Open Access
Add to BibSonomyAbstract
We developed a hard X-ray split-and-delay optical (SDO) system based on Bragg diffraction in crystal optics for generating two split pulses with a variable temporal separation. To achieve both high stability and operational flexibility, the SDO system was designed to include variable-delay and fixed-delay branches. As key optical elements, we fabricated high quality thin crystals and channel-cut crystals by applying the plasma chemical vaporization machining technique. The SDO system using Si(220) crystals covered a photon energy range of 6.5–11.5keV and a delay time range from a negative value to > 45 ps over the photon energy range (up to 220 ps at 6.5 keV). A simple alignment method for realizing a spatial overlap between the split pulses was developed. The SDO system was tested at a SPring-8 beamline in combination with a focusing system. We achieved an excellent overlap with an accuracy of 30 nm for ∼ 200 nm focused beams in both the horizontal and vertical directions. This achievement is an important progress towards the realization of time-resolved studies using multiple X-ray pulses with a time range from femtosecond to subnanosecond scales at X-ray free-electron laser facilities.
© 2016 Optical Society of America
1. Introduction
Capturing ultrafast motion of electrons and atoms presents significant knowledge in many fields of science. X-ray free-electron lasers (XFELs), e.g., the Linac Coherent Light Source (LCLS) [1] in the United States, and SPring-8 Angstrom Compact free-electron LAser (SACLA) [2] in Japan, offer great capabilities for probing ultrafast dynamics with angstrom spatial resolution on femtosecond time scales because of their unique beam characteristics. Furthermore, an ultraintense X-ray pulse with a power density over 1020 Wcm−2 can be generated using focusing optics [3, 4], thus promoting studies of X-ray nonlinear phenomena with anomalous behaviors of electrons [4–7].
Typical time-resolved studies of ultrafast dynamics require multiple optical pulses that are temporally separated. As an experimental scheme for XFELs, the optical pump X-ray probe method has been widely performed. Here, an optical laser triggers a transition of valence electronic states, and an XFEL probe pulse tracks the dynamics as a function of the delay time between the optical and XFEL pulses [8–10]. As another approach, one can consider the X-ray pump X-ray probe scheme using double XFEL pulses. This scheme makes it possible to investigate the ultrafast dynamics associated with core electron excitations and the subsequent multiple ionization initiated by the intense XFEL irradiation, such as Coulomb explosion [11]. Furthermore, the utilization of double XFEL pulses allows the exploration of spontaneous atomic fluctuations in condensed matter and plasma with an ultrafast time scale by means of the X-ray photon correlation spectroscopy (XPCS) technique [12].
We introduce two approaches to generate double XFEL pulses. One is an accelerator based method, whereas the other is an optics based method. In the former approach, a simple and powerful ‘split-undulator’ method that utilizes a small chicane placed between two undulator groups [13, 14] has been developed. Here, an XFEL pulse is generated from an electron bunch in the first undulator group, and subsequently another pulse is produced with the same electron bunch in the second group. A temporal separation between the two XFEL pulses is controlled by electrically changing the deflection angle (i.e., the path length) of the electron beam in the chicane. The separation is therefore controlled without the influence of the acceleration jitter, resulting in an accuracy in the attosecond range. In addition, different colors can be set for the two XFEL pulses by imposing different magnetic field strengths on the two undulator groups. However, the maximum delay time is limited to subpicoseconds because of a small deflection angle in the chicane typically below ∼ 10 mrad for ∼ 10 GeV electron beams.
In the latter approach, a split-and-delay optical (SDO) technique offers distinct opportunities for generating double XFEL pulses. The SDO system consists of a beam splitter, two delay branches, and an optional beam merger. Since the delay time is defined by the difference in path lengths of the delay branches that are mechanically adjusted, a high temporal resolution comparable to the split-undulator method can be achieved. In the hard X-ray regime, several types of the SDO schemes have been proposed by employing various optical elements such as grazing incidence multilayer mirrors [15], diffraction gratings [16], and perfect crystals [17–19]. In particular, a crystal based SDO system facilitates the production of a large temporal separation over ∼ 100 ps by using a large deflection angle (> 10°) of the X-ray beam caused by Bragg diffraction in a crystal lattice. The maximum deflection angle also allows to independently characterize the photon properties of each branch, which are critical parameters required in subsequent data analysis. Furthermore, the crystal based SDO system can be utilized to generate monochromatic XFEL pulses. The latter are preferable for achieving high resolutions in typical X-ray diffraction experiments and for increasing the visibility of speckle patterns in the high wavevector range in XPCS experiments.
Roseker et al. first proposed a crystal based SDO system with the capability of producing a few nanoseconds delay times [17, 18]. They used eight independent silicon crystals, including two thin crystals as the beam splitter and beam merger. In order to reduce the overall motion complexity of all eight crystals and keep the system compact, they adopted a 90° scattering geometry and a single translation for changing the temporal separation. The adopted geometry can, however, operate at specific X-ray energies with small bandwidths, as the result of the high order diffractions needed to obtain a 90° scattering angle in the hard X-ray regime. The simple translation moves four different crystals and therefore changes the path lengths of both branches.
In this paper, we report a new design of the SDO system based on perfect crystal diffraction for providing large operational flexibility while maintaining high stability. For this purpose, we designed the SDO system with an optical configuration combined with a variable-delay branch and fixed-delay branch. Utilization of two channel-cut crystals to the fixed-delay branch made it possible to decrease the degrees of freedom in mechanics for controlling optical elements, thus facilitating alignment and operation while enhancing stability. In the variable-delay branch, two thin crystals and two thick crystals were used. A critical requirement for this scheme is to ensure high perfection for both channel-cut crystals and thin crystals. To achieve excellent diffraction properties for these key optical elements, we used the plasma chemical vaporization machining (PCVM) technique [20] that promotes chemical reactions with radicals generated in a plasma at atmospheric pressure. This method is able to remove defects without inducing strains. The plasma can be localized within a small space that confines a removal footprint within a width of 0.5 mm. Figure errors and thickness variations over a wide range of spatial frequency can be thus reduced with a computer controlled scan of the localized plasma. This SDO design covers a wide photon energy range because of the tunability of diffraction angles. In this study we employed Si(220) crystals, resulting in a photon energy range from 6.5 to 11.5 keV and a delay time range from −50 to +47 ps at 10 keV.
This paper is organized as follows. In section 2, we introduce the detailed design of the SDO system. The optical arrangement, delay time range, and temporal resolution are described. In section 3, we discuss the alignment of the system for achieving a spatial overlap between the split beams. A simple alignment method and requirements on the mechanics of the system are described. In section 4, we present experimental results of a performance test conducted at the 1-km-long beamline BL29XUL of SPring-8 in combination with a focusing mirror system. Finally, we provide the conclusion and future prospects.
2. Design of the SDO system
2.1. Optical arrangement
Figure 1(a) depicts a schematic of the SDO system assembled with six crystals. In the variable-delay (upper) branch, we utilized two thin crystals as the beam splitter (BS) and beam merger (BM), and two thick crystals as the beam reflectors (BRs). In the fixed-delay (lower) branch, we employed two channel-cut crystals (CCs). The reduction of optical elements with the use of the CCs allows to facilitate alignment and operation of the system, because the direction of the X-ray beam propagating through the CCs in the lower branch is parallel to the original optical axis. The path length of the upper branch is adjusted by a linear translation of the BRs along each 2θ direction. For angular control, we adjusted the angles ω and χ of all optical elements and of the 2θ1- and 2θ2-arms over which the BRs are mounted. This angular flexibility allows for a wide tunability of the photon energy. In addition, different photon energies can be set for the two branches. Note that when the photon energy is adjusted, the delay time is also changed, as indicated below by Eq. (1).
Fig. 1 (a) Schematic of the split-and-delay optics based on Bragg diffraction, utilizing six Si(220) crystals. Crystals BS, BRs, BM, and CCs represent the beam splitter, beam reflectors, beam merger, and channel-cut crystals, respectively. The inset on the upper right of (a) depicts the four general axes in X-ray diffraction. The use of the CCs decreases the degrees of rotational freedom (ω and χ) in the lower branch from 8 to 4, and can produce a robust reference beam for alignment of the upper branch elements because the direction of the exit beam from the lower branch is naturally preserved in parallel to the incident beam direction. (b) Dimensions of CC1 in units of mm, whereas those of CC2 are in mirror symmetry.
To achieve a high throughput, the Si(220) diffraction plane with a bandwidth of ΔE/E0 ≈ 5.6 × 10−5 was selected. A spectral splitting method discussed in [19, 21] that produces two pulses with non-overlapping spectra, was applied for further enhancement of the throughput. The photon energy range for the lower branch was set to 6.5–11.5 keV by the dimensions of the CCs shown in Fig. 1(b), whereas that for the upper branch was above 3.2 keV. The theoretical throughputs of the upper and lower branches for 10 keV monochromatized X-rays are 30% and 20%, respectively, assuming a Si(111) bandwidth (ΔE/E0 ≈ 1.3 × 10−4) and a plane wave, as well as a thin crystal thickness of 10 μm and energy deviations of +0.35 eV and −0.35 eV from 10 keV for the upper and lower branches, respectively.
We have already reported the fabrication of high quality thin silicon crystals with thicknesses down to 4.4 μm by using the PCVM technique [21, 22]. For the SDO system, we fabricated 10-μm-thick Si(220) crystals by adopting an improved PCVM procedure. The thin part was supported by a thick frame part with a thickness of 1.5 mm. The thin crystals were put on the holders without inducing mount stress to avoid the introduction of strains caused by the mounting. Experimental results of topographic and rocking-curve measurements are depicted in Fig. 2. The topographic results shown in Figs. 2(a) and 2(b) confirm the crystal has no defects and only a few strains of several microradians in the active area (> 1.0 mm × 0.5 mm in the horizontal and vertical directions, respectively). The rocking curves in Fig. 2(c) show both the high peak reflectivity and transmissivity out of the Darwin range of > 80%. By applying the PCVM technique to processing the inner wall diffraction surfaces of the CCs, we succeeded in fabricating high quality CCs without inducing wavefront distortion to the reflected XFEL beams [23]. The use of such damage-free optics enables us to conduct coherent diffractive experiments that are an important characteristic of XFEL applications.
Fig. 2 (a) Si(220) reflection topograph of the BM on the Bragg condition measured with 10 keV X-rays at BL29XUL of SPring-8. The X-ray probe was monochromatized with a Si(220) crystal, and collimated into a few microradians during the 1 km beam transport. The relatively dark regions correspond to the edge of the thin part, in which large strains exist. (b) Distribution of the slope error of the crystal lattice. The scale bars in (a) and (b) represent 0.5 mm. (c) Measured rocking curve with a confined X-ray probe into the black dashed area displayed in (a) and (b) with dimensions of 1.0×0.5 mm2 in the horizontal and vertical directions, respectively. Red and blue points represent the normalized reflection and transmission intensities, respectively, and black solid lines represent those of calculation with a crystal thickness of 9.8 μm.
The configuration of the SDO system is shown in Fig. 3. High-precision goniometers mounted on the x-z linear stages were used to control the ω angles of the thin crystals and CCs, whereas rotation stages mounted on the linear translation stages connected to the 2θ arms were utilized to tune the ω angles of the BRs. The translation range of the BRs along each 2θ arm was 70 mm. The angles χ of the elements in the upper branch were adjusted by swivel stages mounted on each ω stage, whereas those of the CCs were adjusted manually because χ errors of the CCs only translate into a small positional shift of the exit beam in the horizontal direction (< 0.1 mm). The stage resolutions are listed in Table 1.
Fig. 3 Picture of the SDO system. Ion chambers are used as the beam intensity monitors (BIMs). A photodiode is utilized as BIM3 to measure the X-ray beam intensity penetrating the BM, while BIM3′ is an optional monitor.
Table 1. Stage resolutions in half-step feed mode
2.2. Delay time ranges and resolutions
In this system, the delay time τ can be written as
where D, D0, , and c are the path length between adjacent thin and thick crystals, between the first and second illumination points on each CC, the Bragg angle for the upper (the lower) branch, and the speed of light, respectively. The minimum value of D was set to 57 mm in the SDO system to avoid mechanical interference. The path length D0 is given by , where g is the gap of the CCs, as shown in Fig. 1. A gap of 30 mm was selected in order to access negative delay times over the photon energy range of interest. The delay time range at 10 keV was given by 97 ps (from −50 to +47 ps). This range is enhanced at large Bragg angles, i.e., low photon energies, as shown in Fig. 4(a). Similarly, a temporal resolution of 0.35 fs at 10 keV was set by using the translation stage with a resolution of 0.25 μm. The resolution for the delay time was less than 1 fs over the photon energy range, as shown in Fig. 4(b). It is noted that this resolution is much smaller than the effective pulse duration of ∼ 3 fs at 10 keV that is determined by the Fourier-limited condition in the Si(220) diffracted beam.Fig. 4 (a) Range of the delay time plotted as a function of the photon energy with the translation range D of the BRs, varying from 57 to 127 mm. Here, the same photon energy for both the upper and lower branches was assumed. (b) Temporal resolution calculated with a resolution for the linear translation of 0.25 μm.
3. Alignment procedure
3.1. Principles
An achievement of a spatial overlap between the split beams is a critical requirement for the SDO technique. It becomes difficult to satisfy this condition when we combine a focusing optical system that is widely used to increase power densities and to enhance the size of individual speckles in XPCS experiments. For this purpose, a spatial deviation and angular mismatch between the two beams should be minimized. Here we discuss an alignment procedure of the optical elements in the SDO system for achieving a spatial overlap in the focal plane based on a geometrical consideration. Since an optical axis of the exit beam from the lower branch (referred to as the lower beam) can be defined as the reference axis, the goal of the alignment is to minimize the positional and angular deviations of the beam from the upper branch (referred to as the upper beam). We define the positional deviation at the BM as Δx,z, and the angular deviation as , where the subscripts denote the directions.
A schematic diagram employing a thin lens as a focusing optics is depicted in Fig. 5. The geometrical magnification M is expressed as
where a and b are the source-to-lens and lens-to-image distances, respectively. The minimum focal size ϕi is given by ϕi = ΦM, where Φ is the source size. In practice, this corresponds to assuming a sufficiently large numerical aperture (NA) of the lens, which is typically satisfied by the focusing optics for XFELs to reduce the loss of flux. The positional deviation of the effective source for the upper beam can be represented by Δ and Δ′. When we assume the effective source lies at the intersection of the source plane and the back-propagating X-ray path from the BM, the positional deviation of the effective source along the horizontal and vertical directions can be geometrically written as where L is the source-to-BM distance. From a geometrical consideration, the upper beam with is focused onto the imaging plane with a spatial deviation of given byFig. 5 Side view of the schematic diagram of source imaging with a thin lens. The lower beam propagates the ideal optical path. Effective source position of the upper beam is assumed to be located at the intersection of the back-propagating beam path from the BM (red dashed line) and the source plane.
From Eqs. (3) and (4), the spatial deviation in the imaging plane can be written as
To achieve a sufficient overlap, should be much smaller than the focused beam size ϕi. We denote this criterion by
As another practical criterion, a spatial acceptance of the focusing optics should be sufficiently larger than the positional deviation of the upper beam at the optics in order to avoid a loss of flux. This criterion can be written as
where Lf is the BM-to-lens distance. From Eqs. (5) and (7), and a = L + Lf, this can be rewritten as where use has been made of the perfect overlapping condition at the focal plane, i.e., di = 0 μm. Hence, we can define the criterion by where Wx,z is the spatial acceptance of the focusing optics.We briefly estimate the tolerance of Δ and Δ′ with focusing systems installed at the hard X-ray beamline BL3 of SACLA [24]. Here two Kirkpatric–Baez (KB) mirror systems with a spatial acceptance Wx,z ∼ 600 μm were used to produce a 1 μm [25] and 50 nm [3] focused beams by applying different demagnifications of the source with a typical size Φ of 60–80μm at full width at half maximum (FWHM). The tolerance of is thus in the range of 15–20 μm for both systems. We here consider two simple cases and Δx,z = 0 μm for the first criterion in Eq. (6). In the former case, Δx,z is equal to , indicating should be smaller than 15 μm. In the latter case, should be smaller than 0.15 μrad with L ∼ 100 m. These values correspond to each resolution requirement, because, according to Eq. (5), the positional and angular errors can be compensated with adjustments of the other parameter. For the second criterion in Eq. (9), we should also satisfy |Δx,z| < 125 μm and with a ∼ 120 m.
3.2. Alignment method and angular resolution requirements
We developed a simple alignment method to achieve a spatial overlap of the split beams. During crystal alignment, spatial deviations and angular mismatches along the horizontal (vertical) direction can be induced by the χ (ω) errors of the elements. Since reflection intensities are sensitive to the ω errors, they can be suppressed within a few microradians, generating a small vertical deviation Δz of only a few micrometers within ∼ 1 m distance. However, a mismatch between 2θ1 and 2θ2 can provide a larger error in Δz [26]. Therefore, adjustments of only the χ angle of the BS χBS and 2θ2 should be sufficient for the reduction of |Δx,z|. An angular deviation can be corrected by adjusting only the χ and ω angles of the BM, χBM and ωBM.
The upper beam directions in the horizontal and vertical axes are modified by the χ and ω rotations, respectively, as
where and are the angular changes of the upper beam in the horizontal and vertical directions, respectively, and δχ, and δω are the rotational deviations of χ and ω, respectively. The positional deviation |Δx,z| can be corrected by adjusting χBS and 2θ2 where δΔx and δΔz are the positional shifts of the upper beam at the BM in the horizontal and vertical directions, respectively, and δ2θ2 is the rotational deviation of 2θ2. The parameter lup is the path length of the upper branch given by where l0 ∼ 1 m is the BS-to-BM distance. According to Eqs. (10)–(12), the resolutions of χBS, χBM, 2θ2, and ωBM stages should be finer than 22 μrad, 0.23 μrad, 66 μrad, and 0.075 μrad, respectively, in order to achieve the resolution requirements. This estimation assumes a maximum D of 127 mm, for which the finest resolutions are required, as indicated by Eqs. (11) and (12). These requirements were in good agreement with results from both ray trace calculations and wave optical simulations based on the Fresnel–Kirchhoff diffraction formula.3.3. Practical alignment procedure and expected temporal accuracy
Here we describe the practical alignment procedure. First, we roughly corrected the positional deviation of the upper beam Δx,z by adjusting χBS and 2θ2 with a two-dimensional (2D) detector placed downstream of the BM. Second, we performed coarse adjustments of χBM and ωBM to transport the upper beam in the aperture of the focusing optics. Finally, both the position and angle of the upper beam were finely tuned to satisfy the two criteria in Eqs. (6) and (9) by observing the spatial deviation in the focal plane .
The temporal accuracy of the delay times was approximately evaluated. According to this alignment procedure, the split beams could propagate out of the ideal beam paths in the delay branches even if a perfect overlap was achieved. This will deviate the temporal separations from the target delay times. In particular, a temporal error of a few picoseconds can be obtained with an angular error of ∼ 0.1° between the 2θ arms and . The slope error of the delay time as a function of D, however, was smaller than 1% with possible angular errors of the upper branch elements at a milliradian level. This also indicates that vibrations of submicroradians can provide up to a few femtoseconds temporal jitters over the delay time range (< 0.001%).
4. Test experiment at SPring-8
4.1. Experimental setup
A test experiment of the SDO system was performed at the 1-km-long beamline BL29XUL of SPring-8 [27] with an experimental setup shown in Fig. 6. A 10 keV X-ray beam from a Si(111) double-crystal monochromator (DCM) [28] was used. As a virtual source (VS), we employed a 4-jaw slit with an aperture of 10 μm × 30 μm (here and henceforth in the horizontal and vertical directions, respectively) placed 52 m downstream of the undulator exit. The SDO system was placed 8.4 m downstream of the VS. The positional and angular deviations of the upper beam were measured with two 2D detectors CCD1 (HAMAMATSU, ORCA-RII, 3.1 μm/pixel) and CCD2 (BITRAN, CS-52M, 1.25 μm/pixel), placed 7.2 m and 36.6 m downstream of the BM, respectively. A KB focusing mirror system [29] was placed approximately 47 m downstream of the VS, which was aligned with the lower beam and then fixed during this experiment. Another 4-jaw slit with an aperture of 180 μm × 380 μm corresponded to the spatial acceptance of the mirrors Wx,z was placed in front of the mirrors. The focal positions and profiles were measured by means of the bright field edge scan method using an Au wire. We set the energy deviations of the upper and lower beams to +0.35 eV and −0.35 eV, respectively. For the distance D, we initially employed a nearly maximum value of 125 mm, corresponding to the delay time τ ∼ +45 ps. Under these conditions, it was difficult to evaluate the accuracy of the delay time because the delay was comparable to the duration of the synchrotron X-ray pulses of 30–50 ps at FWHM.
Fig. 6 Experimental setup at the 1-km-long beamline BL29XUL of SPring-8. Beam intensities inside the spatial acceptance of the focusing mirrors can be measured with BIM6.
The focal size of the lower beam ϕi was ∼ 200 nm × 160 nm at FWHM. In this setup, the geometrical magnifications were ∼ 1/310 and 1/190 in the horizontal and vertical directions, respectively, with L = 9.4 m. The focal size was limited by the NA and residual alignment errors of the focusing system. From a geometrical consideration, thus, the effective source size Φ can be considered ∼ 60 μm × 30 μm. In order to realize the criterion in Eq. (6), the required resolutions δΔx,z and were ∼ 15 μm × 7.5 μm and 1.6 μrad × 0.80 μrad, respectively. These resolutions can be satisfied with the mechanics of the SDO system. From Eq. (9), on the other hand, the deviations |Δx,z| and should be smaller than 9.0 μm × 19 μm and 1.0 μrad × 2.0 μrad, respectively.
4.2. Results and discussions
We first attempted to achieve a spatial overlap between the split beams in the focal plane using the alignment procedure described in section 3.3. Since it was possible to achieve |Δz| < 19 μm by using CCD1 for the vertical direction with a beam size of ∼ 100 μm at FWHM, the spatial overlap in the vertical direction can be easily achieved only with a fine adjustment of ωBM by observing the focal position. However, it was difficult to realize |Δx| < 9.0 μm, because the horizontal beam size was larger than 200 μm at FWHM. Fine tuning of both the position and angle of the upper beam was thus required in the horizontal direction when measuring, in addition to the focal position, the intensity of the upper beam inside the spatial acceptance of the mirrors.
Figure 7 shows focal profiles in both the horizontal and vertical directions after fine adjustments. We succeeded in obtaining a spatial overlap of the split beams in the focal plane with spatial deviations of ∼ 30 nm in both directions. The residual angular mismatch evaluated with two CCDs was 0.03 μrad × 0.24 μrad and much smaller than the criterion value. Note that under this condition, the unfocused split beams were also highly overlapped with each other with an accuracy of 10 μm. Experiments with unfocused beams can be therefore conducted without additional alignment.
Fig. 7 Focal profiles in the (a) horizontal and (b) vertical directions. Red, blue, and black points represent the profiles of the upper, lower, and double beams, respectively.
The focal size and profile of the upper beam in the horizontal direction were similar to those of the lower beam, whereas the size of the upper beam in the vertical direction was twice as large as that of the lower beam. The distortion could originate from a change of the divergence of the upper beam from 2 to 5 μrad at FWHM caused by small lattice distortions remaining in the thin crystals, as shown in Fig. 8. Such a change in the divergence deviates the position of the effective source along the optical axis, thereby defocusing the upper beam. The influence will be more crucial at typical XFEL beamlines with a large L ∼ 100 m. In order to suppress the focal shift within the Rayleigh range for the 1 μm focusing system at SACLA, the criterion for a divergence change was simply estimated to be 0.3 μrad at 10 keV using wave optical simulations. The lower beam profile displayed in Fig. 8(b) was, on the other hand, similar to the incident beam profile shown in Fig. 8(a).
Fig. 8 Unfocused beam profiles measured by CCD2 placed 46 m downstream of the VS. (a) Incident beam measured without the SDO system, (b) lower beam, and (c) upper beam. The origins of the plots are set to the center of mass coordinate of each intensity distribution.
The throughput of each branch in the SDO system was defined by , where the subscript and superscript represent the intensity monitor and object branch, respectively. The throughputs of the upper and lower branches were 12.6% and 13.8%, respectively, which were much higher than the previously reported throughputs of 0.3–0.4% [18]. Nevertheless, these values were lower than the calculated ones of 28.8% and 19.4%, respectively, which assumed a beam divergence of 1.8 μrad at FWHM. The reduction for the upper branch can be mainly attributed to the lattice distortion in the thin crystals described previously. It can be also imposed by the relative angular mismatches on the four independent crystals. The throughput of the lower branch was mainly decreased by a non-ideal angular profile of the incident beam to the SDO system with the Si(220) Bragg diffraction, as shown in Figs. 8(a) and (b).
In order to confirm the influence of the delay change by the linear translation along each 2θ direction, we monitored the position of the upper beam in the focal plane after positional shifts of the BRs. We did not observe horizontal displacement for small shifts in the BRs of ∼ 1 mm, corresponding to a change in the delay time of ∼ 1 ps, while the intensity of the upper beam decreased significantly. This result indicates that the ω angles of the BRs were changed with their translations. However, we can readjust the changes with small angular corrections. For larger translations of > 10 mm, a considerable horizontal deviation, in addition to the vertical one, was observed. Further improvements of the mechanics are required for facilitating operation of the SDO system.
5. Conclusions and future prospects
We developed an SDO system with variable temporal separations and a wide photon energy range of 6.5–11.5 keV. We employed six Si(220) crystals including thin crystals and channel-cut crystals to facilitate alignment and operation of the SDO system. The thin crystals and channel-cut crystals with excellent diffraction qualities were fabricated using the PCVM technique. A simple alignment method was developed for achieving a spatial overlap between the split beams in the focal plane. The SDO system was tested at BL29XUL of SPring-8 with a ∼ 200 nm focusing system. A successful spatial overlap with an accuracy of 30 nm was realized. This result is the first success for the generation of spatially overlapping split beams with a nanofocusing optical system. At the SACLA beamline, the spatial overlap can be achieved more simply because of a large spatial acceptance of the focusing optics ∼ 600 μm and long source-to-BM distance L ∼ 100 m. The throughputs of the upper and lower branches were 12.6% and 13.8%, respectively, for the X-ray incidence within the Si(111) bandwidth. Defocusing of the upper beam due to lattice distortions in the thin crystals was observed. Further improvement of the PCVM technique facilitates the fabrication of thin crystals with few strains, resulting in ideal focusing of the upper beam even at XFEL beamlines.
A self-seeding scheme [30] is able to generate a monochromatic intense XFEL pulse with a bandwidth of ΔE/E0 ≈ 5.0–14 × 10−5 [31], which is comparable with the Si(220) bandwidth. This indicates the SDO system has the potential to generate split pulses with pulse energies a few orders of magnitude higher than the present level. At such high brilliant monochromatic X-ray sources, the thermal load will be a crucial issue, in particular, for thin crystals. The influence of thermal load can be minimized by using diamond crystals [32]. In principle, the PCVM technique can be applied to fabricating thin diamond crystals [20]. The diamond C(111) bandwidth (ΔE/E0 ≈ 5.7 × 10−5) is similar to the Si(220) bandwidth, and, therefore, we can develop a more robust SDO system while producing split pulses with similar beam characteristics. Furthermore, combination of the SDO system with an accelerator-based two-color XFEL generation method, in addition to the self-seeding scheme, offers attractive opportunities for the investigation of ultrafast dynamics triggered by different excitation levels of the electrons. The SDO system can extend the accessible time range for time-resolved studies with multiple X-ray pulses to the subnanosecond scale while maintaining high intensities.
Acknowledgments
The authors wish to thank the staff of SPring-8/SACLA for their continuous support and appreciate valuable discussions with Prof. G. Grübel and Dr. W. Roseker of DESY. Portions of this research were supported by JSPS KAKENHI Grant Number 13J00898 and JSPS Core-to-Core Program on International Alliance for Material Science in Extreme States with High Power Laser and XFEL.
References and links
1. P. Emma, R. Akre, J. Arthur, R. Bionta, C. Bostedt, J. Bozek, A. Brachmann, P. Bucksbaum, R. Coffee, F.-J. Decker, Y. Ding, D. Dowell, S. Edstrom, A. Fisher, J. Frisch, S. Gilevich, J. Hastings, G. Hays, Ph. Hering, Z. Huang, R. Iverson, H. Loos, M. Messerschmidt, A. Miahnahri, S. Moeller, H.-D. Nuhn, G. Pile, D. Ratner, J. Rzepiela, D. Schultz, T. Smith, P. Stefan, H. Tompkins, J. Turner, J. Welch, W. White, J. Wu, G. Yocky, and J. Galayda, “First lasing and operation of an ångstrom-wavelength free-electron laser,” Nat. Photon. 4(9), 641–647 (2010). [CrossRef]
2. T. Ishikawa, H. Aoyagi, T. Asaka, Y. Asano, N. Azumi, T. Bizen, H. Ego, K. Fukami, T. Fukui, Y. Furukawa, S. Goto, H. Hanaki, T. Hara, T. Hasegawa, T. Hatsui, A. Higashiya, T. Hirono, N. Hosoda, M. Ishii, T. Inagaki, Y. Inubushi, T. Itoga, Y. Joti, M. Kago, T. Kameshima, H. Kimura, Y. Kirihara, A. Kiyomichi, T. Kobayashi, C. Kondo, T. Kudo, H. Maesaka, X. M. Maréchal, T. Masuda, S. Matsubara, T. Matsumoto, T. Matsushita, S. Matsui, M. Nagasono, N. Nariyama, H. Ohashi, T. Ohata, T. Ohshima, S. Ono, Y. Otake, C. Saji, T. Sakurai, T. Sato, K. Sawada, T. Seike, K. Shirasawa, T. Sugimoto, S. Suzuki, S. Takahashi, H. Takebe, K. Takeshita, K. Tamasaku, K. Tono, S. Wu, M. Yabashi, M. Yamaga, A. Yamashita, K. Yanagida, C. Zhang, T. Shintake, H. Kitamura, and N. Kumagai, “A compact X-ray free-electron laser emitting in the sub-ångström region,” Nat. Photon. 6(8), 540–544 (2012). [CrossRef]
3. H. Mimura, H. Yumoto, S. Matsuyama, T. Koyama, K. Tono, Y. Inubushi, T. Togashi, T. Sato, J. Kim, R. Fukui, Y. Sano, M. Yabashi, H. Ohashi, T. Ishikawa, and K. Yamauchi, “Generation of 1020 Wcm−2 hard X-ray laser pulses with two-stage reflective focusing system,” Nat. Commun. 5, 3539 (2014). [CrossRef]
4. M. Fuchs, M. Trigo, J. Chen, S. Ghimire, S. Shwartz, M. Kozina, M. Jiang, T. Henighan, C. Bray, G. Ndabashimiya, P. H. Bucksbaum, Y. Feng, S. Hermann, G. A. Carini, J. Pines, P. Hart, C. Kenney, S. Guillet, S. Boutet, G. J. Williams, M. Messerschmidt, M. M. Seibert, S. Moeller, J. B. Hastings, and D. A. Reis, “Anomalous nonlinear X-ray Compton scattering,” Nat. Phys. 11, 964–970 (2015). [CrossRef]
5. K. Tamasaku, E. Shigemasa, Y. Inubushi, T. Katayama, K. Sawada, H. Yumoto, H. Ohashi, H. Mimura, M. Yabashi, K. Yamauchi, and T. Ishikawa, “X-ray two-photon absorption competing against single and sequential multiphoton processes,” Nat. Photon. 8, 313–316 (2014). [CrossRef]
6. H. Yoneda, Y. Inubushi, M. Yabashi, T. Katayama, T. Ishikawa, H. Ohashi, H. Yumoto, K. Yamauchi, H. Mimura, and H. Kitamura, “Saturable absorption of intense hard X-rays in iron,” Nat. Commun. 5, 5080 (2014). [CrossRef] [PubMed]
7. H. Yoneda, Y. Inubushi, K. Nagamine, Y. Michine, H. Ohashi, H. Yumoto, K. Yamauchi, H. Mimura, H. Kitamura, T. Katayama, T. Ishikawa, and M. Yabashi, “Atomic inner-shell laser at 1.5-ångström wavelength pumped by an X-ray free-electron laser,” Nature 524, 446–449.(2015)
8. J. N. Clark, L. Beitra, G. Xiong, A. Higginbotham, D. M. Fritz, H. T. Lemke, D. Zhu, M. Chollet, G. J. Williams, M. Messerschmidt, B. Abbey, R. J. Harder, A. M. Korsunsky, J. S. Wark, and I. K. Robinson, “Ultrafast Three-Dimensional Imaging of Lattice Dynamics in Individual Gold Nanocrystals,” Science 341, 56–59 (2013). [CrossRef] [PubMed]
9. M. C. Newton, M. Sao, Y. Fujisawa, R. Onitsuka, T. Kawaguchi, K. Tokuda, T. Sato, T. Togashi, M. Yabashi, T. Ishikawa, T. Ichitsubo, E. Matsubara, Y. Tanaka, and Y. Nishino, “Time-Resolved Coherent Diffraction of Ultrafast Structural Dynamics in a Single Nanowire,” Nano Lett. 14(5), 2413–2418 (2014). [CrossRef] [PubMed]
10. K. H. Kim, J. G. Kim, S. Nozawa, T. Sao, K. Y. Oang, T. W. Kim, H. Ki, J. Jo, S. Park, C. Song, T. Sato, K. Ogawa, T. Togashi, K. Tono, M. Yabashi, T. Ishikawa, J. Kim, R. Ryoo, J. Kim, H. Ihee, and S. Adachi, “Direct observation of bond formation in solution with femtosecond X-ray scattering,” Nature 518, 385–389 (2015). [CrossRef] [PubMed]
11. I. Inoue, Y. Inubushi, T. Sato, K. Tono, T. Katayama, T. Kameshima, K. Ogawa, T. Togashi, S. Owada, Y. Amemiya, T. Tanaka, T. Hara, and M. Yabashi, “Observation of femtosecond X-ray interactions with matter using an X-ray–X-ray pump–probe scheme,” Proc. Natl. Acad. Sci. USA. 113(6), 1492–1497 (2016). [CrossRef] [PubMed]
12. G. Grübel, G. B. Stephenson, C. Gutt, H. Sinn, and Th. Tschentscher, “XPCS at the European X-ray free electron laser facility,” Nucl. Instrum. Meth. Phys. Res. B 262, 357–367 (2007). [CrossRef]
13. A. A. Lutman, R. Coffee, Y. Ding, Z. Huang, J. Krzywinski, T. Maxwell, M. Messershmidt, and H.-D. Nuhn, “Experimental Demonstration of Femtosecond Two-Color X-Ray Free-Electron Lasers,” Phys. Rev. Lett. 110, 134801 (2013). [CrossRef] [PubMed]
14. T. Hara, Y. Inubushi, T. Katayama, T. Sato, H. Tanaka, T. Tanaka, T. Togashi, K. Togawa, K. Tono, M. Yabashi, and T. Ishikawa, “Two-colour hard X-ray free-electron laser with wide tunability,” Nat. Commun. 4, 2919 (2013). [CrossRef] [PubMed]
15. S. Roling, H. Zacharias, L. Samoylova, H. Sinn, Th. Tschentscher, O. Chubar, A. Buzmakov, E. Schneidmiller, M. V. Yurkov, F. Siewert, S. Braun, and P. Gawlitza, “Time-dependent wave front propagation simulation of a hard x-ray split-and-delay unit: Towards a measurement of the temporal coherence properties of x-ray free electron lasers,” Phys. Rev. ST Accel. Beams 17, 110705 (2014). [CrossRef]
16. C. David, P. Karvinen, M. Sikorski, S. Song, I. Vartiainen, C. J. Milne, A. Mozzanica, Y. Kayser, A. Diaz, I. Mohacsi, G. A. Carini, S. Herrmann, E. Färm, M. Ritala, D. M. Fritz, and A. Robert, “Following the dynamics of matter with femtosecond precision using the X-ray streaking method,” Sci. Rep. 5, 7644 (2015). [CrossRef] [PubMed]
17. W. Roseker, H. Franz, H. Schulte-Schrepping, A. Ehnes, O. Leupold, F. Zontone, A. Robert, and G. Grübel, “Performance of a picosecond x-ray delay line unit at 8.39 keV,” Opt. Lett. 34, 1768–1770 (2009). [CrossRef] [PubMed]
18. W. Roseker, H. Franz, H. Schulte-Schrepping, A. Ehnes, O. Leupold, F. Zontone, S. Lee, A. Robert, and G. Grübel, “Development of a hard X-ray delay line for X-ray photon correlation spectroscopy and jitter-free pump–probe experiments at X-ray free-electron laser sources,” J. Synchrotron Rad. 18, 481–491 (2011). [CrossRef]
19. Y. P. Stetsko, Y. V. Shvyd’ko, and G. B. Stephenson, “Time-delayed beam splitting with energy seperation of x-ray channels,” Appl. Phys. Lett. 103(17), 173508 (2013). [CrossRef]
20. Y. Mori, K. Yamauchi, K. Yamamura, and Y. Sano, “Development of plasma chemical vaporization machining,” Rev. Sci. Instrum. 71, 4627–4632 (2000). [CrossRef]
21. T. Osaka, M. Yabashi, Y. Sano, K. Tono, Y. Inubushi, T. Sato, S. Matsuyama, T. Ishikawa, and K. Yamauchi, “A Bragg beam splitter for hard x-ray free-electron lasers,” Opt. Express 21(3), 2823–2831 (2013). [CrossRef] [PubMed]
22. T. Osaka, M. Yabashi, Y. Sano, K. Tono, Y. Inubushi, T. Sato, S. Matsuyama, T. Ishikawa, and K. Yamauchi, “Fabrication of ultrathin Bragg beam splitter by plasma chemical vaporization machining,” Key. Eng. Mater. 523–52440–45 (2012). [CrossRef]
23. T. Hirano, T. Osaka, Y. Sano, Y. Inubushi, S. Matsuyama, K. Tono, T. Ishikawa, M. Yabashi, and K. Yamauchi, are preparing a manuscript to be called “Fabrication of strain-free channel-cut crystal X-ray monochromators using plasma chemical vaporization machining.”
24. K. Tono, T. Togashi, Y. Inubushi, T. Sato, T. Katayama, K. Ogawa, H. Ohashi, H. Kimura, S. Takahashi, K. Takeshita, H. Tomizawa, S. Goto, T. Ishikawa, and M. Yabashi, “Beamline, experimental stations and photon beam diagnostics for the hard x-ray free electron laser of SACLA,” New J. Phys. 15, 083035 (2013). [CrossRef]
25. H. Yumoto, H. Mimura, T. Koyama, S. Matsuyama, K. Tono, T. Togashi, Y. Inubushi, T. Sato, T. Tanaka, T. Kimura, H. Yokoyama, J. Kim, Y. Sano, Y. Hachisu, M. Yabashi, H. Ohashi, H. Ohmori, T. Ishikawa, and K. Yamauchi, “Focusing of X-ray free-electron laser pulses with reflective optics,” Nat. Photon. 7, 43–47 (2012). [CrossRef]
26. T. Osaka, T. Hirano, M. Yabashi, Y. Sano, K. Tono, Y. Inubushi, T. Sato, K. Ogawa, S. Matsuyama, T. Ishikawa, and K. Yamauchi, “Development of split-delay x-ray optics using Si(220) crystals at SACLA,” Proc. SPIE 9210, 921009 (2014). [CrossRef]
27. K. Tamasaku, Y. Tanaka, M. Yabashi, H. Yamazaki, N. Kawamura, M. Suzuki, and T. Ishikawa, “SPring-8 RIKEN beamline III for coherent X-ray optics,” Nucl. Instrum. Meth. Phys. Res. A 467–468, 678–681 (2001).
28. M. Yabashi, H. Yamazaki, K. Tamasaku, S. Goto, K. Takeshita, T. Mochizuki, Y. Yoneda, Y. Furukawa, and T. Ishikawa, “SPring-8 standard x-ray monochromators,” Proc. SPIE3773 (1999). [CrossRef]
29. S. Matsuyama, H. Mimura, H. Yumoto, Y. Sano, K. Yamamura, M. Yabashi, Y. Nishino, K. Tamasaku, T. Ishikawa, and K. Yamauchi, “Development of scanning X-ray fluorescence microscope with spatial resolution of 30 nm using K-B mirrors optics,” Rev. Sci. Instrum. 77, 103102 (2006). [CrossRef]
30. J. Amann, W. Berg, V. Blank, F.-J. Decker, Y. Ding, P. Emma, Y. Feng, J. Frisch, D. Fritz, J. Hastings, Z. Huang, J. Krzywinski, R. Lindberg, H. Loos, A. Lutman, H.-D. Nuhn, D. Ratner, J. Rzepiela, D. Shu, Yu. Shvyd’ko, S. Spampinati, S. Stoupin, S. Terentyev, E. Trakhtenberg, D. Walz, J. Welch, J. Wu, A. Zholents, and D. Zhu, “Demonstration of self-seeding in a hard X-ray free-electron laser,” Nat. Photon. 6, 693–698 (2012). [CrossRef]
31. L. B. Fletcher, H. J. Lee, T. Döppner, E. Galtier, B. Nagler, P. Heimann, C. Fortmann, S. LePape, T. Ma, M. Millot, A. Pak, D. Turnbull, D. A. Chapman, D. O. Gericke, J. Vorberger, T. White, G. Gregori, M. Wei, B. Barbrel, R. W. Falcone, C.-C. Kao, H. Nuhn, J. Welch, U. Zastrau, P. Neumayer, J. B. Hastings, and S. H. Glenzer, “Ultrabright X-ray laser scattering for dynamic warm dense matter physics,” Nat. Photon. 9, 274–279 (2015).
32. Y. Feng, R. Alonso-Mori, V. Blank, S. Boutet, M. Chollet, T. Brandt van Driel, D. M. Fritz, J. M. Glownia, J. B. Hastings, H. Lemke, M. Messerchmidt, P. A. Montanez, A. Robert, J. Robinson, L. Samoylova, Yu. Shvyd’ko, M. Sikorski, H. Sinn, S. Song, V. N. Srinivasan, S. Stoupin, S. Terentiev, G. Williams, and D. Zhu, “Recent development of thin diamond crystals for X-ray FEL beam-sharing,” Proc. SPIE 8778, 87780B (2013). [CrossRef]
