Normalizing flows represent a class of generative models that exhibit a number of attractive properties, but do not always achieve state-of-the-art performance when it comes to perceived naturalness of generated samples. To improve the quality of generated samples, this thesis examines methods to enhance the expressivity of discrete-time normalizing flow models and thus their ability to capture different aspects of the data. In the first part of the thesis, we propose an invertible neural network architecture as an alternative to popular architectures like Glow that require an individual neural network per flow step. Although our proposal greatly reduces the number of parameters, it has not been done before, as such architectures are believed to not be powerful enough. For this reason, we define two optional extensions that could greatly increase the expressivity of the architecture. We use augmentation to add Gaussian noise variables to the input to achieve arbitrary hidden-layer widths that are no longer dictated by the dimensionality of the data. Moreover, we implement Piecewise Affine Activation Functions that represent a generalization of Leaky ReLU activations and allow for more powerful transformations in every individual step. The resulting three models are evaluated on two simple synthetic datasets – the two moons dataset and one generated from a mixture of eight Gaussians. Our findings indicate that the proposed architectures cannot adequately model these simple datasets and thus do not represent alternatives to current stateof-the-art models. The Piecewise Affine Activation Function significantly improved the expressivity of the invertible neural network, but could not make use of its full potential due to inappropriate assumptions about the function’s input distribution. Further research is needed to ensure that the input to this function is always standard normal distributed. We conducted further experiments with augmentation using the Glow model and could show minor improvements on the synthetic datasets when only few flow steps (two, three or four) were used. However, in a more realistic scenario, the model would encompass many more flow steps. Lastly, we generalized the transformation in the coupling layers of modern flow architectures from an elementwise affine transformation to a matrixbased affine transformation and studied the effect this had on MoGlow, a flow-based model of motion. We could show that McMoGlow, our modified version of MoGlow, consistently achieved a better training likelihood than the original MoGlow on human locomotion data. However, a subjective user study found no statistically significant difference in the perceived naturalness of the samples generated. As a possible reason for this, we hypothesize that the improvements are subtle and more visible in samples that exhibit slower movements or edge cases which may have been underrepresented in the user study. / Normalizing flows representerar en klass av generativa modeller som besitter ett antal eftertraktade egenskaper, men som inte alltid uppnår toppmodern prestanda när det gäller upplevd naturlighet hos genererade data. För att förbättra kvaliteten på dessa modellers utdata, undersöker detta examensarbete metoder för att förbättra uttrycksfullheten hos Normalizing flows-modeller i diskret tid, och därmed deras förmåga att fånga olika aspekter av datamaterialet. I den första delen av uppsatsen föreslår vi en arkitektur uppbyggt av ett inverterbart neuralt nätverk. Vårt förslag är ett alternativ till populära arkitekturer som Glow, vilka kräver individuella neuronnät för varje flödessteg. Även om vårt förslag kraftigt minskar antalet parametrar har detta inte gjorts tidigare, då sådana arkitekturer inte ansetts vara tillräckligt kraftfulla. Av den anledningen definierar vi två oberoende utökningar till arkitekturen som skulle kunna öka dess uttrycksfullhet avsevärt. Vi använder så kallad augmentation, som konkatenerar Gaussiska brusvariabler till observationsvektorerna för att uppnå godtyckliga bredder i de dolda lagren, så att deras bredd inte längre begränsas av datadimensionaliteten. Dessutom implementerar vi Piecewise Affine Activation-funktioner (PAAF), vilka generaliserar Leaky ReLU-aktiveringar genom att möjliggöra mer kraftfulla transformationer i varje enskilt steg. De resulterande tre modellerna utvärderas med hjälp av två enkla syntetiska datamängder - ”the two moons dataset” och ett som genererats genom att blanda av åtta Gaussfördelningar. Våra resultat visar att de föreslagna arkitekturerna inte kan modellera de enkla datamängderna på ett tillfredsställande sätt, och därmed inte utgör kompetitiva alternativ till nuvarande moderna modeller. Den styckvisa aktiveringsfunktionen förbättrade det inverterbara neurala nätverkets uttrycksfullhet avsevärt, men kunde inte utnyttja sin fulla potential på grund av felaktiga antaganden om funktionens indatafördelning. Ytterligare forskning behövs för att hantera detta problem. Vi genomförde ytterligare experiment med augmentation av Glow-modellen och kunde påvisa vissa förbättringar på de syntetiska dataseten när endast ett fåtal flödessteg (två, tre eller fyra) användes. Däremot omfattar modeller i mer realistiska scenarion många fler flödessteg. Slutligen generaliserade vi transformationen i kopplingslagren hos moderna flödesarkitekturer från en elementvis affin transformation till en matrisbaserad affin transformation, samt studerade vilken effekt detta hade på MoGlow, en flödesbaserad modell av 3D-rörelser. Vi kunde visa att McMoGlow, vår modifierade version av MoGlow, konsekvent uppnådde bättre likelihood i träningen än den ursprungliga MoGlow gjorde på mänskliga rörelsedata. En subjektiv användarstudie på exempelrörelser genererade från MoGlow och McMoGlow visade dock ingen statistiskt signifikant skillnad i användarnas uppfattning av hur naturliga rörelserna upplevdes. Som en möjlig orsak till detta antar vi att förbättringarna är subtila och mer synliga i situationer som uppvisar långsammare rörelser eller i olika gränsfall som kan ha varit underrepresenterade i användarstudien.

This thesis presents experimental studies of low-temperature atmospheric pressure plasma sources with generic ability to effectively treat large-scale three-dimensional (3D) objects. The reported large-scale plasma sources are developed through parallelisation of single plasma jets. This strategy outshines the other reported strategies for treatment of uneven surfaces by being able to produce spatially extended plasma directly onto the surface of heavily three-dimensional objects. Comparable studies of the design of elemental plasma jets bring out a hybrid electrode configuration, the capillary-ring jet, as the best elemental jet to be used in the parallelisation. It is found that the introduction of a ballast resistor to the individual jet circuit or built-in capacitance is important to assure the jet-to-jet synchronism, stability and uniformity. Electrical and optical analyses of one-dimensional (1D) array of atmospheric pressure plasma jets demonstrate robust temporal and spatial jet-to-jet uniformity both for flat and sloped surfaces. Hexagonally-arranged two-dimensional (2D) arrays of atmospheric pressure plasma jets show good level of insusceptibility to variations of the downstream samples in their physical dimensions as well as structural and material properties. The reaction chemistry impact area of a 2D 37-jet array is estimated to be 18.6 cm2. These confirm the plasma jet arrays as a viable option as large-scale atmospheric plasma sources, well suited for many processing applications including plasma medicine. The spatially separated dual-frequency excitation further benefits the plasma jet in that separate control of different important plasma parameters is possible. Enhanced plasma properties achieved by the dual-frequency offer greater potential to the jet arrays.

Silício amorfo hidrogenado depositado por descarga luminescente tem adquirido posição de destaque entre os vários materiais foto-sensíveis, principalmente devido à sua fácil obtenção e versatilidade em propriedades. Neste trabalho, preparamos amostras de a-Si:H (Silício amorfo hidrogenado) utilizando o método de descarga de RF em atmosfera de gás SiH4, utilizando como substrato vidro e silício cristalino. As amostras depositadas sobre vidro foram utilizadas para estudo de formação de defeitos meta-estáveis no filme morfo devido a exposição à Raios-X. Os mecanismos de formação e recuperação dos defeitos foram analisados através do estudo da fotocondutividade, revelando a possível natureza de tais defeitos. Acreditamos que se trata da quebra de ligações Si-H e Si-Si, formando armadilhas para elétrons condutores. Analisamos a resposta espectral da fotocondutividade e absorção óptica, que forneceram informações a respeito do gap óptico do material. As amostras preparadas sobre c-Si foram utilizadas para a espectroscopia no infra-vermelho, onde analisamos o espectro das vibrações no material, bem como procuramos evidências da existência de hidrogênio molecular. / Hydrogenated amorphous silicon deposited by the glow discharge technique has been prominent among several photosensitive materials, mainly due to their easy obtention process and versatile properties. In this work, we have prepared hydrogenated amorphous silicon (a-Si:H) samples using the RF discharge method in SiH gaseous atmosphere. We have used glass and crystalline silicon as substrate. The samples which were deposited on glasses were used for the study on the creation of metastables defects in amorphous films due to X-Ray exposition. The creation and annealing process were analyzed through photoconductivity measurements which showed the possible origin of such defects. We believe that this behavior takes place due the breaking of Si-Si and Si-H bounds, which build up traps for free electrons. We have analyzed the photoconductivity spectral response and the optical absorption which give us information about the optical gap of this material. The samples which were prepared on crystalline silicon materials were used for infra-red spectroscopy, were we have analyzed the vibration spectra of a-Si:H, and also have been looking for evidences about the existence of molecular hydrogen.

The field of ambient desorption/ionization mass spectrometry (ADI-MS) has existed for over a decade. ADI-MS is a technique that offers benefits including fast analysis time, simple ionization sources that are easily constructed, and little to no required sample preparation. The research presented here describes efforts to better understand plasma-based ADI-MS sources and to explore the use of hydrogen-doped plasma gases with these sources. The use of hydrogen-doped argon (H2/Ar) and hydrogen-doped helium (H2/He) as plasma gases for a dielectric barrier discharge (DBD) and an AC glow discharge is presented first. When using the DBD, the intensity of the signal obtained when analyzing organic molecules in positive ion mode was increased by factors up to 37 times. In negative ion mode, only H2/Ar was shown to enhance the signal of an organic analyte. The limits of detection for caffeine when using hydrogen-doped plasma gases were found to decrease by factors of 78 and 1.9 for H2/Ar and H2/He respectively. The same phenomenon was observed when using H2/Ar with the AC discharge, but no signal enhancement was observed when using hydrogen-doped helium with the AC discharge. Similarly, if the DBD was allowed to ground through a wire rather than through the air, no signal enhancement was observed for H2/He. Using H2/Ar with metal samples is presented second. By using the metal sample as the grounded electrode for the AC glow discharge, many different metals could be detected directly with a time-of-flight mass spectrometer (TOF-MS) in the form of atomic ions both on their own and in combination with water and ammonia from the discharge. Any refractory metals tested did not yield signal. In addition to direct analysis with a TOF-MS, the AC discharge was used as a sampling method for an inductively coupled plasma mass spectrometer (ICP-MS). When coupled with an ICP-MS, the AC glow discharge was found capable of sampling even refractory elements, though the power of the ICP was required for ionization and detection. Scanning electron microscope (SEM) images of a copper surface exposed to the plasma discharge showed signs of melting when using the H2/Ar. Finally, a computer simulation of the chemistry and flow dynamics of a DC glow discharge generated in helium is presented. The simulation explores many of the fundamental processes at work and how they depend on the composition of the plasma gas. The generation of important species in the plasma was found to depend more on the amount of N2 and H2O impurities in the plasma gas rather than on the humidity or air pressure.

This thesis deals with a one and two dimensional numerical modeling of a low-pressure DC glow discharge in argon gas. We develop two-component fluid model which uses the diffusion-drift theory for the gas discharge plasma and consists of continuity equations for electrons and ions, as well as Poisson equation for electric field. Numerical method is based on the control volume technique. Calculations are carried out in MATLAB environment. Computed results are compared with the classic theory of glow discharges and available experimental data.

Plasma enhanced chemical vapor deposition systems are massively deployed to grow numerous thin film coatings including hydrogenated amorphous silicon. A new deposition chamber was designed, procured, and constructed to investigate the plasma properties of a 100% silane (SiH4) glow discharge with varying chamber pressure and inter-electrode spacing. A Hiden EQP1000 ion mass spectrometer sampled the plasma from the substrates point of view. Ion energy distributions were obtained using four different excitation sources +DC, –DC, radio frequency (at 13.56 MHz), and the DC Saddle Field (DCSF) in the tetrode configuration. The shape of the ion energy distributions was constant for the capacitively coupled +DC, –DC, and rf (at higher pressures of 75 and 160 mTorr) glow discharges. The shape of the ion energy distributions for the DCSF plasma exhibited a double peak or saddle structure analogous to radio frequency plasmas. The width between the peaks (peak separation) was controlled by the pressure and the semi-transparent cathode to semi-transparent anode distance. Ion energy distributions from the DCSF plasma concurred with rf and +DC ion energy distributions at specific pressures and inter-electrode distances. This result demonstrates the versatility of the DCSF glow discharge system. Moreover, control of the peak separation is modeled to be iii equivalent to controlling the critical ratio (ion transit time in the sheath to the electron oscillating period), and/or the inferred electron oscillating sheath potential. The DCSF possesses a fusion of rf and +DC methods. The long high energy tail or constant background are indicative of a +DC high voltage sheath in which there is an increasing fraction of collisionless ions as the anode-cathode distance increases. These collisionless ions are provided by the oscillating electrons (or rf nature) of the DCSF method. Higher order silane (silicon containing) ions increase in relative intensity with increasing inter-electrode spacing for the +DC, –DC, and rf plasmas. These higher order silane ions are also detected in the DCSF plasma, and can be reduced at either lower pressure or lower cathode to anode or cathode to substrate distances.

Plasma enhanced chemical vapor deposition systems are massively deployed to grow numerous thin film coatings including hydrogenated amorphous silicon. A new deposition chamber was designed, procured, and constructed to investigate the plasma properties of a 100% silane (SiH4) glow discharge with varying chamber pressure and inter-electrode spacing. A Hiden EQP1000 ion mass spectrometer sampled the plasma from the substrates point of view. Ion energy distributions were obtained using four different excitation sources +DC, –DC, radio frequency (at 13.56 MHz), and the DC Saddle Field (DCSF) in the tetrode configuration. The shape of the ion energy distributions was constant for the capacitively coupled +DC, –DC, and rf (at higher pressures of 75 and 160 mTorr) glow discharges. The shape of the ion energy distributions for the DCSF plasma exhibited a double peak or saddle structure analogous to radio frequency plasmas. The width between the peaks (peak separation) was controlled by the pressure and the semi-transparent cathode to semi-transparent anode distance. Ion energy distributions from the DCSF plasma concurred with rf and +DC ion energy distributions at specific pressures and inter-electrode distances. This result demonstrates the versatility of the DCSF glow discharge system. Moreover, control of the peak separation is modeled to be iii equivalent to controlling the critical ratio (ion transit time in the sheath to the electron oscillating period), and/or the inferred electron oscillating sheath potential. The DCSF possesses a fusion of rf and +DC methods. The long high energy tail or constant background are indicative of a +DC high voltage sheath in which there is an increasing fraction of collisionless ions as the anode-cathode distance increases. These collisionless ions are provided by the oscillating electrons (or rf nature) of the DCSF method. Higher order silane (silicon containing) ions increase in relative intensity with increasing inter-electrode spacing for the +DC, –DC, and rf plasmas. These higher order silane ions are also detected in the DCSF plasma, and can be reduced at either lower pressure or lower cathode to anode or cathode to substrate distances.

Στα πλαίσια της παρούσας διπλωματικής εργασίας μελετήθηκε, σχεδιάστηκε και κατασκευάστηκε ένα σύστημα ελέγχου ενός φασματοσκοπίου εκπομπής υψηλής φασματικής ανάλυσης. Η συσκευή αυτή, γνωστή και ως μονοχρωμάτορας, χρησιμοποιήθηκε για την καταγραφή φασμάτων εκπομπής από μία εκκένωση αίγλης σε διάκενο ακίδα-πλάκα σε συνθήκες χαμηλής πίεσης και θερμοκρασίας. Ο μονοχρωμάτορας χρησιμοποιεί ένα διάφραγμα περίθλασης που αναλύει μια πολυχρωματική δέσμη που προσπίπτει σε αυτό, στα επιμέρους μήκη κύματος Ο έλεγχος του συνόλου των λειτουργιών του συστήματος καθώς και η καταγραφή των φασμάτων πραγματοποιείται με τη χρήση λογισμικού που αναπτύχθηκε υπό την πλατφόρμα Labview. Στα πλαίσια αυτής της εργασίας υλοποιήθηκε: Η σύνδεση της συσκευής οδήγησης του βηματικού κινητήρα που περιστρέφει το διάφραγμα με κατάλληλο τροφοδοτικό. Η ενίσχυση των σημάτων ελέγχου της συσκευής οδήγησης. Η ενίσχυση και καταγραφή του ρεύματος του φωτοπολλαπλασιαστή που καταγράφει τη φωτεινή δραστηριότητα. Η βαθμονόμηση του μονοχρωμάτορα. Η ανάπτυξη λογισμικού, φιλικού προς τον χρήστη, με το οποίο ελέγχεται ο μονοχρωμάτορας, καταγράφεται το φάσμα και αποθηκεύεται σε αρχείο για περαιτέρω επεξεργασία. Η συγκέντρωση του φωτός που εκπέμπεται από την εκκένωση και η οδήγησή του στην είσοδο του μονοχρωμάτορα με τη χρήση κατάλληλης οπτικής διάταξης. Η καταγραφή φασμάτων του παραγόμενου πλάσματος της εκκένωσης κατά μήκος του διακένου. Η μελέτη και ανάλυση των καταγεγραμμένων φασμάτων. / The main purpose of this thesis was to construct a control system for a high precision spectrometer. Using this spectrometer, also known as monochromator, spectra was recorded from a low temperature – low pressure, pin to plate, glow discharge. The monochromator uses a diffraction grating to separate spatially a polychromatic beam of light available at its input. The stepper motor control, along with the recording of spectra, is being conducted by suitable software developed under the Labview platform. Among the challenges of the thesis was to: Αssemble a stepper motor controller which rotates the diffraction grating with a suitable power supply. Αmplify the control signals of the controller. Αmplify the photomultiplier signals which records the luminous intensity. Calibrate the monochromator. Develop a user-friendly software which can control the monochromator, record spectra and export a file with the recorded spectra for further use. Concentrate the emitted light of the discharge and guide it to the monochromator's input slit, using a UV lens and an optical fiber. Capture emitted spectra of the produced plasma, at different positions along the pin to plate gap. Analyze the recorded spectra.

Η παρούσα διπλωματική εργασία έχει ως αντικείμενο την μελέτη της επίδρασης της μορφής της ανόδου στην παραγωγή εκκενώσεων τύπου αίγλης. Για τον σκοπό αυτό μελετώνται οι εκκενώσεις που παράγονται στο εσωτερικό ενός αντιδραστήρα παραγωγής ψυχρού πλάσματος, μεταξύ μιας μη συμμετρικής διάταξης ηλεκτροδίων (διάκενο ακίδας-πλάκας), ο οποίος βρίσκεται στο εργαστήριο Ηλεκτροτεχνικών Υλικών του τμήματος Ηλεκτρολόγων Μηχανικών και Τεχνολογίας Υπολογιστών. Ειδικότερα, μελετώνται οι εκκενώσεις για τρεις διαφορετικούς τύπους ανόδου, κατασκευασμένες από διαφορετικό υλικό κάθε φορά, πραγματοποιώντας μια σειρά μετρήσεων από τις οποίες εξάγονται οι χαρακτηριστικές τάσης-ρεύματος για διάφορες τιμές της πίεσης εντός του αντιδραστήρα και για δύο διαφορετικές τιμές του μήκους του διακένου. Επιπρόσθετα, για τις παραπάνω συνθήκες, λαμβάνονται παλμοί ρεύματος και φωτός(με την βοήθεια ενός φωτοπολλαπλασιαστή) της εκκένωσης, οι οποίοι απεικονίζουν την φωτεινή δραστηριότητα κατά μήκος του διακένου. Από την επεξεργασία όλων των παραπάνω μετρήσεων, γίνεται προσπάθεια εξαγωγής συμπερασμάτων σχετικά με την επίδραση που δύναται να έχει η μορφή της ανόδου, στην παραγωγή μιας εκκένωσης αίγλης. Ειδικότερα: •Στο πρώτο κεφάλαιο, γίνεται μια σύντομη ιστορική αναφορά και αναφέρονται κάποιες βασικές γνώσεις και χαρακτηριστικά που αφορούν στο πλάσμα. •Στο δεύτερο κεφάλαιο, παρατίθενται συνοπτικά τα καθεστώτα λειτουργίας μιας εκκένωσης αίγλης και συγχρόνως αναλύεται η εκκένωση, από φαινομενολογικής άποψης. •Στο τρίτο κεφάλαιο, περιγράφεται αναλυτικά η πειραματική διάταξη στην οποία έγινε το σύνολο των μετρήσεων. Παρουσιάζονται τα επιμέρους όργανα, που χρησιμοποιούνται για την μέτρηση συγκεκριμένων μεγεθών και ταυτόχρονα παρατίθενται διάφορες φωτογραφίες που απεικονίζουν οπτικά τα επιμέρους τμήματα που την απαρτίζουν. •Στο τέταρτο κεφάλαιο, παρουσιάζεται το σύνολο των μετρήσεων, το οποίο περιλαμβάνει χαρακτηριστικές τάσης-ρεύματος, παλμούς ρεύματος και παλμούς φωτός για τις τρεις διαφορετικές ακίδες, για μια σειρά συνθηκών καθώς και συγκρίσεις διαφόρων μεγεθών που εξάγονται από τα παραπάνω. / In this thesis, the effect of the anode’s geometry on the production of glow discharges is studied. For this purpose, glow discharges produced inside a cold plasma reactor, between a non-symmetrical set-up of electrodes, are investigated. In particular, we study glow discharges for three types of anode, which are made of different materials. Initially, a series of measurements of the gap voltage and current is taken and the V-I characteristics are extracted, each for different values of pressure inside the reactor and two different values of gap length (d=1cm, d=2cm). Furthermore, for the above conditions, several current and light impulses (using a photomultiplier) are recorded, demonstrating the light distribution and light emission among the gap. Using all the above measurement, the effect of the anode’s geometry on the glow discharge production is analyzed. In summary, the following subjects are treated: • Basic characteristics of plasma and short historical reference. •Short reference to different functional regimes of a glow discharge and phenomenological analysis of the discharge. •Description of the experimental set-up and measuring instruments. •Experimental measurements of current and gap voltage (V-I characteristics), current pulses, light pulses, conclusions.

Σκοπός της παρούσας διπλωματικής εργασίας είναι η πειραματική μελέτη ηλεκτρικής εκκένωσης σε χαμηλή πίεση αερίου σε ανομοιογενές ηλεκτρικό πεδίο (διάκενο ακίδα-πλάκα) και η σύγκριση με την ηλεκτρική εκκένωση στο ομογενές πεδίο. Βάση για όλες τις θεωρίες που πραγματεύονται την παραγωγή πλάσματος συνιστά η θεωρία Townsend. Σύμφωνα με αυτή, τα αέρια κατά την εφαρμογή μιας κρίσιμης τιμής τάσης παύουν να είναι μονωτές, υφίστανται ηλεκτρική διάσπαση και μετατρέπονται σε αγωγούς. Προτείνεται έτσι μια πολύ καλή εξήγηση των μηχανισμών των ηλεκτρικών εκκενώσεων που αφορά όμως, μόνο τα ομογενή πεδία. Η πλήρης ερμηνεία τους επιτυγχάνεται με τη θεωρία των κυμάτων ιονισμού που λαμβάνει υπόψην τα μη αμελητέα φορτία χώρου στα οποία εξάλλου και αποδίδει τη μεγάλη ταχύτητα διάδοσης του φαινομένου. Έπειτα, παρουσιάζεται συνοπτικά η εκκένωση αίγλης που συνιστά κατηγορία φαινομένων που αναπτύσσονται όταν το ρεύμα που διαρρέει το διάκενο γίνεται σημαντικό. Στο 1ο κεφάλαιο αναφέρονται και αναλύονται όλες οι έννοιες που εμπίπτουν στην ευρεία έννοια της ηλεκτρικής εκκένωσης. Τονίζονται μεγέθη και παράμετροι της εκκένωσης όπως ο συντελεστής δευτερογενούς παραγωγής ηλεκτρονίων γ που συνδυαζόμενα με τα πειραματικά αποτελέσματα, θα μας δώσουν τελικά χρήσιμα συμπεράσματα. Στο κεφάλαιο 2 περιγράφεται πλήρως η πειραματική διάταξη που χρησιμοποιήθηκε, τα όργανα των μετρήσεων καθώς και τα αέρια που χρησιμοποιήθηκαν. Κατόπιν, αναφέρονται συνοπτικά τα στάδια που πειράματος. Στο κεφάλαιο 3, παρουσιάζονται τα πειραματικά αποτελέσματα για τις ηλεκτρικές εκκενώσεις του αέρα φιάλης για πιέσεις 5-50mbar. Δίνονται οι χαρακτηριστικές καμπύλες V-I, μεμονωμένες και ομαδοποιημένες αλλά και τα αντίστοιχα παλμογραφήματα ρεύματος και φωτός για τις τρεις διαφορετικές πλάκες που χρησιμοποιήθηκαν (ατσαλιού, αλουμινίου και ορείχαλκου). Συμπληρωματικά, σχολιάζονται τα ηλεκτρικά και τα οπτικά χαρακτηριστικά τους και επιχειρείται σύγκριση και αναδεικνύεται το αν οι συνθήκες και το υλικό της καθόδου επηρεάζουν τα αποτελέσματα. Στο κεφάλαιο 4 γίνεται μια ανακεφαλαίωση των παρατηρήσεων που προέκυψαν από το πείραμα και παρουσιάζονται κάποιες σκέψεις. Στο παράρτημα, παρουσιάζεται η αρχή λειτουργίας του φωτοπολλαπλασιαστή. Τέλος, παρουσιάζεται η βιβλιογραφία. / The purpose of this thesis is the experimental study electric dis¬charge at low pressure gas in inhomogeneous electric field (clearance pinplate) and the comparison with the electric discharge in a homogeneous field. Specifically, we look with the help of our experimental results it affects the conditions of the experiment (pressure gap) and the cathode material and highlighting the c factor. Basis for all theories dealing with the production of plasma is the theory Townsend. According to this, the gases in implementing a critical voltage ceases to be insulators, no electrical breakdown and converted pipelines. Thus proposed that a very good explanation of mechanisms of electric prods on, however, only the homogeneous fields. The full interpretation accomplished by the theory of wave ionization (streamer) which takes into consideration the non-negligible space charge in which, moreover, gives the rapidly spreading phenomenon. Then summarizes the glow discharge which is matter phenomena that occur when the current flowing through the gap is significant. In the first chapter reported and analyzed all the concepts covered in the broad sense of the electric discharge. Aggregates are highlighted as the factor c, which combined with the experimental results, will ultimately give us useful conclusions. In Chapter 2 describes the full experimental setup used, the instruments of measurement and the landfill gas were used. Then the steps experiment is been summarized. In Chapter 3 presents the experimental results for electrical discharges in air bottle for pressure 5 -50mbar. We present the characteristic curves V-I, isolated and grouped and the corresponding «photos» of power and light for three different plates were used (steel, aluminum and brass). Additionally, we comment electrical and optical characteristics and attempt to compare and highlights whether the conditions and the material of the cathode affect the results. In Chapter 4 is a summary of comments obtained from the experiment and presents some thoughts. The Annex shows the operating principle of the photomultiplier. Finally, we present the literature.