research papers
Rotated-inclined focusing monochromator with simultaneous tuning of asymmetry factor and radius of curvature over a wide wavelength range
aPhoton Factory, Institute of Materials Structure Science, High Energy Accelerator Research Organization, Tsukuba, Ibaraki, Japan, bMechanical Engineering Center, High Energy Accelerator Research Organization, Tsukuba, Ibaraki, Japan, and cTsukuba Advanced Research Alliance (TARA Sakabe Project), University of Tsukuba, and Foundation for Advancement of International Sciences, Tsukuba, Ibaraki, Japan
*Correspondence e-mail: nobuhisa.watanabe@kek.jp
A new single-crystal focusing monochromator for protein crystallography has been developed. In order to achieve simultaneous tuning of the beam demagnification rate and the radius of curvature for focusing over a wide wavelength range, the surface of an inclined monochromator was cylindrically bent. The monochromator is able to tune these two parameters simultaneously with a single rotation of the crystal about the azimuthal axis. A new monochromator incorporating this idea has been developed for beamline BL6B at the Photon Factory. The monochromator was designed for use in the wavelength range 0.87–1.90 Å, and beam focusing was tested at wavelengths of 1.04, 1.38 and 1.74 Å.
Keywords: focusing monochromators; protein crystallography.
1. Introduction
Many beamlines for protein crystallography use single bent asymmetric-cut silicon or germanium crystals to achieve a focused monochromatic beam (Lemonnier et al., 1978). The demagnification rate of a beam passing through an asymmetric-cut crystal monochromator depends on the angle between the beam and the crystal surface with the asymmetry factor, b, given by
where is the et al., 1978). If one uses a conventional single bent asymmetric-cut monochromator, the asymmetry factor or beam demagnification rate alters as shown in (1) with when the crystal is rotated to change wavelength. The ideal asymmetry factor, b, for a monochromator under the Guinier focusing condition should be
and is the angle between the Bragg plane and the crystal surface (Kohrawhere p and q are the distance from the light source to the monochromator and from the monochromator to a focal point, respectively. For this reason, several monochromator crystals of different asymmetric angle, , are usually necessary to cover a wide wavelength range. A rotated-inclined, or rotated Fankuchen, monochromator can cover a wide wavelength range keeping the beam demagnification because it is able to change the asymmetric angle by rotation about an azimuthal axis [see equation (8) below]. However, such a monochromator has no focusing capability. An asymmetrically cut focusing monochromator should be bent to the radius of curvature, R, given by
if the monochromator is used under the Guinier condition. Thus we have conceived a new design in which the surface of a rotated-inclined monochromator can be variably bent into a cylindrical figure.
In this paper, we describe the newly developed monochromator which allows tuning of the asymmetry factor, or beam demagnification rate, and radius of curvature of the crystal simultaneously through rotation about the azimuthal angle only.
2. Theoretical considerations
In order to achieve tunability of the two parameters, namely asymmetry angle, , and radius of curvature, R, simultaneously, we have improved a rotated-inclined monochromator to have a focusing capability.
As shown in Fig. 1, the origin of the azimuthal angle, , is defined as the direction which gives the usual asymmetric angle, , of a Fankuchen monochromator, and the offset angle, , is the angle between the slope of the inclined surface of the asymmetrically cut crystal and the axial direction of the cylinder. This offset angle is necessary in order to make possible the simultaneous tuning of the two parameters with only a -axis rotation.
The radius of curvature, R, of the rotated cylindrical surface at any can be calculated as follows. If a cylinder of radius R0 is cut by a plane at angle as in Fig. 2(a), the intersecting section is an ellipse given by
and the radius of curvature at any point is given by
The radius of curvature of the crystal surface along the beam is the radius of curvature at v = 0 of the ellipse shown in Fig. 2(a). Therefore, if there is an offset of as shown in Fig. 1, the radius of curvature is described by the function
Furthermore, the asymmetric angle at any is determined as follows. Consider a plane inclined by relative to the x-y plane as shown in Fig. 2(b). Then z is given by
Therefore the azimuthal angle, , and asymmetric angle, , are related according to
3. Design and applications
The monochromator was designed for use at the TARA beamline, BL6B, a bending-magnet beamline at the Photon Factory. Beamline BL6B is equipped with a bent plane platinum-coated silicon mirror providing vertical focusing of the beam located at 19.5 m from the source. The monochromator is situated 23.0 m from the bending magnet. The focus position of the beam is 24.38 m from the source. Therefore the optimal asymmetry factor, b, should be 16.7 for this beamline as given by equation (2). The theoretical value of at any wavelength can be calculated using equation (1) with this b value.
The parameters , and R0 were determined to optimize and R between 0.87 Å (Kr K-absorption edge) and 1.90 Å (Mn K-absorption edge), which correspond to 8.0 and 17.6° of , respectively, for Si(111). The parameters optimized for BL6B were = 19.7°, = 20.9° and R0 = 37.9 m. Plots of and R calculated using these values are shown in Fig. 3. This figure shows that it is possible to tune both asymmetry angle, , and radius of curvature, R, simultaneously with only one -axis rotation. Calculated values of , R and at the three wavelengths where the performance of the monochromator was tested are also shown in Table 1. The error of the curvature, , is expected to be less than 1% for all three wavelengths.
The cylindrical surface of the copper block, having a radius of 37.9 m, was fabricated using a high-precision diamond-cutting machine (Higashi et al., 1992). A 1 mm-thick silicon wafer was cut from a 5 inch ingot with an asymmetric angle of 19.7° from the (111) plane and finished by mechano-chemical polishing. In order to fit the crystal to the curvature of the copper block, the crystal was stuck onto the copper cylinder surface using mineral oil (ALDRICH 16,140-3) and the assembly placed into a polyethylene bag. The bag was then evacuated so that the crystal was pushed onto the copper cylinder by atmospheric pressure. Once the crystal was shaped to the copper cylinder and held by the of the mineral oil, the monochromator was used without any other fixing mechanism (see Fig. 4).
Schematic drawings of the monochromator showing the layout of the goniometers are shown in Fig. 5.
4. Results
Three wavelengths (1.07, 1.38 and 1.74 Å) were chosen to demonstrate the performance of the monochromator. The horizontal slit in the beamline was opened to accept 1.0 mrad of the X-ray beam from the bending magnet. The tuning scheme of the monochromator was as follows. First, in order to set the wavelength, the monochromator was rotated around the axis which is parallel to the (111) Bragg plane. The crystal was then rotated about the axis to tune the asymmetry angle, , and the radius of curvature, R, simultaneously as described in the previous section. The two-dimensional intensity profile of the focused beam at several angles was recorded using imaging plates (Miyahara et al., 1986), and the result for 1.38 Å is shown in Fig. 6. To reduce the beam intensity, 6 mm of aluminium plate was placed between the monochromator and the imaging plate. The horizontal intensity profile was also measured by horizontal scanning of a 0.2 mm-wide slit which was placed at the focus position. The results for all three wavelengths are shown in Fig. 7. The expected beam profile for 1.38 Å calculated using a ray-tracing program (Takeshita, 1995) is also shown in Fig. 7(d).
5. Discussion
As shown in Figs. 6 and 7, a well compressed and focused beam was obtained at all three wavelengths by rotation about the axis only. The monochromator could focus the X-ray beam to less than 0.8 mm FWHM in the horizontal direction. The usual width of the final slit or collimator in a protein crystallography experiment is about 0.1 or 0.2 mm. Therefore, a well compressed and focused beam is essential to obtain high throughput, especially at bending-magnet beamlines at second-generation storage rings such as the Photon Factory. The single-crystal monochromator described in this paper is convenient to use and makes the task of obtaining high fluxes easier compared with triangular bent crystal monochromators because it is able to cover a wide wavelength range with a single crystal. If a triangular single bent asymmetric-cut crystal monochromator is used, five or more monochromator crystals of different asymmetric angle, , would be necessary to cover the wavelength range mentioned above.
Since this test, this monochromator has been used as the standard monochromator at BL6B. The output beam is quite stable and there have been no problems during conventional collection of protein data for over half a year. It is necessary, however, to find a more sophisticated method of fixing the crystal onto the cylinder surface. As shown in Fig. 7, the observed profiles have an asymmetric aspect. This asymmetry may be caused by an error in the cylinder surface. Of those tested, mineral oil was the best material for fixing the crystal. Greasy glue, such as liquid InGa, affected the focused beam profile due to difficulties in obtaining uniform thickness. We are currently investigating other methods of fixing the crystal, such as a hot isostatic pressing bonding method or an electrostatic bonding, to reduce non-uniformity of the curvature. As shown in Table 1, this monochromator has the potential to achieve a good energy resolution over a wide wavelength range if an ideal cylinder surface can be made. The monochromator is always used near the Guinier condition.
We plan to construct a new undulator beamline for protein crystallography at the 6.0 GeV PF-AR ring on the same campus. In order to realize this type of monochromator for use with the undulator beamline there, cooling techniques to cope with the heat-load problem are necessary. The mineral oil used in this study is not applicable because of its poor heat conductivity and resistance to synchrotron radiation. Once we can establish a new crystal-fixing method, many indirect cooling techniques which have already been developed will be utilized. This type of monochromator with horizontal dispersion geometry is also useful in the construction of a branch beamline if there are spatial restrictions. Spatial problems may be especially serious on compact storage rings.
Acknowledgements
The authors would like to thank Mr Toshikazu Takatomi for his work on the copper cylindrical base. This work was supported by JSPS under the new research support program `Research for the Future', JSPS-RFTF96R14501.
References
Higashi, Y., Koike, S., Takatomi, T. & Koizumi, S. (1992). Proc. SPIE, 1720, 44–49. CrossRef Google Scholar
Kohra, K., Ando, M., Matsushita, T. & Hashizume, H. (1978). Nucl. Instrum. Methods, 152, 161–166. CrossRef CAS Web of Science Google Scholar
Lemonnier, M., Fourme, R., Rousseaux, F. & Kahn, R. (1978). Nucl. Instrum. Methods, 152, 173–177. CrossRef CAS Web of Science Google Scholar
Miyahara, J., Takahashi, K., Amemiya, Y., Kamiya, N. & Satow, Y. (1986). Nucl. Instrum. Methods, A246, 572–578. CrossRef CAS Web of Science Google Scholar
Takeshita, K. (1995). Rev. Sci. Instrum. 66, 2238–2240. CrossRef CAS Web of Science Google Scholar
© International Union of Crystallography. Prior permission is not required to reproduce short quotations, tables and figures from this article, provided the original authors and source are cited. For more information, click here.