Schott Zerodur

SCHOTT LITHOTEC ZERODUR®

ZERODUR® Zero thermal expansion glass ceramic 10152 e ba/kn/wo Printed in Germany Schott Lithotec Shapes and d ZERODUR®

For many years ZERODUR® zero thermal The t expansion material has provided for reliable rods designs in precision optical applications. charg The homogeneity of the material properties Even of ZERODUR® enables opto-mechanical engineering solutions with long-term Shap mechanical and thermal stability. The easy- to-achieve optical figure and improved Disks microroughness have contributed to the success of this material.

ZERODUR® is widely used as a mirror substrate material and for precision frames in current state-of-the-art microlithography equipment.

Rods Specific structuring of the material allows superior stability of designs in ambitious dynamic environments, such as stages in wafer steppers and scanners.

With its ability to match zero thermal Plate expansion very closely, ZERODUR® is one of the prime material candidates for substrates in lithography applications at Polis wavelengths around 13 nm (EUVL). Light

Applications support from Schott Lithotec is available for material properties and specific customers designs of structured Reticle stage made of ZERODUR ® components. Mechanical P

Crite Youn Poiss Knoo 8 inch Wafer Stage Dens made of ZERODUR ® SCHOTT LIT

Thermal properties

Mean coefficient of The most important and significant properties of the optical glass ceramic linear thermal expansion ZERODUR® are the extremely small coefficient of linear thermal expansion as well as the homogeneity of this coefficient throughout the entire piece. Individual pieces of ZERODUR® (discs, plates, rods) can be supplied with a mean coefficient of linear thermal expansion α in the temperature range 0°C to 50°C in three expansion classes as follows:

Expansion class 0 0± 0.02·10-6/K Expansion class 1 0± 0.05·10-6/K Expansion class 2 0± 0.10·10-6/K

Material up to Expansion class 2 will be supplied as a standard. Closer tolerances will be supplied upon request. The graph below illustrates the typical coefficient of linear thermal expansion α. ZERODUR® exhibits a very slight linear expansion over the entire temperature range. It is especially low in the room temperature range.

0,6 /K] -6 0,4 [10

α 0,2

0,0 -0,2 -0,4 -0,6 0 °C100 °C -0,8 0 100 200 300 400 500 600 700 800 900

Temperature [K]

ZERODUR® M If even higher thermal expansion quality is required and cooling rates other than

0.1K/min (or the initial cooling rate) can not be avoided in the critical temperature 10152 e ba/kn/wo Printed in Germany region during processing or use of ZERODUR®, we recommend considering the use of ZERODUR® M as a material variation of ZERODUR®. Schott Lithotec AG Sales Office Wilh.-Th.-Roemheld-Strasse 32 55130 Mainz Germany Phone: +49(0)6131/8 01 26 - 12 Fax: +49(0)6131/8 01 26 - 27 E-mail: [email protected] www.schott.com/lithotec Shapes and dimensional tolerances Internal quali

The table shows the typical and improved tolerances for disks, rectangular blocks, The i rods and complex shapes. Improved tolerances can be provided for an additional Inc charge. Stri Bul Even tighter tolerances can be provided upon special request.

Inclusions As a r Shape Criterion Dimensions Typical Improved Inclus tolerance tolerance surfac Disks Diameter < 500 mm ± 0.3 mm ± 0.1 mm Diameter < 1000 mm ± 0.4 mm ± 0.2 mm Inclus Diameter < 4200 mm ± 0.5 mm ± 0.3 mm ”Spec Thickness – ± 0.3 mm ± 0.05 mm Profile tolerance < ø 2000 mm ≤ 0.2 mm ≤ 0.1 mm zone of radius Max Profile tolerance > ø 2000 mm ≤ 0.3 mm ≤ 0.2 mm of in zone of radius in m ≤ ± ± Rods Diameter L 500 mm and/ 0.3 mm 0.01 mm diam ≤ or ø 80 mm of th ± Diameter L > 500 mm and/ 0.3 mm – In th or ø > 80 mm < 5 ≤ ± ± Length ø 80 mm 0.3 mm 0.2 mm In th ± ± Length ø > 80 mm 0.4 mm 0.3 mm < 5 Plates Length/thickness – ± 0.3 mm ± 0.05 mm Flatness < 500 mm ≤ 0.4 mm ≤ 0.05 mm Striae Striae Flatness ≥ 500 mm ≤ 0.4 mm ≤ 0.2 mm slight Polishing/ Detailed specification required Lightweighting The s given

Stre DUR ® by s with < 5

Mechanical Properties Bulk stress All ZE perm

Bulk path Criterion Young’s modulus (at 20 °C, GPa) – mean value 90,3 Bulk Poisson’s ratio µ 0,243 [nm Knoop hardness HK 0,1/20 acc. to ISO 9385 620 diam tage Density ρ (g/cm3) 2,53 DUR ® < 5 lerances Internal quality

s for disks, rectangular blocks, The internal quality of ZERODUR® is essentially determined through be provided for an additional Inclusions Striae Bulk stress al request.

Inclusions As a rule inclusions are mainly bubbles and to some extent individual particles. Typical Improved Inclusions have no effect on the function of ZERODUR® as a substrate material for tolerance tolerance surfaces of the highest quality in so far as they lie completely within the volume. ± 0.3 mm ± 0.1 mm ± 0.4 mm ± 0.2 mm Inclusions in ZERODUR® are distinguished by the designations “Standard“ and ± 0.5 mm ± 0.3 mm ”Special“ and also draw on the Inclusion Classes 3 to 0 which apply to optical glass. ± 0.3 mm ± 0.05 mm ≤ 0.2 mm ≤ 0.1 mm StandardSpecialClass 3Class 2 Class 1Class 0 Maximum diameter ≤ 0.3 mm ≤ 0.2 mm of individual inclusions in mm for different ± ± d/ 0.3 mm 0.01 mm diameters or diagonals of the ZERODUR® part ± d/ 0.3 mm – In the critical volume < 500 mm 1.4 1.2 1.0 0.8 0.6 0.4 ± ± 0.3 mm 0.2 mm In the uncritical volume ± ± 0.4 mm 0.3 mm < 500 mm 3.0 2.0 1.5 1.0 0.8 0.6 ± 0.3 mm ± 0.05 mm ≤ 0.4 mm ≤ 0.05 mm Striae Striae are locally very limited transparent regions with composition differing only ≤ 0.4 mm ≤ 0.2 mm slightly from the basic material. required The striae values listed below are maximum values for five quality levels up to a given diameter or diagonal of ZERODUR® parts.

Stress birefringence caused Standard Special Class 3 Class 2 Class 1 by striae [nm/stria] for parts with diameters or diagonals < 500 mm 60 45 30 5 –

Bulk stress All ZERODUR® parts are subjected to precision optical annealing in order to achieve permanent bulk stress, which is low and symmetrically distributed.

Bulk stress causes optical birefringence. The stress birefringence is measured as a path difference.

90,3 Bulk stress birefringence Standard Special 0,243 [nm/cm] for parts with 620 diameters or diagonals 2,53 < 500 mm 6 4 ® Environmental Stability Flexure Strength (MPa) ZERODUR Bonded Components 70

Un-treated All samples made with polished 60 After 85/85 for 1 week General Low Temperature Bonding (LTB) ZERODUR® surfaces, regardless of information curing temperature, show no strength 50 degradation after a 1-week treatment 40 Narrative adapted from telecom write-up previously prepared for OFC. under 85/85 conditions. 30 Bonded ZERODUR® components are mechanically robust and can be Samples made using fine ground sur- 20 machined, chemically durable in traditional grinding and polishing faces, regardless of curing temperature, environments, and are thermally stable in environments ranging from show a significant strength degrada- 10 room temperature to at least 600ºC. 42 45 27 38 24 10 40 17 tion after the same 85/85 treatment. 0 Polished Polished Polished Polished Ground Ground Ground Ground RT Cure RT Cure Std. Cure Std. Cure RT Cure RT Cure Std. Cure Std. Cure Shapes and Dimensional Information

LTB can be applied to solid parts as well as light-weighted (ribbed) structures. Vacuum Compatibility Solid ZERODUR® components have been fabricated up to 2 inches in Vacuum levels attained: diameter with high strength suitable to subsequent post-processing (core drilling, blocking, grinding and polishing). 10 –12 mbar/liter/sec, meaning 1 liter of He leaking per second at 10 –12 mbar pressure, equivalent to ~3 x 107 atoms/sec (ultra-high Lightweighted structures have been fabricated from ZERODUR® up vacuum range) to sizes of 265 mm diameter with rib structures as thin as 2 mm. An Leak rate below current detection limit; lower than that typically example of a light-weighted ZERODUR® sample is shown at the left. specified by manufacturers of UHV valves Same results on both heat-treated and non-heat treated samples; no signs of degassing in either sample, though samples had been Visual Appearance and bonded ~1 month prior to test Bond Thickness 10 –9 torr also demonstrated at Honeywell on non-λ/3 parts: both cured and non-cured Bonds formed between ZERODUR® material are optically transparent and can be tailored to fill gaps of defined thickness. Bond thicknesses can be varied by modifying processing conditions and part attributes. Typical FIB/SEM results are shown at the right for thick bonds as well as thin bonds. [T %] 100 Transmission Thermal Shock Resistance 90

80 Bonds are transparent from 160 nm Bonded samples were submerged in liquid nitrogen for nearly 70 Bonded (5mm) to 2 µm (see figures) and have a Mono 1 (5mm) 3 minutes. In all cases, there was no significant change in measured nominal refractive index (25°C) of 60 Mono 2 (5mm) bond strength after thermal schock. 1.47. 50 40 Flexure Strength (MPa) 70 30 Not Thermally Shocked 20 60 N2 Thermally 10 50 0 160 170 180 190 200 Lambda/nm 40 [T %] 100 30

90 20 80

® 70 Bonded ZERODUR 10 Monolithic ZERODUR® 60 52 50 32 40 24 20 35 31 0 50 Polished Polished Polished Polished Ground Ground Ground Ground RT Cure RT Cure Std. Cure Std. Cure RT Cure RT Cure Std. Cure Std. Cure 40

20 10 Low Thermal Expansion 0 0 500 1000 1500 2000 2500 Wavelength (nm) Bonded ZERODUR® components exhibit CTE performance Bonded components contain minimal residual stress (3nm/cm). (RT-300 ºC) commensurate to monolithic ZERODUR® for bonding perpendicular to bonding axis.

Flexure Strength (MPa) Average CTE from 0–50°C 70 Mechanical Strength 0 CTE Rod #2 60 –10 CTE Rod #1 Bonded ZERODUR® components are 50 mechanically robust, showing MOR’s –20 comparable to that of monolithic 40 ZERODUR® as determined from 4 pt –30 30 bend flexure testing. –40 20

10 –50

52 32 24 35 43 0 –60 Polished Surfaces Polished Surfaces Ground Surfaces Ground Surfaces Monolithic 20 40 60 80 100 120 140 RT Cure Std. Cure RT Cure Std. Cure Cure Temperature Measured on core drilled, fine ground specimens [T %] 100 Transmission Thermal Shock Resistance 90