Back to search

National Technical University of Athens (NTUA)

Προβλήματα αλληλεπίδρασης μη γραμμικών οπτικών παλμών

Abstract

dc:description

The current doctoral thesis consists of four problems in the field of Nonlinear Optics, which we have studied in the Plasma Electron Beam and Nonlinear Optics Laboratory of NTUA. All of the problems have to do with interactions of optical pulses in the presence of nonlinearity. This means that the response of the medium depends on the pulses themselves. Those pulses (or beams) that we deal with, can be of solitonic nature or not, but in any case the indicated model, which describes their evolution, is the Nonlinear Schrödinger Equation (NLS). Our method of study changes according to the problem. In some cases we follow mainly an analytical approach (or rather semi-analytical) while in other, our study is mainly a numerical one.Pulses interactions can be coherent or incoherent, depending on their dependence on pulses’ relative phase. Those pulses we deal with are often called “nonlinear”, since they trigger nonlinear phenomena. Depending on the nature of the interaction (coherent or incoherent), there are different phenomena that are dominant each time. Thus the equations that model them have to change accordingly. Another variation, relative to the physics and the methodology of the problem is imposed by its dimensions. For example, the temporal pulses propagating along a standard monomode fiber are one dimensional entities, because they are trapped along the spatial dimensions and it is only their temporal width, at a certain point along the propagation line, that is subjected to nonlinear effects and dynamical evolution. Such a pulse, with the proper power, can relatively easy create a soliton and propagate undisturbed. A pulse propagating along a planar waveguide is a problem of two dimensions, while another propagating in a bulk medium is a problem of three dimensions.The introduction of the work, found in Chapter 1, is relatively extended, in order for all the above issues to be included. We give some historical elements, concerning the progress of the research on nonlinear waves in general, and also in Optics more specifically. Some brief descriptions about solitons and the nature of pulse and beam interactions are also given. Furthermore, we present the main phenomena that appear in he problems that we study, and the more usual analytical and numerical methods of study followed in relative problems in Optics. Some of these methods are employed in this work. There is also a brief reference on the mathematical background of our study, the NLS and some conserved quantities that are needed in our analysis.In Chapters 2 and 3 we study two one dimensional problems. In both cases we deal with the incoherent interaction of solitonic pulses and the model used is a pair of NLS coupled mainly by the term of cross phase modulation. Also, in both problems we use perturbative methods. More specifically, in Chapter 2 we study the influence of periodically modulated dispersion in pulse interactions. Firstly we deal with the effect that this periodic modulation has on the coupling of two orthogonally polarized modes of a pulse propagating inside a fiber with strong birefringence. The modulation is actually no more than a perturbation, having small amplitude. Nevertheless, it is the frequency of modulation that is most important and we study a wide spectrum of values. We use the Variational Method to transform the problem of two coupled PDEs to one of a system of ODEs. The variables of the system are chosen to be the modes’ characteristics. We perform dynamical analysis of the system in order to retrieve the eigenfrequencies of it, and we use those as guidance through our study. In order to capture a general picture of the system, for a wide range of initial values and of the frequency and amplitude of the modulation, we employ a Poincaré surface of section analysis. Through it we can unveil characteristics of the system, like resonances, chaotic areas, initial values that lead to escapes and generally we can check its stability. Since we end up with systems of three degrees of freedom, the surfaces of section are four-dimensional. Thankfully, there is some not that extended bibliography concerning exactly those subjects, which mainly comes from the field of Astrophysics. Hence by creating three dimensional projections of the surfaces of section we can still comment on system stability. Also, we can create even the four dimensional surfaces themselves by employing color as the fourth dimension. We also solve the coupled PDEs numerically and compare the results with those of the ODEs. Doing systematic comparisons we can comment not only on the stability of the system, meaning the resilience of mode coupling, but also we can test the validity of the Variational Method, which appears to fail for certain values of modulation’s frequency.Another problem that can be modeled with the same equations is the interaction of solitonic pulses of different frequency in the presence of Dispersion Management (DM). Such interactions can happen in a WDM communication system. In this case the frequency of dispersion modulation is quite larger than the characteristic frequencies of the pulses. By employing a simple algebraic procedure and dynamical analysis of the ODEs system, we are able to choose the values of energy enhancement and pre-chirp of the pulse and create a DM-soliton. It is also shown that the sifting of frequency, experienced by the DM-solitons after their collisions, can be nullified through the proper selection of the period and local values of the dispersion.At Chapetr 3 we study yet another problem concerning the interaction of solitonic pulses with different frequencies. However, this time we deal with ultrashort pulses, being of width lesser than 1ps. For this reason, higher order effects, like Raman scattering and dispersion of the third order cannot be ignored. Most importantly, the Raman effect causes downshift of pulse frequency, as well as energy transfer between the pulses, during their collision. The model used is a pair of NLS, with added terms of higher order. We approach the problem through the Direct Perturbation Method, which also gives us a system of ODEs to solve. Our main concern is the possibility of balancing the temporal displacement of pulses, caused by the frequency shift, by forcing collisions with pulses of lower frequency (another channel), which cause opposite temporal displacement. It is shown that this can be possible, but the energy transfer that takes place at every collision can be destructive for the solitonic pulse, in the long run.The rest of the thesis concerns problems of coherent interaction of two-dimensional (spatiotemporal) wavepackets. Our study is mainly numerical and the model that we use is the two-dimensional NLS. Since the pulses are now considered to have the same frequency, they can be presented as a superposition of one over the other and their evolution can be modeled by one equation.The first problem concerns the possibility of spontaneous nonlinear X-wave generation. The specific wavepackets are the analogue of the linear X-waves and are intensively studied for a few years now because they appear to travel through slab waveguides or bulk media without distortion. A necessity for their existence is the medium to be of normal dispersion and the reason behind their specific shape is the spatiotemporal modulation instability (MI) which amplifies specific sidebands. Of great interest is the possibility of creating such wavepackets from simple Gaussian spatiotemporal pulses. We show that this could be made possible through the interaction of such a pulse with a continuous plane wave (CW). The presence of the CW enhances the MI and can reshape the pulse to create an “X-like” wavepacket which does not have the resilience of the linear counterpart, but can propagate for longer distances than the Gaussian pulse, before succumbing to dispersion. The amplitude and most of all the initial phase of the CW are the critical controlling factors. By employing a mostly qualitative analysis we can choose the initial characteristic of the Gaussian pulse and the CW, in order to avoid the temporal splitting or the dispersion of the former and reshape it to an X-like wavepacket.The second problem, which is also the last problem of this thesis, presented at the 5th Chapter, concerns coherent interactions of spatiotemporal Gaussian pulses, in the presence of CW. The medium is again considered to be of normal dispersion. We examine the possibility of them fusing together and even amalgamate. The amalgamation is the state where the pulses merge and create one wavepacket which evolves dynamically without dispersing outright, as the individual pulses would. Hence in the normal dispersion regime, if that wavepacket has the proper “mass” (energy) will experience temporal splitting. That could create tow wavepakets with temporal displacement. In that case we have a situation of spatiotemporal reallocation, where two pulses spatially displaced, interact and create two new wavepackets, which are temporally and spectrally displaced.

Degree

thesis:*
Grantor dc:publisher
National Technical University of Athens (NTUA)
Year dc:date
2012

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Moschonas, Nikolaos
  • Μοσχονάς, Νικόλαος

Subjects

dc:subject × 20

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/27473

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Moschonas, Nikolaos; Μοσχονάς, Νικόλαος. Προβλήματα αλληλεπίδρασης μη γραμμικών οπτικών παλμών. National Technical University of Athens (NTUA), 2012. http://hdl.handle.net/10442/hedi/27473