193 nm immersion lithography optical projection systems using conventional UV optical materials and water as the immersion fluid, with planar lens/fluid interfaces, have a practical numerical aperture (NA) limit near 1.3. Higher-index resists and immersion fluids are being developed, but the bottleneck for pushing the NA further is the relatively small refractive index of the final lens element material. We have been exploring novel high-index materials that have the potential to be used in the last lens element to get around this bottleneck and to push the NA limit to at least 1.5, while containing the lens system size and complexity. We present here preliminary index and intrinsic birefringence results for four classes of high-index (n>1.8), wide-band-gap, oxide-based materials that have the potential for being fabricated with optical properties appropriate for lithography optics: group-II oxides, crystalline magnesium-aluminum-spinel-related materials, ceramic forms of spinel, and aluminum garnets. More details are given in Ref. [1].

I. Alkaline Earth Oxides

The Group II elements Mg, Ca, Sr, and Ba form a series of high-index, rock-salt-structure oxides, MgO, CaO, SrO, and BaO, with decreasing band gap energies, respectively. Magnesium oxide (MgO) with a band gap near 7.6 eV [2], and should be able to be made highly transparent at the wavelength 193 nm (6.41 eV). Our measurements made by spectroscopic ellipsometry, shown in Fig. 1, give a refractive index of 1.96(±0.01) at 193 nm. We consistently report uncertainties as one standard deviation. Our measurements of the intrinsic birefringence at 365.1 nm and 253.7 nm are 3.60(±0.2) nm/cm and 16.0(±0.5) nm/cm, respectively.

Fig. 1. Measurement of the index dispersion of MgO for wavelengths 800 nm to 140 nm using spectroscopic ellipsometry.

Mixed solid solutions can be formed, e.g., Mg_xCa_1-xO, for some cation ratios, which may enable tailoring of some of the optical properties. We are exploring whether these degrees of freedom can be exploited to lower the intrinsic birefringence, because of our calculations of the opposite sign of the intrinsic birefringence for MgO and CaO [3].

II. Crystalline Spinel

Crystalline magnesium aluminum spinel (MgAl₂O₄) is a face-centered cubic material with a band gap near 7.75 eV [4]. Our measurements by spectroscopic ellipsometry, shown in Fig. 2, give a refractive index of 1.87(±0.01) at 193 nm. Our measurements of the intrinsic birefringence, shown in Fig. 3, give a value 50.6(±0.1) nm/cm at 195 nm.

Fig. 2. Measurement of the index dispersion of MgAl₂O₄ for wavelengths 800 nm to 140 nm, from spectroscopic ellipsometry.

Fig. 3. Measurement of the intrinsic birefringence of MgAl₂O₄ for wavelengths down to 195 nm.

In addition to the stoichiometric form of magnesium aluminum spinel, by changing the composition of the melt, spinel can be synthesized in forms from Mg-rich to highly Al-rich. Also, with appropriate growth temperature and conditions, spinel can be grown with an inverse structure, where Mg and Al swap positions at some of the crystal sites. These composition and atomic-site-position variations in the spinel system potentially allow some tailoring of the optical properties, which may be of some benefit, e.g., to minimize the intrinsic birefringence. We are exploring the optical properties of these spinel variations.

III. Ceramic Spinel

Magnesium aluminum spinel can also be made into an isotropic polycrystalline ceramic by high-pressure/high-temperature fusing of spinel powder [5]. This material has index, absorption, and other optical properties similar to those of crystalline spinel, except that it has no systematic intrinsic birefringence, due to the random orientation of the 10 µm to 100 µm size spinel crystals.

IV. Yttrium Aluminum Garnet

Yttrium aluminum garnet (Y₃Al₅O₁₂) (YAG) is a high-quality optical material with an extensive history and manufacturing base, used extensively as a host material for rare-earth-doped solid state laser applications in the infrared. It has an electronic band edge near 6.45 eV, about 40 meV higher than the 193.4 nm photon energy. Our index measurements by spectroscopic ellipsometry give an index at 193 nm near 1.9. Measurements of the intrinsic birefringence, shown in Fig. 4, give a value of 24.7(±0.2) at 195 nm. This value is about a factor of 2 lower than that of MgO and crystalline MgAl₂O₄ at this wavelength, which makes this material an attractive candidate high-index material.

Fig. 4. Measurement of the intrinsic birefringence of Y₃Al₅O₁₂ for wavelengths down to 195 nm.

The material we have measured is partially transmitting at 193 nm (absorption coefficient base 10 of A₁₀=3). However, the closeness of the band edge to the 193 nm photon energy makes substantial improvement of the transmission of YAG difficult. However, YAG is known to form alloys with other members of the yttrium family, e.g., scandium and lanthanum. In particular, since scandium is directly above yttrium in the periodic table, partial substitution of yttrium with scandium should increase the band edge, and thus potentially increase the transmission at 193 nm. This appears to be the most promising of the candidate crystalline materials we are investigating. Consequently, we propose Y_3-xSc_xAl₅O₁₂, for 0 < x ≤ 3, as a class of candidate high-index materials, with potentially low intrinsic birefringence and high transmission, for use as the final lens element material for 193 nm immersion lithography systems, to enable numerical apertures near 1.5. We are presently working with several companies to grow samples of these materials, by Czochralski and gradient furnace methods, for us to characterize for this application.

V. Germanate Garnets

This suggests investigating 2-3-4 garnets, a different class of compounds that feature 2-, 3- and 4-times ionized species along with doubly-negative oxygen anions. (Correspondingly, garnets like Y₃Al₅O₁₂ and Lu₃Al₅O₁₂ are called 3-3 garnets.) Well-known silicate 2-3-4 garnets include pyrope and grossularite. These are difficult to grow, so we propose that the following germanate garnets could be used as high-index, low-IBR lithography-grade materials for 193 nm lithography and 193 nm immersion lithography optics: Mg₃Al₂Ge₃O₁₂, Mg₃Lu₂Ge₃O₁₂, Mg₃Y₂Ge₃O₁₂, and (CaY₂)Mg₂Ge₃O₁₂. In order to characterize their properties for the above lithography applications, our collaboration has grown Mg₃Al₂Ge₃O₁₂, Mg₃Lu₂Ge₃O₁₂ and (CaY₂)Mg₂Ge₃O₁₂, and we are attempting to grow Mg₃Y₂Ge₃O₁₂. A further reason for our selection of these compounds include the relatively short cut-off wavelengths (large band gaps) of some of the parent compounds, MgO, Al₂O₃, Y₂O₃, and GeO₂. This demonstrates a large bonding-antibonding splitting between occupied and unoccupied electron band states in compounds containing oxygen and the above elements, and we presume that this splitting will also occur in germanate garnets. In order to tailor the optical properties and optimize growth characteristics, some combination of Mg and Ca could be used for the +2 site and some combination of Al, Sc, Y, and Lu could be used for the +3 site, satisfying the constraints of ionic radii for the site.

We have now found by measurement that (CaY₂)Mg₂Ge₃O₁₂ has an intrinsic birefringence at 193 nm that is opposite in sign to that in Y₃Al₅O₁₂, and that Mg₃Al₂Ge₃O₁₂ has a short cut-off wavelength like Lu₃Al₅O₁₂. This suggests that one has a variety of garnet materials that are transmissive at 193 nm, and that have opposite signs of intrinsic birefringence. We propose using an appropriate mixture of Ca and Mg as doubly-ionized species and an appropriate mixture of Al, Y, Sc and Lu as triply-ionized species to form an aluminum-, germanium- or aluminum/germanium-based garnet(s) that are transmissive at 193 nm and have a small intrinsic birefringence.

1. J.H. Burnett, S.G. Kaplan, E.L. Shirley, P.J. Tompkins, and J.E. Webb, "High-Index Materials for 193 nm Immersion Lithography," in Optical Microlithography XVIII, Proc. SPIE 5754 611-621 (2005).
2. David M. Roessler and Donald R. Huffman, in Handbook of Optical Constants of Solids II, edited by E.D. Palik (Academic Press, New York, 1998), p. 919.
3. Eric L. Shirley and John H. Burnett, "Intrinsic Birefringence of Cubic Oxides," to be published.
4. William J. Tropf and Michael E. Thomas, in Handbook of Optical Constants of Solids II, edited by E.D. Palik (Academic Press, New York, 1998), p. 883.
5. Mark C.L. Patterson, Anthony A. DiGiovanni, Larry Fehrenbacher, and Don W. Roy, in Window and Dome Technologies VIII, edited by Randal W. Tustison, Proc. SPIE Vol. 5078 (SPIE, Bellingham, WA, 2003), p. 71.