Microfabrication technologies of X-ray optical elements

Research output: Book/ReportPh.D. thesisResearch

31 Downloads (Pure)


Over the last 15 years, X-ray phase-contrast imaging has shown to overcome the limitations of traditional X-ray absorption imaging. Amongst other benefits of this technique, we find the imaging of weakly absorbing materials. This method was made possible thanks to better detectors as well as improved microfabrications methods. An X-ray phase contrast setup needs one or several optical elements with high absorption contrast. Ideally, the absorbing element is made of alternated absorbing and transparent areas exhibiting high aspect ratio. Despite important research efforts, the microfabrication of these absorbing elements still suffers from limitations. These limitations are partly caused by the high aspect ratio structures required to selectively absorb the photon energy to provide sufficient signal-to-noise ratio. The goal of this thesis was to provide an understanding of the microfabrication limits, and to developed reliably methods to produce these absorbing elements. Silicon is often chosen as an X-ray transparent mould, into which the X-ray absorbing material is placed. Combining silicon with absorbing metal filling, is a common fabrication process that we have investigated. In addition, we proposed a new manufacturing possibility, involving only the absorbing metal eliminating the silicon mould.
First, we proposed a fabrication method to manufacture absorbing linear gold gratings for X-ray Talbot interferometry imaging. The manufacturing process was straight forward, and consisted of an anisotropic dry etching of silicon and gold electroplating. The highest aspect ratio of the manufactured absorbing elements was approximately 12:1. Perhaps the most critical outcome of this study was the understanding of the plating dynamics. Existing literature related to absorbing gratings rarely reports the plating dynamics in deep trenches, and our investigations demonstrated that gold concentration in the electrolyte is a critical factor to consider.
Second, we investigated the use of a short-pulse UV laser to make micron-sized holes in bulk tungsten to fabricate a two-dimensional absorbing mask for X-ray edge illumination imaging. Tungsten is a metal which has similar X-ray absorbing properties to that of gold, yet is significantly cheaper. The intensity of the laser has direct influence on the diameter of the hole. Thus, by adjusting the intensity, micron-sized holes with an aspect ratio as high as 44:1 could be achieved. Arguably, this method is a serial process and does not scale easily without affecting significantly the processing time. On the other hand, it has the non-negligible advantage of involving very few fabrication processes, and only uses the absorbing material, thus removing the need for a silicon mould.
Finally, we proposed a method to fabricate silicon moulds for two-dimensional gratings. The moulds consisted of pillar arrays made using silicon dry etching. This study is slightly different to that of the two previous, as it does yield to a finished absorbing element, but rather proposes a reliable method to fabricate silicon moulds for a wide range of dimensions and shapes. More particularly, this study examines the silicon micro-grass formation. Silicon micro-grass is a common, often unwanted effect, which occurs principally in large open area with a low aspect ratio. This issue is often addressed by modifying the etching parameters to prevent the grass from forming. Unfortunately, the modification of the parameters quickly becomes a time-consuming task, especially when different etching machines are involved. Instead, we investigated the critical dimensions at which the micro-grass forms, and proposed a generic method, based solely on sacrificial geometries to suppress the grass.
Original languageEnglish
PublisherDTU Nanolab
Number of pages152
Publication statusPublished - 2020


Cite this