short communications\(\def\hfill{\hskip 5em}\def\hfil{\hskip 3em}\def\eqno#1{\hfil {#1}}\)

Journal logoJOURNAL OF
SYNCHROTRON
RADIATION
ISSN: 1600-5775

Design of a high-flux and high-resolution VUV bending-magnet beamline

aSynchrotron Radiation Research Center, Hsinchu 300, Taiwan
*Correspondence e-mail: cfs@bl11b.srrc.gov.tw

(Received 4 August 1997; accepted 11 November 1997)

A high-flux and high-resolution VUV beamline (4–40 eV) has been designed and is under construction at SRRC. This beamline, which collects 50 mrad of horizontal radiation, uses a 6 m cylindrical-grating monochromator with an incident angle of 70° instead of the conventional normal-incidence-monochromator (NIM) design. Special features, such as movable entrance slit, bendable vertical focusing mirror and movable curved exit slit, are employed to enhance greatly the beamline performance. With both slit openings set at 10 µm, the energy-resolving power can reach as high as 70000. Photon fluxes of 1 × 1013 and 1 × 1010 photons s−1 are calculated for energy-resolving powers of 1000 and 40000, respectively. The best image size at the sample position is smaller than 0.45 × 0.2 mm.

1. Introduction

There is one operational low-energy (6–40 eV) beamline, the 1 m Seya-Namioka monochromator (1 m-SNM), at SRRC (Tseng et al., 1995[Tseng, P.-C., Hsieh, T.-F., Song, Y.-F., Lee, K.-D., Chung, S.-C., Chen, C.-I., Lin, H.-F., Dann, T.-E., Huang, L.-R., Chen, C.-C., Chung, J.-M., Tsang, K.-L. & Chang, C.-N. (1995). Rev. Sci. Instrum. 66(2), 1815-1817.]). However, the total flux of this beamline is less than 1 × 1012 photons s−1, even when both the entrance and exit slits are fully opened. To facilitate experiments of extremely dilute systems or of very low signal-to-noise ratio, it is necessary to construct a high-flux low-energy beamline that can deliver a photon flux greater than 1 × 1013 photons s−1. In addition, a beamline capable of delivering low-energy photons with ultrahigh-energy resolution is also needed for certain types of experiments. In order to meet both requirements, we have decided to construct a high-performance low-energy beamline with the capabilities of providing photons with very high photon flux and ultrahigh-energy resolution. The main design goals of this beamline are (i) it should cover photon energies from 4 to 40 eV, (ii) it should be able to deliver a photon flux greater than 1 × 1013 photons s−1 and to provide an ultrahigh-energy resolution capability over the entire energy range, (iii) the image size at the sample position should be less than 0.5 × 0.5 mm, (iv) adequate floor space should be available for various types of experiment stations, and (v) two separate experiment stations should be operated at the same time to utilize fully the beamtime.

The normal-incidence-monochromator (NIM) design is generally considered for high-flux beamlines in low-energy regions due to its good imaging properties, high-energy resolution and effective rejection of high-order scattered light (Samson, 1967[Samson, J. A. (1967). Techniques of VUV Spectroscopy, pp. 43-84. Lincoln, Nebraska: Pied.]; Wicke et al., 1983[Wicke, H., Bohmer, W. & Schwentner, N. (1983). Nucl. Instrum. Methods, 204, 533-542.]; Eyers et al., 1983[Eyers, A., Heckenkamp, Ch., Schafers, F., Schonhense, G. & Heinzmann, U. (1983). Nucl. Instrum. Methods, 208, 303-305.]; Lama et al., 1994[Lama, F., Derossi, A., Piacentini, M. & Zema, N. (1994). Nucl. Instrum. Methods Phys. Res. A, 339, 610-616.]; Suits et al., 1995[Suits, A. G., Heimann, P., Yang, X., Evans, M., Hsu, C.-W., Lu, K. & Lee, Y. T. (1995). Rev. Sci. Instrum. 66(10), 4841-4844.]; Heimann et al., 1997[Heimann, P. A., Koike, M., Hsu, C. W., Blank, D., Yang, X. M., Suits, A. G., Lee, Y. T., Evans, M., Ng, C. Y., Flaim, C. & Padmore, H. A. (1997). Rev. Sci. Instrum. 68, 1945-1951.]). However, at normal incidence the reflectivity of all materials is extremely low at photon energies above ∼25 eV (Windt et al., 1988[Windt, D. L., Csah, W. C., Scoot, M., Arendt, P., Newnam, B., Fisfer, R. F., Swartzlander, Takacs, P. Z. & Pinneo, J. M. (1988). Appl. Opt. 27(2), 279-295.]). Besides, it is very difficult for an NIM to provide sufficient floor space for various types of experimental stations due to its awkward geometric arrangement. On the other hand, the Dragon-type SGM beamline (Chen, 1987[Chen, C. T. (1987). Nucl. Instrum. Methods, A256, 595-604.]) equipped with a proper gas cell system can meet all our expectations. Therefore, we have chosen the 6 m cylindrical-grating monochromator (6 m-CGM) design with an incident angle of 70° instead of the conventional NIM design. A gas cell with a special differential pumping system will be used to reject the high-order scattered light. A detailed description of the differential pumping system will be presented elsewhere (Chen et al., 1998[Chen, C. T. et al. (1998). In preparation.]).

2. Optical system

The optical layout of the 6 m-CGM high-flux beamline is shown in Fig. 1[link] and the parameters of the optical components are listed in Table 1[link]. Similar to the Dragon-type beamlines, a 6 m-CGM is arranged between pre-focusing and post-focusing mirrors. In order to maximize the beamline performance, we have adopted several special features including a modified plane-elliptical horizontal pre-focusing mirror (HFM), a bendable vertical pre-focusing mirror (VFM), two movable slits, and a curved exit slit. The beamline components and the special features are described in the following.

Table 1
Optical parameters of the 6 cm-CGM high-flux beamline

  Horizontal pre-focusing Vertical pre-focusing Entrance slit S1   Vertical refocusing Horizontal refocusing
  mirror HFM mirror VFM Entrance slit S2 Grating G mirror VRFM mirror HRFM1/2
r1 (mm) 3200 5100   1929 5006 8397
r2 (mm) 8729 3000 ± 500   4071 3250 2250
Deviation angle (°) 160 154   140 166 160
Type Plane ellipse (modified) Cylinder (bendable) Bilaterally adjustable Cylinder Sphere Plane ellipse
  (water-cooled)   Circularly curved S2      
Size (mm) 100 × 75 × 100 200 × 150 × 50 H: 5–70, V: 0–3 120 × 50 × 30 550 × 180 × 70 1000 × 60 × 100
Coating Al + MgF2 Al + MgF2   Al + MgF2 Al + MgF2 Al + MgF2
Substrate Glidcop ULE   ULE ULE ULE
Curvature (mm) Polynomials a + a′, b + b 16793.8 (14900–18500) 5200 (S2) 8510 32340 a = 5323.3, b = 754.8
Ruling profit       Laminar mask 1:1    
Ruling density (mm−1)       450 800 1600    
Ruling depth (Å)       800 450 250    
[Figure 1]
Figure 1
Optical layout of the 6 m-CGM high-flux beamline. S, radiation source; HFM, horizontal pre-focusing mirror; VFM, vertical pre-focusing mirror; S1, entrance slit; G, gratings; S2, exit slit; VRFM, vertical refocusing mirror; HRFM, horizontal refocusing mirror; D, sample position.

The HFM, which is located at 3.2 m from the source, collects 50 mrad of the horizontal radiation fan. Since a thermal loading of 200 W will be deposited on this mirror, the HFM is internally water-cooled. The photon beam is horizontally focused onto a location between the entrance slit (S1) and the exit slit (S2) in order to keep the image size at either slit less than 7 cm. The HFM is a specially designed plane-ellipse. The shape of the ellipse is calculated directly from the geometry of the optical arrangement and the actual shape of the radiation source (Chen et al., 1998[Chen, C. T. et al. (1998). In preparation.]). The radiation source is a 20 cm-long arc strip which is a section of the electron orbit in the bending magnet. Unlike the normal plane-elliptical surface which focuses a point source into a line, the modified plane-elliptical surface can focus an arc-strip source into a line perpendicular to the plane of the arc. This modified plane-elliptical surface can thus eliminate the large aberration effect caused by the large horizontal source size when using a normal plane-elliptical surface and hence provide perfect focusing in the horizontal plane. The surface figure of the modified plane-ellipse is defined by an eighth-degree polynomial surface height function, [y=\sum_{n=2}^8c_nx^n]. Thanks to recent advances in beamline optical technology, the modified plane-ellipse is now commercially available.

The VFM is a bendable cylindrical mirror, whose radius can be continuously changed so that the photon beams can always be focused vertically onto the movable entrance slit. The entrance slit is made movable to enhance greatly the energy resolution. Three cylindrical gratings, 450, 800 and 1600 lines mm−1, are used to cover photon energies from 4 to 40 eV. Gratings with a cylindrical reflection surface are chosen to reduce the aberration caused by the large horizontal angular acceptance (50 mrad). A laminar ruling profile is used for each grating to reduce the efficiency of high-order scattered light. Due to the large horizontal angular acceptance, the focused image is curved for the entire energy range, but can be represented by a simple circular arc with a radius of 5.2 m. Therefore, a curved slit jaw is used to improve further the flux and the resolution. One spherical re­focusing mirror, VRFM, is used to focus the beams vertically onto the sample. A regular plane-elliptical mirror, HRFM, located after the VRFM is employed to focus from the HFM focal point onto the sample. The best focused spot at the sample position is smaller than 0.45 × 0.2 mm. The vertical spot size at the sample position changes from 0.2 mm to 0.6 mm with different exit-slit positions. Two HRFMs are used for delivering the photon beam to two separate experimental stations.

3. Estimated performance

Following the Dragon beamline design criteria (Chen, 1987[Chen, C. T. (1987). Nucl. Instrum. Methods, A256, 595-604.]), the performance of the 6 m-CGM is optimized. An incident angle of 70° is chosen and the sum of the entrance arm and exit arm is set at roughly 6 m for the monochromator to maximize the energy resolution and photon throughput. In order to reduce the resolution limits, cylindrical gratings are used to minimize the astigmatic coma term (C12) and astigmatic term (C02), and a movable exit slit is used to eliminate the defocus term (C20). A radius of 8.51 m is chosen for the grating to match the monochromator parameters and optimize the resolution. If the entrance slit is kept fixed, the resolution may be limited by the coma term (C30) in some energy range. The resolving power of the fixed-entrance-slit 6 m-CGM calculated using the SHADOW ray-tracing programs (Lai & Cerrina, 1986[Lai, B. & Cerrina, F. (1986). Nucl. Instrum. Methods A, 246, 337-341.]) is shown in Fig. 2[link](a). In this calculation, the opening of both slits is 10 µm and the slope error of the grating surface is ±0.1 arcsec. The resolving power of each grating has a maximum in the middle and falls off on both sides. The resolving power can be further improved by eliminating the C30 term. By moving both the entrance and exit slit to satisfy the Rowland-circle focusing conditions (Samson, 1967[Samson, J. A. (1967). Techniques of VUV Spectroscopy, pp. 43-84. Lincoln, Nebraska: Pied.]; Hogrefe et al., 1986[Hogrefe, H., Howells, M. R. & Hoyer, E. (1986). Proc. SPIE, 733, 274-285.]), i.e. r1 = Rcosα and r2 = Rcosβ, both the C20 and C30 terms can be eliminated simultaneously. Fig. 2[link](b) displays the resolving powers under this condition, showing a maximum value of 70000.

[Figure 2]
Figure 2
Calculated resolving power of each grating, (a) with a fixed entrance slit and a movable exit slit, and (b) with both slits movable to meet the Rowland-circle conditions. In these calculations, both slits are set at 10 µm.

The photons from the bending-magnet radiation source are composed of p-polarization and s-polarization components. The reflectance of each polarization component is different, and thus it should be considered separately. For example, the reflectivity of the p-polarization component at photon energy 10 eV is about 20% lower than that of the s-polarization component, if an aluminium-coated mirror faces horizontally, such as the HFM, and the incident angle of the photon beam is 80°. The polarization-dependent reflectance of all the optical surfaces, mirrors and gratings can be calculated using the Fresnel equations and the tabulated refraction coefficients, n and k (Marr, 1987[Marr, G. V. (1987). Handbook on Synchrotron Radiation, Vol. 2, pp. 41-52. Amsterdam: North-Holland.]; Windt et al., 1988[Windt, D. L., Csah, W. C., Scoot, M., Arendt, P., Newnam, B., Fisfer, R. F., Swartzlander, Takacs, P. Z. & Pinneo, J. M. (1988). Appl. Opt. 27(2), 279-295.]). The calculated photon fluxes for different gratings and slit openings are shown in Fig. 3[link]. With a storage-ring current of 200 mA, photon fluxes of 1 × 1013 and 1 × 1010 photons s−1 are expected for energy-resolving powers of 1000 and 40000, respectively.

[Figure 3]
Figure 3
Calculated photon flux for each grating and various slit openings.

4. Summary

A very high photon flux and ultrahigh-resolution low-energy beamline, covering photon energies from 4 to 40 eV, has been designed. In addition to the conventional 6 m-CGM design, several special features, such as the specially designed plane-ellipse mirror, the bendable vertical focusing mirror, the movable entrance slit and exit slit, and the curved exit slit jaw, are utilized to enhance greatly the beamline performance. The expected photon fluxes are greater than 1 × 1013 and 1 × 1010 photons s−1 for energy-resolving powers of 1000 and 40000, respectively, even after taking into account the polarization effect on reflectance. The best focused spot at the sample location is less than 0.45 × 0.2 mm. This beamline is currently under construction at SRRC and will be ready for commissioning in the spring of 1998.

Acknowledgements

We would to thank Dr Li-Jin Lu at HESYRL for useful discussions on the polarization effect.

References

First citationChen, C. T. (1987). Nucl. Instrum. Methods, A256, 595–604.  CrossRef CAS Web of Science
First citationChen, C. T. et al. (1998). In preparation.
First citationEyers, A., Heckenkamp, Ch., Schafers, F., Schonhense, G. & Heinzmann, U. (1983). Nucl. Instrum. Methods, 208, 303–305.  CrossRef CAS Web of Science
First citationHeimann, P. A., Koike, M., Hsu, C. W., Blank, D., Yang, X. M., Suits, A. G., Lee, Y. T., Evans, M., Ng, C. Y., Flaim, C. & Padmore, H. A. (1997). Rev. Sci. Instrum. 68, 1945–1951.  CrossRef CAS Web of Science
First citationHogrefe, H., Howells, M. R. & Hoyer, E. (1986). Proc. SPIE, 733, 274–285.  CrossRef
First citationLai, B. & Cerrina, F. (1986). Nucl. Instrum. Methods A, 246, 337–341.  CrossRef Web of Science
First citationLama, F., Derossi, A., Piacentini, M. & Zema, N. (1994). Nucl. Instrum. Methods Phys. Res. A, 339, 610–616.  CrossRef CAS Web of Science
First citationMarr, G. V. (1987). Handbook on Synchrotron Radiation, Vol. 2, pp. 41–52. Amsterdam: North-Holland.
First citationSamson, J. A. (1967). Techniques of VUV Spectroscopy, pp. 43–84. Lincoln, Nebraska: Pied.
First citationSuits, A. G., Heimann, P., Yang, X., Evans, M., Hsu, C.-W., Lu, K. & Lee, Y. T. (1995). Rev. Sci. Instrum. 66(10), 4841–4844.  CrossRef Web of Science
First citationTseng, P.-C., Hsieh, T.-F., Song, Y.-F., Lee, K.-D., Chung, S.-C., Chen, C.-I., Lin, H.-F., Dann, T.-E., Huang, L.-R., Chen, C.-C., Chung, J.-M., Tsang, K.-L. & Chang, C.-N. (1995). Rev. Sci. Instrum. 66(2), 1815–1817.  CrossRef Web of Science
First citationWicke, H., Bohmer, W. & Schwentner, N. (1983). Nucl. Instrum. Methods, 204, 533–542.
First citationWindt, D. L., Csah, W. C., Scoot, M., Arendt, P., Newnam, B., Fisfer, R. F., Swartzlander, Takacs, P. Z. & Pinneo, J. M. (1988). Appl. Opt. 27(2), 279–295.  CrossRef

© 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.

Journal logoJOURNAL OF
SYNCHROTRON
RADIATION
ISSN: 1600-5775
Follow J. Synchrotron Rad.
Sign up for e-alerts
Follow J. Synchrotron Rad. on Twitter
Follow us on facebook
Sign up for RSS feeds