Summary

My research grant will end tomorrow... So now I try to overview the achievements of my work in this period.

The period of my grant started in September 2013. It seems to be a long time since then (16 months), but for a researcher, and mostly for an experimentalist, time is going very fast. 16 months in experimental nuclear physics is especially a short term: sometimes we need to wait a whole year to get a beamtime at an accelerator research facility to perform our experiment, and if something happens to this quite complicated machine, we have to wait another year... Something the same happened to us a year ago in Munich as I wrote in a previous entry.

Despite of the technical problems I think this period was very successful. My research work was very productive, which is best represented by my six scientific publications in high-rank journals. I feel that I could also fulfil the requirements of dissemination as well with a couple of informative manuscripts for the open public and with this blog. Besides, I was involved "seriously" in teaching activities of the University of Debrecen. At the Informatics Faculty I was the lecturer of a course with the title "Artificial Intelligence" which is a basic course for engineers. I have also started a new special course "Experimental Planning and Data Analysis".

But the most important achievements of the research grant period is collaboration agreement and research project contract that was signed on 14th of December 2014 by MTA Atomki and ELI-NP. This contract is based on the research project that was initiated and elaborated by my research group here at MTA Atomki.  I will be one of the principal investigators of the research project. I think this research project contract both ensures our scientific contribution in the EU funded ELI-NP project in a long term and also grant a nice research fund for our scientific work.

As a summary I think that this research grant gave me an excellent support for my research and also for all the other activities during this period. This grant has a large contribution in the realization of the collaboration between ELI-NP and MTA Atomki. The benefit of having a formal collaboration with world-leading research facility is obvious....

The fission barrier of Neptunium-238

In a recent entry I have already introduced our measurement on Neptunium-238 and the first steps of the data analysis. In the last couple of weeks I could proceed so now I want to write some details about it

a) Fission probability of Neptunium-238 in the function of excitation energy. b) Sharp resonances in the fission probability and its rotational structure

In Figure (a) the data points of our measurement are indicated by blue open triangles, and open circles represent a previous, low resolution measurement. The continuous red line indicates the result of our model calculations, in which we altered the free parameters (actually the fission barrier parameters) until the theoretical line fits to the data points. As a result we got a quite good description with a triple-humped fission barrier having barrier heights of 5.5 MeV, 6.2 MeV and 6.3 MeV.

 

This result shows that the structure of Neptunium-238 and the previously examined Protactinium-232 (both odd-odd nucleus) is quite similar. The fission resonance structure which is shown in Figure (b) also supports this similarity: these resonances are in the same energy region as in the case of Protactinium-232. Actually, this means that their 3rd minimum in the potential barrier have the same depth.

The manuscript on these results have already been accepted by Acta Physica Polonica B for publication, and will be published expectedly in April.

 

Design and simulation

As I presented in previous entries, we develop two different detector arrays for our research projects at ELI-NP. The first step of the development is planning, which means we sketch our idea on paper according to the requirements and the scientific goals of the project. We have to decide the dimensions, the geometry and the materials. During the planning we also have take our financial possibilities into consideration. It also a great advantage if the tolerance of the mechanical and electric parts allows to use the infrastructure of our institute (e.g.: the mechanical and electronic workshop).

A very important step is to estimate the response of the detector to the investigated phenomenon and also to estimate the possible background signals having particular physical conditions. For these estimations we perform simulations with a software environment developed at CERN (GEANT4).

It is especially important to perform such simulations for the Bragg ionization chamber. For this we have to know all the relevant environmental conditions of the experiments. The time structure of the photon beam seems to be particularly important. And we are going to have serious technological challenges due to the time structure of the beam at ELI-NP: photons will come in little bunches and not as a continuous beam. We are going to have 100 big bunches in one second each having 32 small bunches. The time interval of one small bunch is 1-2 picosecond (!!), in which we get 100000 photons. The problem is obvious: reactions induced by individual photons will "overlap" since the time resolution of our detector limited resulting in a distortion of the energy measurement. This background we have to minimize.

According to our very recent GEANT4 simulations we can minimize the effect down to 1% of the effective signal if we design a chamber taking some consideration on the materials and dimensions. For example we have to use entrance and outgoing window on the wall of the chamber for the beam covered by a very thin (Kapton) foil in order to minimize the scattering of photons.

Foton optics at very high energy

In experimental nuclear physics quite often we need to localize very precisely the position of the nuclear reaction within the target, or, equivalently, the place of origin of the emitted reaction products. For that it is elementary to reduce the size of the target and the beam diameter as much as possible. The later can be done by focusing the beam.

At a conventional particle accelerator the charged particle beam can be focused on the target by quadrupole magnets. Submillimeter beam diamater can be easily achieved. For the focusing of photon beams one can use optical lenses: the focal distance of the lense is determined by the index of refraction of its material (n), its thickness at the middle (d) and the two radius of curvature (R1,R2). However, the index of refraction depends on the photon energy: n(E) = 1 − δ(E) + iβ(E), where iβ(E) stands for the absorption, and δ(E) drops very fast with energy! So at very high energy the focal distance (f) is divergent: for X-rays typically δ=10^-5 - 10⁻⁶, and in the γ energy region δ → 0. By increasing the number of lenses f can be reduced since f=n/(2Nd), where N is the number of lenses (see Figure). But for the energetic γ photons it does not help...

Array of lenses for focusing X-rays

Very recently a series of experiment performed by a group of physicists from the Ludwig Maximilians University and from Institut Laue-Langevin showed signatures on a deviation from the above mentioned tendency. The index of refraction can be different from 1 even for γphotons! A possible explanation is based on the so-called Delbrück scattering, which means an interaction between the photons and the electron-positron pairs those appear in the vicinity of nucleus with high atomic numbers.

We hope that in the near future already at ELI-NP the γ beam can be focused by using this effect with lenses having high atomic numbers (e.g. gold lenses).

Conference in Zakopne

I am going to participate the fourth time now at the International Nuclear Physics Conference in Zakopene (Poland), which has been organized every second year. Conferences, organized by the Polish physical society, are always pretty good due to the Polish hospitality (especially towards the Hungarians) and the high scientific quality. From these the Zakopane Conference is one of the best: the site is located in the middle of the High Tatras... Participants and organizers are neither hiding one of main goal of the conference: hiking at the mountains discovering the beauty of the nature.

At the conference I am going to give a presentation about the topic and the scientific results discussed in my last entry. Before the conference I have lot of questions to myself: who will be there? In which field they are professionals at? Can I give a presentation that can be understood by everyone? Will be any relevant questions, comments, remarks after the talk?

Anyway, the weather seems to be perfect so my hiking boot goes to my bag...

Summertime...

... in our research institute is usually the time for maintenance so a researcher must have a quite long break in his experimental work. This is the time for a little rest, but also for have an overview on the latest publications in the field, and for having a "look" at the data of previous experiments...

Very recently we performed an experiment at the Tandem Accelerator of the Technical University (Munich). In this experiment we measured the fission probability of 238Np as a function of the excitation energy by employing the 237Np(d,pf) reaction. However, the data analysis of this experiment is complicated by the very large level density which originates from the odd-odd nature of this nucleus: it has an odd number of neutrons and protons. Now, I have the time for the careful analysis of the data of this experiment...

Fission probability of 238Np as a function of excitation energy (blue spectrum), and a zoomed spectrum on the individual resonances.

Investigation of the photofission of Thorium

As a first step of our experimental program, we now plan to perform high resolution transmission resonance spectroscopy of ²³²Th via photofission to resolve, for the first time, the isomeric shelf at lower (E=4-4.5 MeV) energies. After resolving the isomeric shelf into individual resonances, one has to identify and assign them to the corresponding minimum in the potential energy surface. One characteristic property of the octupole- and quadrupole-deformed 3^rd minimum is the alternating-parity bands (0⁺,1^-,2⁺,3^-,…), while in the 2^nd minimum, which is only quadrupole deformed, the mass symmetric and asymmetric bands separately appears with a structure of (0⁺,2⁺,4⁺,...) and (1^-,3^-,5^-,...). Since the sub-barrier photo-excitation is very spin-selective, we can expect to observe always doublets of 1^- and 2⁺ states with a spacing of about ~6 keV in the 3^rd minimum, providing the moment of inertia, while in the 2^nd minimum, completely independent 1^- and 2⁺ resonances are expected. This represents a very clear criterion to discriminate between resonances in the 2^nd and 3^rd minimum. Therefore, in these studies, the measurement of the fission fragment angular distributions is essential in order to identify the spins, the parity, and the K values of the resonances. We can exploit the fact that the electromagnetic interaction follows well-established selection rules, leading to the very spin-selective photofission (1^-,1⁺,2⁺) together with typical angular distributions due to the Bohr's picture of transmission resonances with well-defined values of spins.

We can also determine the spin-dependent inner barrier heights for dipole and quadrupole excitations E_A(1-) and E_A(2+), respectively, and the depth E_II of the 2^nd minimum from the SD states via level density arguments. Once E_A and E_II is known, the lifetime of the so far unobserved fission isomer in ²³²Th will be estimated for the first time. In addition, identifying the multi-phonon β vibrational excitation pattern over a wide energy range from the isomeric ground state to the region of the barrier top will provide valuable insight into the harmonicity of the 2^nd potential well.

As a long term perspective, the γ-decay in the 2^nd minimum of thorium and uranium isotopes with its predominant γ back-decay to the 1^st minimum can also be studied with high resolution. Therefore, it will become possible to measure the isomeric excitation energies with an unprecedented resolution (ΔE/E=10^-3). Measurements of the ground state of the 3^rd minimum via its γ decay will also be enabled. Moreover, a detailed γ spectroscopy of the HD potential minimum will be possible for the first time due to the enhanced E1 strengths in the higher-lying minima and due to the strong spin selectivity of the photo-nuclear reactions.

On the other hand, theoretical considerations predicted that the hyperdeformed minimum in a cluster description consists of a rather spherical ¹³²Sn-like component with magic neutron and proton numbers of N=82 and Z=50, respectively, complemented by an attached elongated second cluster of nuclei [6]. Since the fission mass distribution is distinctly determined by the configuration at the scission point, and the 3^rd minimum is very close to the scission configuration, we expect that the mass distributions originating from the 3^rd minimum exhibit a much more pronounced asymmetric mass distribution with a much larger peak-to-valley ratio as compared to fission from the 1^st minimum. However, this effect has never been observed due to the limited mass resolution (~5 amu) of the fission fragment detection systems and the large non-resonant fission background. Therefore, we plan to study the fission fragment mass distribution following the fission decay of the HD states of ²³²Th. The precise knowledge of the fission barrier parameters of ²³²Th, and more generally of the light actinides, are of great importance even for designing more efficient nuclear power plants (cross-section inputs for IV. generation reactors), for performing astrophysical network calculations (heavy element production in the r-process), and also for testing the available theoretical models. 
Moreover, the resonances of the 3^rd minimum are expected to have much larger population width than those from the 1^st minimum. The very large deformation causes the giant dipole resonance (GDR) to split up into two components, with a low-lying oscillation along the long symmetry axis with a typical excitation energy of E=4-5 MeV. The long-term idea towards an application of brilliant γ beams is to use selected transmission resonances in the 3^rd minimum to transmute minor actinides by irradiating radioactive waste containers with deeply penetrating γ beams tuned to these special resonance energies in such a way that fission of these potentially harmful radioactive waste components can be induced non-invasively!

Gamma beams with lasers vs. Bremsstrahlung

By going through the previous entries, I have just realized that no explanation was given on the reason of our enthusiasm about the γ beam of ELI-NP... So here, in a few sentence, I will give a short comparison on the two ways of producing energetic fotons.

In the past decades many experiments have been performed on photon-induced reactions all over the world by using the so-called bremsstrahlung beams of electron accelerator facilities. These energetic photons were produced by the deceleration of the accelerated electrons in a gold target foil/plate. The photon spectrum is continuous, having a maximum energy which is equal to the energy of the electron beam. The difficulty of these experiments is to give the cross section of a photon-induced reaction at a given energy (E0). The procedure was the following: 1) measuring the reaction yield at E1=E0+dE than 2) measuring the yield at E2=E0-dE and finally 3) subtract the second from the first. If we do the same procedure on the energy spectrum of the two bremsstrahlung beams (E1 and E2) then we get the energy spectrum of the "effective" beam which is "responsible" to the measured yield, cross section at E0. However, as you see, this effective beam is not well-defined, a long tail is present in the spectrum at low energies, which is not negligible. So due to the experimental procedure described here, a very large uncertainty is present in the extracted results of such an experiment.

Energy spectra of the bremsstrahlung beam at two energies (top) and the difference "effective" beam

 

The situation is getting better, if we measure the energy of the decelerated electron so we can "tag" the photon and assign an energy to the bremsstrahlung photon. Of course we have to be sure in this case that we produce only one photon by the deceleration of the electron. This can be assured by choosing the right gold-target (so-called radiator) thickness. On one hand the problem of the energy determination is solved, on the other hand we face to a new problem: due to the coincidence technique involved in such an experiment, there is a strong limitation on the photon flux we can use... This is typically very low, too low for us. We want to measure very low photofission cross sections....

This is the reason for our interest in Compton backscattered γ beam facilities. In such a facility a conventional, high power, optical laser beam is pointed to a relativistic electron beam. The photons backscatters on the electrons resulted in an up-shifted energy of the photons. This up-shift is very effective: we can have photons in the γ (MeV) region! The beam has a Gaussian energy profile and the intensity is not limited. We can measure the cross sections directly having low uncertainty. This is what we were waiting for!

Detector development II.

In the previous blog a newly developed detector array was presented that devoted to the measurement of the cross-section of photofission and the angular distribution of the fission fragments. However, measuring the atomic and mass number as well as the energy of the fragments can also serve very important information on the fission process and the structure of the fissioning nucleus. Thus, we started the development of a more complicated detector system which will be dedicated to measure all these quantities.

Gaseous ionization chambers are ideal devices to measure the total kinetic energy (TKE) of the fission fragments with high resolution and large solid angle when compared to solid state detectors. The measurement of fragment kinetic energies allows to deduce the mass distribution using the 2E technique. Moreover, the so-called cold fission events (the fragments are in the ground state) can be selected by setting a proper lower limit on the energy.

The proposed design (Figure) is based on a twin gridded ionization chamber. Frisch-gridded, twin-ionization chambers permit simultaneous measurements of the TKE and the emission angles of both fission fragments in coincidence. The energy of the fragment is determined from the anode pulse height, while the sum of the grid and anode signals is used to deduce the fragment emission angle with respect to the symmetry axis of the chamber. Thus, a precise angle-dependent energy loss correction can be performed. The proposed multi-target detector layout, 5 small area (1 cm2) and thin (200-300 μg/cm2) actinide targets will be placed into the center of each cathode to increase the effective target thickness, and thus the fission yield of the (γ,f) reaction at deep sub-barrier energies.

This chamber will be a versatile tool which will allow measurements of mass distribution of the fission fragments induced by brilliant γ beams on Th, U and other actinide isotopes. Due to its capability of determining the particle’s specific ionization and localization of their traces by using a digitizer and digital signal processing (DSP) techniques, the chamber can provide also information about charge distribution of the fission fragments. In order to increase the capability of the chamber to detect also the process of nuclear fission accompanied by emission of light charge particles (LCP) ΔE-E telescopes will be placed inside the chamber in a transversal position between the electrodes.

The first test experiment of the prototype ionization chamber, which was  performed in November 2013 at MTA Atomki using a 252Cf fission source, was devoted to acquire knowledge on the digital signal processing of fission events in
particular extracting atomic number related information of the signal. 

Schematic design of the proposed detector system

Detector development I.

For the fission-fragment angular distribution measurements, a multi-target detector array will be developed, consisting of position sensitive gas detector modules based on the state-of-the-art THGEM technology. THGEM is a robust, simple to manufacture, high-gain, gaseous electron multiplier. Its operation is based on gas multiplication within small, sub-millimeter to millimeter diameter holes, in standard double-face Cu-clad printed circuit boards (PCB). The hole structure of the THGEM together with a segmented readout electrode provides a true pixelated radiation localization. The electron multiplication of the THGEM is based on the large potential difference between the two sides of the board resulting in a strong dipole field within the holes. Electrons, deposited by ionizing radiation in a conversion region above the THGEM, are focused into the holes by the dispersed electric field. Then, the electrons are multiplied within the holes under the high electric field (25–50 kV/cm). A small fraction of the resulting avalanche electrons are collected on a bottom electrode, while the significant part is transferred to a collecting anode or to a second multiplier element. Each hole acts as an independent multiplier.

At the low-pressure operation mode the signals are very fast, having a rise time of t ~ 3-4 ns. With the standard design (1 mm holes pitch), a position resolution of 2 mm can be achieved by using segmented anode planes with 1-2 mm wide anode pads. In our design the detector covers almost a full solid angle and has an angular resolution of about 5 degrees. The background sensitivity and the radiation damage is negligible, however, the extremely low counting rates in photofission experiments at deep sub-barrier energies require sufficient α particle discrimination. This goal can be achieved by the coincident detection of both fission fragments.

It has to be emphasized, that the foreseen unprecedented sub-millimetre γ beam-spot size allows to develop considerably more compact photofission detectors than those of before when only bremsstrahlung γ sources were available with a beam spot diameter of ~4-5 cm. Furthermore, due to the small diameter of the targets, we can even use highly-radioactive target materials (e.g. ²³⁹Pu) without encountering radiation safety problems, which was not possible before. The well-focused γ beam also defines a distinct fission position, so a remarkably improved angular resolution can be achieved compared to previous bremsstrahlung photofission experiments.

The photo of such a TH-GEM board (manufactured at CERN) is presented at the bottom of the page, as well as the results of the first tests by using a 252Cf fission source.

TH-GEM board manufactured at CERN

2D position distribution of the fission fragment from a 252Cf fission source

Photofission studies of extremely deformed nuclear states: a new era in nuclear physics

Photofission measurements enable selective investigation of the extremely deformed nuclear states in the light actinides and can be utilized to better understand the landscape of the multiple-humped potential energy surface in these nuclei. The selectivity of these measurements originates from the low and reasonably well-defined amount of angular momentum transferred during the photoabsorption process. So far, fission resonances have been studied primarily in light-particle-induced nuclear reactions. These studies do not benefit from the same selectivity found in photonuclear excitation and consequently they are complicated by statistical population of the states in the 2^nd and 3^rd minima with a very limited probability of. These measurements have also suffered from dominating prompt-fission background.

Until now, sub-barrier photofission experiments have been performed only with bremsstrahlung photons and have determined only the integrated fission yield. In these experiments, the fission cross section is convolved with the spectral intensity of the photon beam, resulting in a typical effective γ-ray bandwidth ΔE/E between only 300-400 keV. These experiments observe a plateau, referred to as the "isomeric shelf", in the fission cross section, presumably as a result of the competition between prompt and delayed photofission. However, due to the lack of high resolution photofission studies in the corresponding energy region (E≈4-5 MeV), no experimental information exists to confirm this concept.

Higher-resolution studies could be performed at tagged-photon facilities (e.g. NEPTUN at Darmstadt, Germany), though only with marginal statistics, due to the limited beam intensities realizable through tagging, ~10⁴ γ/(keVs) [13]. This beam intensity cannot be significantly improved, since it is determined by the random coincidence contribution in the electron-tagging process. Thus, high statistics photofission experiments in the deep sub-barrier energy region, where cross sections are typically as low as σ=1 nb-10 μb, cannot be performed with tagged-photon beams.

The relatively recent development of inverse-Compton scattering γ-ray sources, capable of producing tunable, high-flux, quasi-monoenergetic γ-ray beams by Compton-backscattering of eV-range photons off a relativistic electron beam, offers an opportunity to overcome previous limitations. It has to be emphasized that a measurement of the photofission cross section in the deep sub-barrier energy region will be a crucial step towards a reliable characterization of the PES, including unambiguous determination of the double- or triple-humped nature of the surface and precise evaluation of the barrier parameters.

Currently, the most intense Compton-backscattered γ-ray source is the High Intensity γ-ray Source (HIγS) at the Duke University (USA) with a bandwidth of ΔE=150-200 keV and a spectral flux of 10² γ/(eVs). Next-generation Compton-backscattering γ-ray sources, such as the upgrade of HIγS (HIγS2), and the Extreme LIght Infrastructure - Nuclear Physics (ELI-NP, Bucharest, Romania), are anticipated to provide beams with spectral fluxes up to ~10⁶ γ/(eVs) and energy resolutions down to ΔE~1 keV, far superior to those currently available at presently available γ sources. The capabilities of these next-generation sources allow one to aim at an identification of low-amplitude fission resonances. The narrow energy bandwidth expected for the new γ beam facilities will also allow for a significant reduction of the presently dominant background from non-resonant processes. Thus, next-generation γ-ray sources are expected to enable observation of the fine structure in the isomeric shelf. This may open the perspective towards a new era of photofission studies.

Experimental studies of strongly deformed nuclei

Present experimental nuclear research has three main frontiers. The search for anticipated new phenomena requires the study of exotic nuclear states, where either the excitation energy, the deformation or the so-called isospin takes extreme values. The recently developed, highly-efficient Germanium detector arrays have provided great opportunities for the high-resolution studies of nuclear states associated with extremely elongated nuclear shapes. In the mass region of A=130, more than a hundred superdeformed (with a nuclear axis ratio of 2:1) rotational bands have been identified so far. In this mass region, the centrifugal force related to the fast rotating motion is responsible for the large deformation which can be stabilized by strong nuclear shell effects (like in the case of the atomic shell closures in the noble gases). Contrary to the observations of discrete superdeformed (SD) transitions, the identification of discrete γ rays from hyperdeformed (HD) nuclear states, having an axis ratio of 3:1, represents one of the last frontiers of nuclear physics. Although a large community with such arrays have been searching for HD states in very long experiments, no discrete HD states have been identified so far.

In the actinide (Uranium-Plutonium-Thorium) region, however, the appearance of fission resonances gives a unique possibility for the identification and examination of extremely deformed nuclear states. The appearance of a deep and local 2^nd minimum in the potential energy surface (PES) of the nucleus at large quadrupole deformations (β₂=0.6) is a typical feature of the actinide nuclei. It can be described within the macroscopic-microscopic theoretical framework, in which the liquid drop deformation energy of the fissioning nucleus is superposed by periodically varying shell corrections. The ground state of the 2^nd potential minimum is called the fission isomeric state. So far 33 fission isomers have been identified in the region between U (Z=92) and Bk (Z=97) (''island of fission isomers'') having a decay lifetime in the ps-ms range [2].

Soon after the double-humped fission barrier concept was well-established having two minima in the PES, unexpected features of the fission cross-sections of the light actinides (called ''Thorium-anomaly'') pointed to extend the existing picture and introduce the triple-humped fission barrier concept with a shallow 3^rd (hyperdeformed) minimum in the PES at very large quadrupole and large octupole deformations. Almost two decades later we found experimentally in several measurements on ²³⁴U, on ²³⁶U and most recently on ²³²U, that the 3^rd minimum is in fact as deep as the 2^nd minimum for the uranium isotopes.

Our experimental approach to investigate extremely deformed nuclear states of the light actinides is based on the observation of resonances in the fission cross section. Resonances appear when states in the 1^st minimum energetically coincide with (β vibrational) states in the 2^nd (SD) or in the 3^rd (HD) potential minimum. Observing resonances as a function of the excitation energy caused by resonant tunnelling through excited states in the 3^rd minimum of the potential barrier, allows us to identify the excitation energies of the HD states. Moreover, the observed states can be ordered into rotational bands, with moments of inertia proving that the underlying nuclear shape of these states is indeed a HD configuration. For the identification of the rotational bands, the spins and their projections onto the nuclear symmetry axis (K values) can be obtained by measuring the angular distribution of the fission fragments. Furthermore, by analysing the overall structure of the fission cross section and by fitting it with nuclear reaction code calculations, the fission barrier parameters can be extracted very precisely.

Introduction: a portrait

Welcome Dear Reader! My name is Lorant Csige, and I would like to present my research work which is supported (and financed) by European Union and the State of Hungary, co-financed by the European Social Fund in the framework of TÁMOP-4.2.4.A/ 2-11/1-2012-0001 ‘National Excellence Program’. First, let's start with a short introduction.

I graduated at the University of Debrecen as a Master of Science in Physics in 2004. I had my first contact to nuclear physics in the 4th year of my studies, when I started to work on a student project at the Institute of Nuclear Research of the Hungarian Academy of Sciences (MTA Atomki). My main research interest is the nuclear fission process, and the structure of nuclear states having extreme deformation. I use particle accelerators for the production of such states and to induce fission, and quite some time I have to design, contruct, and build particle detectors and other devices "with my bare hands".

From 2004 to 2009 I was a PhD student at MTA Atomki. I defended my theses in 2009 with the best, summa cum laude qualification. In January 2009 I left Hungary, and the alma mater to be a post-doc at the Ludwig Maximilians University in Munich, Germany. I spent 4 years here, and after another one year at the Max-Plack Institute of Quantum Optics (where I was a research fellow), I came back to Hungary in 2013 with the help of the above mentioned research grant.