For many applications there is a requirement for nondestructive analytical investigation of the elemental distribution in a sample. With the improvement of X-ray optics and spectroscopic X-ray imagers, full field X-ray fluorescence (FF-XRF) methods are feasible. A new device for high-resolution X-ray imaging, an energy and spatial resolving X-ray camera, is presented. The basic idea behind this so-called "color X-ray camera" (CXC) is to combine an energy dispersive array detector for X-rays, in this case a pnCCD, with polycapillary optics. Imaging is achieved using multiframe recording of the energy and the point of impact of single photons. The camera was tested using a laboratory 30 μm microfocus X-ray tube and synchrotron radiation from BESSY II at the BAMline facility. These experiments demonstrate the suitability of the camera for X-ray fluorescence analytics. The camera simultaneously records 69,696 spectra with an energy resolution of 152 eV for manganese K(α) with a spatial resolution of 50 μm over an imaging area of 12.7 × 12.7 mm(2). It is sensitive to photons in the energy region between 3 and 40 keV, limited by a 50 μm beryllium window, and the sensitive thickness of 450 μm of the chip. Online preview of the sample is possible as the software updates the sums of the counts for certain energy channel ranges during the measurement and displays 2-D false-color maps as well as spectra of selected regions. The complete data cube of 264 × 264 spectra is saved for further qualitative and quantitative processing.
We report on a new camera that is based on a pnCCD sensor for applications in scanning transmission electron microscopy. Emerging new microscopy techniques demand improved detectors with regards to readout rate, sensitivity and radiation hardness, especially in scanning mode. The pnCCD is a 2D imaging sensor that meets these requirements. Its intrinsic radiation hardness permits direct detection of electrons. The pnCCD is read out at a rate of 1,150 frames per second with an image area of 264 × 264 pixel. In binning or windowing modes, the readout rate is increased almost linearly, for example to 4,000 frames per second at 4× binning (264 × 66 pixel). Single electrons with energies from 300 keV down to 5 keV can be distinguished due to the high sensitivity of the detector. Three applications in scanning transmission electron microscopy are highlighted to demonstrate that the pnCCD satisfies experimental requirements, especially fast recording of 2D images. In the first application, 65,536 2D diffraction patterns were recorded in 70 s. STEM images corresponding to intensities of various diffraction peaks were reconstructed. For the second application, the microscope was operated in a Lorentz-like mode. Magnetic domains were imaged in an area of 256 × 256 sample points in less than 37 seconds for a total of 65,536 images each with 264 × 132 pixels. Due to information provided by the two-dimensional images, not only the amplitude but also the direction of the magnetic field could be determined. In the third application, millisecond images of a semiconductor nanostructure were recorded to determine the lattice strain in the sample. A speed-up in measurement time by a factor of 200 could be achieved compared to a previously used camera system.
Up to now in a typical transmission electron microscope (TEM) camera, the incident electrons are indirectly detected. They are converted into photons in a phosphorous layer, which are then guided via fiber optics to a CCD or CMOS imager for detection. This results in inherent disadvantages of conversion efficiency, scattering of photons, reflections at optical interfaces and absorption losses. Contrary, this can be avoided in a direct detector which directly converts the incident electrons into a spatially resolved signal. This brings an increase in resolution and sensitivity, minimizing specimen damage. Therefore, in past years various systems compromising direct detectors have been launched, however mostly CMOS imagers [1, 2]. In contrast, we present first results of a new ultrafast TEM camera based on a pnCCD detector [3].A pnCCD is a radiation hard, back-illuminated device, which is fully depleted over the whole thickness of 450µm thick, high-purity n-type silicon, which is also the sensitive region. In this version it features a 48x48µm² pixel size with 264x264 pixels. The multiparallel readout of the pnCCD enables a frame rate of up to 1000 full frames per second offering increased temporal resolution. The sensitivity is such that each incident primary electron can be easily distinguished from noise, including primary electrons that are scattered out of the detector again. With an appropriate low-dose (100-1000 incident electrons per frame), single individual primary electrons can be separately detected, enabling further analysis and processing of these electron events.The final image can be obtained by simply integrating multiple frames and their intensities, which results in a conventional intensity image (called intensity mode). Even more, by using the potential of electron event processing, the image can formed by taking the individual events with their reconstructed point of entry and count them in a subpixel grid. A simple method to determine the point of entry is to calculate the center of gravity of each event (called CoG method). Since this is done with a subpixel precision, the grid for the final image can be chosen much finer than the device pixel size of 48µm (see Figure 1). Additionally, the frames and events used for image formation can be chosen after the exposure to correct for specimen drift and selecting only those images which have been recorded before the specimen was damaged. Also temporary changes of the sample, e.g. oxidation processes, can be nicely observed with a time resolution of 1000 images per second.The pnCCD TEM camera was installed at a TEM (Titan80/300) for evaluation of the imaging capabilities at electron energies of 80, 120, 200 and 300keV. The modulation transfer function was calculated with the slanted knife edge method for the intensity mode and with event processing with no subpixel and 5x5 subpixel grids (see Figure 2). Generally, the contrast increases with lower electron energy. Looking at the higher energies of 200 and 300keV in intensity mode, the contrast fell...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.