An Efficient Algorithm for Automatic Structure Optimization in X-ray Standing-Wave Experiments

Karslıoğlu, Osman; Gehlmann, Mathias; Müller, Juliane; Nemšák, Slavomír; Sethian, James A.; Kaduwela, A. P.; Bluhm, Hendrik; Fadley, C. S.

doi:10.1016/j.elspec.2018.10.006

Cited by 15 publications

(10 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The other layers were not varied independently to prevent overfitting, after checking the lower layer consistency by the STEM results, which were within experimental error. Fitting was done using a standard R-factor, least squares, comparison of experiment and theory, and also making use of a new, more rapid and accurate search algorithm based on a versatile Black Box Optimizer [39]. The total R factor is the sum of each least squares fit of the YXRO computed and normalized RC with the relevant normalized experimental RC.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Hard x-ray standing-wave photoemission insights into the structure of an epitaxial Fe/MgO multilayer magnetic tunnel junction

Conlon

Conti

Nemšák

et al. 2019

Journal of Applied Physics

View full text Add to dashboard Cite

The Fe/MgO magnetic tunnel junction is a classic spintronic system, with current importance technologically, and interest for future innovation. The key magnetic properties are linked directly to the structure of hard-to-access buried interfaces, and the Fe and MgO components near the surface are unstable when exposed to air, making a deeper probing, non-destructive, in-situ measurement ideal for this system. We have thus applied hard x-ray photoemission spectroscopy (HXPS) and standing-wave (SW) HXPS in the few keV energy range to probe the structure of an epitaxially-grown MgO/Fe superlattice. The superlattice consists of 9 repeats of MgO grown on Fe by magnetron sputtering on an MgO (001) substrate, with a protective Al 2 O 3 capping layer. We determine through SW-HXPS that 8 of the 9 repeats are similar and ordered, with a period of 33 ± 4 Å, with minor presence of FeO at the interfaces and a significantly distorted top bilayer with ca. 3 times the oxidation of the lower layers at the top MgO/Fe 2 interface. There is evidence of asymmetrical oxidation on the top and bottom of the Fe layers.We find agreement with dark-field scanning transmission electron microscope (STEM) and x-ray reflectivity measurements. Through the STEM measurements we confirm an overall epitaxial stack with dislocations and warping at the interfaces of ca. 5 Å. We also note a distinct difference in the top bilayer, especially MgO, with possible Fe inclusions. We thus demonstrate that SW-HXPS can be used to probe deep buried interfaces of novel magnetic devices with fewangstrom precision. INTRODUCTION:Since they were first realized in the late 1980s, multilayer magnetic heterostructures showing the giant magneto-resistance (GMR) effect have led to great leaps in the understanding of coupled magnetic systems and magnetic data storage technology [1,2,3]. Metallic GMR structures were then augmented by multilayer oxide/metal magnetic tunnel junctions (MTJ's) that are now ubiquitous in spintronic devices such as read heads and magnetic random access memory (MRAM) [4,5,6,7,8,9,10]. Intense efforts have been devoted to magnetic nanostructures to explore perpendicular magnetic anisotropy, novel geometries and optimal dopant materials at the heterostructure interfaces, and to high quality epitaxial superlattices to create smaller, faster, and more energy efficient spintronic devices. One classical system is the Fe/MgO/Fe MTJ, with recent developments proving additions of new dopants [11], novel structures [12], or new growth techniques [13] can result in tunneling magnetoresistance (TMR) ratios of >600% [11] or emergent magnetic properties such as layer-by-layer magnetic switching in multilayer superlattices [13].Over the decades of study of the Fe/MgO/Fe system, a few key structural components have been realized to be crucial to TMR and magnetic properties of the MTJ, including layer thickness [14], atomic order [11,15,16], and oxygen concentration [9,10,11,12,14].

show abstract

Section: Resultsmentioning

confidence: 99%

“…We used 10,000 sample calculations, and found a good surface estimation at around 7,000 calculations. The Black Box method utilized [39] avoids local minima that often arise with other fitting approaches, and has been found to speed up data analysis by 10-100 times.…”

Section: Resultsmentioning

confidence: 99%

Hard x-ray standing-wave photoemission insights into the structure of an epitaxial Fe/MgO multilayer magnetic tunnel junction

Conlon

Conti

Nemšák

et al. 2019

Journal of Applied Physics

View full text Add to dashboard Cite

show abstract

“…Quantitative structural and interface information can be determined from the experimental rocking curves (RCs) by matching them to simulated RCs using YXRO and Black Box Optimizer programs. 8,11 2.2 Sample Preparation ½Si∕Mo 80 multilayer mirrors with a period of 3.4 nm were prepared by magnetron sputtering and used as standing-wave generators. The ½Si∕Mo 80 multilayer mirror terminates with the Si layer on top, and the native SiO 2 was not removed.…”

Section: Standing-wave X-ray Photoelectron Spectroscopymentioning

confidence: 99%

Chemical and structural characterization of EUV photoresists as a function of depth by standing-wave x-ray photoelectron spectroscopy

Conti

Martins

Cordova

et al. 2021

J. Micro/Nanopattern. Mats. Metro.

Self Cite

View full text Add to dashboard Cite

The success in the miniaturization of the electronic device constituents depends mostly on the photolithographic techniques. Recently, to achieve patterning at the sub-10-nm node, extreme ultraviolet (EUV) lithography has been introduced into high volume production. Continued scaling of EUV via increased numerical aperture to achieve nodes at 3-nm and below requires the development of fundamentally new patterning materials and new characterization methods. Current EUV-resist film thicknesses are in the 20-to 40-nm range, and further thickness reduction is required for the next generation. Therefore, interfaces become exceedingly important, and the properties of the resist film would be dominated by top and bottom interfacial effects. X-ray photoelectron spectroscopy (XPS) combined with standing-wave excitation (SW-XPS), a fairly new method in the EUV lithography field, previously had been largely applied in multilayers and superlattices for characterizing the composition and electronic structure of buried layers and interfaces as a function of depth. We applied the SW-XPS method to organic/inorganic photoresists to provide depth-selective information on their structural and chemical conditions of as a function of temperature, EUV exposure, different underlayers, and other fundamental parameters. As a first attempt, we perform an SW-XPS feasibility study on self-assembled monolayer (SAM) films after exposure to an electron beam. By SW-XPS, we determined that the interface between the Al 2 O 3 underlayer and the SAMs is smooth, with a mean roughness of about 0.2 nm. Moreover, we determined that the SAM chains are, on average, tilted by ∼30 deg off the sample normal. The SW-XPS results also suggest that the SAM is not a perfectly aligned and uniform monolayer, with some areas having thickness higher than a single monolayer. We demonstrated that SW-XPS can provide useful information on ultrathin materials with high potential for being used as a characterization method of organic/inorganic photoresists.

show abstract

“…Many difficult optimization problems have been tackled successfully using surrogate model methods, e.g., in the structure optimization of nanomaterials [36], in cloud simulations [37], in climate modeling [38]; in watershed water quality management [22], and in aerodynamic design [39].…”

Section: Surrogate Models For Efficient Hyperparameter Optimization (...mentioning

confidence: 99%

“…Radial Basis Functions and Stochastic Sampling Our first approach ("RBF approach") to the HPO of learning models is based on RBF models with a cubic kernel. These models have previously been shown to perform well for various optimization problems including problems with integer constraints, see e.g., [22,36,44]. RBFs have the general form…”

Section: Steps 4-9: the Adaptive Sampling Loopmentioning

confidence: 99%

Surrogate Optimization of Deep Neural Networks for Groundwater Predictions

Mueller¹,

Park²,

Sahu³

et al. 2019

Preprint

View full text Add to dashboard Cite

Sustainable management of groundwater resources under changing climatic conditions require an application of reliable and accurate predictions of groundwater levels. Mechanistic multi-scale, multi-physics simulation models are often too hard to use for this purpose, especially for groundwater managers who do not have access to the complex compute resources and data. Therefore, we analyzed the applicability and performance of four modern deep learning computational models for predictions of groundwater levels. We compare three methods for optimizing the models' hyperparameters, including two surrogate model-based algorithms and a random sampling method. The models were tested using predictions of the groundwater level in Butte County, California, USA, taking into account the temporal variability of streamflow, precipitation, and ambient temperature. Our numerical study shows that the optimization of the hyperparameters can lead to reasonably accurate performance of all models, but the "simplest" network, namely a multilayer perceptron (MLP) performs overall better for learning and predicting groundwater data than the more advanced long short-term memory or convolutional neural networks in terms of prediction accuracy and time-to-solution, making the MLP a suitable candidate for groundwater prediction.

show abstract

An Efficient Algorithm for Automatic Structure Optimization in X-ray Standing-Wave Experiments

Cited by 15 publications

References 29 publications

Hard x-ray standing-wave photoemission insights into the structure of an epitaxial Fe/MgO multilayer magnetic tunnel junction

Hard x-ray standing-wave photoemission insights into the structure of an epitaxial Fe/MgO multilayer magnetic tunnel junction

Chemical and structural characterization of EUV photoresists as a function of depth by standing-wave x-ray photoelectron spectroscopy

Surrogate Optimization of Deep Neural Networks for Groundwater Predictions

Contact Info

Product

Resources

About