Atomic Structure-Composition of Materials




The Atomic Structure-Composition of Materials group focus on the study of the atomic structure, atomic composition and defect behavior of nanomaterials, through in-situ TEM, high-resolution TEM, aberration-corrected TEM/STEM, precession microscopy and EELS/EDS techniques. In particular, the group is interested in understanding the relationships between the atomic structure, composition and the properties of nanomaterials, and the fundamental underlying mechanisms of structural and property changes induced by crystalline defects. The material systems of interest include Li-ion oxides for batteries, proton exchange membranes fuel cells, catalyst nanoparticles and nanoscale particles, wires and thin films.


Funding: Portuguese Foundation for Science – FCT


ZnFe Bimetallic Nanoparticles a) Phase contrast Image, b) FFT of a), c) HAADF STEM image d) EDX spectrum images acquired for Zn and Fe.

The development of materials sensitive to oxygen and able to detect it is one of the main aims in the packaging industry, even 2% residual oxygen concentration may alter food organoleptic properties. The presence of oxygen in the packaging, as well as the permeation of oxygen during storage is avoided by using oxygen scavenging materials and controlled by using packaging materials with low permeability to gases. Most of the commercial active agents used on packaging materials are dispersed in the film matrix or/and used as sachet, cards, and self-adhesive labels that can lead to an inefficient activity of the active compound. Thus, the incorporation of the oxygen scavengers in the packing materials avoids the necessity of additional components. However, the use of these materials do not give information of the level of oxygen inside the packaging or avoid the microbial growth during storage that can happens in the presence of oxygen, and thus acting as intelligent and active packaging.

This project addresses the development of a new generation of active and intelligent packaging film for foods through the dispersion of nano-sized multifunctional particles in a bio-polymeric matrix. Enhanced oxygen absorption capacity, as well as high absorption kinetics are expected by alloying Fe and Zn nanoparticles (NPs) by exploiting the larger reduction potential of Fe and by inducing a controlled oxidation of Zn through consumption of oxygen and water present in the package material. After Zn consumption, Fe can also convert into oxide consuming any remaining oxygen inside the package. The nano-sized multifunctional particles confer to the package not only the ability to absorb oxygen permeating from the environment, but also produces ZnO, an antimicrobial agent extensively used in the food industry, that prevents microorganism proliferation. The formation of ZnO is critically dependent on the amount of oxygen and humidity, and is thus automatically controlled by the environment. The increasing number of oxidized NPs in the package induces a change in colour, producing a chromatic effect as a function of oxygen absorbed and antimicrobial agent produced.

Funding: NASA and US Department of Energy

The surface of materials is one of the most important aspects of electrochemistry. The surface is where all critical charge transfers and catalytic interactions occur. For Li-ion batteries, the surface of the electrodes dictates the reactivity with the electrolyte, the ability for Li-ions to shuttle between the bulk and the electrolyte, and the rate at which the ions can transfer, all of which has an effect on rate-capability and cyclability. LiMn2O4 (LMO) and LiNi0.5Mn1.5O4 (LNM) are both promising cathode materials with high energy density and high rate capability, but both are plagued with cyclability problems based on surface effects. In the LMO system the main contributor to cycling degradation is the Mn disproportionation reaction (2Mn3+ = Mn2+ + Mn4+) which creates soluble Mn2+ that is lost to solution. In the LNM system, the redox-active Ni reacts with the electrolyte at the surface of the cathode leading to cathode electrolyte interphase (CEI) formation, which reduces cyclability by creating an increasingly thicker interphase that slows Li diffusion to and from the cathode. Between LMO and LNM, LNM is the more desirable cathode material due to the higher voltage (~4.7 V vs Li0/+ as compared with ~4 V vs Li0/+ for LMO) of the Ni2+/3+ and Ni3+/4+ redox couple with equivalent capacity.


HAADF STEM image of LMO viewed along the [110] zone axis. A blue diamond, indicative of the LiMn2O4 cubic spinel structure, is shown in the bulk while a Mn3O4 phase (red diamond) is visible at the surface. A fast Fourier transform (FFT) of the original full image is included to indicate the crystal.

Since LMO has fewer cations than LNM while still containing the cubic spinel structure, we first studied the LMO cathode system using a combination of high-angle annular dark-field (HAADF) aberration-corrected scanning transmission electron microscopy (STEM) (Figure) and electron energy loss spectroscopy (EELS) to confirm the underlying spinel structure, and we have found, in as-processed LMO, a surface structure composed of Mn3O4 and a lithium-rich Li1+xMn2O4 subsurface layer which occurs as a result of surface reconstruction[1]. We have also identified that oxygen deficiency is the mechanism by which the surface reconstruction occurs.

This research project seeks to study the surface of LNM using a similar approach as that used for the study of LMO and identify the role that Ni plays in the surface reconstruction of LNM. INL is the ideal place to study LNM with their image and probe corrected FEI Titan TEMs, which have HAADF STEM  and EELS capabilities. We will use HAADF STEM techniques to identify atomic surface structures and EELS to identify Ni and Mn oxidation states. With the collected information, inferences will be made as to why and how the surfaces of these spinel cathode materials reconstruct and conclusions may be made about their electrochemical performance once inside a battery.

[1] Nano Lett., 2016, 16 (5), pp 2899–2906

Funding: CNPq and CAPES

With the advancements of nanotechnology, particles can be synthesized in a controlled way, for example with crystallographic morphologies/faces and desired sizes, in order to improve the performance of these materials in a specific applications. Photocatalysis is a promissing technology application in a wide variety of chemical and environmental technologies, for example in the conversion of solar energy into chemical energy and to remove pollutants on liquid and gaseous environmental. In this technique semiconductors are use as photocatalysts and their photocatalytic activity and the mechanism of the photocatalytic reaction are influenced by the crystalline structure, defects and impurities, surface morphology and interface (photocatalyst/environment), among other factors. In this context, nanostructured TiO2 has been prominent in photocatalytic applications where it is a crystal structure, size (surface area) and morphology (exposed facets) are important. The materials studied in this work are synthesized through calcination of trititanate nanotubes (TTNT) previous synthesized by alkaline hydrothermal route. The morphology of the synthesized materials does not always assume a form of equilibrium of its natural phase and is not completely know, being this an important characteristic for a photocatalytic activity.


TEM image from one particle of TiO2 nanorod synthesized at 550°C calcination of TTNT (a) with FFT square (b) and a HRTEM image of the same particle showing two planes and angle between them (c).

The higher the surface energy of the exposed faces of the material, the greater its photocatalytic activity, consequently better results of photocatalysis. The relationship between agent of a police material and its nanostructure can be well elucidated through different electron microscopy techniques. Transmission electron microscopy (TEM) is a more complete technique for a nanoscale characterization of crystal structure and morphology. From this, grain size, morphology, crystallography, chemical composition, phase determination, particle coalescence, etc. can be obtained. The main objective of the project is to use TEM advanced techniques to characterize TiO2-based nanomaterials synthesized from post-heat treatment of TTNTs, allowing a study and understanding of the entire synthesis process, such as nanomaterials and correlating the morphology of each nanomaterial with their photocatalytic properties.

Funding: Portuguese Foundation for Science (FCT), PhD fellowship SFRH/BD/98199/2013

Dental implants are usually fabricated using titanium (Ti) based materials due to its biocompatibility and good corrosion resistance. However, the low capacity to form a strong chemical bond with living tissue, known as bioactivity, is one of drawbacks of Ti dental implants. Tantalum (Ta) and tantalum oxide coatings have been proven bioactive materials and recently proposed to enhance osseointegration and performance of medical devices such as dental implants. Ta is bioactive and exhibits high wettability and high surface energy, which promotes high osseointegration and good corrosion resistance. On the other hand, the higher surface energy of tantalum oxides stimulates the regeneration process in living tissues, and thus, increases the osseointegration efficiency. Also, nanostructured surfaces enhance surface-protein interaction, bioactivity and osteoblast adhesion that play a fundamental role in bone ingrowth.


(a) Top surface (SEM) and (b) cross-section (STEM) of nanoporous tantalum oxide produced by anodization process.

In this work, Ta-based coatings are deposited by DC magnetron sputtering onto Ti CP substrates in an Ar+O2 atmosphere. Moreover, nanostructured anodic tantalum oxide is successfully prepared by electrochemical deposition onto Ta sheets. Ta1-xOx coatings with different oxygen content were successfully deposited by reactive magnetron sputtering. The increase of oxygen content in the coatings change the morphology from columnar to featureless. Likewise, oxygen incorporation inhibits the grain growth and thus the initial formation of a metallic crystalline structure changes to an amorphous phase.  On the other hand, the anodisation process reveals that the electrolyte, composed by H2SO4 and HF, in a 15-30V potential range, allows us to control the Ta interconversion from nanopores assemblies to nanotubes array. The control of the nanostructures (nanopores or nanotubes) size is critical for their final properties.

Funding: US Department of Energy


Example of segmentation results from our state of the art particle picker.

Nanoparticles play important roles in a number of different fields, and many of their properties show a strong dependence on size and shape (i.e., their morphology). There are numerous analytical methods used to characterize their morphologies, and transmission electron microscopy represents a highly attractive option because it allows for the real space visualization of NPs. However, subsampling presents a large source of uncertainty when using this method, as many particle size distributions (PSDs) are drawn from a small sample size typically on the order of 100 particles. Subsampling arises primarily from the complexity of TEM micrographs, which often precludes the automated segmentation and sizing of NPs. The consequent need to manually segment NPs in TEM micrographs represents a bottleneck that must be overcome to address the crucial problem of subsampling. We have developed a state of the art particle picker to address this gap. The Figure above contains an example of segmentation results from this particle picker.

Funding: Portuguese Foundation for Science (FCT), PhD fellowship


Structural analysis of Ataxin-3 aggregation pathway. SDS-resistant elongated mature fibrils form by pathogenic Ataxin-3 visualized by Transmission Electron Microscopy. Cartoon representation of Ataxin-3 globular Josephin Domain (JD) involved in the first Atx3 aggregation step. JD aggregation prone region highlighted in red

Ataxin-3 (Atx3) is one of nine proteins containing an expandable polyglutamine (polyQ) repeat segment that are associated with late-onset human neurodegenerative diseases. Aggregation of the non-pathological and polyQ-expanded Atx3 is well characterized and is critically dependent on early self-assembly events modulated by its globular Josephin domain. PolyQ expansion beyond a certain threshold elicits a second (polyQ-dependent) aggregation step that is critical to mature fibril generation. Biophysical studies unveiled the Atx3 multistep aggregation pathway, but structural data on the nature of the intermediates formed is still lacking. To uncover the structural aspects of Atx3 aggregation intermediates, advanced electron microscopy and optical microscopy approaches will be used.

Funding: US Department of Energy and General Motors

Proton exchange membrane fuel cells (PEMFCs) are promising power sources for transport and stationary applications. Pt and Pt-alloy nanoparticles are currently used as the catalyst to promote the kinetics of the hydrogen oxidation and oxygen reduction reactions in the anode and cathode of the fuel cell, respectively. Yet, the efficiency of PEMFC is largely restricted by the instability of catalyst nanoparticles during fuel cell operation.  Due to their large surface area-to-volume ratio, Pt and Pt-alloy nanoparticles have a strong tendency to grow in size over short time scales, which lead to a reduction in their electrochemically active surface area, and consequently to an undesired catalyst deactivation and reduction in cell performance after several cycles. In this context, it is still challenging to fully understand 1) the interactions between the carbon support and Pt and Pt-alloyed nanocatalysts in the cathode and 2) the degradation mechanisms of the Pt and Pt-alloy nanocatalysts under the fuel cell environment.


HAADF STEM image of Pt on functionalized carbon and corresponding EDS mapping

To address these issues, the first goal of this work is to understand the effects of different parameters, such as the potential cycling profile, temperature, and atmosphere on the degradation of carbon supports. Different kinds of carbon supports, including amorphous carbon, carbon nanotubes and graphene, with the following characteristics: i) with and without functionalization combined with ii) loaded and unloaded with Pt and Pt-alloyed nanocatalysts, will be first studied by high resolution transmission electron microscopy (HRTEM), scanning transmission electron microscopy (STEM) and EDS (Fig.1). Subsequently, the aforementioned samples will be characterized before and after cycling, in the same exact location, using identical location transmission electron microscopy (ID-TEM). In addition, to better understand the interaction between the carbon supports and catalysts, we will perform 3-D TEM tomography. The second goal of this work is to distinguish between the roles played by particle migration and electrochemical Ostwald ripening as degradation mechanisms. In this regard, a special MEMS chip will be designed to be used inside an electrochemical cell TEM holder to perform, for the first time, in-situ TEM experiments of a PEMFC in real time. In summary, the work proposed above will provide a systematic study on the effects of fuel cell operation conditions on the degradation of carbon supports, as well as on the degradation mechanisms of Pt or Pt-alloyed catalysts in the cathode.

Funding: Ferespe, Fundição de Ferro e Aço Lda

Cast duplex stainless steels (DSS) have been applied in highly corrosive conditions, such as marine environments (oil and gas), chemical and petrochemical industry, as well as pulp and paper industry. The superior performance of these material is achieved by a balanced microstructure, consisting of approximately equal amounts of ferrite (α) and austenite (γ) in the as-heat treated condition. In terms of mechanical properties, the balanced microstructure is beneficial to obtain high toughness and high strength, when compared to austenitic stainless steels with same corrosion resistance.

In the most severe applications, DSS grades with high Cr, Ni, Mo and N are required, to ensure the best performance in terms of pitting, stress and inter-crystalline corrosion. Among these DSS, the so-called super (SDSS) and hyper grades (HDSS) contain even higher Cr and N contents, when compared with other DSS grades. As a consequence, these grades are prone to the formation of detrimental secondary phases, in particular sigma (σ) and chi (χ). The destabilization of duplex microstructure is more pronounced on thick areas of casting components. The primary objective of this project is to characterize among several different secondary phases that may form during casting of large DSS components and to establish kinetic models for their formation, by means of correlate the information collected on optical microscopy, as well the scanning and transmission electron microscopy techniques.


BSE-SEM image of DSS containing α, γ, χ- and σ-phases, ascending ordered by brightness levels. On right side, the distribution of principal thermal residual stresses. The maximum stresses are concentrated on interfaces between α (lower coefficient of thermal expansion) and the γ, χ- and σ-phases areas.

Currently, a methodology to identify χ- from σ-phases has been presented, which allow us to measure their volume percent. Therefore, volume percent measurements are critical data to establish the kinetic models of each phase, to tailor the appropriate cooling cycle. The methodology is based on matching between different the grey levels of χ- from σ-phases on BSE-SEM images with semi-quantitative chemical analysis with aid of EDS. Therefore, the semi-quantitative analysis on SEM are compared with measurements by means of TEM-EDS. In fact, TEM plays a key role on characterization methodology, due to electron diffraction capability in order to unequivocally identify χ- and σ-phases. Finally, as the identification is complete, an image segmentation procedure can be applied to BSE-SEM images.

Apart from the kinetics description of secondary phases formation, being able to detect and identify all the phases that may be present on DSS castings allows us to study how their distribution can affect the magnitude and gradient of thermal residual stresses. A finite element method has been used, where BSE-SEM images are inputs and it is possible to assign the elastic properties (Young’s modulus and Poisson’s ration) and coefficient of thermal expansion to individual phases. The application of a thermal gradient, the software calculates maps of distribution of principle stresses. In fact, due to significant differences in terms of coefficient of thermal expansion, thermal residual stresses arise and interfaces acts as concentration areas of stresses.

Funding: Sandia National Laboratories

The mechanical properties of nanocrystalline thin films are strongly related to their grain size according to the Hall-Petch equation. However, in materials with nano grain sizes, rapid and abnormal grain growth may occur when subjected to heating during manufacture or usage, which will decrease their strength. In addition, as grain growth occurs, the local texture may evolve, which also affects the mechanical properties. In this context, the goal of this work is to develop a thorough study on how grain growth and local texture of copper and nickel thin films may be affected by the deposition method, substrate, film thickness and annealing temperature.


Orientation maps combined with reliability maps. (a) as-deposited 30 nm films and annealed for (b) 20min, (c) 40 min, (d) 60 min and (e) 80min. (f) as-deposited 120nm film and annealed for (g) 20min (h) 40 min, (i) 60min and (j) 80 min

In order to determine the grain size and texture in nano grain size materials, traditional techniques such as Electron Backscattered Diffraction (EBSD) can no longer provide the resolution required. Therefore, a technique called Precession Electron Diffraction Microscopy, which provides a spatial resolution of 3-5nm and minimizes dynamical diffraction effects in TEM, will be performed. This technique will allow us to investigate the grain orientation, average grain size and grain boundary information in a fully automated fashion. Furthermore, to improve the Precession analysis, a newly developed method will be used where the diffraction patterns are filtered for noise threshold, spot enhance loop, gamma, spot radius and softening loop. In this fashion, the reliability of the results will be improved. Finally, the orientation images acquired after indexing will then exported to the TexSEM Laboratories Orientation Imaging (TSL OIMTM) software for further filtering.

Two sets of samples will be studied by Precession Electron Diffraction in TEM. The first set of samples correspond to nickel and copper thin films with thicknesses of 30nm and 120 nm deposited on NaCl crystals at room temperature by pulsed laser deposition (PLD) and sputtering. The second set of samples are copper thin films with thicknesses of 30nm, 120nm and 900 nm deposited on Ta/SiO2/Si substrates by using sputtering. The first set of samples will be subjected to an annealing temperature of 350 C for various times to monitor the evolution of grain size and texture. The second set of samples will be subjected to two treatments of 700 C and 900 C and three different times.

In summary, this work will enable us to correlate between grain size, local texture, deposition method, substrate, film thickness and annealing temperature in Ni and Cu nanocrystalline thin films.


Paulo Ferreira


Enrique Carbó-Argibay
Facility Manager

Sebastian Calderon Velasco
Research Fellow

Cinthya Blois
Ph.D Student

Kang Yu
Ph.D Student

More info will be available soon.