The imaging beamline, IMX, at LNLS extracts synchrotron radiation from bending magnet D6 with magnetic field of 1.67 T and bending radius of 2.736 m. It has an electron source size of 391 $ \rm \mu m$ x 97 $ \rm \mu m$ and beam divergence of 808 $ \rm \mu rad$ x 26 $ \rm \mu rad$ and was designed to operate in either white beam or monochromatic beam. Monochromatic elements are Si(111) and a Ru/B4C Multilayer pair, located 12 m downstream from the source, and range in energy spectrums from 5 keV to 14 keV. Monochromatic elements can be removed from the beam to allow white light to enter the hutch. Pink beam energy spectrum ranges from 5 keV to 20 keV.
On entering the beamline, the beam is conditioned by white beam slits, before passing through a water-cooled Beryllium window (125 $ \rm \mu m$ thick) positioned in front of the monochromator. Following this, the beam passes through the final Beryllium window which separates the vacuum from the air in the experimental hutch then, a small nitrogen filled ion chamber. The fast shutter controls the dosage on the sample, before the beam passes through a small vacuum section, provided by two Kapton windows (25 $ \rm \mu m$ thick). A set of carefully optimised filters are available, which can be inserted into the beam path to remove low energies. These are particularly useful when imaging heavier or fragile samples where the lower energies either significantly damage the sample or result in beam hardening. The set consists of 4 highly polished Si filters of 200 and 350 $latex \mu$m thickness that can be combined to offset the average energy. Finally, a set of high precision slits defines the beam profile just before the sample.
The sample stage is located a few centimeters downstream of this set of slits. This stage allows for 2-dimensional translation (in the plane perpendicular to x-ray beam propagation) with sub-0.1 micron precision, to allow positioning of the sample within the x-ray beam, or to allow tiling of images. An air-bearing rotation stage, used for tomography scans, sits on the translation stage. Two high precision linear stages, mounted transversely to each other, allowing positioning of the sample over the axis of rotation. A magnetic sample mount is placed on top of the rotation stage, to locate the sample on the stage.
After passing through the sample, the X-rays pass through a thin vitrous carbon glass cover to enter a light-tight camera box. The x-rays reach a scintillator, which produces visible light; the local contact will select from one of 6 available scintillators; scintillator materials include Tb:LSO, LuAG, GGG, and Yag:Ce, of varying thicknesses, each optimized for work at different resolutions. Directly behind the scintillator is a mirror, set at a 45 degree angle. A turret containing long working distance lenses can then be used to image the scintillator onto a pco.2000 CCD camera. The objective is mounted on a stage which controls the image focus. The image recorded by the camera corresponds to a parallel projection of the sample onto the scintillator by the x-ray beam. In tomographic scans, a series of projection images are taken while the sample is usually rotated 180 degrees. The entire camera box move on two stages, which allows the sample-to-detector distance to be varied from a minimum of approximately 1 mm to a maximum of 300 mm and height to be adjusted given the monochromatic beam offset.
The IMX Beamline also has an automatic sample exchange system, which allow the creation of experiment queues, improving the efficiency and automatization of the beamline. This system uses a Mitsubishi RV-2F-D1 robot (CR750-D1 controller) to do the sample exchanging. Sensors were implemented on sample table to give a feedback to the system about the sample existence or not on the position, avoiding problems on queue creation and robot movement.