Júlio Criginski Cezar
LNLS

Yb based ternary compounds presenting cubic crystal structure have drawn the attention of the solid state community due to the variety of phenomena they present, mostly related with their heavy fermion behavior. In these compounds Yb sits in a cubic environment, which splits the J=7/2 angular momentum of the 4f13 electrons in three manifolds of energy levels. For more than one decade there is a debate about the ground state of these energy levels, with different experimental techniques arriving to contradictory conclusions. In particular YbInNi4 has been presented as a model system given its stable valence state, but still, the ground state of the Yb crystal field configuration is a disputed issue. We show that simulating the experimental magnetic linear dichroism at the Yb M5 edge (~1500 eV) we can indicate the most probable scenario for the Yb crystal field in this compound. We expect that these conclusions could be applied to other Yb compounds of the same family. Furthermore, this work shows the potential of soft x-ray absorption to the study of crystal field related problems.

February 23^th, at 2:00 pm

Local: Room 50 – Upper floor of the Administrative building (Yellow)
Campus – Centro Nacional de Pesquisa em Energia e Materiais – CNPEM
Rua Giuseppe M. Scolfaro, 10.000 – Guará
Phone: (19) 3512-1010

Artistic representation (oilified) of the phosphate-dependent Glutaminase C (GAC) crystal structure on top of tumor tissue slice stained for this protein

In a recent paper published in PNAS 109, 1092 (2012)*Alexandre Cassago et al. present a collection of results, ranging from tissue and cell culture studies to x-ray crystallography and biochemistry, that collectively suggest that the lesser studied of the three so far identified mammalian glutaminases, GAC, is the key glutaminase isozyme to supply for the increased tumor metabolic needs.

Though at very distinct rates, all cells in a living body grow and multiply. This process, called proliferation, must be tightly regulated in order for a healthy tissue – an organ like the lungs for example – to maintain its correct size and function. This if often accomplished through a carefully controlled event of production and release of proliferation-promoting signals, such as hormones, within a cell and its neighbors.

In order for a eukaryotic cell like ours to reproduce, it must duplicate its macromolecular biomass (genome, proteins, cell membrane) before they go ahead and divide. The sugar glucose is one of the main nutrients that will fuel this progression. Through a series of multistep processes performed by proteins generally termed metabolism, glucose is broken down and/or condensed into new molecules that will result in the generation of both energy (in form of ATP) and the synthesis of building blocks (nucleic acids, other amino acids, lipids) for the assembly of the macromolecules mentioned above.

Tumors are often a big ugly-looking deformed mass of cells that looks nothing like the organ it was removed from. Oncologists know, for quite some time now, that this is in great part due to the deregulation of the proliferation process. Cancer cells grow at their own will, at faster rates and at the expense of everything in its surroundings, like parasites. Therefore, cancer is generally defined as a group of diseases characterized by uncontrolled growth and spread of abnormal cells.

In order to provide for this seemingly uncontrolled growth, the energetic and biosynthetic metabolism of cancer cells ought to be readjusted. Tumors must then capture great amounts of other extracellular nutrients, such as the amino acid glutamine, and quickly and efficiently metabolize them for shunting into the appropriate pathways.

Glutamate production by mitochondrial glutaminase (GA), the first enzyme in glutaminolysis, is a key process for body homeostasis, and a crucial carbon donor for amino acid and lipid synthesis in tumor cells. To date, three GAs have been identified in humans: the Liver-type (or simply LGA), the Kidney-type (or KGA) and Glutaminase C (GAC), a splice variant of KGA (both usually referred to as GLS1).

In this extensive study, the authors present a collection of results, ranging from tissue and cell culture studies to x-ray crystallography and biochemistry, that collectively demonstrate that the lesser studied of the three so far identified mammalian glutaminases, Glutaminase C is the better adapted isozyme to supply for the increased tumor metabolic needs.

They believe that a clear distinction of the molecular and structural specifics of the three isozymes, especially between the two isoforms encoded by the gene gls, Glutaminase C and Kidney-type Glutaminase, is mandatory in the context of both understanding the mitochondrial glutamine-based metabolism of cancer and the future development of target-specific therapeutics.

In order to contribute to this distinction, they have introduced a number of original observations. They first show that although protein levels of the two kidney-type isoforms are increased in tumor tissues versus normal, only Glutaminase C is compartmentalized in the mitochondria, where glutaminolysis takes place. This is indeed surprising, as they both contain the canonical sequence that targets to the mitochondria. It seems this is the first study in the literature where isoform-specific antibodies are used exploring their unique C-termini, which could explain these new findings.

It has been known for several decades now that activity levels of the mammalian glutaminases respond to the presence of inorganic phosphate, though it is the first time that the three known isozymes are comprehensively studied together. The authors performed kinetic analysis of the three, and the outcome clearly shows that Glutaminase C most responsive to increasingly concentrations of the activator inorganic phosphate. This might be of particular importance in the context of glutaminolytic rates in tumors since hypoxic conditions may result in accumulation of this ion the mitochondria.

Furthermore, by solving Glutaminase C crystal structure in three different states, i.e., ligand-free, and either bound to L-glutamate or inorganic phosphate, they offer the structural basis for protein tetramerization-induced lifting of a “gating loop” as essential for the phosphate-dependent activation process. Amongst several new structural observations, they show that phosphate binds inside the catalytic pocket rather than the oligomerization interface, resulting in allosteric stabilization of tetramers and at the same time mediating substrate entry by competing with glutamate, therefore guaranteeing enzyme cycling. A higher tendency to oligomerize differentiates GAC from its splicing isoform and phosphate cycling in and out of the active site tells GAC specifically apart from the liver-type isozyme, known as inhibited by this ion. Besides, they believe the structural information will be particularly valuable in helping the future development of novel target-specific drug-based therapies to fight the aberrant cancer metabolism.

 X-ray diffraction data sets were obtained at beamlines X12-C at NSLS (ligand- free crystals), F-1 at CHESS (phosphate-bound crystals) and W02B-MX2 at LNLS (L-glutamate-bound crystals).  SAXS data for GAC, in the presence and absence of phosphate, were collected at the D01A-SAXS1 beamline at LNLS.

*PNAS 109, 1092 (2012)- Alexandre Cassago, Amanda P. S. Ferreira, Igor M. Ferreira, Camila Fornezari, Emerson R. M. Gomes, Kai Su Greene, Humberto M. Pereira, Richard C. Garratt, Sandra M. G. Dias and Andre L. B. Ambrosio.

For more informations, one can contact: andre.ambrosio@lnbio.org.br or sandra.dias@lnbio.org.br.

January 30, 2012

Focal adhesion kinase (FAK) is a tyrosine kinase that localizes to focal adhesions in adherent cells. Through phosphorylation of proteins assembled at the cytoplasmic tails of integrins, FAK promotes signaling events that modulate cellular growth, survival, and migration. The amino-terminal region of FAK contains a region of proteins termed a FERM domain. FERM domains are thought to mediate intermolecular interactions with partner proteins and phospholipids at the plasma membrane and intramolecular regulatory interactions.

An intramolecular cleft of the FAK FERM domain mediates interaction with sarcomeric myosin. Chemical cross-linking, SAXS and mutational analyses confirm the interaction, and inhibiting the interaction with a peptide activates FAK and promotes cardiac myocyte hypertrophic response.

Under prolonged stress, the heart can change its mass and diameter (i.e. remodeling) to cope with alterations in workload. This results from a variety of physiologic and pathologic conditions such as regular exercise training, hypertension and myocardial infarction. Because adult cardiac myocytes have little proliferative capacity, the structural remodeling of the heart primarily reflects the increase in size of individual myocytes (i.e. hypertrophy) in patterns that are unique to the inciting mechanical stress. In spite of its adaptive nature, in the context of pathological triggers myocyte hypertrophy sets in motion insidious processes that impair normal contractile function and shorten cell survival. These are seminal components in the development of heart failure, a condition in which hypertrophic growth is no longer able to compensate for increased workload.

The fate of cardiac myocytes challenged by sustained mechanical stress has been shown to be dependent on multifaceted cellular programs, which involve the action of diverse signaling  molecules.  A  central  challenge  in  this  context  is  to  define  the  precise regulation of molecular signals triggered by mechanical stress.

Several lines of evidence indicate that Focal Adhesion Kinase (FAK) functions to promote cardiac myocyte hypertrophy1. Accumulated data on FAK show that it is a highly versatile scaffold and tyrosine kinase protein which can act in various subcellular contexts, particularly in modulating cellular functions related to cell adhesion such as migration, proliferation and survival2. These broad range of functions place FAK at center stage in embryonic development, and pathological conditions, including cancer and cardiovascular diseases.

The 125-kDa FAK consists of an N-terminal band 4.1, ezrin, radixin, moesin (FERM) domain, followed by a tyrosine kinase domain and a C-terminal FA targeting (FAT) domain3. While none of these folded domains is unique to FAK, they are specifically modified in FAK resulting in a number of unique features. The FAK FERM domain4 is arranged in a globular clover-leaf structure formed by three independently folded subdomains F1, F2 and F3. The F2 subdomain interacts with the kinase C lobe, leading to an autoinhibited conformation that impedes access of ATP and substrates to the active site [Figure 1]. Partnering with other proteins and lipids are thought to modulate the interaction between the FERM F2 subdomain with the kinase domain and thereby the inactive and active state of FAK5.

Figure 1. Overall structure of autoinhibited FAK including the FERM, linker, and kinase regions. In the autoinhibited state, the FERM domain (blue) binds to the kinase domain (red) through an interaction between the FERM F2 lobe and the kinase C-lobe. A section of the linker that contains the autophosphorylation site Tyr397 (orange) is located between the FERM F1 lobe and the kinase N-lobe (Lietha et al., 2007).

FAK is promptly activated by mechanical stress and initiates intracellular signaling processes that coordinately regulate the hypertrophic response of cardiac myocytes6. However, the molecular mechanisms that regulate FAK activity in cardiac myocytes remain unclear. Previously, the authors reported evidence for a distinct pool of FAK, one that is linked to and may be negatively regulated by an interaction with the C-terminal tail of myosin heavy chain in the sarcomere, the contractile unit of cardiac myocytes7. In a paper that appeared in the January 2012 number of Nature Chemical Biology*, they performed extensive biochemical and structural studies to demonstrate that the FAK FERM domain interact with myosin through a wide cleft on its surface. The studies were extended to show that the interaction with sarcomeric myosin inhibits FAK activity, which was demonstrated to be critical to maintain FAK inactive in cardiac myocytes while they are not being overstretched by mechanical stimuli.

1.         Franchini KG. Focal adhesion kinase – the basis of local hypertrophic signaling domain. Journal of Molecular and Cellular Cardiology. 2011.

2.         Schaller MD.  Cellular functions  of  FAK  kinases: Insight  into  molecular mechanisms and novel functions. Journal of Cell Science. 123,1007-1013, 2010.

3.         Arold ST.How focal adhesion kinase achieves regulation by linking ligand binding, localization and action. Current Opinion in Structural  Biology. 21,808-813, 2011.

.

4.         Lietha D, Cai X, Ceccarelli DF, Li Y, Schaller MD, Eck MJ. Structural basis for the autoinhibition of focal adhesion kinase. Cell. 129,1177-1187, 2007.

5.         Frame MC, Patel H, Serrels B, Lietha D, Eck MJ. The ferm domain: Organizing the structure and function of FAK. Nature Reviews. Molecular Cell Biology.11,802-814, 2010.

6.         Torsoni AS, Constancio SS, Nadruz W, Jr., Hanks SK, Franchini KG. Focal adhesion kinase is activated and mediates the early hypertrophic response to stretch in cardiac myocytes. Circulation Research. 93,140-147, 2003.

7.         Fonseca PM, Inoue RY, Kobarg CB, Crosara-Alberto DP, Kobarg J, Franchini KG. Targeting to c-terminal myosin heavy chain may explain mechanotransduction involving focal adhesion kinase in cardiac myocytes. Circulation Research. 96,73-81, 2005.

*Santos AM, Schechtman D, Cardoso AC, Clemente CF, Silva JC, Fioramonte M, Pereira MB, Marin TM, Oliveira PS, Figueira AC, Oliveira SH, Torriani IL, Gozzo FC, Neto JX, Franchini KG. Ferm domain interaction with myosin negatively regulates FAK in cardiomyocyte hypertrophy. Nature Chemical Biology. 8,102-110, 2012

For more informations, contact kleber.franchini@lnbio.org.br

The superconducting state of solid state systems has been one of the most highly investigated phenomena since its discovery in elemental mercury around 100 years ago, which characterized by zero electrical resistance and the expulsion of an external magnetic field. Since then, many nonmagnetic metallic and intermetallic compounds were found to superconduct. The superconducting transition temperatures (Tc) of these materials were not higher than 25 K. However, since 1986 Tc as high as 140 K was discovered in a new class of antiferromagnetic copper-oxides compounds, known as high-Tc cuprates. It is now well established that by introducing charge carrier doping (via chemical substitution) these insulating ceramics became superconductors. In 2008, new intermetallic Fe-based superconductors were reported with Tc between 25 and 55 K. Like the cuprates, the undoped compounds presented an antiferromagnetic ordering transition and a superconducting state emerges within a finite range of substitutions (see Fig. 1). Though, the magnetic ordering transition is connected with a tetragonal-to-orthorhombic structural phase transition. These novel materials provided a whole new family of compounds to test the theories of the yet unknown microscopic mechanism of superconductivity in magnetic superconductors, like cuprates and heavy fermions. Technologically, they offer a new route to find room temperature superconductors. Interestingly, superconductivity without charge carrier doping was observed for Fe-based compounds, which certainly differs them from the cuprates. These observations made some wonder the role of chemical doping for the superconductivity in these materials.

Figure 1. BaFe2-xCoxAs2 phase diagram from electrical resistivity (!), specific heat (Cp) and magnetic susceptibility (") measurements. TN, TE and Tc are the antiferromagnetic, structural and superconducting transition temperatures, respectively. AFM stands for the antiferromagnetic phase. The shaded areas are guides to the eyes.
 

The most widely investigated Fe-based superconductor system is BaFe2-xCoxAs2, partly due to the availability of high-quality homogeneous single crystals. Since Co has one electron more than Fe it is usually interpreted as electron doping. This terminology brings an implicit assumption, still not verified experimentally by an element-specific probe, that Co substitution is able to tune the Fe electronic occupation. However, recent theoretical calculations predicted that the extra electrons have little effect on the charge density distribution of the rest of the material. Researchers at the Brazilian Synchrotron Light Laboratory (LNLS), UNICAMP, UFPel and UFRGS have investigated the Fe K xray absorption near edge structure (XANES) of BaFe2-xCoxAs2 superconductors, with measurements performed at the XAFS-2 beamline (LNLS). The results have been published recently in Phys. Rev. Lett. (107, 267402, 2011)*. They observed no appreciable alteration in shape or energy position of the Fe edge with Co substitution, indicating no change in the electronic occupation of the Fe ions, providing experimental support to the previous theoretical calculations. Superconductivity in this system may emerge due to bonding modifications induced by the substitute atom that destabilizes the magnetic ground state.

A convenient approximation to investigate the effects of Co-substitution on the electronic structure of BaFe2-xCoxAs2, under density functional theory (DFT), has been the virtual crystal approach, in which the extra nuclear charge of Co in the Fe crystallographic site is averaged out without the need of a supercell [1]. Under this approximation, a shift of the chemical potential with increasing Co content is computed, indicating that the extra Co electron is released to the Fermi sea, qualifying Co substitution as an “electron doping” of the FeAs layers. Angular resolved photoemission spectroscopy (ARPES) experiments show an evolution of the hole and electron pockets of the Fermi surface with x that is consistent with this shift. In fact, the Co substituted system has been widely referred to as electron-doped BaFe2As2. Alternatively, the real space density distribution of the extra d electrons brought by Co substitution cannot be estimated under the virtual crystal approximation. Rather, a supercell approach to DFT, leading to the prediction that the excess d electrons from the impurity are actually concentrated at the substitute Co site [2] with little effect on the charge density distribution of the rest of the material. Whether or not Co substitution is able to charge dope the Fe ions in this system is a major issue that may guide the identification of the mechanism of superconductivity in the Fe pnictides.

XANES is a classic probe to determine element-specific electronic ground states. In the electric dipolar absorption process involved in the Fe K near edge, for instance, a photon-induced electronic transition from the Fe 1s core level to Fe 4p unoccupied states takes place. The energy of core level and end states are modified in distinct ways by local changes in electronic occupation due to the characteristic Coulomb interactions of each level with the doped electron or hole, causing a shift of the threshold absorption energy.

The normalized Fe K edge XANES spectrum μ(E) of BaFe2-xCoxAs2 is given in Fig. 2 (a). Figure 2(b) shows the first derivative spectrum dμ(E)/dE. Six distinct peaks or shoulders are noticed in the spectral region of interest and are labeled as A-F in Fig. 2 (a). To each peak or shoulder in this figure a corresponding maximum and a minimum are identified in the first derivative spectrum of Fig. 2(b) as A'-F' and A''-F'', respectively. The ab initio calculation for the XANES spectrum of BaFe2As2, obtained using the FEFF8 code, and its energy derivative are also shown in Fig. 2 (solid line).

Figure 2. (a) Normalized Fe K edge XANES spectra of BaFe2-xCoxAs2 at room temperature. Prominent peaks and shoulders are indicated A-F. (b) First derivative of the XANES spectra in (a). Derivative maxima and minima

Figure 3 shows the x dependence of the positions of the A', B', C', D' and E' features of the spectra. These results indicate no observable change in the Fe K edge XANES spectra of BaFe2As2 under Co substitution.

Figure 3. Fe K edge XANES spectra first derivative A’, B’, C’, D’ and E’ feature (see Fig. 2) position as a function of Co substitution in BaFe2-xCoxAs2. The solid spline lines and open symbols are the expected red shifts by electron doping, obtained by simulated XANES spectra of Ba1-xLaxFe2As2 (x = 0.00, 0.25 and 0.50) model compounds. The dashed lines indicate the feature position for x = 0.

Previous K edge x-ray absorption spectroscopy studies on electron Nd2-xCexCuO4-y and hole La2-x(Sr,K)xCuO4 doped cuprate superconductors showed a Cu K edge shift as a function of carrier concentration. For the cuprates XANES can indeed probe changes of the local electronic structure of the absorbing atom. In general, positive (negative) shifts are observed for electron (hole) doping. The lack of a Fe K edge absorption threshold energy shift in Co substituted BaFe2As2 implies that Co is not charge doping the Fe ions which are thought to be responsible for the superconductivity.

In order to quantify the expected effects of true Fe electron doping on the XANES spectrum of BaFe2As2, they computed the relative shifts of the simulated B'-E' features in Fig. 2(b) for the model system Ba1-xLaxFe2As2. The calculated energy shifts of these features, as a function of x, are displayed in Fig. 3 as empty symbols and solid spline lines. These shifts were obtained relative to the Fe K edge XANES simulation for pure BaFe2As2. It is clear that a negative shift of some of the features of the XANES spectrum was computed under electron doping, most notably the D' and E' features. The contrast between the observed constant positions of the D' and E' features under Co substitution and the calculated negative shifts of up to ~ 1 eV for x = 0.50 electron doping (see Fig. 3) unambiguously demonstrates that Co substitution does not induce Fe valence changes in BaFe2As2.

The rigid Fe valence under Co substitution raises the question on the nature of the perturbation that weakens the magnetic ground state allowing the emergence of superconductivity in BaFe2-xCoxAs2 and other related systems. The researchers argue that despite not provoking significant local electronic charge instability in the Fe ions, Co substitution does alter the local atomic structure. In a previous work [3] they showed a slight decrease of the As-(Fe,Co) bond length, which is possibly connected with a reduction of the Fe local moments. Also, they state that the Co ions have enhanced hybridization with As with respect to Fe, which may further increase the elastic energy penalty of the orthorhombic distortion associated with the magnetic order and help destabilizing the magnetic ground state with respect to the superconducting one.

The authors acknowledge the financial support of FAPESP and CNPq. For more details, you can contact: eduardo.bittar@lnls.br

[1] A. S. Sefat et al., Phys. Rev. Lett. 101, 117004 (2008).
[2] H. Wadati et al., Phys. Rev. Lett. 105, 157004 (2010).
[3] E. Granado et al., Phys. Rev. B 83, 184508 (2011).

* Phys. Rev. Lett. 107, 267402 (2011) E. M. Bittar, C. Adriano, T. M. Garitezi, P. F. S.
Rosa, L. Mendonça-Ferreira, F. Garcia, G. de M. Azevedo, P. G. Pagliuso, and E. Granado.

Ricardo Donizeth dos Reis
LNLS/UNICAMP

Este seminário será dividido em duas partes:

Na primeira apresentaremos um trabalho sobre as propriedades eletrônicas e magnéticas do composto Eu0.5Yb0.5Ga4 como função da pressão externa. Foram realizadas medidas de XANES e XMCD nas bordas Eu- L3 e Yb L3 com a aplicação de pressões de até 30GPa. Nossos resultados mostram que a pressão provoca uma mudança na valência de ambas as terras raras. Também observamos que para pressões maiores do que 8GPa existe mudança do tipo de ordenamento magnético do composto, passando de um estado antiferromagnético para ferromagnético. Com o auxílio de cálculos de primeiros princípios (DFT) nós observamos que a hibridização entre os orbitais do Eu-5d e Ga-3p é a principal responsável por determinar qual o tipo de interação magnética deste composto.

Na segunda parte iremos descrever o projeto de doutorado em andamento no LNLS sobre o estudo da valência e do magnetismo orbital nos compostos da família RT2X2 (R=Eu, Yb e T=Re, Ir, Pt e X=P e Si). Estes compostos são conhecidos por apresentarem características bastante peculiares como: polimorfismo, diferentes tipos de ordenamento magnético e transições de valência em função da temperatura. Nosso objetivo neste estudo é investigar, como os estados de valência da terra rara e o magnetismo orbital dos metais 5d variam como função da contração da rede cristalina sob o efeito de uma pressão externa.

19 de Janeiro de 2012, Quinta-feira, às 14 horas

Local: Sala 50 – Andar superior do Prédio Amarelo (Administrativo)
Campus – Centro Nacional de Pesquisa em Energia e Materiais – CNPEM
Rua Giuseppe M. Scolfaro, 10.000 – Guará
Telefone: (19) 3512-1010

Raios X de altas energias, entre 40 a 60 keV, foram utilizados com sucesso na determinacao de estruturas moleculares de proteinas. A utilizacao de raios X de comprimento de ondas curtas (0.317 a 0.207 A) minimizam danos de radiacao em cristais organicos, e abrem a possibilidade entre outras da utilização de elementos pesados para extensão de fases e cristalografia de alta pressão. Mapas de densidades eletrônica de alta qualidade e resolução foram obtidas utilizando-se detetores convencionais e programas de análise padrão. A estrutura de macromoleculas biológicas conhecidas e novas foram determinadas por um ou uma combinação dos métodos disponíveis atualmente: substituição molecular (molecular replacement; MR), substituição isomorfa (molecular replacement; MIR, SIR) ou dispersão anômala (multiple or single anomalous diffraction; MAD, SAD). Nesta apresentação discutiremos os desafios para a implementação de uma linha dedicada para a cristalografia de proteínas de altas energias e os passos de pesquisa e desenvolvimento necessárias para que essa seja possível. 

Local: Sala 50 – Andar superior do Prédio Amarelo (Administrativo)

Campus – Centro Nacional de Pesquisa em Energia e Materiais – CNPEM

Rua Giuseppe M. Scolfaro, 10.000 – Guará

Telefone: (19) 3512-1010

Nowadays, extensive efforts have been devoted in developing new classes of powerful antibacterial agents due to the epidemics caused by different pathogenic bacteria. Within this context, nanomaterials have arisen as new promising antimicrobial agents due to their high surface area to volume ratio as well as their unique chemical and physical properties. Among inorganic structures, silver nanoparticles (AgNP) are well-known for their antibacterial properties and several works have been dedicated to explain this phenomenon.

Due to antibacterial properties of AgNPs, these structures have been used to prevent bacterial growth in wound dressings, textile fabrics and wood flooring. In particular, this irreversible bacterial inhibition is suitable to prevent bacterial colonization on silver-coated medical devices, such as catheters, where bacteria-killing activity is highly desired.

The relationship between the biological activity and size/shape of AgNPs has been studied. As a general trend, it has been found that smaller particles generally present an enhanced antibacterial power if compared with their larger counterparts. However, the antibacterial efficacy of AgNPs has been compared taking into account distinct reactions to synthesize nanoparticles with different sizes or shapes. Although similar trends have been obtained by different groups, the comparability of the biological activities of nanoparticles obtained from different syntheses is rather limited since the results may be influenced by hidden artifacts carried over from the syntheses procedures.

Mateus Cardoso and his co-workers present the first selective size-fractionation process of silver nanoparticles obtained from the same reaction batch followed by successive bactericidal tests. Separation of AgNPs was achieved through selective precipitation by tuning the amount of surfactant (polyvynilpyrrolidone – PVP) in solution. Thanks to this process, they were able to compare the biological activity of nanoparticles obtained from the same reaction. Along this work, silver nanoparticles obtained in the same reaction batch were size-fractionateed in two distinct populations. The larger silver nanoparticles (LAgNP) tend to aggregate and can be easily separated from the smaller particles (SAgNP) through centrifugation. The aggregated fraction can be redispersed in ethanol afterward with ultrasonic treatment.

Transmission electron microscopy (TEM) was used to investigate the shape and size of fractionated AgNPs. TEM images of small and large colloid particles are shown in Figure 1, in which can be confirmed that their sizes are in the nano range. The overall dimension of SAgNP ranges between 10 and 30 nm, while the diameter of LAgNP vary from 15 to 50 nm.

Figure 1. TEM images of (A) small and (B) large silver nanoparticles.

As microscopy is a local technique and requires specific sample preparation, particle size distribution was obtained through solution small-angle X-ray scattering (SAXS). This technique offers the possibility of rapid determination of particle size and polydispersity due to the high statistical averaging of the nano order structure. In addition, there is no limitation imposed by drying or special care with sample preparation which prevents nanoparticles aggregation. Figure 2 shows the experimental SAXS curves for (A) SAgNP and (B) LAgNP solutions and their corresponding best fittings considering polydisperse homogeneous sphere model.

Figure 2. SAXS patterns for (A) small and (B) large silver nanoparticles and their corresponding fittings (solid line).

Log-normal size distribution of SAgNPs and LAgNPs obtained from SAXS fittings are shown in Figure 3. The obtained distributions are shifted with respect to each other and the maximum for the solution of large Ag nanoparticles is at a greater particle size (20.8 nm) than that for the solution of the smaller ones (15.6 nm). Also, for small silver nanoparticles, polydispersity is larger (σ = 0.3) than for LAgNP (σ = 0.23). This difference may be due to the fact that during centrifugation some of the larger nanoparticles or small aggregates remained in solution.

Figure 3. Particle size distribution of small (red) and large silver nanoparticles (blue) obtained from SAXS fittings presented in Figure 2.

The bactericidal effect of different size-fractionated silver nanoparticles was compared for various microorganisms using the diameter of inhibition zone (DIZ) in the disk diffusion test. In this method, DIZ reflects the magnitude of susceptibility of the microorganism and illustrates, comparatively, the potential antibacterial activity of nanoparticles to different microbial strains. The strains which are more susceptible to antibiotics and/or nanoparticles exhibit larger DIZ, whereas resistant strains present smaller DIZ.

Figure 4 shows images of disk diffusion tests of small and large silver nanoparticles against Staphylococcus epidermidis. No significant differences were observed in the antibacterial effect when the efficacy of either SAgNP or LAgNP was compared across four different bacteria strains. Instead, a substantial discrepancy was noted when SAgNP and LAgNP were directly compared within the same strain. Small Ag nanoparticles presented bigger DIZ than that of large nanoparticles. However, a small inhibition halo could also be seen around papers with LAgNP proving that bigger particles also had antibacterial activity. A quantitative analysis was performed for determining the inhibition zone around the papers for four different bacteria strains. These results can be seen in the original published paper.

Figure 4. Images of disk diffusion tests for small (A) and large (B) silver nanoparticles against Staphylococcus epidermidis. Numbers represent the loaded silver mass (µg) in every paper disk.

The original work was published in the Journal of Materials Chemistry 21, 12267 (2011) by Virginia Dal Lago, Luciane França de Oliveira, Kaliandra de Almeida Gonçalves, Jörg Kobarg and Mateus Borba Cardoso. This manuscript was also featured on the cover of the journal.

All figures were taken from the original article above.

Centro de Desenvolvimento Mineral – CDM – VALE

About Mineral Technology

Since few years ago, GETEK is promoting mineral technology workshops that, by integrating a team of experts from Vale and internationally renowned professionals, promote the dissemination and expansion of knowledge generated within Vale, as well as the acquisition of new knowledge of the global advances of technologies and developments in their state of the art. Thus, continuing this initiative, we are pleased to invite you to the event "Workshop on Applications of Synchrotron Light in Mineralogy and Mining".

Workshop Applications of Synchrotron Light in Mineralogy and Mining

A synchrotron light is an electromagnetic radiation produced by a specialized accelerator of electrons. Once the high-energy electron beam has been generated, it is directed into storage rings under strong magnetic fields in order to convert the high electron energy into light, X-ray or some other form of radiation. The experiments involve probing the structure of matter from sub-nanometer to micrometer and millimeter levels, and the major applications of synchrotron light are in condensed matter physics, materials science, biology and medicine. Applications in mining industry are relatively recent and still little explored in Brazil, with a great potential for growth demanded mainly by multicommodity projects with high mineralogical complexity.

Program
28th of November 2011

8:15 – 8:30 Welcome
- Techniques of spectroscopic characterization -
8:30 – 9:00 Spectroscopy of X-ray Absorption
9:00 – 9:30 X-ray Micro-Fluorescence
9:30 – 10:00 Spectroscopy of photo-emission with UV and X-ray
10:00 – 10:30 Cofee Break
- Techniques of structural characterization -
10:30 – 11:00 X-ray powder diffraction
11:00 – 11:30 Low-angle X-ray scattering
11:30 – 12:00 X-ray micro-tomography
12:00 – 13:30 Lunch
13:30 – 15:00 Tour to LNLS and associated facilities
15:00 – 15:30 Cofee Break
15:30 – 17:30 Important concepts in mineralogy

29th of November 2011
8:30 – 10:00 Case studies of applications in mineralogy
10:00 – 10:30 Cofee Break
10:30 – 12:00 Discussion of tests in Vale samples
12:00 – 13:30 Lunch
13:30 – 15:00 Sirius and new perspectives with Synchrotron Light
15:00 – 15:30 Coffee Break
15:30 – 17:30 Brainstorm about potential applications in Vale BUs

Location: Auditorium – LNLS
Campus – Centro Nacional de Pesquisa em Energia e Materiais – CNPEM
Giuseppe M. Scolfaro St., 10.000 – Guará
phone: (19) 3512-1010

Microelectronics was the standard of high technology during the XXth century last decades. Today, on the other side, photonics is becoming more and more important, with optical fibers, lasers and so on. But the coexistence of these two paradigms creates problems. Microelectronics is almost entirely based on silicon, but this semiconductor does not have good optical properties (it is a very poor emitter of light) and so it is not suitable for photonics. So the interface between the communication technologies based on photonics and the microprocessors based on microelectronics is an active area of research in basic and applied science. One strategy is to search for silicon-based materials that do have the required optical properties simultaneously with electronic characteristics similar to pure silicon, so that they could substitute the latter in the chips. The most promising candidate is beta-iron disilicide (β-FeSi2).

The problem is how to produce β-FeSi2 films with properties suitable for industrial applications. Among the requisites, the samples should have low strain and should be homogeneous, avoiding metastable phases commonly observed in this material. After trying several techniques in the last years, a team of scientists from the Federal University of Minas Gerais, in Belo Horizonte, Brazil, led by Prof. Juan Carlos González Pérez, obtained high quality β-FeSi2 nanodots and discovered a method to control their areal densities, sizes, shapes, crystalline phases and strain.

The “magical” phase transition

The technique used to synthesize the samples was reactive deposition epitaxy (RDE). First, a thin film of iron (thickness from 2 to 12 nm) was deposited on a silicon substrate. Then, the samples were annealed at 700°C for two hours in ultra high vacuum. As a result, iron and silicon formed compounds like -FeSi and β-FeSi2 and agglutinated spontaneously in nanoparticles spread over the silicon surface.

This method alone didn’t produce sufficiently homogeneous nanoparticles (that is, with one of the phases – -FeSi or β-FeSi2 – in much higher proportion than the other). But that could be done by taking advantage of a peculiar physical phenomenon related to a phase transition that happened as the iron thin film built at the beginning of the synthesis process was chosen to be increasingly thicker.

Indeed, the authors observed a relationship between the iron film thickness and several properties of the resulting system. One of them was the nanoparticles shape – more specifically, their aspect ratio (the volume to base area ratio). This property is not distributed uniformly in the material. Far from this: an analysis with atomic force microscopy (AFM) revealed two clearly distinct phases in the nanoparticles, as shown in Fig. 1. In the graphs, the particles are distributed in two sets, with different aspect ratios (the two distinct inclinations of the sets). One can also see that there is a phase transition almost completed when the thickness is about 5.5 nm. Below this limit, the predominant phase has an aspect ratio lower than 1; above, higher aspect ratios are dominant. Interestingly enough, this point coincides with the moment the particles begin to coalesce (Fig 1a-c).

Figure 1 – Left: Atomic force microscopy images of the samples after annealing, with the thickness of the initial iron coverage being of (a) 2.4 nm, (b) 3.8 nm and (c) 5.5 nm. Right: Statistical analysis of the nanoparticle shapes; iron coverage thickness of (d) 2.4 nm, (e) 3.8 nm and (f) 5.5 nm. Two sets of points are distinguishable in each figure, colored with red and blue. As the aspect ratio is the volume to base area ratio, the inclination of the sets is a measurement of this parameter. Low aspect ratio predominates for low thickness iron coverage while high aspect ratio dominates for high thickness. The values inside the graphs d-f indicate the nanoparticle mean volumes.

To investigate what that transition meant, the authors did X-ray diffraction measurements at the Brazilian Synchrotron Light Laboratory (LNLS) in Campinas, Brazil. Analysis of the data could extract information about the concentration of the different crystalline phases (ε-FeSi and β-FeSi2), the strains (along the growth direction) to which the nanoparticles were submitted and the grain sizes along the growth direction. Fig 2 shows that these parameters suffered strong changes around the 5.5 nm limit. These results indicated that the two shape phases in Fig. 1d-f (low and high aspect ratios) correspond to the two crystalline phases, ε-FeSi and β-FeSi2, the latter having high aspect ratio.

The most interesting phenomena happened with the concentrations. As the thickness of the initial iron thin film approached 5.5 nm, the β-FeSi2 concentration rose steeply and reached a maximum at that value, while the ε-FeSi reached a minimum. At 5.5 nm, about 95% of the material consisted of the β phase. Additionally, the strain decreased substantially. The nanoparticle mean size, however, remained relatively small (around 30 nm). This is a very happy combination, because homogeneity, low grain sizes and low strain are properties required from technological applications. So the authors could build β-FeSi2 samples with high quality just by selecting the amount of iron to deposit on the silicon substrate at the beginning of the process of synthesis.

Figure 2 – Analysis of the X-ray diffraction data of the samples as a function of the original iron coverage thickness θFe. (a) Relative concentrations of the ε-FeSi and β-FeSi2 phases; (b) strain along the growth direction; (c) grain size along the growth direction.

Why does it work?

The last step was to understand the physical mechanisms involved in the transition. A model for the relaxation mechanisms in epitaxial layers was developed by Tersoff and LeGoues in 1994[1]. According to it, there are two concurrent influences in the process of nanoparticle formation: (1) a increase of the energy due to surface and interface energies necessary to form a nanoparticle; and (2) a decrease due to elastic relaxation. These two phenomena together form a potential barrier that precludes the nucleation of the nanoparticles with volumes lower than a certain critical value. The Brazilian scientists improved this model so as to describe also the observed phase transition. Just after the volume critical value, ε-FeSi nanodots are energetically more favorable to be formed and they predominate. However, at some particular volume, the situation is inverted and β-FeSi2 becomes energetically more favorable. Beyond this point, this phase tends to be dominant, as can be seen in Figs. 1d-f and 2a.

There may be other mechanisms involved in the process. As the transition coincides with the coalescence of the nanodots (Fig. 1), the interaction between them must play a role, as well as kinetic aspects. Also, the presence of dislocations in the β-FeSi2 nanodots must reduce even more the excess energy of the nanoparticles. These features should be the object of further research that should help to make the method increasingly more efficient.

[1] J. Tersoff and F. K. LeGoues, “Competing relaxation mechanisms in strained layers”, Phys. Rev. Lett. 72, 3570 (1994).

The original work was published in Physical Review B 81, 113403 (2010) by González, J. C.; Miquita, D. R.; da Silva, M. I. N. da; Magalhães-Paniago, R.; Moreira, M. V. B.; and de Oliveira, A. G.

All figures were taken from the original article above, except when indicated.