Computational X-ray Microscopy of Functional Materials and Devices

The project CompMic was a collaboration between NTNU (Department of Physics) and USN (Department of Microsystems). The project concerned the use of computers with sophisticated algorithms as an integral part of the imaging process in light- and X-ray microscopy. The project had as an ambition to contribute to the fast development that takes place within microscopy towards higher resolution in space and time, including better contrast, to investigate materials, components, and 3D processes. For example, based on a series of exposures under coherent light, one may derive quantitative phase contrast images. This implies that the phase change of the light that passes through materials can be used for detailed measurements of the physical properties of the material under study. In CompMic it was a specific aim to do highly resolved imaging of surfaces and thin films. Such knowledge is important, for example for developing paint with improved wear resistance and improved optical properties. A highlight from 2018 was an article by Mürer et al published in Scientific Reports about hydroxyapatite (HA) structures in fossil tissue, more specifically a 300-million-year-old tetrapod with the name Discosauriscus Austriacus. We showed, aided by CT based on diffraction contrast, that there is orientation of HA even in such ancient samples. An invited presentation was given at the MAX-IV User Meeting in Lund about X-ray microscopy studies of soft materials. In 2019 we also gave invited presentations at both the Nordic Polymer Days and a «keynote» at «Coherence at ESRF-EBS» in Grenoble. Several conference contributions were given in 2020, including a study of wavelets for image analysis by Hussain. Several MSc projects were finished as part of CompMic, including on the topic of «high-performance computing» (HPC). By using a dome-shaped illumination source we achieved extraordinarily high resolution, cf. Aldabbagh et al (OSA, 2020). In 2021 we published an article (Mürer et al, Scientific Reports) concerning the tissue structure in the transition zone between bone and cartilage, studied using X-ray microscopy based on diffraction contrast. We also continued our work on optical microscopy where we have developed new setups for precision measurements of surfaces, partly motivated by industrial collaborations. New algorithms to calculate phase contrast from light microscopy under oblique reflection geometry have been developed (Hasanzade et al, 2021) and presented at international conferences. A highlight from 2022 was a thorough study that describes many of the mathematical and technical details of our Fourier ptychographic microscope (Hasanzade et al, Results in Optics, 2022). This article culminates with a wide field-of-view microscope image (~3 x 5 mm2) with high resolution. Using phase-retrieval algorithms the resolution is significantly better than what the diffraction limit of the microscope conventionally dictates. In 2022 we continued working also with highly resolved polarized Fourier ptychography, published in Optics Express (Gholamimayani et al), see also (Gholamimayani et al, Electronics Letters, 2022). Also in 2023 we have published several articles as an integral part of CompMic. We would in particular like to mention the study of glass surfaces, published in Applied Physics Letters (Tekseth et al). This work was carried out in collaboration with SFI CASA and will be continued in new projects. We are very content with Dr. Madathiparambil’s compression studies of clay shales published in Physical Review Applied (Madathiparambil et al, 2023). The main highlight is nonetheless the study of liquid flow in porous media, published in prestigious PNAS (Tekseth et al, 2023). A particularly important contribution of the CompMic project is that it has help building competence in a series of related research efforts that we are involved in. Most importantly, it has been pivotal in seeding new projects as it has allowed us to explore many novel “hot topics” in physical optics. Several Ph.D. students have been associated with the project and will defend their theses in the near future. The project has created many nice and original research results. We see it as a quality mark that Norwegian and European industry have shown strong interest in several of the problems CompMic has addressed.

The CompMic project (financed through NANO2021 – SYNKNØYT) grant agreement was redefined already during the first months after acceptance, because we also received a FRINATEK grant on a related topic. Additional adjustments had to be made because of the rapid strides of the international research front in this competitive field. Yet more changes came about owing to both the corona pandemic and the postdoc draught which made it difficult to find/hire/keep qualified personnel. The project started as planned in 2018. Here is a list of the most relevant main activities as originally planned: M 1. Article on quantitative FP microscopy M 2. Article on FPM applied to surfaces M 3. Article on quantitative computational X-ray microscopy M 4. Article on X-ray microscopy applied to time resolved processes Workshop with invited industry M5. Article summarizing CompMic The milestones about optical microscopy (M1 and M2) have been carried out largely by USN, where two PhD students, a postdoc, and several MSc students have been working associated with CompMic under the guidance of Prof. Akram. These former Ph.D. students are now preparing to defend their theses and are working in Norwegian industry. Two of the former MSc students have continued as Ph.D. students at other institutions. The main optical microscopy study under CompMic at NTNU was an investigations into surface defects in glass, published in reputable Applied Physics Letters (Tekseth et al, 2023). Several other peer reviewed journal and conference papers have been published. X-ray microscopy (M3 and M4) has also been pushed forward by CompMic, as demonstrated by several articles – most prominently with the recent publication in PNAS on drainage dynamics in porous media (Tekseth et al, 2024). Our results from CompMic have been presented at conferences, both nationally and internationally, including a keynote lecture at ESRF. A workshop with industry was arranged at NTNU in May 2023. Amongst the participating companies were Zeiss, Equinor and SINTEF. We currently have a new project running, directly funded by Equinor, which can be considered an offspring of the CompMic project. We are also in contact with other European companies with the ambition of setting up committing research collaborations. These potential collaborations are a result of the fact that industry is evidently highly interested in methods for non-destructively studying dynamic processes and structures buried inside materials and devices. We gratefully acknowledge RCN for financial funding and the officers at RCN for their helpful and positive attitude throughout the project. We conclude that CompMic has been a successful project.

Microscopy has until recently been synonymous with expensive, heavy and bulky hardware. New computational imaging techniques are capable of retrieving a quantitative high-resolution image from a series of low-resolution images captured at slightly different imaging conditions. In the X-ray regime, the lack of high-quality lenses has led to lensless imaging techniques, which rely on numerically retrieving real-space images from coherent diffraction patterns. In this project, we shall further develop computational imaging in the visual and X-ray regimes, exploring the potentialities of working in reflection-geometry. We will construct a visual-light testing setup for oblique-incidence reflection-geometry microscopy algorithms. We further aim to advance time-resolved 3D (=4D) X-ray microscopy towards dose-efficient, fast, reliable and high-resolution operation. The computational imaging variant known as Fourier ptychography, employing objective lenses combined with illumination diversity, has a robustness owing to the hardware image formation that we hypothesize will render this technique superior to other high-resolution X-ray microscopy techniques - in particular for low dose measurements. A top international team with complementary knowhow in the fields of X-ray physics and optics has been assembled, including expertise on image formation and hard X-ray surface diffraction. Literally, the project aims to shed light on the subtle interplay between (X-ray) optical hardware and clever software with the aim of developing readily applicable new microscopy schemes - a topic with profound scientific and societal consequences.