flat Si wafer

Download Report

Transcript flat Si wafer

Novel X-ray Optics
with Si Wafers
and Formed Glass
IBWS Vlasim 2006
R. Hudec, L. Pína, A. Inneman, V. Semencová,
M. Skulinová, L. Švéda, V. Brožek, J. Šik, M.
Míka, R. Kačerovský, J. Prokop
Astronomical Institute, Academy of Sciences of the Czech Republic,
Ondřejov, Czech Republic
Czech Technical University, Prague, Czech Republic
Centre for Advanced X-ray Technologies, Reflex sro, Czech Republic
Institute of Chemical Technology, Prague, Czech Republic
ON Semiconductor, Rožnov pod Radhoštěm, Czech Republic
Long Term ISS Utilisation
ESA XEUS
XEUS will have a very large collecting area and fine angular resolution
of 2..5" so very light-weight novel X-ray optics is to be developed
Czech participation
• ESA PECS project accepted for funding
from Dec 2006, PI L. Pina Reflex, CI ASU,
VSCHT, FJFI CVUT. Duration 4 years.
• Preliminary results described in this talk
were obtaained WITHOUT any funding
thanks to the enthuniasmus of
participants.
Our approach to X-ray optics
based on Si wafers
1. the Si wafers parameters are optimized
already at the production stage
2. the Si wafers are shaped to precise optical
surfaces/shapes
3. the internal stress is minimized. Important.
The standard Si wafers are not stressfree.
4. the shaped/bent Si wafers are stacked to
form the Multi Foil Optics (MFO)
Measuring the quality of flat
Si wafers
 the production of Si wafers is a complex
process. The recent Si wafers are optimized
for semiconductor industry but not for X-ray
optics applications
 we use Si wafers already at production
optimized for X-ray optics application
 precise measurements and optimization
already at production stage are important for
X-ray optics based on Si wafers
Comparison of properties of Si
wafers produced by different
manufacturers
standard Si wafers
CZ2 – ON Semiconductor Czech Republic
MEMC – MEMC Electronic Materials
SHE – Shin Etsu SEH
WW – Wafers Works
Thickness homogeneity of
standard Si wafer
produced by ON Semiconductor Czech Republic
min. thickness
max.thickness
686.07 mm
687.75 mm
ave. thickness
687.18 mm
cen. thickness
686.92 mm
TTV Total Thickness Variation 1.68 mm
TIR Total Indicated Reading
1.81 mm
Flatness and thickness uniformity of a standard Si
wafer (diameter 150 mm). Expected to further improve
due to technological innovations planned.
6“ Silicon Wafer Manufacturing Process Flow Chart
SINGLE CRYSTAL
GROWING
INGOT SHAPING
EVALUATION
FLATTING
SLICING
EDGE GRINDING
LASER
MARKING*
LAPING AND
CLEANING
ACID ETCH
ANNEALING*
BACKSIDE
TREATMENT*
BACKSEAL OXIDE,
POLYSILICON*
New step:
GRINDING
POLISHING
CHEMICAL
CLEANING
MECHANICAL
CLEANING
INSPECTION
* Optional
Diamond Grinding Technology for Ideal Flatness
Ground Si wafer
TTV 0.37 mm
6“ Silicon Polished Wafer: Flatness Evolution (TTV)
TTV < 2 mm (2005)
process optimization
TTV = 1.68 mm
TTV < 1 mm (2006)
grinding technology
TTV < 0.5 mm (2007)
new polisher (ordered)
TTV = 0.76 mm
measured Total Thickness Variation
Measuring of shape
Still optical profilometer – 3D chart
flat Si wafer (dopant As), D = 150 mm, t = 0.625 mm
Special very flat Si wafers have been developed in the
collaborating industry for use in X-ray optics
Flat Si wafer, D = 150 mm, t = 0.625 mm
ON Semiconductor Czech Republic,
Taylor-Hobson profilometer, 2 perpendicular axes
Flatness better than 1 mm
Measuring of microroughness
AFM microscope, Ra ~ 0.1 nm in 10 x 10 mm
Flat Si wafer
D=150 mm
t=0.625 mm)
ON SEMICONDUCTOR
Measuring of roughness
Interferometer Zygo, Ra ~ 0.2 nm in 1.4 x 1.1 mm
dopant B, D = 150 mm, t = 0.625 mm
ON Semiconductor, Czech Republic
Measuring of roughness
Interferometer Zygo
effect of various dopants on roughness
(boron-B and phosphorus-P)
SHAPING OF SI WAFERS
 the goal is to precisely shape Si wafers
to desired optical shape and to remove
the internal stress
 3 different technologies tested (I, II,
III)
 so far, Si wafers have been shaped to
test cylindrical and parabolic surfaces
both in 1 and in 2 dimensions
Si wafers shaping
test cylindrical samples
gold-coated, D=100-150 mm, t=0.8-1.3 mm, R=1.5 m
Measuring of roughness after
shaping
concave side
Interferometer Zygo
convex side
Bent (cylinder) Si wafer
R = 1650 mm, D=150 mm, t = 1.3 mm
Measuring of shape
Still optical profilometer – 3D chart
flat Si wafer (dopant P)
D = 150 mm
t = 0.625 mm
bent Si wafer (dopant P)
D = 150 mm
t = 1.3 mm
R = 1650 mm
Deviation bent Si wafers
before
processing
(deviation
from plane)
± 2 µm
after
processing
(deviation
from
cylinder)
± 2.5 µm
Parabolically shaped Si wafer, D = 150 mm, t = 0.625 mm
profile measurement in 2 perpendicular axes
Measuring of shape
Taylor-Hobson profilometer – deviation from ideal shape
D = 150 mm, t = 0.625 mm, parabolic shape
Except edge effects PV < 0.5 mm
Thermally formed Si wafers
thermally formed Si wafer to test cylinder
(R = 150 mm, 72 x 23 x 0.325 mm)
thermally formed Si wafers to test cylinder
(R = 150 mm, 50 x 7 x 0.625 mm)
Optimizing parameters of thermal
forming of Si wafers
The effect of elastic tension on deviation from ideal surface
(thermal forming of Si wafers)
SUMMARY Si wafers
1. Interdisciplinary co-operation (team with 11
members from 5 Institutions) created within the
Czech Republic with experienced teams including
researchers at the large company producing Si
wafers.
2. Si wafers successfully bent to desired geometry
by 3 different techniques with PV ~ 1 mm in the
best case.
3. The bending before stacking is advantageous eg.
to avoid increase of internal stress and to allow
very long-term stability of the mirror array.
4. The production of Si wafers very complex, need
to modify and optimize the parameters at the
production stage.
X-RAY OPTICS BASED ON
GLASS THERMAL
FORMING (GTF)
alternative glass technologies represent glass forming
avoiding heat
The parameters of the GTF may be improved by:
 optimization of the glass material (limited)
 optimization of the mandrel material/design
 optimization of the GTF process
 optimization of the GTF temperature and duration
Expectations (goals)
microroughness of float - glass not degraded
~ 0.5 nm RMS
deviation PV < 0.02 mm
Various approaches in Glass
Thermal Forming
 low-cost design needed (the goal is to produce
very large number of shells at a low cost)
 expensive production/material are to be avoided
 the mandrel material/design is important
 recent design: proprietary technology (composite)
Glass Thermal Forming – one of
studied approaches
parabolic profile
Glass thermal forming (GTF)
cylinder profile
75 x 25 x 0.7 mm
the largest samples so far
300 x 300 mm
parabolic profile
100 x 150 x 0.7 mm
Measuring float glass
(flat glass substrates)
used for GTF
Measuring of roughness before
slumping
Interferometer Zygo
flat thin glass , 100 x 70 x 0.75 mm
Measuring of roughness after
sluming
Interferometer Zygo
bent glass , R = 150 mm, 75 x 25 x 0.75 mm
Measuring of the roughness after slumping
Interferometer Zygo, bent glass, 75 x 25 x 0.75 mm,
optimization using > 100 samples formed at different conditions
minimal
value
Optimizing
the
parameters
of GTF
Ra [nm]
RMS[nm]
Waviness of the surface as function
of time and temperature of GTF
based on TH profilometer measurements of numerous samples
(75 x 25 x 0.75 mm, R = 150 mm) - optimization
Measuring of shape
Still optical profilometer – 3D chart
thermally formed glass, parabolic profile
R = 150 mm, 100 x 150 x 0.75 mm, PV from ideal
shape ~ 0.7 mm in the best case recently
SUMMARY GTF
 surface microroughness not degraded down to
measuring accuracy ~ few 0.1 nm
 profile deviations 0.7 µm (peak-valley) in the best
case recently, expectations < 0.02 µm
 sensitive to T and other parameters of the TGF
process = need for optimization
 sensitive to mandrel design / material / process =>
need for optimization
SUMMARY
 Samples of test X-ray mirrors
produced using novel technologies.
have
been
 Shaped thin glass mirrors and Si mirrors have
been successfully produced.
 Both approaches show promising results justifying
further efforts in these directions.
We acknowledge collaboration with ON Semiconductor CR, Rožnov pod
Radhoštěm, Optical Development Workshop of the AS CR Turnov, and TTS sro
Prague
Contact: [email protected]
The End
