| BZs >gnuplot.ps

Acta Polytechnica Vol. 51 No. 2/2011

Czech Participation in International X-Ray Observatory (IXO)

R. Hudec, L. Ṕına, V. Marš́ıková, A. Inneman, M. Skulinová, M. Mı́ka

Abstract

Here we describe the recent status of Czech participation in the IXO (International X-ray Observatory) space mission,
with emphasis on the development of new technologies and test samples of X-ray mirrors with precise surfaces, based
on new materials, and their applications in space. In addition, alternative X-ray optical arrangements are investigated,
such as Kirkpatrick-Baez systems.

Keywords: X-ray satellites, X-ray telescopes, X-ray optics.

1 Introduction
The design and development of X-ray optics has a
long tradition in the Czech Republic (e.g. Hudec et
al., 1991, 1999, 2000, 2001, Inneman et al. 1999,
2000). A range of various related technologies have
been exploited and investigated, including technolo-
gies for future large, light-weight X-ray telescopes.

Future large space X-ray telescopes (such as
IXO considered by ESA or IXO/Constellation X
by NASA) require precise, light-weight X-ray optics
based on numerous thin reflecting shells. Novel ap-
proaches and advanced technologies need to be devel-
oped and exploited. In this paper, we refer to Czech
efforts in connection with IXO (now Athena), focus-
ing on the results of test X-ray mirror shells produced
by glass thermal forming (GTF) and by shaping Si
wafers. Both glass foils and Si wafers are commer-
cially available, have excellent surface microrough-
ness of a few 0.1 nm, and low weight (the volume
density is 2.5 g · cm−3 for glass and 2.3 g · cm−3
for Si). Technologies need to be exploited for shap-
ing these substrates to achieve the required precise
X-ray optics geometries without degrading the fine
surface microroughness.

Although glass, and more recently, silicon wafers
have been considered the most promising materi-
als for future advanced large aperture X-ray tele-
scopes, other alternative materials are also worth
further study, such as amorphous metals and glassy
carbon (Marsch et al., 1997). In order to achieve
sub-arsec angular resolutions, the principles of ac-
tive optics need to be adopted. The International
X-ray Observatory (IXO) is a new X-ray telescope
with joint participation of NASA, the European
Space Agency (ESA) and the Japan Aerospace Ex-
ploration Agency (JAXA). This project supersedes
both NASA’s Constellation-X and ESA’s XEUS mis-
sion concepts. In mid-2008, officials from ESA,
NASA and JAXA headquarters agreed to conduct
a joint study of IXO with a single merged set of
top-level science goals. This agreement established

the key science measurement requirements (White et
al., 2009). The spacecraft configuration for the IXO
study is a mission featuring a single large X-ray mir-
ror, an extendible optical bench with a focal length
of ∼ 20 m and a suite of five focal plane instruments.
The X-ray instruments under study for the IXO con-
cept include: a wide field imaging detector, a high-
spectral-resolution imaging spectrometer (calorime-
ter), a hard X-ray imaging detector, a grating spec-
trometer, a high timing resolution spectrometer and
a polarimeter. The IXO mission concept was sub-
mitted to the U.S. Decadal Survey committee and to
ESA’s Cosmic Vision process.
Note added in proofs: As a consequence of

US Decadal Survey, on European side (ESA) the
IXO will be replaced by new project Athena. As
the Athena will also use the imaging X-ray Optics
similarly to IXO, the developments described in this
paper refer now to Athena.

2 Czech involvement in
IXO-related studies

At the moment, the Czech participation in IXO con-
centrates on: (1) participating in defining scientific
goals, justification and project preparation, (2) par-
ticipating in the design and development of mirror
technologies. The first author of this paper was del-
egated as a member of the IXO Telescope Working
Group. In the mirror development, we focus on sup-
porting ESA ESTEC micropore silicon technology
design and also on designing and developing alter-
native background technologies discussed in greater
detail below.

3 The glass foil alternative for
IXO

Glass science and technology has a long tradition in
the Czech Republic. At the same time, glass technol-

Acta Polytechnica Vol. 51 No. 2/2011

ogy is one of most promising technologies for produc-
ing mirrors for IXO, as the volume density of glass
is nearly four times less than the volume density of
electroformed nickel layers. Glass foils can be used
as flats or may be shaped or thermally slumped to
achieve the required geometry. Thermal forming of
glass is not a new technology, and it has been used in
various sectors of the glass industry and in glass art,
as well as in the production of Cherenkov mirrors.
However, the application of this technology in X-ray
optics is related with the need to improve accuracy
significantly and minimize errors. As the first step,
small (various sizes typically less than 100×100 mm)
glass samples of various types provided by various
manufacturers were used and thermally shaped. The
geometry was either flat or curved (cylindrical or
parabolic). The project continued with larger sam-
ples (up to 300 × 300 mm) and further profiles. Re-
cent efforts have focused on optimizing the relevant
parameters of both glass material and substrates, as
well as the parameters of the slumping process.

Various approaches have been investigated (Fi-
gure 1). We note that these are not quite identical
with efforts by another teams (e.g. Zhang et al., 2010,
Ghigo et al., 2010). The glass samples were thermally

formed at Rigaku, Prague, and also at the Institute
of Chemical Technology in Prague. For large samples
(300×300 mm), facilities at the Optical Development
Workshop in Turnov were used. The strategy is to
develop a technology suitable for inexpensive mass
production of thin X-ray optics shells, i.e., to avoid
expensive mandrels and techniques that are not suit-
able for mass production or that are too expensive.
Numerous glass samples have been shaped and tested
in order to find out the optimal parameters. The
shapes and profiles of both mandrels, as well as the
resulting glass replicas, have been carefully measured
using metrological devices. The results show that
the quality of the thermal glass replica can be signif-
icantly improved by optimizing the material and im-
proving the design of the mandrel, by modifying the
thermal forming process, as well as by optimizing the
temperature (Figure 2). After the modifications and
improvements, some of them significant, we obtained
the resulting deviation of the thermally formed glass
foil from the ideal designed profile less than 1 μm
(peak to valley value) in the best case. This value
is however strongly dependent on the exact temper-
ature, so we believe that further improvements are
still possible.

Fig. 1: The three investigated glass thermal arrangements: convex and concave mandrels and a double mandrel

Fig. 2: An example of the optimization studies performed for glass thermal forming: optimization map for waviness
(wavinessWaas function of forming process parameters i.e. duration of the thermal forming process (τ) and temperature
(T), Tg means glass transformation temperature)

Acta Polytechnica Vol. 51 No. 2/2011

The fine original microroughness (typically better
than 1 nm) of the original float glass foil was found
not to be degraded by the thermal forming process.
We note that our approach in thermal glass forming is
different from the approaches used by other authors.
Recent efforts have been devoted to optimizing the
whole process, using and comparing different forming
strategies etc., as the final goal is to further improve
the forming accuracy to less than 0.1 μm values. For
the near future, we plan to continue these efforts
together with investigations of computer-controlled
forming of glass foils (according to the principles of
active optics).

4 The silicon wafer
alternative

Silicon is a relatively light material and already dur-
ing the manufacturing process it is lapped and pol-
ished (either on one side or on both sides) to very fine
smoothness (better than a few 0.1 nm) and thickness
homogeneity (of the order of 1 μm). Another obvi-
ous alternative, recently considered as one of most
promising for high-precision X-ray optics for IXO, is
the use of X-ray optics based on commercially avail-
able silicon wafers manufactured mainly for the pur-
poses of the semiconductor industry.

The main advantages of the application of Si
wafers in space X-ray optics are (i) the volume den-
sity, which is more than 4 times lower than the elec-
troformed nickel used in the past for galvanoplas-
tic replication of multiply nested X-ray mirrors, and
slightly less than the alternative approach of glass
foils, (ii) very high thickness homogeneity, typically
less than 1 μm over 100 mm, and (iii) very small
surface microroughness either on one side or on both
sides (typically of the order of a few 0.1 nm or even
less, e.g. Figure 3). Silicon wafers were expected to
be used in the ESA XEUS project and are still un-
der consideration for the IXO (now Athena) project.
The recent baseline optics for the IXO X-ray tele-
scope design is based on X-Ray High Precision Pore
Optics (X-HPO), a technology currently under de-
velopment with ESA funding (RD-Opt, RD-HPO),
in view of achieving large effective areas with low
mass, reduced telescope length, high stiffness, and a
monolithic structure, favoured for handling the ther-
mal environment and for simplifying the alignment
process (Bavdaz et al. 2010). In addition, due to
the higher packing density and the associated shorter
mirrors required, the conical approximation to the
Wolter-I geometry becomes possible. The X-HPO
optics is based on ribbed Si wafers stacked together.
The Si wafers to achieve the conical approximation
are formed by stacking large number of plates to-

gether using a mandrel. The typical size of the Si
wafers is 10 × 10 cm.

There are also alternative X-ray optics arrange-
ments with the use of Si wafers. In this paper, we
refer to the development of an alternative design of
innovative precise X-ray optics based on Si wafers.
Our approach is based on two steps, namely (i) devel-
oping dedicated Si wafers with properties optimized
for use in space X-ray telescopes and (ii) precisely
shaping the wafers to the optical surfaces (Figure 4).
Stacking to achieve nested arrays is performed af-
ter the wafers have been shaped. In this approach,
Multi Foil Optics (MFO) is thus created from shaped
Si wafers (Figure 5). For more details on MFO, see
Hudec et al. (2005).

Fig. 3: AFM mesurement results for Si wafers

Fig. 4: Taylor-Hobson profilometric measurement of a
bent Si wafer

Fig. 5: Multi Foil Optics (MFO) in the Kirkpatrick-Baez
(K-B) arrangement

Acta Polytechnica Vol. 51 No. 2/2011

This alternative approach does not require the Si
wafers to have a ribbed surface, so problems with
transferring any deviation, stress, and/or inaccuracy
from one wafer to the neighbouring plates or even to
the whole stacked assembly will be avoided. However,
suitable technologies for precise stacking of optically
formed wafers to a multiple array have to be devel-
oped.

The Si wafers available on the market are de-
signed for use mainly in the semiconductor industry.
It is obvious that the requirements of this industry
are not the same as the requirements of precise space
X-ray optics. Si wafers are a monocrystal (single
crystal) with some specifics, and this must also be
taken into account. Moreover, Si wafers are fragile,
and it is very difficult to bend and/or shape them pre-
cisely (for thicknesses required for X-ray telescopes,
i.e. around 0.3–1.0 mm. An exception is thin Si
wafers below 0.1 mm in thickness. However, these
can be hardly used in this type of X-ray optics be-
cause of diffraction limits. Also, while their thickness
homogeneity is mostly perfect, the same is not true
for commercially available wafers for their flatness
(note that we refer here to the deviation of the up-
per surface of a free-standing Si wafer from an ideal
plane, while in the semiconductor community flatness
is usually represented by a set of parameters).

In order the achieve the very high accuracy re-
quired by future large space X-ray telescopes like
ESA/NASA/JAXA IXO, now Athena by ESA, the
parameters of the Si wafers need to be optimized
(for application in X-ray optics) at the production
stage. For this purpose we have established and de-
veloped a multidisciplinary working group including
specialists from the development department of the Si
wafer industry with the goal to design and manufac-
ture Si wafers with improved parameters (mostly flat-
ness) optimized for application in X-ray telescopes.
It should be noted that the manufacture of silicon
wafers is a complicated process with numerous tech-
nological steps and with many free parameters that
can be modified and optimized to achieve optimal
performance. This can also be useful for further im-
proving the quality of X-HPO optics. As we are deal-
ing with high-quality X-ray imaging, the smoothness
of the reflecting surface is important. The standard
microroughness of commercially available Si wafers
(we have used the products of ON Semiconductor,
Czech Republic) is of the order of 0.1 nm, as con-
firmed by several independent measurements by var-
ious techniques including the Atomic Force Micro-
scope (AFM). This is related to the method of chem-
ical polishing used in the manufacture of Si wafers.
The microroughness of Si wafers exceeds the micro-
roughness of glass foils and most other alternative
mirror materials and substrates. The flatness (in the
sense of the deviation of the upper surface of a free-

standing Si wafer from a plane) of commercially avail-
able Si wafers was however found not to be optimal
for use in high-quality (order of arcsec angular reso-
lutions) X-ray optics. Most Si wafers show deviations
from the plane of the order of a few tens of microns.
After modifying the technological process during Si
wafer manufacture, we were able to reduce this value
to just a few microns. Also, the thickness homogene-
ity was improved. In collaboration with the man-
ufacturer, further steps are planned to improve the
flatness (deviation from an ideal plane) and the thick-
ness homogeneity of Si wafers. These and planned
improvements introduced at the Si wafer manufac-
ture stage can also be applied for other designs of Si
wafer optics including X-HPO, and can play a crucial
role in the IXO project.

The X-ray optics design for IXO (now Athena) is
based on the Wolter 1 arrangement, and hence re-
quires curved surfaces. However, due to the material
properties of monocrystalline Si, Si wafers (except
very thin ones) are extremely difficult to shape. It is
obvious that we have to overcome this problem in or-
der to achieve the fine accuracy and stability required
for future large X-ray telescopes. The final goal is to
provide optically shaped Si wafers with no or little
internal stress. Three different alternative technolo-
gies for shaping Si wafers have been designed and
tested to achieve precise optical surfaces. The sam-
ples shaped and tested were typically 100 to 150 mm
large, typically 0.6 to 1.3 mm thick, and were bent
to either cylindrical or parabolic test surfaces.

The development described here is based on a sci-
entific approach, and hence the large number of sam-
ples formed with different parameters must be pre-
cisely measured and investigated in detail. Especially
precise metrology and measurements play a crucial
role in this type of experiment. The samples of bent
wafers with the investigated technologies have been
measured, including Taylor-Hobson mechanical and
STILL optical profilometry, as well as optical inter-
ferometry (ZYGO) and AFM (Atomic Force Miscro-
scope) analyses (Figure 3). It has been confirmed
that all these three technologies do not degrade the
intrinsic fine microroughness of the wafer. While
the two physical/chemical technologies exploited give
peak-to-valley (PV) deviations (of the real surface of
the sample compared with the ideal optical surface)
of less than 1 to 2 μm over the 150 mm sample length,
as preliminary values, the deviations of the first ther-
mally bent sample are larger, of the order of 10 μm.
Taking into account that the applied temperatures,
as well as other parameters, were not optimized for
this first sample, we anticipate that the PV (peak
to valley) value can be further reduced down to the
order of 1 μm and perhaps even below. Fine adjust-
ments of the parameters can also further improve the
accuracy of the results for the other two techniques.

Acta Polytechnica Vol. 51 No. 2/2011

5 K-B alternative for IXO
Although the Wolter systems are generally well
known, Hans Wolter was not the first to propose X-
ray imaging systems based on reflection of X-rays.
In fact, the first grazing incidence system to form a
real image was proposed by Kirkpatrick and Baez in
1948. This system consists of a set of two orthogonal
parabolas of translation. The first reflection focuses
to a line, which is focused by the second surface to a
point. This was necessary to avoid the extreme astig-
matism suffered from a single mirror, but it still was
not free of geometric aberrations. Nevertheless, the
system is attractive because it is easy to construct the
reflecting surfaces. These surfaces can be produced
as flat plates and then mechanically bent to the re-
quired curvature. In order to increase the aperture
a number of mirrors can be nested together, but it
should be noted that this nesting introduces addi-
tional aberrations. This configuration is used mostly
in experiments not requiring a large collecting area
(solar, laboratory). Recently, however, large modules
of KB mirrors have also been suggested for stellar X-
ray experiments.

As mentioned above, Si wafers are difficult to
shape, especially to small radii. To overcome this
difficulty, another X-ray optics arrangement can be
considered, namely the Kirkpatrick-Baez (KB) sys-
tem. Then the curvature radii are much larger, of the
order of a few km, while the imaging performance is

similar. For the same effective area, however, the fo-
cal length of the KB system is about twice as large as
the focal length of the Wolter system. Nevertheless,
KB systems represent a promising alternative to the
classical Wolter systems in future large space X-ray
telescopes.

A very important factor is the ease (and hence
the reduced cost) of constructing highly segmented
modules based on multiply nested thin reflecting sub-
strates in comparison with the Wolter design. While
e.g. the Wolter design for IXO requires the substrates
to be precisely formed with curvatures as small as
0.25 m, the alternative KB arrangement uses almost
flat or only slightly bent sheets. Hence the feasibil-
ity of constructing a KB module with the required
5 arcsec FWHM at an affordable cost is higher than
for the Wolter arrangement.

Advanced KB telescopes are based on the Multi
Foil Optics (MFO) approach (X-ray grazing incidence
imaging optics based on numerous thin reflecting sub-
strates/foils). The distinction between MFO and
other optics using packed or nested mirrors is that
MFO is based on numerous very thin (typically less
than 0.1 mm) substrates. The MFO KB test modules
were recently designed and constructed at Rigaku In-
novative Technologies Europe (RITE) in Prague, and
2 modules were tested in full aperture X-ray tests in
the test facility of the University of Boulder, with pre-
liminary results of FWHM 26 arcsec for a full stack
of 24 standard Si plates at 5 keV (Figures 6, 7).

Fig. 6: Left: Test K-B modules assembled in Rigaku RITE in Prague, Right: K-B module during full aperture X-ray
tests at Boulder University

Fig. 7: The measurement results of the K-B test module with 24 Si wafers, full aperture tests at 5 keV at University of
Colorado at Boulder. The estimated FWHM is 26 arcsec

Acta Polytechnica Vol. 51 No. 2/2011

6 Conclusion
Suitable technologies for future large X-ray telescopes
require extensive research work. Two promising tech-
nologies suitable for future large-aperture and fine
resolution X-ray telescopes, such as IXO, were ex-
ploited and investigated in detail, namely Glass Ther-
mal Forming and Si wafer bending. In both cases,
promising results have been achieved, with peak-to-
valley deviations of the final profiles from the ideal
profiles being of the order of 1 μm in the best cases,
with space for further essential improvements and
optimization. In the Czech Republic, an interdisci-
plinary team with 10 members is cooperating closely
with experienced specialists, including researchers
from a large company producing Si wafers. Si wafers
have been successfully bent to the desired geometry
by three different techniques. In the best cases, the
accuracy achieved for a 150 mm Si wafer is 1–2 μm for
deviation from the ideal optical surface. Experiments
are continuing in an attempt to further improve the
forming accuracy.

Acknowledgement

We acknowledge the support provided by the Grant
Agency of the Academy of Sciences of the Czech Re-
public, grant IAAX 01220701, by the Ministry of Ed-
ucation and Youth of the Czech Republic, projects
ME918, ME09028 and ME09004. The investiga-
tions related to the ESA IXO project were supported
by ESA PECS Project No. 98038. M.S. acknowl-
edges support from a junior grant from the Grant
Agency of the Czech Republic, grant 202/07/P510.
We also acknowledge collaboration with Drs. J. Sik
and M. Lorenc from ON Semiconductor Czech Re-
public, and with the team of Prof. Webster Cash from
University of Colorado at Boulder for X-ray tests of
KB modules in their X-ray facility.

References

[1] Hudec, R., Valníček, B., Červencl, J., et al.:
SPIE, 1991, 1343, 162.

[2] Hudec, R., Pína, L., Inneman, A.: SPIE, 1999,
3766, 62.

[3] Hudec, R., Pína, L., Inneman, A.: SPIE, 2000,
4012, 422.

[4] Hudec, R., Inneman, A., Pína, L.: Lobster-Eye:
Novel X-ray Telescopes for the 21st Century,
New Century of X-ray Astronomy, ASP Conf.
Proc., 2001, 251, 542.

[5] Hudec, R., Pína, L., Inneman, A., et al.: SPIE,
2005, 5900, 276.

[6] Inneman, A., Hudec, R., Pína, L., Goren-
stein, P.: SPIE, 1999, 3766, 72.

[7] Inneman, A., Hudec, R., Pína, L.: SPIE, 2000,
4138, 94.

[8] Kirkpatrick, P., Baez, A. V.: J. Opt. Soc. Am.
38, 766 (1948).

[9] Marsch, H., et al.: Introduction to Carbon Tech-
nologies, University of Alicante, 1997.

[10] White, N. E., Hornschemeier, A. E.: Bulletin
of the American Astronomical Society, 2009,
Vol. 41, p. 388.

[11] White, N. E., Parmar, A., Kunieda, H., Interna-
tional X-ray Observatory Team, 2009, Bulletin
of the American Astronomical Society, Vol. 41,
p. 357.

[12] http://ixo.gsfc.nasa.gov

[13] Bavdaz, M., et al.: Proceedings of the SPIE,
2010, Vol. 7732, pp. 77321E-77321E-9.

[14] Zhang, W. W., et al.: Proceedings of the SPIE,
2010, Vol. 7732, pp. 77321G-77321G-8.

[15] Ghigo, M., et al.: Proceedings of the SPIE, 2010,
Vol. 7732, pp. 77320C-77320C-12.

René Hudec
E-mail: r.hudec@asu.cas.cz
Astronomical Institute
Academy of Sciences of the Czech Republic
CZ-25165 Ondřejov, Czech Republic
Czech Technical University in Prague
Faculty of Electrical Engineering
Technická 2, CZ-166 27 Prague, Czech Republic

Ladislav Ṕına
Czech Technical University in Prague
Faculty of Nuclear Engineering
Břehová 78/7, CZ-110 00 Prague, Czech Republic

Veronika Marš́ıková
Adolf Inneman
Rigaku Innovative Technologies Europe, s. r. o.
Novodvorská 994, CZ-142 21 Prague 4,
Czech Republic

Michaela Skulinová
Astronomical Institute
Academy of Sciences of the Czech Republic
CZ-251 65 Ondřejov, Czech Republic

Martin Mı́ka
Institute of Chemical Technology
Technická 5, CZ-166 28 Prague, Czech Republic