вход по аккаунту



код для вставкиСкачать
Laboratory-size three-dimensional water-window x-ray microscope with Wolter type I
mirror optics
Shinji Ohsuka, Akira Ohba, Shinobu Onoda, Katsuhiro Nakamoto, Tomoyasu Nakano, Motosuke Miyoshi, Keita
Soda, and Takao Hamakubo
Citation: AIP Conference Proceedings 1696, 020009 (2016);
View online:
View Table of Contents:
Published by the American Institute of Physics
Articles you may be interested in
Laboratory-size three-dimensional x-ray microscope with Wolter type I mirror optics and an electron-impact water
window x-ray source
Review of Scientific Instruments 85, 093701 (2014); 10.1063/1.4894468
Three-dimensional structure of brain tissue at submicrometer resolution
AIP Conference Proceedings 1696, 020004 (2016); 10.1063/1.4937498
Layout and first results of the nanotomography endstation at the P05 beamline at PETRA III
AIP Conference Proceedings 1696, 020008 (2016); 10.1063/1.4937502
Development of high-resolution x-ray CT system using parallel beam geometry
AIP Conference Proceedings 1696, 020007 (2016); 10.1063/1.4937501
Laboratory-size three-dimensional water-window x-ray
microscope with Wolter type I mirror optics
Shinji Ohsuka∗,† , Akira Ohba∗ , Shinobu Onoda∗ , Katsuhiro Nakamoto∗ , Tomoyasu
Nakano∗,∗∗ , Motosuke Miyoshi‡ , Keita Soda‡ and Takao Hamakubo‡
Hamamatsu Photonics K.K., 5000 Hirakuchi, Hamakita-ku, Hamamatsu-City, 434-8601, Japan
The Graduate School for the Creation of New Photonics Industries, 1955-1 Kurematsu-cho, Nishi-ku,
Hamamatsu-City, 431-1202, Japan
Ray-Focus Co. Ltd., 6009 Shinpara, Hamakita-ku, Hamamatsu-City, 434-0003, Japan
Research Center for Advanced Science and Technology, The University of Tokyo, 4-6-1 Komaba, Meguro-ku,
Tokyo 153-8904, Japan
Abstract. We constructed a laboratory-size three-dimensional water-window x-ray microscope that combines wide-field
transmission x-ray microscopy with tomographic reconstruction techniques. It consists of an electron-impact x-ray source
emitting oxygen Kα x-rays, Wolter type I grazing incidence mirror optics, and a back-illuminated CCD for x-ray imaging. A
spatial resolution limit better than 1.0 line pairs per micrometer was obtained for two-dimensional transmission images, and
1-μm-scale three-dimensional fine structures were resolved.
Keywords: x-ray microscopy, tomography, water-window, Wolter mirror
PACS: 68.37.Yz
X-ray microscopy combined with tomographic reconstruction techniques has become a powerful tool for threedimensional (3D) observation of biological samples such as cells and tissues, especially with water-window (284-543
eV) x-rays. Recent advancement of x-ray microscopes at synchrotron radiation facilities has enabled observation
of biological samples under the cryogenic sample condition with resolution of around 50 nm[1], and biological
studies have been reported[2, 3]. Although the performance of x-ray microscopes at synchrotron radiation facilities
is excellent, access is limited; thus laboratory-size x-ray microscopes have been developed[4, 5] to provide highresolution 3D imaging utilizing laboratory-size water-window x-ray sources such as laser-produced[4] and gasdischarge[5] plasma x-ray sources.
In this article, we present the potential of combining Wolter type I mirror optics and a conventional electron-impact
water-window x-ray source, which is simpler and easier to use than plasma x-ray sources.
Figure 1 shows a schematic picture of our laboratory-size 3D x-ray microscope. It consists of an electron-impact x-ray
source with a Cr2 O3 target emitting oxygen Kα (525 eV) x-rays, a condenser and an objective Wolter type I grazing
incidence mirror fabricated by a glass replication method, a rotating sample stage with a three-axis translation stage,
and a back-illuminated CCD with 1024 × 1024 pixels at a physical pixel pitch of 13 μm for x-ray imaging. The
focus of the x-ray source is demagnified by a factor of 3.5 and imaged on the sample plane by the condenser mirror.
The magnification factor of the objective mirror was designed to be 100, and thus the pixel pitch of the image on the
sample plane is 0.13 μm. The distance from the focus of the x-ray source to the CCD is only 3.4 m, so that the 3D
x-ray microscope can be placed in a typical laboratory. Further details of the x-ray source and the mirror optics are
described in a previous article [6].
XRM 2014
AIP Conf. Proc. 1696, 020009-1–020009-4; doi: 10.1063/1.4937503
© 2016 AIP Publishing LLC 978-0-7354-1343-6/$30.00
Condenser mirror
Objective mirror
Vacuum chamber
Rotating sample stage
FIGURE 1. Schematic picture of 3D x-ray microscope consisting of electron-impact water-window x-ray source, Wolter type I
grazing incidence mirror optics, rotating sample stage, and back-illuminated CCD.
Two-dimensional (2D) imaging performance of our 3D x-ray microscope was examined using an x-ray resolution
chart made of 500-nm-thick tantalum (XRESO-50HC, NTT Advanced Technology Corp., Japan). Figure 2(a) shows
the central part of the radial pattern of the resolution chart observed by the 3D x-ray microscope with an exposure
time of 75 min. The x-ray source was operated at an acceleration voltage of 15 kV and current of 400 μA. At this
operating condition, the intensity of x-ray photons entering the CCD without any samples was determined to be 0.1
photons/s/pixel using photon counting mode operation of the CCD. From this image, it is obvious that the 0.5 μm
line-and-space (L&S) pattern could be resolved. The image contrasts at several spatial frequencies measured from Fig.
2(a) are plotted as black circles in Fig. 2(c). The image contrast at the spatial frequency of 0.87 lp/μm (0.58 μm L&S)
was measured as 11.4% ± 2.6%. From Figs. 2(a) and (c), the 2D spatial resolution limit of the 3D x-ray microscope
was evaluated to be better than 1.0 lp/μm.
The image contrast obtained with the objective Wolter type I mirror is deteriorated by the surface roughness and
the figure error of the mirror surface, which are around 2 nm (rms) and less than ±0.1 μm, respectively. To restore
the image contrast, we used the Richardson-Lucy (RL) method [7, 8], which is a Bayesian-based iterative scheme for
image restoration that deconvolves the point spread function (PSF) of the imaging system. The PSF was estimated by
rotating the line spread function (LSF), and the LSF was measured using a knife-edge method with a 50-μm-diameter
platinum wire. Figure 2(b) shows a restored image with 8-times iteration of the RL method. The image contrasts after
restoration are plotted in Fig. 2(c) as white circles, and it can be seen that the image contrast in the spatial frequency
range between 0.3 and 0.8 lp/μm was largely improved.
3D tomographic imaging performance was evaluated using a micro glass capillary for cell manipulation (Femtotips,
Eppendorf AG, Germany). Figure 3(a) shows an x-ray transmission image at 0◦ rotation angle. Transmission images
were acquired at 121 rotation angles in 1.5◦ increments with an exposure time of 20 min. for each image. The x-ray
source was operated at 15 kV and 280-350 μA during image acquisition. All transmission images were restored with
the RL method and negative logarithms of the transmissivity images were calculated before 3D reconstruction, which
used the maximum likelihood expectation maximization (MLEM) algorithm. Figure 3(b) shows the negative logarithm
image of the transmissivity of Fig. 3(a). A reconstructed slice image parallel to the capillary axis is shown in Fig. 3(c).
Also reconstructed slice images perpendicular to the capillary axis at positions indicated by arrows (A, B, C) in Fig.
3(c) are shown in Fig. 3(d) with profiles of the linear attenuation coefficient (LAC) along the dashed line in the slice
images. Even at position A, where the distance between the peaks of the LAC profile is 1.17 μm, wall and hollow
structure of the micro glass capillary is visible in the reconstructed slice image. As a result, the 3D x-ray microscope
has the capability of resolving 1-μm-scale 3D structures.
Since 3D tomographic reconstruction from x-ray transmission images calculates the LAC of each voxel, segmentation of the sample based on the LAC of such substances as proteins and lipids is possible [9] if the quantitativity
of the LAC is sufficient. The peak value of the LAC profile at position C in Fig. 3(d) is 0.4 μm−1 , while the LAC of
Image contrast (%)
1 μm L&S
0.5 μm L&S
Spatial frequency (lp/μm)
FIGURE 2. Radial pattern images of x-ray resolution chart (a) before and (b) after image restoration using Richardson-Lucy
method. (c) Image contrasts at several spatial frequencies measured from radial pattern images (a) and (b).
(d) A
1.17 μm
2.34 μm
3.38 μm
LAC (μm-1)
2 μm/div.
FIGURE 3. 3D tomographic imaging of micro glass capillary. (a) X-ray transmission image and (b) negative logarithm of
transmissivity image of (a) after image restoration. (c) Reconstructed slice image parallel to the capillary axis. (d) Reconstructed
slice images perpendicular to the capillary axis at positions A-C indicated by arrows in (c) with profiles of linear attenuation
coefficient along the dashed line. The distances between the peaks of the LAC profiles at positions A-C are 1.17, 2.34, and 3.38
μm, respectively. All scale bars are 5 μm.
SiO2 with a density of 2.2 g/cm3 is calculated as 0.98 μm−1 at 525 eV [10]. Even though the composition of the glass
material of the micro capillary is unknown, the calculated LAC value was rather smaller than the expected value of
around 1 μm−1 . This discrepancy is probably due to the limited resolving power of the 3D x-ray microscope, which
blurs the thickness of the wall of the capillary. It is important to improve the spatial resolution for achieving not only
good quality but also high quantitativity of reconstructed images.
We observed biomedical samples to demonstrate the potential of the 3D x-ray microscope as a useful tool for the
life science research field. Figures 4(a) and (b) show an optical microscope and an x-ray transmission image of
a glomerulus observed in a 5-μm-thick dehydrated mouse kidney slice placed on a 0.1-μm-thick silicon nitride
membrane. The glomerulus is a tuft of capillaries and a functional unit of the kidney for filtering wastes from the
blood. X-ray transmission images were obtained in rotation angle ranges of 0◦ -63◦ and 118.5◦ -180◦ in increments of
1.5◦ . Exposure time was 15 min. for each image and the x-ray source was operated at 15 kV and 250-270 μA. Figures
4(c)-(g) show the reconstructed slice images with 1-μm spacing. Complex 3D structures of the glomerulus can be
-2 μm
Reconstructed slice images
-1 μm
±0 μm
+1 μm
+2 μm
FIGURE 4. Images of dehydrated mouse kidney slice. (a) Optical microscope image. (b) X-ray transmission image. (c)-(g)
Reconstructed slice images perpendicular to the optical axis with 1-μm spacing. The 3D structure of a glomerulus can be observed.
All images are 100 μm × 100 μm.
A laboratory-size 3D x-ray microscope using Wolter type I grazing incidence mirror optics and an electron-impact
water-window x-ray source was presented. It was developed to provide high accessibility and easy use. Imaging
performances for both 2D and 3D observation were examined, and a dehydrated mouse kidney tissue sample was
observed. Although further improvements such as higher spatial resolution and ability to observe frozen hydrated
samples are necessary, these results demonstrate the potential of the 3D x-ray microscope as a useful tool for life
science research fields.
This work was carried out under collaboration between the Research Center for Advanced Science and Technology of
the University of Tokyo and Hamamatsu Photonics K.K. The authors gratefully acknowledge Prof. Y. Nakano of the
University of Tokyo and A. Hiruma and T. Hara of Hamamatsu Photonics K.K. who initiated this collaboration.
J. L. Carrascosa, F. J. Chichón, E. Pereiro, M. J. Rodríguez, J. J. Fernández, M. Esteban, S. Heim, P. Guttmann, and
G. Schneider, Journal of Structural Biology 168, 234 – 239 (2009).
2. E. M. Duke, M. Razi, A. Weston, P. Guttmann, S. Werner, K. Henzler, G. Schneider, S. A. Tooze, and L. M. Collinson,
Ultramicroscopy 143, 77 – 87 (2014).
3. E. A. Smith, G. McDermott, M. Do, K. Leung, B. Panning, M. A. Le Gros, and C. A. Larabell, Biophysical Journal 107, 1988
– 1996 (2014).
4. H. Hertz, O. von Hofsten, M. Bertilson, U. Vogt, A. Holmberg, J. Reinspach, D. Martz, M. Selin, A. Christakou,
J. Jerlström-Hultqvist, and S. Svard, Journal of Structural Biology 177, 267 – 272 (2012).
5. D. B. Carlson, J. Gelb, V. Palshin, and J. E. Evans, Microscopy and Microanalysis 19, 22–29 (2013), ISSN 1435-8115.
6. S. Ohsuka, A. Ohba, S. Onoda, K. Nakamoto, T. Nakano, M. Miyoshi, K. Soda, and T. Hamakubo, Review of Scientific
Instruments 85, 093701 (2014).
7. W. H. Richardson, J. Opt. Soc. Am. 62, 55–59 (1972).
8. L. B. Lucy, Astron. J. 79, 745 (1974).
9. G. McDermott, M. A. L. Gros, C. G. Knoechel, M. Uchida, and C. A. Larabell, Trends in Cell Biology 19, 587 – 595 (2009).
10. B. Henke, E. Gullikson, and J. Davis, Atomic Data and Nuclear Data Tables 54, 181 – 342 (1993).
Без категории
Размер файла
186 Кб
Пожаловаться на содержимое документа