Modular instrument mounting system for variable ...

0RGXODULQVWUXPHQWPRXQWLQJV\VWHPIRUYDULDEOHHQYLURQPHQWLQRSHUDQGR ;UD\H[SHULPHQWV &0)RONPDQ0-+LJKODQG(3HUUHW6..LP77)LVWHUHWDO &LWDWLRQ5HY6FL,QVWUXP GRL 9LHZRQOLQHKWWSG[GRLRUJ 9LHZ7DEOHRI&RQWHQWVKWWSUVLDLSRUJUHVRXUFH56,1$.YL 3XEOLVKHGE\WKH$PHULFDQ,QVWLWXWHRI3K\VLFV

5HODWHG$UWLFOHV 5HDFWLYHVSXWWHUPDJQHWURQUHDFWRUIRUSUHSDUDWLRQRIWKLQILOPVDQGVLPXOWDQHRXVLQVLWXVWUXFWXUDOVWXG\E\;UD\ GLIIUDFWLRQ 5HY6FL,QVWUXP $FRPSDFWKLJKYDFXXPKHDWLQJFKDPEHUIRULQVLWX[UD\VFDWWHULQJVWXGLHV 5HY6FL,QVWUXP 3ODVPDLRQVRXUFHIRULQVLWXLRQERPEDUGPHQWLQDVRIW[UD\PDJQHWLFVFDWWHULQJGLIIUDFWRPHWHU 5HY6FL,QVWUXP 1HZ3URGXFWV 5HY6FL,QVWUXP 2QOLQHLQVLWX[UD\GLIIUDFWLRQVHWXSIRUVWUXFWXUDOPRGLILFDWLRQVWXGLHVGXULQJVZLIWKHDY\LRQLUUDGLDWLRQ 5HY6FL,QVWUXP

$GGLWLRQDOLQIRUPDWLRQRQ5HY6FL,QVWUXP -RXUQDO+RPHSDJHKWWSUVLDLSRUJ -RXUQDO,QIRUPDWLRQKWWSUVLDLSRUJDERXWDERXWBWKHBMRXUQDO 7RSGRZQORDGVKWWSUVLDLSRUJIHDWXUHVPRVWBGRZQORDGHG ,QIRUPDWLRQIRU$XWKRUVKWWSUVLDLSRUJDXWKRUV

Downloaded 15 Feb 2013 to 164.54.122.182. Redistribution subject to AIP license or copyright; see http://rsi.aip.org/about/rights_and_permissions

REVIEW OF SCIENTIFIC INSTRUMENTS 84, 025111 (2013)

Modular instrument mounting system for variable environment in operando X-ray experiments C. M. Folkman,1,a),b) M. J. Highland,1,b) E. Perret,1 S. K. Kim,1 T. T. Fister,2 H. Zhou,3 P. M. Baldo,1 S. Seifert,3 J. A. Eastman,1 P. H. Fuoss,1 and D. D. Fong1

1 Materials Science Division, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439, USA 2 Chemical Sciences and Engineering, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439, USA 3 X-ray Science Division, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439, USA

(Received 12 December 2012; accepted 29 January 2013; published online 15 February 2013) In the growing field of in operando and in situ X-ray experiments, there exists a large disparity in the types of environments and equipment to control them. This situation makes it challenging to conduct multiple experiments with a single mechanical interface to the diffractometer. Here, we describe the design and implementation of a modular instrument mounting system that can be installed on a standard six-circle diffractometer (e.g., 5021 Huber GmbH). This new system allows for the rapid changeover of different chambers and sample heaters and permits accurate sample positioning (x, y, z, and azimuthal rotation) without rigid coupling to the chamber body. Isolation of the sample motion from the chamber enclosure is accomplished through a combination of custom rotary seals and bellows. Control of the pressure and temperature has been demonstrated in the ranges of 10−6 – 103 Torr and 25◦ C–900◦ C, respectively. We have utilized the system with several different modular instruments. As an example, we provide in situ sputtering results, where the growth dynamics of epitaxial LaGaO3 thin films on (001) SrTiO3 substrates were investigated. © 2013 American Institute of Physics. [http://dx.doi.org/10.1063/1.4791799] I. INTRODUCTION

As the use of synchrotron X-ray sources increases, so does the diversity of scientific problems to which they are applied. Commonly used surface-sensitive approaches include surface X-ray diffraction (SXRD),1 grazing incidence small angle X-ray scattering (GISAXS),2 and total reflection X-ray fluorescence (TXRF),3 among many others. Because of the weakly interacting nature of X-rays, these techniques can probe a wide array of experimental conditions, thus allowing the in operando and in situ study of many important surface and interfacial processes, including thin film growth,4 nanostructure synthesis,5 geochemical processing,6 and heterogeneous catalysis.7 While some measurements require unique instrumentation permanently affixed to the diffractometer, e.g., a load-locked ultrahigh vacuum (UHV) chamber,8 such systems greatly complicate the use of the station for other experiments. For instance, many materials cannot be introduced into a UHV chamber for contamination reasons; also, equipment flexibility in UHV systems is limited by significant vacuum evacuation delays. Hence, for the rapidly expanding synchrotron surface and interface community, the design of an alternative modular system able to handle a broad array of chambers and sample environments on a single diffractometer is highly desirable. We describe a modular instrument mounting system (MIMS) compatible with a surface diffraction geometry tailored for single crystal surface, thin film and interfacial a) Author to whom correspondence should be addressed. Electronic mail:

[email protected].

b) C. M. Folkman and M. J. Highland contributed equally to this work.

0034-6748/2013/84(2)/025111/7/$30.00

studies. The design allows the rapid installation of different chambers and sample heaters, utilizing a differentially pumped rotary seal to permit rotation about the sample normal and x, y, and z translation without mechanical coupling to auxiliary equipment affixed to the chamber. This system could potentially be applied at a number of synchrotron facilities because of its integration with the common Huber six-circle type diffractometer (5021 Huber GmbH). Currently, it is in use at Sector 12ID-D of the Advanced Photon Source and has been utilized for studies of in operando studies of model catalysts and studies of oxygen exchange processes in oxide thin films under controlled electrical and chemical boundary conditions. We demonstrate the capabilities of this instrument with an example of in situ off-axis magnetron sputter deposition of an epitaxial oxide heterostructure.

II. EXPERIMENTAL APPARATUS A. Diffraction geometry

Bright synchrotron X-ray sources have led to the development of a variety of mature X-ray scattering diffraction geometries, which makes it possible to select a configuration that best suits the needs of the experimental station. A well-known geometry typically employed in labsource diffractometers is the four-circle, with three sample degrees of freedom (3S) and one detector degree of freedom (1D), or written more concisely as “3S+1D” in a notation introduced by Evans-Lutterodt and Tang.9 However, studies specific to X-ray surface scattering have produced alternative geometries to the standard four-circle diffractometer

84, 025111-1

© 2013 American Institute of Physics

Downloaded 15 Feb 2013 to 164.54.122.182. Redistribution subject to AIP license or copyright; see http://rsi.aip.org/about/rights_and_permissions

025111-2

Folkman et al.

FIG. 1. General functions of the MIMS shown schematically in two offspecular X-ray scattering geometry diagrams. Sample motions integrated with the MIMS are x, y, z, and χ (a). The sample surface normal is parallel to z and the sample surface or xy-plane defines the X-ray incident (α) and X-ray take-off (β) angles. The scattering vector Q is defined by the incident and final wave vectors ki and kf , respectively, (orange dotted lines). All MIMS motions are independent of a fixed platform, labeled P. This platform can be used to mount a variety of enclosures or chambers that share the MIMS controlled environment (blue dotted lines). The diffractometer sample motions are χ , η and µ; while detector motions are δ and ν (b). The laboratory frame is defined by x0 , y0 and z0 , where y0 is parallel to the incident X-ray propagation direction.

and have become more numerous, for example, the so-called z-axis diffractometer “2S+2D” and the six-circle diffractometer “4S+2D.”9, 10 These geometries have the advantage of two independent detector degrees of freedom, resulting in precise control of the approach angle of the X-ray beam to the sample surface (α) and constant sample illumination, with minimal restrictions on Q [Q = kf − ki , where ki and kf are the incident and exit wave vectors, respectively, as shown in Fig. 1(a)]. The number of sample degrees of freedom is also an important consideration. The advantage of three or more sample degrees of freedom as in the “3S+1D” or “4S+2D” diffractometers is the ability to set the reference vector onthe-fly, where in most cases the reference vector is coincident with the sample surface normal.10 In comparison, the z-axis

Rev. Sci. Instrum. 84, 025111 (2013)

“2S+2D” diffractometer is limited in its ability to set the reference vector and instead relies on accurate positioning between the sample and diffractometer during sample mounting, which is often achieved with a goniometer. The benefit of the z-axis “2S+2D” diffractometer is its relative mechanical simplicity and higher load bearing capabilities, which shifts overall flexibility from the diffractometer to the auxiliary instrumentation. This shift is desirable in more complex in situ type X-ray experiments as illustrated by Fuoss et al.4 for metal organic vapor phase deposition of ZnSe thin films where heaters, gas lines, pumps, and other equipment are required in conjunction with the diffractometer setup. For this reason, we also use a variation of the z-axis diffractometer to implement a modular sample-positioning instrument with integrated environmental control. The variation of the z-axis diffraction geometry used in this work positions the sample approximately horizontally with respect to the laboratory Cartesian coordinate frame shown in Fig. 1(b). This geometry opens a large space near the sample, where auxiliary equipment can be conveniently mounted. The geometry is also conducive to vertical X-ray beam focusing (Pd- or a Pt-coated Si bent mirror). The two detector motions (2D) include δ, set to the take-off angle β of the X-rays from the sample surface and ν, the rotation of the detector about zo in the laboratory frame (useful for offspecular crystal truncation rod measurements). A Pilatus 100 k area detector is mounted on the diffractometer arm.11 The primary sample motions (2S) are comprised of η and χ . In this arrangement, η = 0◦ is defined when the reference vector, or sample surface normal is vertical. Large motions in η are available, such as necessary for specular crystal truncation rod measurements. The χ -circle rotates (±180◦ ) about the sample reference vector (analogous to the ϕ-circle in the four-circle geometry). The details, including exact formulas for transforming between the crystal reference frame and the laboratory reference frame are given by Fister et al.;12 also described therein is the use of a third sample degree of freedom µ [Fig. 1(b)]. Although µ is partially redundant with χ , it can correct for small sample tilt misalignments. The MIMS was integrated with this diffraction geometry to permit a high level of flexibility for installation of equipment and the addition of accompanying sample positioning capabilities. Similar instruments have been implemented by various synchrotron X-ray facilities, notably an interchangeable chamber mounting system by Headrick and Zhou,13 installed at the Cornell High Energy Synchrotron Source and National Synchrotron Light Source. The diffractometer and MIMS described here is distinguishable by the horizontal sample geometry, which allows for the mounting of relatively heavy equipment with reduced stress on the diffractometer. An important attribute of the MIMS in this design is the decoupling of the sample χ -rotation axis from a fixed top platform, while achieving environmental isolation with a custom rotary seal. A schematic illustrating this concept is shown in Fig. 1(a). The instrument is also capable of moving the sample position along the Cartesian coordinates x, y, and z. Here, z is parallel to zo in the laboratory frame when η = 0◦ , and y is parallel to yo in the laboratory frame when η = 0◦ and χ = 0◦ . The z-motion is required for aligning the sample height with

Downloaded 15 Feb 2013 to 164.54.122.182. Redistribution subject to AIP license or copyright; see http://rsi.aip.org/about/rights_and_permissions

025111-3

Folkman et al.

Rev. Sci. Instrum. 84, 025111 (2013)

respect to the X-ray beam, which for experiments with significant pressure and temperature changes, must be frequently scanned and positioned. The x and y sample motions change the X-ray position on the sample without changing the diffraction condition and is advantageous when mapping regions of the sample or mitigating beam damage effects. All motions described are controlled from SPEC software.14

B. MIMS component overview

The design consists of three main sections: (1) a fixed platform with chamber mount (2) a χ -motion and z-motion assembly, and (3) an xy-platform with sample mount. All three can be independently assembled allowing for ease of maintenance. The sections are shown in Figs. 2(a)–2(c). A compact cylinder shape was chosen as the support structure for the instrument because of the high mechanical rigidity per unit mass, lowering its torque and moment of inertia on the diffractometer system (η-circle). The fixed platform provides a vertical mounting surface for equipment and is separated from motions in x, y, z, and χ . To increase working space below the X-ray beam, a two-tier platform was implemented as shown in Fig. 2(a). The large upper tier is bolted to the non-rotating outer-section of the χ -circle (5021 Huber GmbH), while the inner platform (1.5 in. below the top tier) provides a vacuum-compatible chamber-mounting surface and acts as the top section of a custom vacuum rotary seal responsible for decoupling the sample rotation (χ ) from the top fixed platform. The rotary seal consists of a stack of three flexible seals creating two environmentally separate stages. These stages are connected to two separate pumping lines that exit through the top platform to external pumping equipment (Sec. II C 4). The seal design was inspired by Fuoss and Robinson,15 where a custom rotary seal was utilized in a UHV apparatus for X-ray diffraction studies. Directly beneath the top fixed platform are components needed for χ - and z-motion [Fig. 2(b)]. This assembly also includes the bottom section (tube) of the custom rotary seal, completing the opposing vacuum sealing surface. The χ - and z-motion are accomplished with two mating cylinders, a large outer cylinder and smaller inner cylinder. The χ -motion from the diffractometer is coupled to the MIMS with brackets connecting the rotating part of the χ -circle to the large outer cylinder. Therefore, all components intimately connected to the outer cylinder rotate with χ -motion and are shown in Figs. 2(b) and 2(c). Within the outer cylinder is a smaller cylinder that translates along their shared central axis parallel to z. The inner cylinder is constrained by three linear guide rails to restrict the deflection perpendicular to the z-axis. In addition, three ball screws are used to drive and hold the small inner cylinder at a specified z-position. The three ball screws rotate simultaneously by means of a spring-tensioned belt connected to a vertically mounted stepper motor. At the bottom of the MIMS instrument is an xy-stage as shown in Fig. 2(c). This stage consists of two pairs of orthogonal linear rails (Thomson, Inc.) holding a plate for x- and y-motion. The plates are driven at their centerline by stepper

FIG. 2. Model of instrument showing three sub-assemblies: the fixed platform (a), χ -z system (b) and xy-stage (c). Select individual components shown include: brackets (1) securing the upper tier (2) to the lower tier (3); the flexible PTFE seals (5) and outer mating tube (4); the mechanical rotation of the diffractometer χ -circle is coupled with brackets (9) to a large outer cylinder (8); an inner cylinder (7) is translated concentrically in z by three ball screws and belt driven by the z-motor (6). The x-motor (12) and y-motor (11) translate on orthogonal rail pairs where the translational motion is accommodated by bellows (10). A cross section of the instrument (d) illustrates the top ISO-160 mating chamber flange (13) and bottom CF-4.5 in. mating sample flange (14) and dimensions relative to the global Cartesian coordinate system where z0 is normal to the sample surface and y0 is parallel to the propagating X-ray beam at η = 0◦ .

motors. Each plate has a 4.5 in. diameter opening in the center from which a bellows traverses and connects from the base of the rotary seal to the outer plate of the xy-stage. The bellows allows motion of the sample mount in x, y, and z directions and provides a vacuum flange at the underside of the instrument for custom sample mounting flanges. C. MIMS instrument specifications

1. Working space and fittings

Enabled by the MIMS and diffraction geometry is a large working space above the top platform that is important for increasing the variety of equipment that can be integrated with

Downloaded 15 Feb 2013 to 164.54.122.182. Redistribution subject to AIP license or copyright; see http://rsi.aip.org/about/rights_and_permissions

025111-4

Rev. Sci. Instrum. 84, 025111 (2013)

Folkman et al.

the X-ray experiment [see Figs. 1(b) and 6(a)]. A cross section of the assembled instrument is shown in Fig. 2(d); the center of the goniometer lies at the origin of the y- and z-axes, above the two-tiered platform. The top surface of the bottom tier is 3.5 in. below the X-ray beam position (at η = 0◦ ). The top plate surface is 2 in. from the X-ray beam (also at η = 0◦ ) and has a face diameter of 21 in. (not shown). Both tiers of the fixed platform have an array of tapped 14 in.-20 holes and 1/4 in. slip pin holes with 0.5 in. and 1 in. hole spacing for the convenient attachment of non-vacuum equipment. For attachment of a vacuum chamber, the bottom tier has a centered ISO-160 retaining-ring groove and tapped mounting holes (M10×1.5) as labeled in Fig. 2(d)-13. Opposite the X-ray beam side of the instrument is the mount for the sample flange assembly. Here, the mating flange is a 4.5 in. CF with tapped holes as labeled in Fig. 2(d)-14. The sample flange assembly can be easily customized for a specific experiment. Design constraints include the distance from beam position to the sample mating flange, which is equal to 10.25 in. at the midpoint of the z-range [Fig. 2(d)]. Additionally, the diameter of the mounting post that extends along the z-direction must fit through a centerline aperture with a diameter of 1.75 in. The size of this aperture in conjunction with the diameter of the sample mounting post, sets the maximum range of motion in x and y directions. We often employ a mounting post with a diameter of 0.75 in., giving a range of ± 0.5 in. in x and y. If required, the sample mount near the beam position can exceed the 1.75 in. diameter, assuming it can be assembled in two parts (connecting a custom spacer above the fixed platform to a smaller post underneath). This is a good option if using a small auxiliary sample positioner or a goniometer head. 2. XYZ translation

The vertical z-motion is necessary for maintaining alignment of the sample surface with the center of rotation of the diffractometer and is especially critical when operating at grazing incidence angles. This requires a motor with reproducible positioning better than 1 µm. This was accomplished using a z-stepping motor with 0.18◦ /step resolution and an integrated 10:1 gearbox (Oriental Motor, Inc., 44 mm). A ball screw of 1 mm lead was employed (NSK, Inc., 8 mm nominal diameter) resulting in a vertical resolution of 0.5 µm/step. With ×16 micro-stepping, available from the motor driver (Galil, Inc.), the resolution can be increased to