An All-reflective Integral Field Spectrograph for Far Ultraviolet Astrophysics


Nov 18, 2013 (4 years and 6 months ago)



An All
reflective Integral Field Spectrograph for Far Ultraviolet Astrophysics

Stephen E. Kendrick
, Dennis Ebbets
, Chuck Hardesty
, Ken Sembach
, Matt Beasley
, Bruce Woodgate


Ball Aerospace & Technologies Corp.,
Space Telescope Science Institute,
University of Colorado,
NASA Goddard Space Flight Center


This paper overviews the supporting optical technologies for an ultraviolet integral field
spectrograph (IFS) that will be used for future space astrophysics missions. The new technology
is an all
reflective image slicer that directs light to an array of imaging diffraction gratings.
Previous UV instruments recorded the spectra of point sources or spatially resolved elements
along a long slit. Our IFS has nominally only one reflection more than the Cosmic Origins
Spectrograph for Hubble Space Telescope, which is the most sensitive UV spectrograph yet
built, but is limited to point sources. An efficient UV IFS enables simultaneous spectroscopy of
many spatially resolved elements within a contiguous two dimensional field of view in
diagnostically important ultraviolet lines. The output is thus a data cube having one spectral and
two spatial coordinates. This is the astrophysical analog to hyperspectral imaging in Earth

The scientific benefits of such an instrument were developed during Vision Missions, Origins
Probes, and Astrophysics Strategic Mission Concept Studies between 2004 and 2009.
Implementation can be scaled for a small payload such as a sounding rocket or Explorer
mission, leading to a flight experiment within the next few years. Of particular interest would be
the application of this technology for an instrument on a version of the Advanced Technology
Aperture Space Telescope (ATLAST) which will have an 8+
m aperture. A 4
m version
would be applicable to missions such as the Telescope for Habitable Exoplanets and
Interstellar/Intergalactic Astronomy (THEIA).

We will focus on the spectral region near Lyman alpha, but the all
reflective approach is
applicable to other spectral regions when matched with wavelength appropriate gratings and
detectors. Our project is a collaboration between Ball Aerospace & Technologies Corp., the
University of Colorado, NASA Goddard Space Flight Center, and the Space Telescope Science
Institute, all of which have extensive experience with the science and instrumentation for UV


Multiple gratings and detectors provide field coverage without the use of multiple
beamsplitter optics

reduced optics in path means increased throughput to the detector.

Increased packaging flexibility.

Separated paths allow increased straylight suppression and control.

Instrument approach is scalable for different systems from small breadboard to
suborbital flight to 4 to 8 meter telescopes.


The UV is a spectral region rich in diagnostic emission and absorption lines that provide
information about temperature, metallicity, ionization and excitation processes, and
interaction of matter with radiation fields, shocks, winds and other processes. Species
such as H I, C IV, N V, O VI, Si III and S IV are prominent in the spectra of a myriad of
objects and environments. Planets exhibit UV aurorae when the solar wind interacts

their magnetic fields and atmospheres.
Figure 1

is a beautiful picture of Saturn taken
with the UV imaging mode of STIS, showing both the north and south pole auroral ovals.
Stars produce strong UV features in their photospheres, chromospheres, coronae and
winds. Diffuse gas in the interstellar and intergalactic medium is traced by UV lines. Star
formation, starbursts, jets, fountains and galaxy mergers can be studied.
Figure 2

illustrates how visible long
slit spectroscopy with STIS revealed the kinematics of matter
in the powerful gravity field of the nucleus of M848. With future UV
optical telescopes
and next
generation instrumentation, UV integral field spectroscopy will be a powerful
diagnostic technique.


Additional trades of F#, field of view, and physical size (and number) of the slicers.

Continued development of the microslicer element assembly.

Control of edges and characterization of scattered light.

Fabrication of a brassboard assembly that can be tested and perhaps flown on a
suborbital flight to raise the TRL to an acceptable value for spaceflight programs.


Integral Field Unit (IFU)

The secondary mirror acts as the relay mirror to achieve the appropriate image scale.
Depending on how slow or fast the system F# is chosen, a toroidal relay mirror may be
needed. If packaging allows a slow F#, that additional mirror can be deleted. The image
slicer is extremely small; only a handful of mm in the spatial direction and a mm or so in
the dispersion direction (depending on the number of facets and spatial FOV). To envision
how narrow the individual facets are in the dispersion direction, one could picture the thin
edge of a microscope cover slide. Those small slicer faces must be polished without chips
or artifacts that cause scatter and crosstalk. The facets are offset from each other by a
few degrees, each directing its light to a separate grating. The reflective surfaces must be
polished and coated to operate efficiently at Ly a.
Figure 5

shows one approach to
fabricating the slicer using a stack of thin rectangular plates.

Diffraction Gratings

Our UV IFS concept requires that the diffraction gratings provide both dispersion and
imaging of the spectrum. With the use of toroidal substrates we can produce excellent
imaging over a broad bandpass. Beasley et al., 2004, have investigated and
demonstrated UV holographic gratings with excellent performance over both a wide FOV
and broad spectral coverage. These designs are particularly well suited to changing focal
length and plate scale of a re
imaged beam.


For the desired spectroscopic resolution, individual spectra would cover a few hundred Å,
with a resolution element of a few Å. Since the spectral formats are small, only modest
detectors are required. Each spectrum will be flat, but the five spectra may not fall in the
same plane. Separate detectors will be employed for each spectrum.

For additional information, contact Steve Kendrick (303
1108; or

Dennis Ebbets (303
5964; at Ball Aerospace & Technologies Corp.

113 mm

800 mm

Telescope Axis
(Incoming Light)

Relay Mirror

4x Image on
(Image Slicers)

~80 mm

5 Toroidal

5 FPAs

330 mm

Figure 3
. Integral Field Spectrometer

Top and Side Views

Figure 4.
IFS Instrument Perspective View

Figure 5.
Individual Slicers Direct Sectors of the FOV to Individual
Gratings and Detectors

Spatial direction = long dimension of facets

Dispersion direction = short dimension of facets.

Each facet reflects light to a separate grating and detector.

Relay Mirror

Toroidal Grating (5x)

FPA (5x)

Integral Field Unit (5 Image Slicers)

Figure 1.
Planets, satellites and comets all interact
with the solar wind, producing conspicuous
emission in the ultraviolet. A UV IFS could record
the spectrum of every spatially resolved point in the
region of interaction.

Figure 2.
STIS observations revealed velocities of
400 km s
1 for matter orbiting the nucleus of M84,
and inferred a mass of 300 million solar masses for
the central object. A UV IFS will map the kinematics
of hot stars and ionized gas in great detail.