Design and Fabrication of a MEMS X-ray Optic using

GEO-X : Geospace X-ray imager
Yuichiro Ezoe1, Yoshizumi Miyoshi2, Satoshi Kasahara3,
Hiroshi Hasegawa3, Tomoki Kimura3, Takaya Ohashi1, Yoshitaka Ishisaki1, Ikuyuki Mitsuishi1,2,
4
1
4
4
Ryuichi Fujimoto , Kazuhisa Mitsuda , Atsushi Noda , Kunitoshi Nishijo , the GEO-X team
1
Tokyo Metropolitan Uni.,
2
Nagoya Uni.,
3
ISAS/JAXA,
A12105
solar wind plasma close to Earth. Similar to modeled
observations made by a hypothetical X-ray detector on
IMAGE, the production rate is clearly at a maximum in
the cusp region, with a secondary maximum in the subsolar
region. In the flanks and tail region, where the solar wind
density is much lower, the production rates are significantly
lower too.
[24] Integration along each of the 101 ! 101 lines of sight
produces the image of SWCX soft X-ray emission (see
Figure 4, left side). The largest X-ray intensities are for the
cusps, followed by a secondary maximum in the subsolar
region (40% less than in the cusps). Because the lines of
sight intersect the flanks, the bowshock and the magnetopause are more diffuse, but can still be seen clearly. The
maximum subsolar X-ray intensity is twice the intensity that
would be seen from inside the magnetosheath due to the
geometrical effects.
[25] Robertson and Cravens [2003] showed images from
‘‘the outside’’ for a subsolar magnetopause distance of
9.5 RE (see right side of Figure 4). The upstream solar
wind density for that paper was 7 cm"3 and the speed was
400 km/s. The maximum intensity obtained for those
conditions was about 8 keV cm"2 s"1 sr"1, which was in
the subsolar region. On the other hand the maximum
predicted intensity for the subsolar region for the
31 March 2001 conditions is about 160 keV cm"2 s"1 sr"1,
which is a factor of 20 greater than the maximum values
Robertson and Cravens [2003] obtained for average solar
wind conditions. This clearly shows the time variability of
X-ray emissions produced by the SWCX mechanism as well
as the highly nonlinear response of the intensity as the
magnetopause moves closer to the Earth. This behavior is
similar to that actually observed in the low-energy neutral
atom data and simulations.
[26] Charge transfer between solar wind alpha particles
and neutral hydrogen produces He+ 30.4 nm emission
[Gruntman, 2001]. This process is essentially the same as
the soft X-ray process except that the He++ charge exchange
cross sections are more velocity dependent and somewhat
smaller than the heavier ion cross sections. The intensity
maps in Figures 2 and 4 can easily be converted to
Each XTU is composed of an X-ray telescope and a detector. We
are currently developing a new X-ray telescope, called MEMS Xray optics for GEO-X as well as X-ray astronomy missions. We
have constructed a Wolter type-I optic optimized for soft X-rays.
(m-sphere) via Charge eXchange (CX) between solar wind and
exospheric neutrals.
Figure 3. X-ray production plot in the GSM x-z plane.
Earth is at (0,0). Axes are in distance to the subsolar
magnetopause (D).
the subsolar region. Clearly, a suitably instrumented spacecraft should be able to detect the cusps in soft X rays.
3.2. X-Ray Observations From Outside the
Magnetosheath
[22] Robertson and Cravens [2003] created images of
SWCX X-ray emission as would be seen from an observation point 50 RE removed from Earth. The observation point
was along the y axis perpendicular to the x-z plane, where
the x axis is directed from the Earth to the Sun. Average
solar wind conditions were used. The cusps were not
included in this earlier study. We have now created a similar
image, but for the 31 March 2001 event and including the
cusps. Figure 3 shows predicted X-ray production rates in
the GSM x-z plane (y = 0). Earth is located at the center of
the image. The axes are in units of D, which is the subsolar
distance to the magnetopause (6 RE in our case). The
resolution is 101 ! 101 pixels, which is sufficiently high
for the cusps to be evident.
[23] The cusps do not extend all the way to Earth because
in our current simulations it is impossible to identify purely
Magnetic Field
Sun
cusp
sheath
Solar Wind
O6+
O7+
e-
Earth
X-ray
H
MDSG/JAXA, E-mail : [email protected]
3. Science Payload
1. GEO-X
Recent observations with Earth-orbiting X-ray astronomy satellites
mission
revealed time-variable X-ray emission from Earth s magnetosphere
ROBERTSON ET AL.: MAGNETOSHEATH X-RAY EMISSIONS
A12105
4
The X-ray intensity is
expected to be strong
at m-spheric boundaries
such as magnetosheath
and cusps.
20 µm-width curvilinear pores
Ezoe+12
A MEMS X-ray optic is made from a few
hundreds µm thick Si wafers with metal
coating. It is ultra light in weight (x >40
lighter than past X-ray telescopes).
X-ray
100 mm
<Expected performances>
Diameter : 100 mm
2013 Oct
Al Kα 1.49 keV
Expected X-rays as seen at GSE
Focal length : 250 mm
Figure 2. Process flow of MEMS X-ray optics.
Y = 60 RE (Robertson+06)
Exosphere
FOV : 4 in diameter
Table 1. Development items reported in this paper and their correspondence to each process step shown in figure 2.
2 @ 0.6 keV
On-axis
area
:
4
cm
Figure 4. X-ray intensities for the 31 March 2001 event, as observed from the GSM y axis 50 R
removed from Earth (left). Units are in D, distance to the subsolar magnetopause. The right side was
process step
1
2
3
4
5
2 deg2 @ 0.6 keV
√ √ √
modeled by Robertson and Cravens [2003] for average solar wind conditions. Note that the color scales
Grasp
(SΩ)
:
16
cm
of the two panels differ. To convert this map to 30.4 nm emission (photons/cm /s/sr), multiply the X-ray
§3. single-stage
We have begun concept study
of a new mission GEO-X, aiming at
Assembly
of optics
√ √
√
intensities by 2.5 (see text for details).
§4. coating of a heavy metal √
Weight : 5 kg (mirrors are only 4 g)
√
√
Wolter
type-I
§5. Wolter
type-I optics
imaging of Earth s m-sphere in X-rays for the ﬁrst time. Our
FWHM 4 arcmin
optic
Prototype
Macropixel devices Ogawa+14
initial study conﬁrms that such a mission is achievable in JAXA sThis method hold many advantages
over the traditional mirrors and the other
micro pore optics. The matured
E
2
5 of 8
1501 tens
2402
2701
1801µm. 2100
photolithography technology allows accurately-arranged0micro299pores598with a900width1199of a few
High aspect
ratio more than 10 is possible. The stiffness of silicon and nickel micro structures enables a large open area ratio
more than 10 %. Consequently,
the mass
area ratio shown
in figure
1 can be extremely small, ∼1 kg for 1000
独toPNsensor,
MPE
が開発
Demonstrator
cm2 . Also, a large, more than
4 inch in500diameter,
single-piece optic can be fabricated by use of a commercially
pixel
x 500 µm²
Calibration
of the
DEPFET
Detectors
for the
MIXSdegradation
Instrument on
available large wafer. A short
focal length
of <1 m
is possible
without
of BepiColombo
angular resolution 9
format
64 x 64 optic
pixels
3.2 x 3.2 cm²
due to the negligible conical approximation.
Assuming that the side walls of the micro pores are flat and the
mirror arrangement is perfect,
a theoretical
on the angular resolution arises from X-ray diffraction within a
frametime
450 seclimit
(ASTEROID)
narrow micro pore. If the pore
width d-80is…20
and the X-ray wavelength λ is 1.24 nm (1 keV), the theoretical
temperature
-90 µm
°C
scalable
results
limit is represented as ∼ λ/d representative
∼13 arcsec.
Our
final goal is this X-ray diffraction limited micro pore X-ray optics.
small scientiﬁc satellite program.
DepFET sensors developed by MPE are low noise, high frame
rate, and extremely radiation hard, ideal for astronomy and
planetary missions. We plan to use this detector for GEO-X.
SDS-4 Spacecraft Overview
50kg class
zero momentum 3-axis stabilized
Sun oriented
SSO altitude approx. 677km
Piggyback payload of H-IIA LV ( FY2011 )
Weight
Size
Power
Orbit : two candidates
(a) HEO, (b) Earth-Moon L1
approx. 48kg
approx. 500 500 450 mm
approx. 120 W
Attitude Control
Three-axis Sun-oriented
Communication
S-band
1 Mbps (mission downlink)
16 kbps (downlink)
4 kbps (uplink)
Orbit
SSO, altitude approx. 677km
②
2
9
1
Moon
Lunar swingby
Earth
We started our development of this novel method with miniature optics (7.5×7.5 mm2 ) made of silicon and
500 µm pixel
<Expected performances>
32 mm
Pixel size：300 μm square
Launch from USC
Array：64 x 64
(Uchinoura Space Center)
.
.
.
.
.
.
.
.
.
Size：1.92 cm x 1.92 cm
(a)
(c)
17
Johannes Treis MPI Halbleiterlabor
Perigee
1 770[m/sec]
200km×378,370km
Frame time：1 ms
Apogee
Temperature ：< -40 deg C
Launch : JAXA s Epsilon rocket
E reso.：<100 eV @ 0.6 keV
239,000km×378,370km
Spacecraft
: 330 kg (wet), 235 kg (dry)
Perigee
2
3000
134[m/sec]
239,000[
O Kα
0.5 keV
Global&distribu7on&of&So:;Xray
]
Science payload : 40 kg, 40 W
4
Team GEO-X
futureof observations
Estimation
X-ray emission
Attitude : 3-axis stabilized (accuracy 1 arcmin/min) Several targets for
Data rate : 30 kbps (mainly science data)
Power & weight : 10 W & 3.5 kg
Majewski+13
(incl. readout electronics)
Team GEO-X
Global&distribu7on&of&So:;Xray
X-ray
power
density
(b)
Estimation
of X-ray emission
P = α nsw g n(d)H (eVcm−3s −1 )
1: global imaging of magnetosphere
r2
1 for all three detector modules
Fig.
Spectroscopic
Performance.
X-ray 2
power
density
P = α nsw (a)
g nHSpectra
(eVcm−3s −1 )−of
2 single
−1
−1events
B
(
eVcm
s
str
)
=
Pxray dr visible. The energy
r2
recorded at a beam energy of 3314 −eV.
Peak
and
escape
peak
are
clearly
1
4π r1
2 −1
−1
B(eVcm
s str
= single
Pxray dr events and
∫
resolution is 103 eV ≤ FWHM
≤ 104
eV) for
107 eV ≤ FWHM ≤ 108 eV
4π r1
for all valid events. (b) Spectra of single events for all three detector modules recorded at a
beam energy of 500 eV. The energy resolution is 64 eV ≤ FWHM ≤ 67 eV for single events
high-charge
state
andabundance
66 eV of≤
FWHM
≤population
69state
eV population
for all valid events. (c) Energy resolution FWHM (single
abundance
of
high-charge
Assumption: O7+ is 0.1 % of H+
events) vs. photon energy for all three Assumption:
modules over
the
full%energy
O7+
is 0.1
of H+ range. As a reference
thesolar
theoretical
as a dashed line. For completeness, also the result from
nsw
wind densityFano limit is shown
the reference measurement with an 55 Fen source
From which was recorded during the spectroscopic
solar
wind
density
sw
MHD model
2
2
measurements
before the transfer
BESSY
is indicated. (d) Energy resolution FWHM for
total relative speed
< g >= uto
sw + vthermal
single events and all valid events vs. photon energy for module
From DPA F10.
A ∫typical X-ray ﬂux in large CME and
geomagnetic storm events (Dst < -100
nT) is 30 and 10 ph/cm2/s/str @
0.6 keV for cusp and sheath. With one
Our science goal is visualization of m-spheric boundaries (cusp,
MHD model
total relative speed
XTU,
this promises S/N = 10 in 1 ks
< g >= u + v
sheath, LLBL etc..) in soft X-rays (< 1.5 keV). Requirements to
the theoretical Fano limit, which is shown in the graph
At an
andas a dashed
3 ksline.
exposure
times, satisfying
energy
of 1820
eV, just
below the silicon
absorption
From edge (1839 eV), a dip in
geocoronal
hydrogen
density
! 10 ( Re ) "
X
(GSM)
= 25 # energy
meet this science goal are deﬁned as follows.
$
geocoronal
model
the data is visible. As photons ofn this
have
a significantly
larger time
penthe
required
cadence (<1 hr).
% r &
spherically symmetric
etration Miyoshi+13
depth
to photons at an energy
of 1860 eV, just above the
- Observations for understanding
solar compared
wind – magnetosphere
interactions.
Si-K absorption edge, signal charge produced by these
photons is2020
significantly
In early
s, we expect that such
- Field of view:
20 Re x 20 Re
less affected by effects of charge recombination in the vicinity of the dead layer
Requirements (at 60 RE)
Comment
- View points:at the entrance window region, hence the FWHM isevents
narrower. For
the occur
module
will
5-10 per year.
Z (GSM)
2. Science goals & requirements
geocoronal hydrogen density
! 10 ( Re ) "
nH = 25 #
$
% r &
2
sw
3
2
thermal
From
geocoronal model
spherically symmetric
3
H
Item
fromDPA
side:F09
visualization
themeasurements
cusp
additional flatof
field
were done to study the properties
fromclose
topto
: the
visualization
of KHI
thetheflank
Al-K edge (1560
eV)atand
same effect - but to a much lower ex-
・C, N, O, Fe, Ne, Mg emission lines (< 1.5 keV)
Energy band
0.3 - 2 keV
・Estimate NXB by impacts of ions using X-ray
photons above 1 keV
Spatial reso.
Energy reso.
Time cadence
FOV
<0.1 RE (<6 arcmin)
•
tend -platform
was shown.
is caused
by thedistributions.
thin Al layer on the entrance window
- Lunar orbit should be a good
toThis
observe
global
Resolve ﬁne structures of cusp and sheath
<100 eV @ OVII Kα 0.6 keV
•
<1 hr (> 20 cm2 deg2 @ OVII
・Detect changes of m-sphere in size and shape
Kα 0.6 keV)
> 5 RE x 5 RE (5 x 5 )
4. Future prospects
side of the detector chip, which serves as an intrinsic light filter. In addition
to the data points derived from the calibration measurements at BESSY-II,
the energy resolution measured with an 55 Fe source (Mn-Kα: 5.9 keV) sev-
We plan to propose GEO-X for a future call of JAXA s small
science satellites. The earliest opportunity will be early 2020 s,
corresponding to the next solar maximum phase. We envision that
the XTU will be used in various future exploration missions.
Resolve C, N, O, Fe, Ne Mg emission lines
・Expected X-ray ﬂux for a large magnetic storm
event 0.1 cps @ OVII Kα
mid 2010 s
early 2020 s
Figure
1 – Science
objectives
of the
JUXTA
・Grasp
cusp and
sheath at
same(see
timeEzoe et al., 2013, Adv. Space Res., 51, 1605 and references therein).
late 2020 s
To this end, we plan to have a
Technology
development
2x2 array of compact X-ray
telescope units (XTUs).
Spacecraft Overview
Prototype Macropixel SDS-4
devices
Demonstrator
<Resources per XTU>
Mass : 10 kg
Power : 10 W
Size : 30 cm cubic
FOV : 4 x 4
pixel
500 x 500 µm²
format
64 x 64 pixels
3.2 x 3.2 cm²
frametime
450 sec (ASTEROID)
temperature
-80 … -90 °C
representative scalable results
Jupiter, Mars, Moon
Earth s m-sphere
JMO
GEO-X
50kg class
zero momentum 3-axis stabilized
Sun oriented
SSO altitude approx. 677km
Piggyback payload of H-IIA LV ( FY2011 )
Weight
Size
Power
approx. 48kg
approx. 500 500 450 mm
approx. 120 W
Attitude Control
Three-axis Sun-oriented
Communication
S-band
1 Mbps (mission downlink)
16 kbps (downlink)
4 kbps (uplink)
Orbit
SSO, altitude approx. 677km
9
Johannes Treis MPI Halbleiterlabor
>
17
Solar Sail

Download Report