Dan Jaffe

1
Silicon Diffractive Optics for Infrared
Spectroscopy

• Dan Jaffe

2
Why Infrared? Why Spectroscopy? Tools of the
trade How to make a better mousetrap out of
silicon
3
Infrared lets you see through dust into star
forming regions. It also gives you access to
physical processes you cannot study in the
visible.
4
Diffraction gratings bend light like prisms, but
through a greater angle for a given wavelength
separation. This larger dispersion allows you
to resolve more closely spaced spectral features.
5
Grating Equation mld(sinasinb)
The grating equation gives allowable
directions. The blaze tells you how much power
goes to which.
6
Key Science for High Resolution Near-IR
Spectroscopy
Star Formation
Probe the kinematics, densities, and chemical
properties of unresolved disks. Molecular lines
at 3-5 mm
7
Pre-Main Sequence Stellar Astrophysics

• Rotational Properties (vsini)
• Magnetic Field Strength (Zeeman splitting)
• Accretion Rates (Brackett Line Profiles)
• Cluster Kinematics (radial velocities)
• Fundamental Parameters (Teff, log-g, abundances)
• Courtesy K. Covey (U. Washington/CfA)

8
Enable planet searches around low mass stars and
brown dwarfs. Sensitivity for cooler, low mass
stars is much better than in the visible.
Put color argument here

9
Determine the chemical evolution of the Galaxy.
Study isotopic ratios such as 18O/16O
10

• The problem is that everything radiates in the
  infrared and so your entire spectrograph must be
  cooled to almost absolute zero.

11
Immersion Gratings are the Key to the New
Spectrographs
An immersion grating is a grating in which
grooves are immersed in a medium with an index
of refraction n.
12
An immersion grating gains you a factor of n2-n3
in spectrograph volume.
Idea dates back to Fraunhofer (1822). Reinvented
in 1954 by Hulthén and Neuhaus Patented in
1984 by Sica. Never made to work because of
production difficulties.
13
How precise does your grating have to be?
Assuming random errors in groove spacing,
allowable wave front RMS error is
Dsrms25nm
14
We produce immersion gratings by a process of
photolithography and chemical micromachining.
Step 1. Purchase a boule of high purity
material Step 2. Cut the boule into
disks Material needs to be oriented to achieve
various blaze angles
6.16o
54.7o
63.4o
15
Production
Step 3. Polish disks using chemical mechanical
polish (CMP). Step 4. Deposit passivation
layer. The passivation layer will be patterned
into an etching mask which defines the grating
period.
16
Production

• Step 5. Deposit photoresist.
• Spin-on photoresist at 3500 rpm

Photoresist layer
Passivation layer
Silicon substrate
17
Production
Production
Step 6. UV exposure through contact
photolithography mask (contact is
a critical issue) Step 7. Develop exposed
photoresist We have the image of the mask in
photoresist
18
Production
Production
Step 8. Etch the passivation layer Si3N4 is
etched via reactive ion etching (RIE).

Substrate
19
Production
Production
Step 9. Photoresist removal Positive image of
grating mask pattern in the passivation layer.
20
Production
Production
Step 10. Anisotropic silicon etch in a KOH
solution Si 2OH- 2H2O ? SiO2(OH)2-- 2H2
21
Production
Production
The exposed (111) crystal planes are smooth on
an atomic scale.
22
Production
Production
Step 11. Remove the remaining passivation
layer Remaining Si3N4 is etched in concentrated
phosphoric acid at 150oC.
23
Production
Production
Step 12. Cut the disk into a prism
24
Production
Production
Step 13 anti-reflective coating on the entrance
face and reflective coating on the groove surfaces
25
Si Immersion Grating Production

Grating etched into silicon puck
Puck cut into prism and then polished
Flat entrance face antireflection coated
Device completed by aluminizing the grooves along
the hypotenuse
Note Only the bottom of the coating matters
26
Evaluation

• Combination of tests
• efficiency measurements
• interferometric measurements
• point-spread function (PSF) measurements
• analysis of grating defects and aberrations
  (ghosts, scattered light)
• Tests give us consistent results on grating
  performance and help us analyze the sources of
  errors.

27
Evaluation
Sample 1D monochromatic spectra
28
Evaluation
Interferometric tests done at 632.8 nm.
29
Evaluation

• Point spread function measurements also test
  diffraction limited performance.
• Compare to optical performance of a flat
  mirror
• Analyze errors (ghosts) and compare to
  periodic errors observed with interferometric
  tests
• Determine resolving power from FWHM
• Demonstrated resolving power of up
• to 75,000 with G1

30
Diffuse scattered light is caused by surface
microroughness and various macro defects.
Atomic force microscopy of a 5 ?m by 5 ?m area
of G2. RMS roughness is 1.6 nm.
Current grooves are 80 mm wide and 40 mm
long. If this were a 1m wide sidewalk, it would
be 0.5 km long. The bump in the picture would
be 60 mm.
31
Actual Flight Hardware for SOFIA
32
Planned Instrument for NASA IRTF
33
GMTNIRS
34
Planned instrument for the Giant Magellan
Telescope