The process of ion bombardment is investigated for the fabrication of Mo/Si multilayer x-ray mirrors using e-beam evaporation. The ion treatment is applied immediately after deposition of each of the Si layers to smoothen the layers by removing an additional thickness of the Si layer. In this study the parameters of Kr ϩ ion bombardment have been optimized within the energy range 300 eV-2 keV and an angular range between 20°and 50°. The optical performance of the Mo/Si multilayers is determined by absolute measurements of the near-normal-incidence reflectivity at 14.4 nm wavelength. The multilayer structures are analyzed further with small-angle reflectivity measurements using both specular reflectivity and diffuse x-ray scattering. The optimal smoothening parameters are obtained by determining the effect of ion bombardment on the interface roughness of the Si layer. The optimal conditions are found to be 2 keV at 50°angle of incidence with respect to the surface. These settings result in 47% reflectivity at 85°( ϭ14.4 nm͒ for a 16-period Mo/Si multilayer mirror, corresponding to an interface roughness of 0.21 nm rms. Analysis shows that the interface roughness is determined by ion induced viscous flow, an effect which increases with ion energy as well as angle of incidence. In order to determine the effect of intermixing of the Si and Mo atoms, the penetration depth of the Kr ϩ ions is calculated as a function of ion energy and angle of incidence. Furthermore, the angular dependence of the etch yield, obtained from the in situ reflectivity measurements, is investigated in order o determine the optimal ion beam parameters for the production of multilayer mirrors on curved substrates. © 1997 American Institute of Physics. ͓S0021-8979͑97͒04716-6͔

Ultrahigh vacuum deposition-reflectometer system for the in situ investigation of Y/Mo extreme-ultraviolet multilayer mirrors

ASML has built and shipped to The College of Nanoscale Science and Engineering of the University at Albany (CNSE) and IMEC two full field step-and-scan exposure tools for extreme ultraviolet lithography. These tools, known as Alpha Demo Tools (ADT), will be used for process development and to set the foundation for the commercialization of this technology. In this paper we will present results from the set-up and integration of both ADT systems, status of resist and reticles for EUV, and the plans for using these tools at the two research centers. We will also present the first resist images from one of the tools at the customer site, and demonstrate 32nm half-pitch dense lines/spaces printing as well as 32nm dense contact hole printing.

No abstract

We discuss a new method to characterize multilayer structures with grazing-incidence reflectivity measurements using hard x-ray radiation, such as Cu-Kα or Mo-Kα radiation. The method is based on the analysis of the reverse Fourier transforms of the reflectivity at the Bragg peaks in q-space, the reflectivity data being obtained from an angular scan (θ−2θ). This method is faster than curve fitting of the reflectivity data, results in an accurate value of the density and thickness of both materials, and needs no pre-assumptions about the material composition and the parameters of the multilayer. The method makes a distinction between interface roughness and layer thickness errors, and is independent of measurement of the critical angle. A minor disadvantage is that only an average value of the layer thickness is determined, rather than the individual layer thicknesses. As an example our method is used to analyze small-angle reflectivity measurements of Mo/Si and Co/C multilayers. The parameters thus obtained are used to model a structure, which is subsequently used to predict the near-normal incidence reflectivity at soft x-ray wavelengths. The accuracy thus found corresponds to a relative error of 5%. The densities of a Mo/Si multilayer are determined independently with critical angle measurements to verify the values obtained from the grazing incidence reflectivity measurements. The results of the analysis of the Co/C multilayer are compared to values obtained using a conventional method based on the kinematical theory. However, the conventional method requires additional data of soft x-ray reflectivity measurements. The new method yields the same values for the multilayer parameters but does not require the extra soft x-ray measurements.