotron [8]) and the interferometer itself. No complex optics are needed, the system is simple and robust, and the pattern is formed over depths of hundred of microns. The high power allows for fast exposures, so that the throughput is high, while the small beam emittance reduces image blurring [8].
3. Second Order Diffraction Transmission Gratings
Various interferometric lithography techniques based on diffraction gratings have been described and used in the UV/deep-UV regions [5]. Recently, we have extended this technique to EUV at 13.4 nm [6, 9]. A conceptual design is shown in Fig. 1(a). All of these techniques use transmission grating’s 1st diffraction order to produce an interferometric fringe pattern, which means the period of the produced gratings is reduced by a factor of 2. Here, we report the extension of the interferometric lithography technique to produce gratings using the 2nd diffraction order of the transmission gratings, which gives a reduction factor of 4X. Thus, the period of the interference fringe is 4X smaller than transmission mask gratings period, considerably simplifying the fabrication of the original gratings.
The interference of two plane waves creates a standing wave pattern with a period p given by p = λ/2sinθ, where θ is equal to half of the angle between the propagation directions of the
#94978 - $15.00 USD Received 14 Apr 2008; revised 21 May 2008; accepted 23 May 2008; published 4 Jun 2008
(C) 2008 OSA 9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 9107
two beams. The fringe pattern formed is independent of the wavelength of the illumination; hence, the name of “achromatic interferometric lithography.” The ultimate resolution limit achievable in interference imaging is equal to one-quarter of the wavelength, corresponding to θ = π/2. When higher order (m) diffracted beams from two gratings overlap in the central area at a certain distance from the gratings, they form additional interference patterns, with the period pm = pgr/2m, where pgr is the period of the original grating and m is the diffraction order. These fringe patterns are achromatic as well. We stress that the period of imaged fringes is reduced by a factor m, or equivalently the spatial frequency is multiplied by a factor of 2m.
Fig. 1. (a). Sample figure EUV-IL detail showing the regions where the beams overlap for the synthesis of 1st and 2nd order. Right, (b) SEM images of the mask grating structure, (c) 1st and (d) 2nd order diffraction exposure interference fringes recorded in PMMA resist. Notice the relative period of the images. The 1st order working distance (WD) is 400 µm and the 2nd order optimal WD is 132 µm. The scale bar indicates 200 nm.
One of the most important parameters of the transmission gratings is the diffraction efficiency, which is analyzed in detail in [10–13]; here we limit the discussion to the case of binary gratings, since at a 13.4 nm wavelength, all materials are highly absorptive. The maximum theoretical diffraction efficiency for a binary grating is 1/m2π2, where m is the diffraction order and equal to ±1, ±3, ±5, … . One can see that the “usable” diffraction order for interferometric imaging is the 1st diffraction order and the diffraction efficiency is equal to ~10%. All other high orders are too weak for imaging (e.g., the next 3rd order is only ~1.1%). All calculations assume that the grating profile is ideal, and we notice that there are no even diffraction orders when the spacing between bars is equal to a half-period (e.g., 50-50 grating). The absorption caused by the supporting membrane exp(−μt), where μ is the absorption coefficient and t is the thickness of supporting membrane, is not included in the calculations since it can be easily compensated for by increasing the exposure time of the experiment. Varying the spacing between the grating bars (i.e., changing the duty cycle) opens new possibilities for using higher order diffraction. The diffraction efficiency of the grating of
#94978 - $15.00 USD Received 14 Apr 2008; revised 21 May 2008; accepted 23 May 2008; published 4 Jun 2008
(C) 2008 OSA 9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 9108
period p can be written as: sin2(mdπ /p)/m2π2, where d is the spacing between the bars and m = ±1, ±2, ±3 … . The efficiency for the mth diffracted order is dependent on d/p, and when d = (2m-1)p/2m, the efficiency reaches a maximum. Thus, 2nd order diffraction is possible when the spacing between the bars is equal to 3p/4 and 1st order diffraction efficiency is ~5%, 2nd order is ~2.5%, and 3rd order is ~ 0.5%.
With an interferometer, we must carefully consider the superposition of the orders in the propagation region. The geometry of the 4X reduction exposure system is shown in Fig. 1(a). The 1st order diffracted beams overlap together starting from a distance of
( ) λλ 21 2 /p/wp − , where w is the grating width, assumed equal to the separation. The
optimal working distance is λλ /)/(1 2pwp − when the width of overlap region reaches its
maximum, and there is no double overlap between 1st and 2nd orders. This gives us an opportunity to use a mask designed for 2nd order to image 1st order interference fringes as well, with reduced efficiency. On the other hand, the 2nd order working distance at
λλ 2/)/2(1 2pwp − overlaps with the 1st order. It can be easily derived that to avoid
overlap between the 1st and 2nd orders the optimal working distance for the 2nd order should
Fig. 2. SEM image of 17.5nm half-period 4X reduced 2nd order EUV-IL interference fringes recorded in a 50 nm thick hydrogen silsesquioxane (HSQ) resist.
The fabrication process is based on a standard process developed at the Center for NanoTechnology for creating X-ray diffractive optics [7, 14]. A low-stress 200 nm Si3N4 is chosen as the thin membrane. A ~7 nm Cr and a ~15 nm Au layers are deposited on the front-side of the wafer. A JEOL JBX-5DII e-beam lithography tool with 50 keV beam energy is used to pattern the EUV-IL mask structure in 150 nm thick negative tone NEB22A resist. After development, an O2 plasma etch/5 sec is done to completely clean any residue. The EUV-IL mask is then plated for 14 min at 5 mA current using TechnoGOLD 25ES plating solution. The Au is ~100 nm thick. Finally, the EUV-IL mask is etched by O2 plasma for 5 min to remove the NEB22A resist. In this case, an additional stop layer is not needed, due to the high absorption in Au. The fabricated EUV-IL mask contains five different period gratings, starting from 110 nm half-period down to 70 nm half-period at a 400 µm 1st order working distance. The scanning electron microscope (SEM) image of the 110 nm half-period grating is shown on Fig. 1(b). Since the grating spacing bar equals p/4 or 3p/4, the fabrication processes for such masks are sometimes easier.
#94978 - $15.00 USD Received 14 Apr 2008; revised 21 May 2008; accepted 23 May 2008; published 4 Jun 2008
(C) 2008 OSA 9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 9109
4. Exposure Results
Exposures were performed at the Center for Nanotechnology’s EUV-IL exposure tool at the Synchrotron Radiation Center of the University of Wisconsin-Madison [15]. Figures 1(c) and 1(d) show the 55 nm half-period 2X reduced 1st order and 27.5 nm half-period 4X reduced 2nd order EUV-IL printed interference fringes on 55 nm thick PMMA resist. Figure 2 shows 17.5nm half-period 4X reduced 2nd order EUV-IL interference fringes recorded in a 50 nm thick hydrogen silsesquioxane (HSQ) resist. Due to a difference in diffraction efficiency, the dose required to print the 2nd order IL image is 4X greater than that for the 1st order. The exposure time is 40 sec for PMMA, which is acceptable. With the existing mask, we have an exposure area of 80um x 10um (H x L). By modifying the distance between the gratings this area can be increased, with the size being ultimately limited by the spatial coherence of the source. In the specific case of a synchrotron source like ours [8] the vertical spatial coherence allows the exposure of several hundred microns.
It is important to clarify the role of “leakage” through the unpatterned regions, particularly in the area separating the gratings [8]. Because of the geometry of the system, any radiation transmitted through this region will illuminate the interference area. This leakage will introduce a coherent background in the image, reducing the image modulation. A mask plated with 100 nm of Au has ~1% transmission; we note that ~1% leakage is comparable to the diffraction efficiency in the 2nd order. The uniform background introduces an alternating modulation in the fringes. This effect is clearly visible in Fig. 3(a), where the linewidth is modulated. The right panel in Fig. 3(c), shows the simulated image intensity of the fringes taking leakage into account. The leakage can be reduced by increasing the thickness of the absorbing material (Au) in the separation region between the gratings, while keeping the thickness of the absorber in the grating region to its optimal value.
a) b) c)
Fig. 3. SEM image of 27.5 nm half-period grating in PMMA resist: (a) “Leakage” caused distortion of the peaks in the interference pattern; notice the alternating narrower and wider lines. (b) Intensity of the SEM scans; (c) Predicted fringe intensity, explaining the corresponding linewidth modulation. The lines have been added to aid the eye.
5. Conclusion
We have presented the initial results of a 2nd order interferometric lithography technique using the EUV-IL exposure system at the University of Wisconsin-Madison Synchrotron Radiation Center. The techniques we have developed allow patterning with a 4X reduction factor. We have successfully recorded 17.5 nm half-period directly from a 70 nm half-period
#94978 - $15.00 USD Received 14 Apr 2008; revised 21 May 2008; accepted 23 May 2008; published 4 Jun 2008
(C) 2008 OSA 9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 9110
transmission mask in a 50 nm thick HSQ resist using 13.4 nm wavelength EUV radiation from an undulator source. This technique will allow us to print 10 nm half-period interference fringes from 40 nm half-period gratings on the mask.
The analysis presented here is based on a simple binary grating approach. With increasingly finer gratings this hypothesis becomes inadequate, and phase effects must be considered. However, a more accurate analysis will not change the main conclusions. We also note that the calculations presented here assume a scalar diffraction model. A more sophisticated vector model will likely be needed to analyze the larger diffraction angles formed in the 2nd order and at smaller periods, or p/λ value [16]. We plan to address these issues in a later publication.
Acknowledgments
This work was supported in part by Sematech, Albany. It was also supported in part by the Semiconductor Research Corporation under Contract No. 2005-OC-985. It was also funded by the Nanoscale Science and Engineering Center (NSF DMR-0425880), and by the Synchrotron Radiation Center (NSF DMR-0084402), both at the University of Wisconsin-Madison. The fabrication was performed at the Wisconsin Center for Applied Microelectronics (WCAM) and UW Center for Nanotechnology.
#94978 - $15.00 USD Received 14 Apr 2008; revised 21 May 2008; accepted 23 May 2008; published 4 Jun 2008
(C) 2008 OSA 9 June 2008 / Vol. 16, No. 12 / OPTICS EXPRESS 9111
