abstract: Different environmental factors, such as ultraviolet radiation (UV), relative humidity (RH) and the presence of reducing gases (acetone and ethanol), play an important role in the daily life of human beings. UV is very important in a number of areas, such as astronomy, resin curing of polymeric materials, combustion engineering, water purification, flame detection and biological effects with more recent proposals like early missile plume detection, secure space-to-space communications and pollution monitoring. RH is a very common parameter in the environment. It is essential not only for human comfort, but also for a broad spectrum of industries and technologies. There is a substantial interest in the development of RH sensors for applications in monitoring moisture level at home, in clean rooms, cryogenic processes, medical and food science, and so on. The concentration of acetone and other ketone bodies in the exhaled air can serve as an express noninvasive diagnosis of ketosis. Meanwhile, driving under the influence of alcohol is a serious traffic violation and this kind of deviant behavior causes many accidents and deaths on the highway. Therefore, the detection of ethanol in breath is usually used as a quick and reliable screening method for the sobriety checkpoint. Traditionally, semiconductor metal oxide sensors are the major candidates employed in the sensing applications mentioned above. However, they suffer from the low sensitivity, poor selectivity and huge power consumption. In this dissertation, Zinc Oxide (ZnO) based Film Bulk Acoustic Resonator (FBAR) was developed to monitor UV, RH, acetone and ethanol in the environment. FBAR generally consists of a sputtered piezoelectric thin film (ZnO/AlN) sandwiched between two electrodes. It has been well developed both as filters and as high sensitivity mass sensors in recent years. FBAR offers high sensitivity and excellent selectivity for various environment monitoring applications. As the sensing signal is in the frequency domain, FABR has the potential to be incorporated in a wireless sensor network for remote sensing. This study extended our current knowledge of FBAR and pointed out feasible directions for future exploration. / Dissertation/Thesis / Ph.D. Electrical Engineering 2011

L’extrême sensibilité des résonateurs mécaniques (NEMS) aux variations physiques à l’échelle atomique a permis le développement d’un nouveau concept de spectrométrie de masse à base de NEMS capable de mesurer la taille d’une particule unique. L’utilisation de large réseau de capteurs doit permettre à terme de palier la faible surface de capture des résonateurs tout en ouvrant de nouvelles perspectives pour les applications qui nécessitent des informa- tions sur la répartions spatiale des particules au sein du faisceau de mesure. Pour réaliser un spectromètre de masse à base de NEMS viable pour des applications de mesures réelles, il est impératif de développer une technologie de co-intégration NEMS CMOS permettant de fortement densifier le niveau d’interconnexion entre le capteur et l’électronique de lecture. Dans ce travail, nous présentons les premiers résultats mettant en oeuvre une telle techno- logie au travers de mesures de laboratoire et de la conception de circuit intégré co-intégré avec les résonateurs mécaniques. L’électronique de lecture capable de suivre la fréquence de nombreux NEMS simultanément est encore un facteur limitant la forte intégration nécessaire à la lecture de grand réseau de NEMS (>1000), les travaux de cette thèse mettent l’accent sur les problématiques liées à la lecture d’un grand nombre de résonateurs en termes de surface de silicium, de consommation et de performances. Nous présentons dans ce manuscrit une nouvelle architecture d’oscillateur hétérodyne bimode qui doit permettre de répondre à la fois au besoin de compacité tout en assurant le suivi simultané des différents modes de résonances des capteurs. Les travaux présentent également l’effort de modélisation et de co-simulation électro mécanique mis en oeuvre pour la conception des trois circuits. Enfin, nous présentons les résultats de mesure physique obtenue avec l’un des circuits revenus de fabrication et testé au sein du banc de spectrométrie de masse mise en place par les équipes du CEA/LETI. / The extreme sensitivity of nano electro mechanical system (NEMS) to atomic scale physical variations has led to the breakthrough development of NEMS- based mass spectrometry sys- tems capable of measuring a single molecule. Parallel sensing using thousands of devices will help to circumvent the small effective sensing area while opening new perspectives for applica- tions which require spatial mapping. While the development of NEMS CMOS co-integration technology is of paramount importance to achieve high density sensor arrays (>1000 devices), the readout circuitry capable of tracking NEMS resonator frequency shifts is still the limiting factor for the very large scale integration of individually addressed sensors. Moreover, in order to resolve the mass and position of an adsorbed analyte, single particle mass sensing appli- cations require to track simultaneously and in real time at least two modes of the resonators. This requirement adds complexity to the design of the overall system. To respond to the size, power consumption and resolution constraints linked to NEMS array measurement, this work focuses on the development of a new readout architecture based upon a dual mode heterodyne oscillator. This work also emphasis the effort made on the modelization and co-simulation of the NEMS devices with their readout electronics. Then, the manuscript describe the first results of the CEA/LETI CMOS co-integraton process developed to tackle the sensor density challenge of mass spectrometry application. Finally, present the two integrated circuit that were designed during this thesis. The first one was a proof of concept for the aforementioned oscillator architecture while the second one combine the architecture with the co-integration processus developed.

Submitted by Claudinei Pracidelli (cpracide@ipen.br) on 2014-11-10T10:46:11Z No. of bitstreams: 0 / Made available in DSpace on 2014-11-10T10:46:11Z (GMT). No. of bitstreams: 0 / Dissertação (Mestrado em Tecnologia Nuclear) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN-CNEN/SP

In this thesis, simulations of two novel features of a serially cascaded micro-ring resonator are presented. The thesis firstly describes the simulation of a novel, silicon on insulator (SOI) method to determine the refractive index change of a covering analyte by the extraction of the refractive index change information in the time domain. Secondly a novel arrangement of the serially cascaded micro-rings has the effect of producing a null instead of a peak in the Vernier enhanced resonant spectrum. The null feature, as well as the enhanced sensitivity of the sensor, allows the sensor to be used as an intensity interrogating device. The development of these applications using ring resonator physics is achievable, out-of-lab, by the application of photonic software. Finite difference time domain (FDTD), beam propagation method (BPM), finite element(FE) and eigenmode expansion (EME) methods were all used in the simulated development of the sensor. As a result of the dual ring resonator arrangement, the temporal output undergoes a wavelength (or frequency) shift from the micrometre (or TeraHertz) to the centimeter (or GigaHertz) range of frequencies. This allows the refractive index information to become available for transmission in the cm wavelength range over a standard wireless network. The latter could be realized by integration of a photo-detector and antenna into the final design. The sensor output is invariant to any structural or temperature changes applied to both rings. Two sensors based on the same design, but having different fabrication methods, are simulated. Models of the rib and ridge structures are realized by using optical simulation software. The data obtained from these simulations are then used to plot the ring resonator outputs in MATLAB. The design can be applied for either bulk (homogeneous) or surface sensing. Only homogeneous sensing, in the form of a uniform refractive index cover change, is simulated in this thesis. The spectral sensitivity of the rib based design, without Vernier enhancement, is 87.65nmRIU-1, while the spectral sensitivity of the ridge waveguide, without Vernier enhancement, is 422nmRIU-1. The Vernier enhanced spectral sensitivity of the rib design is 6415nmRIU-1 and the limit of detection is 12.47x10-6 RIU. The temporal sensitivity of the ridge is 1.9418μsec RIU-1. The rib temporal sensitivity was not calculated but it is expected to be ~ five times less sensitive than the non Vernier enhanced ridge design. Titanium Nitride (TiN) heaters were also included over the coupling regions of the dual ring resonators. The effect of the heaters on the dual ring resonant wavelength and on the single ring spectral shift were also simulated using a multi-physics utility of the applied FEM and BPM software. With the heater at 1.28μm above the resonator coupling waveguides, a single ring spectral shift of 717pm was exhibited by this simulation. For the heater positioned at 250nm above the coupling waveguides, a single ring spectral shift of 2.89nm was exhibited. Finally the fabricated designs, which are based on the models of the simulation data, were characterized and the results compared to the predicted outputs generated by the models of the Temperature Invariant Modulated Output Sensor (TIMOS).

This thesis presents work to help enable the transition of sensitive nanoscale instruments from research laboratory demonstration to societal use. It focuses on nanomechanical resonators made by field-directed assembly, with contributions to understanding effects of materials, clamp geometries and laser measurement of motion, towards their use as commercial scientific instruments. Nanomechanical resonators in their simplest form are cantilevered or doubly- clamped nanowires or nanotubes made to vibrate near one of their resonant frequencies. Their small mass and high frequency enable extraordinary mass sensitivity, as shown in published laboratory-scale demonstrations of their use for detection of a few molecules of prostate cancer biomarker and of their response to mass equal to that of a single proton. However such sensitive devices have been prohibitively expensive for societal use, since the fabrication process cost scales with number of devices and the chip area covered, when they are made using standard electron beam lithography. Our laboratory has published new results for the method of field-directed assembly, in which the nanofabrication process cost is independent of the number of devices. While drastically lowering the cost, this method also broadens the range of device materials and properties that can be used in instrument applications for sensitive mass and force detection. Unanswered questions affecting the performance of devices made by this method are studied in this thesis. Clamping variability can cause uncertainties in the device resonant frequency (effective stiffness), raising manufacturing metrology costs to track reduced homogeneity in performance. Using a numerical model, we quantify how compliant clamp material and insufficient clamp depth reduce the effective stiffness and resonance frequency. Obliquely clamped nanowires and defects at the clamp-nanowire interface break the symmetry and split the resonance frequency into fast and slow modes. The difference of resonance frequency between the fast and slow modes corresponds to the degree of asymmetry and must be controlled in fabrication to keep device error bounded. Optical transduction has been used for measuring the nanoresonator frequency spectrum; however, the influence of the laser in the measurement process is only recently receiving attention and is not well understood. We found that the measured spectrum is significantly influenced by laser-nanowire interaction. Variation of input laser power could result in resonance peak shifts in the kHz range for a resonance frequency in the MHz range, which could reduce device mass resolution by a factor of 100 or greater. As the laser power is increased, the resonance frequency decreases. The heating effect of the laser on temperature-dependent Young’s modulus could explain this phenomenon. To our surprise, we also found that the amplitude and frequency of the resonance peak signal vary significantly with the angle made by the plane of laser polarization with the nanowire axis. Our measurements established that the maximum signal amplitude is seen when the plane of the linearly polarized laser is parallel to SiNW or perpendicular to RhNW. Maximum resonance frequency was found when laser is polarized perpendicular to SiNW or parallel to RhNW. / Graduate / 0537 / 0548 / 0752

Photonic integration and all-optical signal processing are promising solutions to the increasing demand for high-bandwidth and high-speed communication systems. III-V semiconductor materials, specially AlGaAs, have shown potentials for photonic integration and efficient nonlinear processes due to their low nonlinear absorption, flexibility at controlling the refractive index, and mature fabrication technology. In this thesis, we report the designs of AlGaAs microring resonators optimized for efficient four-wave mixing. Four-wave mixing (FWM) is a nonlinear optical phenomenon which can be used to realize many optical signal processing operations such as optical wavelength conversion and optical time division multiplexing and demultiplexing. Our designed AlGaAs microring resonators are expected to have good optical confinement, transmission characteristics, and efficient coupling between the ring and waveguide. Here we also present our fabrication efforts to fabricate the microring resonators device and the insights gained in the process. The microring resonators devices have a potential to be used in optical communication networks for all-optical signal processing operations.

The objective of this dissertation is to investigate several dynamical approaches aiming to improve the sensitivity and performance of microelectromechanical systems (MEMS) resonant sensors. Resonant sensors rely on tracking shifts in the dynamic features of microstructures during sensing, such as their resonance frequency. We aim here to demonstrate analytically and experimentally several new concepts aiming to sharpen their response, enhance the signal to noise ratio, and demonstrate smart functionalities combined into a single resonator. The dissertation starts with enhancing the excitations of the higher order modes of vibrations of clamped-clamped microbeam resonators. The concept is based on using partial electrodes with shapes that induce strong excitation of the mode of interest. Using a half electrode, the second mode is excited with a high amplitude of vibration. Also, using a two-third electrode configuration is shown to amplify the third mode resonance amplitude compared with the full electrode under the same electrical loading conditions. Then, we demonstrate the effectiveness of higher order mode excitation and metal organic frameworks (MOFs) functionalization for improving the sensitivity and selectivity of resonant gas sensors. Also, using a single mode only, we show the possibility of realizing a smart switch triggered upon exceeding a threshold mass when operating the resonator near the dynamic pull-in instability. The second part of the dissertation deals with the dynamics of the microbeam under a two-source harmonic excitation. We experimentally demonstrate resonances of an additive and subtractive type. It is shown that by properly tuning the frequency and amplitude of the excitation force, the frequency bandwidth of the resonator is controlled. Finally, we employ the multimode excitation of a single resonator to demonstrate smart functionalities. By monitoring the frequency shifts of two modes, we experimentally demonstrate the effectiveness of this technique to measure the environmental temperature and gas concentration. Also, we present a hybrid sensor and switch device, which is capable of accurately measuring gas concentration and perform switching when the concentration exceeds a specific (safe) threshold. In contrast to the single mode operation, we show that monitoring the third mode enhances sensitivity, improves accuracy, and lowers the sensor sensitivity to noise.

Automotive turbo compressors generate high frequency noise in the air intake system. This sound generation is of importance for the perceived sound quality of luxury cars and may need to be controlled by the use of silencers. The silencers usually contain resonators with slits, perforates and cavities. The purpose of the work reported is to develop acoustic models for these resonators where relevant effects such as the effect of realistic mean flow on losses and possibly 3D effects are considered. An experimental campaign has been undertaken where the two-port matrices and transmission loss of four sample resonators has been measured without flow and for two different mean flow speeds (M=0.05 & M=0.1) using two source location technique. Models for the four resonators have been developed using a 1D linear acoustic code (SIDLAB) and a FEM code (COMSOL Multi-physics). Different models, from the literature, for including the effect of mean flow on the acoustic losses at slits and perforates have been discussed. Correct modeling of acoustic losses for resonators with complicated geometry is important for the simulation and development of new and improved silencers, and the present work contributes to this understanding. The measured acoustic properties compared well with the simulated model for almost all the cases.

Quantum technologies are transitioning from the laboratories of academia to industry and the commercial sector. Scaling the utilization of such systems will require optical quantum networks capable of interconnecting quantum nodes, as well as transmitting and receiving quantum states without loss of entanglement. In order to enable such quantum networks and pave the way for a quantum internet, “quantum state transceivers” (QSTs) must be developed. QSTs will be systems with mixed classical and quantum capabilities, and are ideally implemented as quantum-capable “systems-on-chip” (QSoCs) to allow for cost and complexity scaling. In this work, we leverage electronics-photonics monolithic integration in silicon CMOS technology to develop a new class of electronic-photonic systems-on-chip with quantum photonic functions -- electronic-photonic quantum systems-on-chip (epQSoCs). As a first demonstrator of such a “quantum” electronics-photonics platform, we implement a single-chip “wall-plug”, high efficiency photon pair source. We demonstrate a first example system-on-chip, as well as a basic building block component that can be utilized in future, more complex electronic-photonic QSoCs, by analogy to the hierarchical design of classical electronics chips, comprising of smaller component building blocks (so called IP blocks). Additionally, we investigate a novel radiation-free dielectric nanocavity with all-evanescent confinement based on perfect mode-matching and minimal use of negative permittivity, with potential for applications in future generations of photonic circuits. / 2024-05-24T00:00:00Z

Starke Licht-Materie-Wechselwirkung in Halbleiter-Mikroresonatoren führt zur Ausbildung von Eigenmoden mit gemischtem Licht-Materie-Charakter, die als Polaritonen bezeichnet werden. Die besonderen Eigenschaften dieser bosonischen Quasiteilchen können zur Realisierung neuartiger Bauteile genutzt werden, wie etwa des Polariton-Lasers, der auf stimulierter Streuung beruht anstatt auf stimulierter Emission, durch die Photon-Lasing ausgelöst wird. Durch den direkten Zugang zu Polariton-Zuständen in spektroskopischen Experimenten, sowie durch die Möglichkeit mit vielfältigen Mitteln nahezu beliebige Potentiallandschaften definieren zu können, eröffnen sich zahlreiche weitere Anwendungsgebiete, etwa in der Quantensimulation bzw. -emulation. Mittels externer elektrischer und magnetischer Felder können Erkenntnisse über Polaritonen gewonnen werden, die in rein optischen Experimenten nicht zugänglich sind. Durch die Felder, die nicht mit rein photonischen Moden wechselwirken, kann auf den Materie-Anteil der Hybridmoden zugegriffen werden. Weiterhin können die Felder zur in-situ Manipulation der Polariton-Energie genutzt werden, was für die Erzeugung dynamischer Potentiale relevant werden könnte. Der Fokus dieser Arbeit liegt daher auf der Betrachtung verschiedener Phänomene der Licht-Materie-Wechselwirkung unter dem Einfluss äußerer Felder. Dazu wurden auf das jeweilige Experiment abgestimmte Strukturen und Bauteile hergestellt und in magneto-optischen oder elektro-optischen Messungen untersucht. Um elektrische Felder entlang der Wachstumsrichtung anlegen zu können, d.h. in vertikaler Geometrie, wurden dotierte Resonatoren verwendet, die mit elektrischen Kontakten auf der Probenoberfläche und -rückseite versehen wurden. In diesen Bauteilen wurde die Energieverschiebung im elektrischen Feld untersucht, der sogenannte Stark-Effekt. Dieser im linearen Regime bereits mehrfach demonstrierte Effekt wurde systematisch auf den nichtlinearen Bereich des Polariton-Lasings erweitert. Dabei wurde besonderes Augenmerk auf die Probengeometrie und deren Einfluss auf die beobachteten Energieverschiebungen gelegt. Die Untersuchungen von Proben mit planarer, semi-planarer und Mikrotürmchen-Geometrie zeigen, dass ein lateraler Einschluss der Ladungsträger, wie er im Mikrotürmchen erzielt wird, zu einer Umkehrung der Energieverschiebung führt. Während in dieser Geometrie mit zunehmender Feldstärke eine Blauverschiebung des unteren Polaritons gemessen wird, die durch Abschirmungseffekte erklärt werden kann, wird in planarer und semi-planarer Geometrie die erwartete Rotverschiebung beobachtet. In beiden Fällen können, je nach Verstimmung, Energieverschiebungen im Bereich von einigen hundert µeV gemessen werden. Die gemessenen Energieverschiebungen zeigen gute Übereinstimmung mit den Werten, die nach einem Modell gekoppelter Oszillatoren berechnet wurden. Weiterhin werden vergleichbare Energieverschiebungen unter- und oberhalb der Schwelle zum Polariton-Lasing beobachtet, sodass der Polariton-Stark-Effekt als eindeutiges Merkmal erachtet werden kann, anhand dessen optisch angeregte Polariton- und Photon-Laser eindeutig unterschieden werden können. Wird das elektrische Feld nicht entlang der Wachstumsrichtung angelegt, sondern senkrecht dazu in der Ebene der Quantenfilme, dann kommt es schon bei geringen Feldstärken zur Feldionisation von Elektron-Loch-Paaren. Um diese Feldgeometrie zu realisieren, wurde ein Verfahren entwickelt, bei dem Kontakte direkt auf die durch einen Ätzvorgang teilweise freigelegten Quantenfilme eines undotierten Mikroresonators aufgebracht werden. Durch das Anlegen einer Spannung zwischen den lateralen Kontakten kann die Polariton-Emission unterdrückt werden, wobei sich die Feldabhängigkeit der Polariton-Besetzung durch ein Modell gekoppelter Ratengleichungen reproduzieren lässt. Die neuartige Kontaktierung erlaubt es weiterhin den Photostrom in den Quantenfilmen zu untersuchen, der proportional zur Dichte freier Ladungsträger ist. Dadurch lässt sich zeigen, dass die zwei Schwellen mit nichtlinearem Anstieg der Emission, die in derartigen Proben häufig beobachtet werden, auf grundsätzlich verschiedene Verstärkungsmechanismen zurückgehen. An der zweiten Schwelle wird ein Abknicken des leistungsabhängigen Photostroms beobachtet, da dort freie Ladungsträger als Reservoir des Photon-Lasings dienen, deren Dichte an der Schwelle teilweise abgeklemmt wird. Die erste Schwelle hingegen, die dem Polariton-Lasing zugeordnet wird, hat keinen Einfluss auf den linear mit der Anregungsleistung ansteigenden Photostrom, da dort gebundene Elektron-Loch-Paare als Reservoir dienen. Mittels angepasster Ratengleichungsmodelle für Polariton- und Photon-Laser lässt sich der ermittelte Verlauf der Ladungsträgerdichte über den gesamten Leistungsbereich qualitativ reproduzieren. Abschließend wird durch ein magnetisches Feld der Einfluss der Licht-Materie-Wechselwirkung auf die Elektron-Loch-Bindung im Regime der sehr starken Kopplung beleuchtet. Durch die Messung der diamagnetischen Verschiebung wird der mittlere Elektron-Loch-Abstand von unterem und oberem Polariton für zwei Resonatoren mit unterschiedlich starker Licht-Materie-Wechselwirkung bestimmt. Bei geringer Kopplungsstärke werden die Hybridmoden in guter Näherung als Linearkombinationen der ungekoppelten Licht- und Materie-Moden beschrieben. Für den Resonator mit großer Kopplungsstärke wird eine starke Asymmetrie zwischen unterem und oberem Polariton beobachtet. Die diamagnetische Verschiebung des oberen Polaritons steigt mit zunehmender Verstimmung auf bis etwa 2,1 meV an, was fast eine Größenordnung über der Verschiebung des unteren Polaritons (0,27 meV) bei derselben Verstimmung liegt und die Verschiebung des ungekoppelten Quantenfilms um mehr als den Faktor 2 übersteigt. Das bedeutet, dass das untere Polariton durch eine Wellenfunktion beschrieben wird, dessen Materie-Anteil einen verringerten mittleren Elektron-Loch-Abstand aufweist. Im oberen Polariton ist dieser mittlere Radius deutlich größer als der eines Elektron-Loch-Paars im ungekoppelten Quantenfilm, was sich durch eine von Photonen vermittelte Wechselwirkung mit angeregten und Kontinuumszuständen des Quantenfilms erklären lässt. / Strong light-matter interaction in semiconductor microcavities leads to the formation of eigenmodes with mixed light-matter characteristics, so-called polaritons. The unique properties of these bosonic quasiparticles may be exploited to realize novel devices, such as polariton-lasers which rely on stimulated scattering instead of stimulated emission, which in turn triggers photon-lasing. Polariton states are directly accessible in spectroscopic experiments and can be subjected to almost arbitrary potential landscapes which could lead to numerous applications, for instance in quantum simulation or emulation. External electric and magnetic fields can be used to gain insights into polaritons that are not available in all-optical experiments. The matter part of the hybrid modes is accessed by the external fields that do not interact with purely photonic modes. Furthermore, in-situ manipulation of the polariton energy by external fields could be used to create dynamic potentials. This thesis is therefore focussed on studying different aspects of light-matter coupling under the influence of external fields. To this end, structures and devices tailored to the specific experiments were fabricated and investigated in electro-optical or magneto-optical measurements. Doped microcavities with electrical contacts on the sample surface and back side were used to apply electric fields along the growth direction, i.e. in vertical geometry. The energy shift in an electric field, the so-called Stark effect, was investigated in these devices. In this work, measurements of the polariton Stark effect, which has previously been demonstrated in the linear regime, were systematically extended to the nonlinear regime of polariton-lasing with special attention paid to the sample geometry and its influence on the observable energy shifts. Investigations of samples with planar, semi-planar and micropillar geometries show that lateral carrier confinement in a micropillar leads to an inversion of the energy shift. While in this geometry a blueshift with increasing field strength is measured, which can be explained by screening effects, the expected redshift is restored in planar and semi-planar geometries. In both cases, detuning-dependent energy shifts of up to hundreds of µeV are observed in good agreement with values calculated with a model of coupled harmonic oscillators. Furthermore, comparable shifts below and above the polariton-lasing threshold are observed both in the semi-planar and in the micropillar geometry. The polariton Stark effect may therefore be considered as criterion to unambiguously distinguish optically excited polariton- and photon-lasers. If the electric field is not oriented along the growth direction but perpendicular to it, i.e. in the plane of the quantum wells, then field ionization of electron-hole pairs occurs already at low field strengths. To realize this field geometry, a process was developed to deposit electrical contacts directly onto the quantum wells of an undoped microcavity which are partially exposed in an etching step. The polariton emission can be suppressed by applying voltage to the lateral contacts and the dependency of the polariton occupation upon the electric field is reproduced using a set of coupled rate equations. This novel contacting technique furthermore allows to measure the photocurrent in the quantum wells which is proportional to the free carrier density. The two thresholds of nonlinear emission, which are commonly observed in similar samples, can then be shown to rely on fundamentally different gain mechanisms. A kink in the power dependence of the photocurrent is observed at the second threshold, where free carriers act as reservoir for photon-lasing which is why their density is partially clamped at threshold. The first threshold on the other hand, which is attributed to polariton-lasing, has no influence on the linear increase of the photocurrent with increasing excitation power, since there bound electron-hole pairs act as reservoir. The experimentally determined power dependence of the photocurrent is reproduced qualitatively over the whole range of excitation powers using adapted rate equation models for polariton- and photon-lasers. Finally, a magnetic field is used to reveal the impact of light-matter interactions on electron-hole coupling in the regime of very strong coupling. By measuring the diamagnetic shift, the average electron-hole separations of lower and upper polariton are determined for two microcavities with different light-matter coupling strengths. At small coupling strength, describing the hybrid modes as linear combinations of uncoupled light and matter modes is a valid approximation. At large coupling strength, significant asymmetries between lower and upper polariton are observed. With increasing detuning, the upper polariton diamagnetic shift increases up to 2.1 meV, almost an order of magnitude larger than the lower polariton shift (0.27 meV) at the same detuning and more than twice as large as the bare quantum well diamagnetic shift. Thus, the lower polariton is described by a wavefunction with a matter part exhibiting a decreased average electron-hole separation. For the upper polariton, this average radius is much larger than that of an electron-hole pair in the uncoupled quantum well which can be explained by photon-mediated interactions with excited and continuum states of the quantum well.