Recent results

Results in 2019

Interface properties of nanoscale layers.  — Our work aimed to study the buried interfaces ex situ in polycrystalline trilayer structures by Mössbauer spectroscopy. In a structurally perfect and chemically sharp layered structure it is only two monolayers of Fe atoms at the interface which “feel the presence of other kind of atoms”, since hyperfine parameters of the Mössbauer atom are basically determined by atoms located in its first two coordination shells even in case of metallic systems. This allows us to reveal differences in width and chemical composition of the bottom and top interfaces of nanoscale Fe layers. For this purpose the Mössbauer parameters of trilayer pairs of equal Fe layer thickness are compared, for example Nb/Fe/Nb and Nb/Fe/Ag pairs. (The layer sequence is given as starting from the bottom, i.e. from the substrate side.)  The Mössbauer parameters of the Fe-on-Nb interface can be derived from the Nb/Fe/Ag samples, since the Ag-on-Fe interface is chemically very sharp and the Mössbauer parameters of two monolayers of Fe atoms at the Fe-Ag interface are well known from the literature. The Mössbauer parameters of the Nb-on-Fe interface can be derived from the comparison of the Nb/Fe/Nb and Nb/Fe/Ag sample pairs of equal Fe layer thickness, as the fitted parameters and the intensity of the different components are compared. For example, if the bottom and top interfaces of the Fe layer are similar, then the hyperfine parameters of the Fe-Nb interface components remain unaltered and the intensity of the Nb-Fe interface sub-spectrum is halved when the top Nb layer is replaced by Ag.

In case of Nb/Fe/Nb vacuum evaporated trilayers on top of Si(111) substrate our Mössbauer analysis indicated that the chemical mixing forms less than 1 nm thick Fe-rich interfaces at both sides of the Fe layer. It is in accordance with our x-ray reflectometry and Rutherford backscattering spectrometry measurements [1], where the evaluations provided slightly larger -  around 2 nm wide and symmetric - interface layers indicating that chemical mixing and geometrical roughness equally contribute to the element distribution across the layers.

(a)	(b)
Figure 1. Schematic structure of the interfaces of Ti/Fe/Ti trilayers in the as-received state (a) and after annealing for 2 hours at 250 °C (b). The color code of the Ti, Fe, Fe-rich magnetic alloy and Ti-rich paramagnetic alloy components is shown in the middle.

The different interface width and alloy composition of the Fe-on-Ti and Ti-on-Fe interfaces was shown to result different evolution of the phase compositions at the two interfaces in Ti/Fe/Ti trilayers (Fig. 1) under heat treatments below 300 °C. At the Ti-on-Fe interface the amount of both the paramagnetic and magnetic alloy components increase, while at the Fe-on-Ti interface the amount of the paramagnetic alloy increases at the expense of the magnetic alloy component. The results are explained by the different thickness of the magnetic alloy at the two interfaces and the consequently different degree of layer formation.

Characterization of protein molecules by ¹H NMR — To characterize aqueous protein solutions, a novel thermodynamic method was applied. It is based on the two-component wide line ¹H NMR signal and provides direct experimental data on the potential energy surface of protein molecules. The temporal behavior of the wide line ¹H NMR signal component coming from mobile hydration water proves clearly that during heating the internal energy changes. The amount of mobile hydration water as a function of temperature gives a melting diagram. Melting diagrams provide special characterization of the proteins.

The usefulness and merits of our approach are demonstrated through four globular structured protein examples . Quantitative conclusions could be deduced about the ratios of the globular (ordered) and the more solvent exposed (disordered) regions of the protein molecules and also about the energy relations of the protein-water interactions. These “globular” proteins are disordered to some definite, though small degree. The extent of disorder ranges from 0.14% to 0.49% in these proteins.

Hydration properties of folded and unfolded/disordered miniproteins were monitored in frozen solution . The amount of mobile water as function of T was found characteristically different for folded, semi-folded and disordered variants. Comparing results of wide-line ¹H-NMR and molecular dynamics simulations, it was found that both the amount of mobile water, as well as the miniproteins’ thaw profiles differ significantly as function of the compactness and conformational heterogeneity of their structure.

Parkinson’s disease is connected with abnormal α-synuclein (αS) aggregation. Energetics of potential barriers governing motions of hydration water is examined . Information about the distributions and heights of potential barriers is gained by a thermodynamical approach. The αS monomer contains 33% secondary structure and is more compact than a random coil. αS monomers realize all possible hydrogen bonds. Oligomers are ordered by 66%. Wild type and A53T amyloids show identical, low-level hydration, and are considered as disordered to 75%.

 

Results in 2018

Non-equilibrium alloys

Magnetic properties of nanoscale Fe-Ag multilayers. — The aim of our work was to give a broad-range map on the variation of the blocking temperature as a function of the Fe-layer (t_Fe) and Ag-layer (t_Ag) thickness and the number (n) of the bilayers in Fe-Ag granular multilayers. The magnetic relaxation of Fe-Ag multilayers with t_Fe in the few monolayers range was first recognized in epitaxial single-crystal multilayers. The relaxation was first attributed to the two-dimensional nature of the ferromagnetic layers but the observation of a linear temperature dependence of the magnetic hyperfine field gave way to explanations based on the island structure of the Fe layers. The deviation of the magnetization of the field-cooled (FC) and zero-field cooled (ZFC) sample undoubtedly indicates a small particle behavior of Fe islands in polycrystalline multilayers. The superparamagnetic properties of such so-called discontinuous or granular multilayers have been the subject of several investigations, similarly to granular alloys prepared by co-deposition. The superparamagnetic properties were shown to be determined by the magnetic anisotropy and the size distribution of the ferromagnetic grains and the strength of the exchange and dipolar interactions between the grains within and between the layers. Our work pointed to the role of the grain-size variation along the subsequent ferromagnetic layers.

The dependence of the blocking temperature (T_B) on the bilayer number is shown in Fig. 1, where solid and open symbols denote two series of samples with t_Ag = 50 and 70Å and T_B is measured by the maximum value of the ZFC magnetization curve. It is evident at a first glance that the scatter of the T_B values is significant, which is basically due to the relatively large error of the nominal layer thickness in the few Å layer-thickness range. In spite of the large experimental errors, one can make two definite observations; (i) the blocking temperature increases monotonically with the bilayer number, showing a tendency to saturation and (ii) the t_Ag= 70 Å points (open symbols) lie systematically below the t_Ag= 50 Å points (closed symbols). Observation (ii) can evidently be explained by the decrease of magnetic interactions between the ferromagnetic layers as the distance between them increases. Dipolar interactions may also contribute to observation (i) as some previous works have already shown in the literature, but the observed changes were smaller.

Figure 1. Blocking temperature (T_B) of (4 Å Fe + t_Ag)_n multilayers as a function of bilayer number for t_Ag = 50 Å (solid symbols) and 70 Å (open symbols). The solid and dashed lines are guides to the eye for the solid and open symbols, respectively.

Another possible cause, namely that the grain size may vary along the multilayer stack, has been demonstrated by our experiment shown in Figs 2 and 3. A composite multilayer pair with bilayer number of n = 20 was fabricated which contained only one Fe-Ag bilayer, the rest being composed of non-magnetic Nb-Ag bilayers, as shown in Fig. 2. The multilayer with the Fe layer on the top has a higher blocking temperature (T_B = 80 K) than that with the Fe layer on the bottom (T_B = 25 K), but the increase is smaller than that what can be seen in Fig. 1. These results suggest that the dipolar interactions between the magnetic layers and the grain-size variation along the layer structure might equally play a role in the relation between T_B and the bilayer number. The respective role of these factors should be further studied.

Figure 2. Schematic view of a special Fe-Ag/Nb-Ag composite multilayer pair with bilayer number of n = 20, containing one Fe layer of 4 Å either in the bottom bilayer (left panel) or in the top bilayer (right panel). The equal thickness of the bottom and top Fe layers was ensured by the simultaneous deposition of the layers.	Figure 3. Magnetization as a function of temperature measured at 10 Oe for the Si/70 Å Ag + 4 Å Fe + (70 Å Ag + 4 Å Nb)19 (bottom) and Si/(70 Å Ag + 4 Å Nb)19 + 70 Å Ag + 4 Å Fe) (top) multilayer. For each sample, the lower curve (with the maximum) belongs to the ZFC condition while the upper curve to the FC condition.

Radiofrequency spectroscopy

Melting diagram of protein solutions and its thermodynamic interpretation — NMR characteristics of frozen aqueous solutions provide direct information on the immobile and partially or fully mobile parts of the molecules. The purpose is the thermodynamic characterization of protein systems. The studies were focused on the proteins ubiquitin (UBQ) and early response to dehydration 10 (ERD10). These proteins are representatives of distinct structural classes, UBQ is a globular protein and ERD10 is intrinsically disordered protein (IDP). The melting diagrams (MDs) of the protein solutions are the amount of the mobile hydration water as a function of temperature or potential barrier for this motion (Fig. 1). The amount of mobile hydration water is measured by the amplitude of the slowly decaying ¹H NMR signal component expressed as a fraction of the total water amount. Temperature is measured as normalized functional temperature T_fn = T/273.15 K (T is absolute temperature) and the potential barrier for the motion of water is E_a = T_fn·6.01 kJ/mol, where T_fn is multiplied with the heat of fusion of water. Hydration is grams water/grams protein.

Figure 1. Melting diagram of proteins ubiquitin (UBQ, green circles) and early response to dehydration 10 (ERD 10, red stars), both dissolved in water (50 mg/mL) and that of frozen water under identical conditions (blue squares).

In aqueous solutions, melting (i.e. beginning of molecular motion) of protein-bound water begins at a much lower temperature than the melting of bulk ice. Each protein has a unique MD that results from its individual thermodynamic characteristics. The MD of globular and ID proteins vastly differ. They can be characterized by temperature-independent sections (globular proteins) or without them (IDPs) or can have a small such section (partly IDP).

For UBQ, at melting (-46 °C), the steep step shows that there are water molecules in the first hydrate shell that are bound almost identically, so the relevant molecular surface is equipotential. The potential field of nearly identical elements resembles the feature of hydrogen bridges. The next wide region is a plateau, in this excitation energy region no new water molecules begin to move, because there are no water molecules that are bound by corresponding energy to the protein. The hydrogen bridges here link the bulk of the molecule to a globule. The water molecules, which are bound to surface areas more accessible to water, begin to move at higher potential barriers. For ERD10, the melting occurs at higher temperature (-42 °C) what it is common in IDPs. The plateau after the melting step is significantly narrower than that observed in globular proteins. Then, a phase of continuous rise in MD is observed indicating that a much larger part of the protein surface is accessible to water in ERD10, in an IDP, than in globular proteins. This means that the solvent-accessible surface of ERD10 is highly heterogeneous energetically which is characteristic to IDPs.

 

Results in 2017

Interface  properties of nanoscale multilayers. — Artificial multilayers and hetero-structures of nanoscale layers are essential parts of information technology as well as many other technical applications. Since these multilayers are prepared by depositing layer by layer the appropriate amount of different atoms, the interface structure is determined by stochastic processes taking place at the actual, continuously changing surface of the structure. This way the width, the crystal structure, the atomic composition and the associated physical properties of the interface can be largely different when A atoms are deposited over a layer of B atoms or, vice versa, when B is deposited over A.

In our study, a significant difference of the Ti-on-Fe and the Fe-on-Ti interfaces are undoubtedly revealed both by molecular dynamics simulations of the layer growth (in collaboration with the Research Institute for Materials Science, Centre for Energy Research, HAS) and by experimental investigations. The calculated concentration depth profiles for different substrate orientations are displayed in Fig. 1. The Fe-on-Ti interfaces are slightly intermixed and the interface broadening depends on the orientation of the substrate, although the variation is small. In the Ti-on-Fe case, no atomic mixing can be seen for the Fe(001) substrate orientation, but the interface is slightly wavy which results in a 0.3 nm intermixing in the concentration depth profile. The largest intermixing takes place in case of the Fe(111) substrate orientation, but the interface width (0.7 nm) still remains below the Fe-on-Ti values. Altogether, from the simulations we can conclude that the intermixing is asymmetric with respect to the interchange of the constituents of the film and the substrate.

The experimental studies were made by conversion electron Mössbauer spectroscopy (CEMS) complemented by cross-sectional transmission electron microscopy (TEM) and X-ray reflectometry (XRR). The Ti/Fe/Ti trilayer samples were prepared by evaporation in high vacuum onto Si single crystal substrate using iron metal highly enriched in the ⁵⁷Fe Mössbauer resonant isotope. In order to determine the phase fractions within the top (Ti-on-Fe) and bottom (Fe-on-Ti) Fe interface, Ti/⁵⁷Fe/Ti/Si and Ag/⁵⁷Fe/Ti/Si sample pairs were compared by exploiting the non-mixing property of Fe and Ag. Analyzing the spectra of Ag/⁵⁷Fe/Ti samples provides information on the Fe-on-Ti interface, since the Ag/Fe interface is chemically sharp and the Ag layer causes only a small and well documented change in the hyperfine parameters of ⁵⁷Fe within the atomic layers nearest and next-nearest to the Fe-Ag interface. Additional spectral components appearing in the Mössbauer spectra of Ti/⁵⁷Fe/Ti samples reveal the specific properties of the Ti-on-Fe interface.

From the Mössbauer results shown in Fig. 2 one can conclude that the bottom and the top interface of a Fe layer in between Ti layers are very different both in the extent and in the ratio of the amorphous and the crystalline alloys appearing at the interface. The Fe-on-Ti interface is about three times thicker than the Ti-on-Fe interface and the ratio of the Fe-rich bcc alloy is larger than that of the Ti-rich amorphous alloy. The Ti-rich amorphous alloy has a higher ratio than the crystalline alloy in the thin Ti-on-Fe interface, which probably does not form a continuous layer.

Surprisingly, the TEM experiments do not show the marked asymmetry of the Fe interfaces, probably due to the averaging of the signal across the specimen thickness in the view direction and a possible modest heating in spite of all the care taken during the sample thinning, but the XRR measurements also indicate a significant asymmetry. The fitted roughness values of 0.86 nm and 0.23 nm result in 2.0 and 0.54 nm “10-90” interface widths for the Fe-on-Ti and Ti-on-Fe interfaces, respectively. The scaling of the interface width may be model dependent, but a strong asymmetry of the Ti-on-Fe and the Fe-on-Ti interface widths is undoubtedly experimentally verified by the Mössbauer and the X-ray reflexivity (XRR) data. The TEM and XRR studies have been carried out in collaboration with the Peter Grünberg Institut, Forschungszentrum Jülich (Jülich, Germany).

 

Results in 2016

Magnetic properties of nanoscale multilayers. — The superparamagnetic nature of ultrathin Fe layers between Ag layers has been investigated for a long time, but the dependence of the superparamagnetic properties on the number of the alternating Fe and Ag layers in a multilayer stack has not yet been studied. Superparamagnetism of a nominally few monolayer thick magnetic layer appears due to the island growth mode of the layer and multilayers of such discontinuous layers are called granular multilayers. The effects of dipole interactions on the superparamagnetic behavior have long been intensively studied, but only a few granular multilayer systems were studied from this point of view experimentally.

We have studied the blocking temperature, T_B, – the temperature above which the relaxation frequency of the magnetic moment exceeds the frequency characteristic to the measurement –  in Fe/Ag granular multilayers with different number (n=1, 2, 5, 10, 20) of Fe/Ag bilayers. T_B was determined from magnetization measurements in a superconducting quantum interference device (SQUID) and it is defined as the cusp temperature of the magnetization measured after cooling the sample from 300K to 5K in zero field. The average size of the Fe granules were estimated from Mössbauer spectroscopy measurements. The role of the dipolar interactions and a perpendicular magnetic anisotropy component in the observed unusually large variation of T_B was investigated by Monte-Carlo (MC) simulations in collaboration with the Democritos Research Center, Athens.

Two representative FC/ZFC magnetization curves are shown for the n=1 and 10 samples in Fig. 1. The T_B values measured in different applied fields are shown for all the samples in Fig. 2. As can be seen from the figures there is a large, almost an order of magnitude increase of T_B as the number of bilayers is increased from 1 to 10. Upon further increasing the number of bilayers (n=20), no change of the blocking temperature is observed.

In case of a non-interacting particle assembly, the increase of T_B can result from the increase of the average grain size and/or the anisotropy energy. Significant changes of these factors for our samples could be ruled out by the Mössbauer spectroscopy measurements. The Monte Carlo simulations indicated that the dipolar interactions between the neighboring layers and the out-of-plane anisotropy should play a decisive role in the observed large variation of the blocking temperature.

Figure 1. Magnetization of granular multi­layers with the indicated number of Fe/Ag bilayers measured after cooling in 10 Oe field (full symbols) and in zero field (empty symbols).

Figure 2. Dependence of the temperature of the maxima of the measured ZFC curves (TB) on the measuring field for different number of Fe/Ag bilayers.