AIXUV's Tools for EUV-Reflectometry Rainer Lebert, Christian Wies AIXUV GmbH, Steinbachstrasse 15, D Aachen, Germany

Transcription

1 AIXUV's Tools for EUV-Reflectometry Rainer Lebert, Christian Wies, Steinbachstrasse 5, D-, Germany and partners developed several tools for EUV-reflectometry in different designs for various types of applications. The range of tools is starting with the affordable, flexible, and compact reflectometers for laboratory use (CEUVR) and is including the high end fully automated high speed reflectometer for EUV mask blanks (MBR) that is fully compliant with the SEMI-standard P38 for EUV mask blank metrology and suited for daily in-fab use. In this paper the different concepts will be discussed and the performance of the systems will be presented. The paper will focus on the upgrades of the mask blank reflectometer for being used on structured masks including smaller spot size as well as improved accuracy and precision. The compliance with the ITRS metrology roadmap is discussed. The work was partially founded by the German Bundesministerium für Bildung und Forschung (BMBF).

2 Polychromatic EUV Reflectometry (Spectrophotometry) Grating CCD-Camera In our tools, we apply Polychromatic Spectrophotometry : EUV-Source Aperture (Virtual) Entrance Slit The whole EUV-spectrum is, spectrally dispersed and detected by a CCD. This allows for minimized measuring spot, high throughput with high numbers of spectral channels and highest DQE. Advantages: Most effective use of photons emitted from the The set-up is stationary With a CCD camera: brilliant DQE and Side-information (scatter, geometry) is obtained The whole spectral reflectance curve is measured simultaneously. Hence, measuring 2000 spectral channels can be accomplished with less than 00 pulses. Measuring system is mechanically stable. With reference mirrors any fluctuations do not influence the relative results. With plasma source wavelength calibration with emission lines is immanent, Disadvantages: Instead of a diode: CCD camera: evaluation. Data transfer from the CCD limits the throughputs. Polychromatic Photometry offers the most effective approach for characterizing spectral reflection, transmission, absorption, scatter etc.

3 EUV-Reflectometer designed for EUV Mask Blanks (MBR) Parameter: Spectral resolution Wavelength calibration: Wavelength reproducibility Measuring spot size Measuring spectral range Time for one spot best quality Angle of incidence reproducibility Absolute accuracy (MADT) Reproducibility of reflectance MBR typical.6 pm ± < 2 pm ± < pm < 00 µm mm.8 to 4.9 nm (2048 channels) 20 s < 00 µrad < 0.5 % absolute < 0. % RMS Reflectance Increment < % Minimum Reflectance Capability < 0. % Dedicated for SEMI compatible Qualification of EUV Mask Blanks (extendable to EUV-Masks, due to extremely small spot and high throughput)

4 MBR: Three Spectra are Measured in Comparison 0000 R = R Cal Signal * Signal CAL@ Calibration Signal Signal Reference@ Calibration Reference@ Measurement CCD Ref Ref 2 Grating Slit Schematic of comparative measurement with reference mirrors Wavelength ( nm in 2048 channels) NB Reference Typical raw CCD-result BB Reference nm Average Counts per Pixel Ref Ref2 Signal, a.u CCD PIXEL number measured spectrum NIST Identified Xe-Lines Dispersion Linear (Dispersion) Spectral Channel Three samples are measured simultaneously with high dynamics y = x R 2 = WL(963) = (3.507 ± ) nm Dispersion := (.576 pm ± 5 fm) /PIXEL Wavelength, nm Immanent wavelength calibration with Xe-emission lines

5 High Reproducibility and Sensitivity R eflectan ce 70% 60% 50% 40% 30% 20% 0% 0% R e fle c ta n c e 3.7% 3.5% 3.3% 3.% 2.9% 2.7% 2.5% Wavelength, nm Layer deviation off-center is detected Peak reflectance = % ± 0.05 % Central wavelength = nm ± 0.4 pm WL of peak FWHM = nm ± 0.8 pm = nm ± 0.4 pm R e fle c ta n c e Wavelength, nm 70.4% 70.3% 70.2% 70.% 70.0% 69.9% 69.8% 69.7% 69.6% R = 0.06 % λ = 8 pm R = 0. % λ = 0 pm 69.5% Wavelength, nm R = 0.55 % λ = 2 pm Off-center results demonstrate Sensitivity b c d Center measurements (with movement after 5) overlaid

6 Absorber Measurements (Down to Substrate).2% 0.00% % 0.009% 0.008% 000 reflectance 0.8% 0.6% 0.4% Reflectance 0.007% 0.006% 0.005% 0.004% 0.003% 00 0 Signals, av. counts per pixel row CXRO_Reflectance REFLECTANCE SIGNAL Substrate SIGNAL Reference 0.2% 0.002% 0.00% 0.0% wavelength [nm] Typical 5 measurements on Absorber overlaid 0.000% Wavelenght, nm Reflectance measurement on uncoated substrate 0000 Peak reflectance =.087 % ± % 9000 Central wavelength = nm ±. pm FWHM = nm ± 2.3 pm 4 bit dynamics is usable Signal Substrate: av. counts per pixel row Signal Broadband: av. counts per pixel row SIGNAL Braodband Signal Substrate Substrate Spectra are of > /0000 contrast to Reference Signals of same measurement and significant to < 0.2 counts (as averaged over 200 pixel columns) Wavelenght, nm Signals from Reference and Substrate compared

7 CPR: Compact Polychromatic R&D Reflectometer Parameter: CPR typical Spectral resolution 0 pm Wavelength calibration: ± < 0 pm Wavelength reproducibility ± < 0 pm Measuring spot size < 00 µm mm Measuring spectral range 0 to 6 nm (024 ch) Time for one spot best quality 5 s (LAMP) 0 min (Tube) Absolute accuracy (MADT) < % absolute Reproducibility of reflectance < 0.5 % RMS Reflectance Increment < 0.02 % Minimum Reflectance Capability < 0. %

8 CEUVR: Flexible Spectrophotometer (Reflectance, Transmission, Scatter.) EUV-Source Kirkpatrik-Beaz Spot forming Spectrograph Position for Calibration and Transmission CCD Grating Manipulator Spectrograph Position for Reflectance Rotation Rail SpeXUV

9 Summary AIXUV and our partners can supply a variety of solutions for Spectrophotometry in EUV spectral range. Ranging from dedicated reflectometer for EUV mask-blanks over highly flexible EUV-Spectrophotmeter to economic solution for Coating R&D. Parameter: MBR CEUVR++ CEUVR CPR Spectral resolution, pm.6 20 Wavelength calibration, pm ± < 2 ± < 0 tube) ± < 40(tube) ± < 4(lamp) ± < 20(lamp) Wavelength reproducibility, pm ± < ± < 6 ± < 30 Measuring spot size 50 µm * 2 mm (down to 250 µm) Measuring spectral range, nm 3 (e.g. 2-5) 0 (e.g.8 8) Spectral channels Time for one spot (tube) 0 min 0 min (Lamp) 20 s 20 s Angle of incidence one fixed variable variable one fixed (e.g. 6 ) (0-85 ) (e.g. 6 ) Alignment triangulation manual reproducibility < 00 µrad < 2 mrad Size, mm 50*50 Ø< 400 Ø< 50 Based on experiments we can suggest best adapted solution for your task.