Doctoral theses at Solid-State Electronics

Abstract

Radio Frequency (RF) power sources are extensively applied in various fields. Radioisotope production, i.e., the production of short-lived radioactive isotopes, for positron emission tomography (PET) is one of the most important applications in the medical and healthcare domains. Full-time operation and substantial maintenance of such systems lead to high operating expenses. Hence, the development of more efficient and reliable RF power amplifiers, which are the main contributors to the energy consumption and maintenance costs of the RF power sources, is a high priority. Solid-state technology has emerged as a viable alternative to conventional vacuum tube based high-power RF/microwave systems, offering advanced control, reliability, and ease of use. Power amplifiers based on solid-state technology enable dynamic adjustment of power to optimize the transmitted energy. Furthermore, solid-state power amplifiers (SSPA) technology shows a longer lifetime leading to increased uptime and lower maintenance costs. Concisely, with the introduction of solid-state technology in high-power RF sources, RF energy can be generated more efficiently and more controllable in a smaller form factor, allowing for more compact systems with less downtime and less maintenance. This thesis is one step further toward demonstrating the feasibility of such systems.

The thesis first introduces the RF measurement setup. It implements automation for quick measurements and supports the evaluation of the high-power RF performance of the developed SSPA modules. Moreover, a novel thru-only de-embedding approach is developed to address the calibration difficulties under multi-port excitation conditions. The second part of the thesis deals with the development and analysis of efficient kilowatt SSPA modules. A multimode SSPA with quasi-static supply control for power regulation is implemented. It achieves more than 90% efficiency over a 5 dB output power back-off range. Another compact and efficient SSPA, implemented in push-pull architecture, adopts harmonic load-pull integrated with the same quasi-static supply modulation which also achieves 90% efficiency over a 5 dB output power back-off range. The implemented SSPAs improve the state-of-the-art in these frequency bands and power ranges.

This thesis broadens RF SSPA theoretical research to the kilowatt power range and provides a new understanding of high-power SSPAs from circuits, design methodologies, and analytical approaches. And it leads to new methods and tools to improve the energy efficiency of high-power RF sources. The knowledge gained and technology developed is not limited to RF power sources in radioisotope production applications, it can also be applied in the communication industry, such as radar systems, and other RF energy systems in industrial, scientific, and medical (ISM) fields, such as particle accelerators, welding, drying, heating, and many more.

Abstract

A biosensor based on streaming current is a new and relatively unexplored subject with significant potential. This thesis attempts to gain a deeper understanding of the governing principles, and then exploit them to further improve its performance as well as develop novel applications. To this end, the underlying theoretical frameworks were examined and two critical parameters of the target: its size and electric charge, influencing the sensor’s sensitivity were identified. This was followed by experimental evaluation of the parameters, using a set of tailor-made proteins, aiming to understand the nature and extent of their influence on the sensor response in relation to simulation performed following an established model.

The dependence of the sensor response on the charge of an analyte, or specifically the charge contrast between the sensor surface and an analyte, opens a new avenue to improve the sensitivity and also to develop novel functionality. First, this aspect was exploited to improve the sensitivity by optimizing the surface functionalization strategy. Three such methods were compared in terms of the resulting zeta potential of the surface. The sensitivity was the highest when the charge contrast was maximum. The optimal functionalization strategy was then used for highly sensitive detection of extracellular vesicles (EVs), where an improvement in the limit of detection by two orders of magnitude over the previously reported results was demonstrated. Two applications of the improved method were then demonstrated: monitoring the effectiveness of targeted cancer medicines and analysis of liquid biopsy of cancer patients via sensitive profiling of EV-membrane proteins.

Improvement in the detection specificity is a critical aspect of biosensing. This was achieved by implementing a sandwich immunoassay and demonstrating the proof of concept using trastuzumab as the target and Z-domain as both the capture and detection probes. Although the improved selectivity came at the cost of a lower sensitivity, this could be mitigated via DNA-conjugation with the detection probes, a novel electrostatic labelling strategy that allows for improvement of the sensitivity by exploiting the electrostatic influence. An application of this method was then demonstrated by detecting the target from a complex medium of E. coli cell lysate. Continuing the prospect of charge engineering of antibodies, a set of positively and negatively charged antibodies were synthesized by conjugating poly-lysine and DNA oligonucleotides, respectively. This enabled stepwise, multiplexed membrane protein analysis of EVs using the alternating charge-labelled antibodies. The method was then applied to investigate EV-heterogeneity.

Abstract

Graphene is widely touted as the thinnest and the most versatile material available. As an atomically thin layer of carbon atoms arranged in a hexagonal configuration, graphene has a combination of technologically important properties, such as thermal and electrical conductivity, mechanical strength, and impermeability to gases. From an industrial perspective on applications, water as a dispersing media for graphene offers safer handling and environmental benefits compared with conventional organic solvents. However, the high surface tension of water and the attractive forces between graphene surfaces drive the sheets to aggregation. Although surfactants have been an important stepping stone in the advancement of aqueous graphene dispersions, these surface-active molecules are often needed in excess and have adverse effects on coatings during film formation. These challenges limit the industrial relevance of graphene as an effective barrier in composites. In general, gas barriers against both oxygen and water vapour, made from a single coating formulation, is seemingly a holy grail for the packaging industry. In this thesis work, the aim was to gain a fundamental understanding of aqueous graphene dispersions for gas barriers used in paper packaging. Biobased materials were systematically investigated as dispersing agents for graphene based on dispersing conditions and functional barrier performance. Flavin mononucleotide (FMN), a food additive, dispersed graphene using a relatively low amount of FMN and showed intriguing spectroscopic signatures of π-π interactions with graphene. Starch nanoparticles (SNPs) realised concentrated and stable aqueous graphene dispersions for composite films. The SNP-stabilized graphene sheets in starch films lowered the gas permeability of both oxygen and water vapour simultaneously by over 70% under all the conditions tested. In general, a combined gas barrier performance is unusual for both bioplastics and common petrochemical-based plastics used in the packaging industry. Motivated by the graphene network leading to the extraordinary barrier performance, the aqueous SNP-graphene dispersion was modified for inkjet printing. The printed patterns were flexible and electrically conductive in the order of 10⁴ S m^-1 that is on par with the highest reported values in the literature. These surfactant-free aqueous SNP-graphene dispersions have the potential and versatility for paper-based gas barriers with integrated electronics. Multifunctional composite films made from these dispersions, when optimized, could become competitive with commercial plastics, and meet the current and future demands of the packaging industry.

Abstract

Silicon nanowire (SiNW) based electronic devices fabricated with a complementary metal-oxide-semiconductor (CMOS) compatible process have wide-range and promising applications in sensing area. These SiNW sensors own high sensitivity, low-cost mass production possibility, and high integration density. In this thesis, we design and fabricate SiNW electronic devices with the CMOS-compatible process on silicon-on-insulator (SOI) substrates and explore their applications for ion sensing and quantum sensing. 

The thesis starts with ion sensing using SiNW field-effect transistors (SiNWFETs). The specific interaction between a sensing layer and analyte generates a change of local charge density and electrical potential, which can effectively modulate the conductance of SiNW channel. Multiplexed detection of molecular (MB⁺) and elemental (Na⁺) ions is demonstrated using a SiNWFET array, which is functionalized with ionophore-incorporated mixed-matrix membranes (MMMs). As a follow-up, polyethylene glycol (PEG) doping strategy is explored to suppress interference from the hydrophobic molecular ion and expand the multiplexed detection range. Then, the SiNW is downscaled to sub-10 nm with a gate-oxide-free configuration for single charge detection in liquid. We directly observe the capture and emission of a single H⁺ ion with individually activated Si dangling bonds (DBs) on the SiNW surface. This work demonstrates the unprecedented ability of the sub-10 nm SiNWFET for investigating the physics of the solid/liquid interface at single charge level.

Apart from ion sensing, the SiNWFET can be suspended and act as a nanoelectromechanical resonator aiming for electrically detecting potential quantized mechanical vibration at low temperature. A suspended SiNW based single-hole transistor (SHT) is explored as a nanoelectromechanical resonator at 20 mK. Mechanical vibration is transduced to electrical readout by the SHT, and the transduction mechanism is dominated by piezoresistive effect. A giant effective piezoresistive gauge factor (~6000) with a strong correlation to the single-hole tunneling is also estimated. This hybrid device is demonstrated as a promising system to investigate macroscopic quantum behaviors of vibration phonon modes.

Noise, including intrinsic device noise and environmental interference, is a serious concern for sensing applications of SiNW electronic devices. A H₂ annealing process is explored to repair the SiNW surface defects and thus reduce the intrinsic noise by one order of magnitude. To suppress the external interference, lateral bipolar junction transistors (LBJTs) are fabricated on SOI substrate for local signal amplification of the SiNW sensors. Current gain and overall signal-to-noise ratio of the LBJTs are also optimized with an appropriate substrate voltage.

Abstract

Nanopores have emerged as a special class of single-molecule analytical tool that offers immense potential for sensing and characterizing biomolecules such as nucleic acids and proteins. As an alternative to biological nanopores, solid-state nanopores present remarkable versatility due to their wide-range tunability in pore geometry and dimension as well as their excellent mechanical robustness and stability. However, being intrinsically incompatible with biomolecules, surfaces of inorganic solids need be modified to provide desired functionalities for real-life sensing purposes. In this thesis, we presented an exploration of various surface engineering strategies and an examination of several surface associated phenomena pertaining specifically to solid-state nanopores. Based on the parallel sensing concept using arrayed pores, optical readout is mainly employed throughout the whole study.

For the surface engineering aspect, a list of approaches was explored. A versatile surface patterning strategy for immobilization of biomolecules was developed based on selective poly(vinylphosphonic acid) passivation and electron beam induced deposition technique. This scheme was then implemented on nanopore arrays for nanoparticle localization. In addition, vesicle rupture-based lipid bilayer coating was adapted to truncated-pyramidal nanopores, which was shown to be effective for the minimizing DNA-pore interaction. Further, HfO₂ coating by means of atomic layer deposition was employed to prevent the erosion of Si-based pores and to shrink the pore diameter, which enabled reliable investigations of DNA clogging and DNA polymerase docking.

For the surface associated phenomena, several findings were made. The lipid bilayer formation on truncated pyramidal nanopores via instantaneous rupture of individual vesicles was quantified based on combined ionic current monitoring and optical observation.  The probability of pore clogging appeared to linearly increase with the length of DNA strands and applied bias voltage, which could be attributed a higher probability of knotting and/or folding of longer DNA strands and more frequent translocation events at higher voltage. A free-energy based analytical model was proposed to evaluate the DNA-pore interaction and to interpret observed clogging behavior. Finally, docking of DNA polymerase on nanopore arrays was demonstrated using label-free optical method based on Ca²⁺ indicator dyes, which may open the avenue to sequencing-by-synthesis enabled by the docked polymerase.

Abstract

The demand for wireless data communications is rapidly increasing due to several factors including increased internet access, increasingly growing number of mobile users and services, implementation of the Internet of Things (IoT), high-definition (HD) video streaming and video calling. To meet the bandwidth requirement of new and emerging applications, it is necessary to move from the existing microwave bands towards millimeter-wave bands. 

This thesis presents different antenna arrays at 60 GHz and 28 GHz that are integrated with the front-end RFIC to steer the beam in ≈ ±50° in the azimuth plane. The 5G antenna arrays at 28 GHz are designed to provide broadband high data rate services to the end users. In order to transport this high-volume data to the core network, a fixed wireless access (FWA) link demands the implementation of a broadband, high gain and steerable narrow-beam array. The 60 GHz antenna arrays, presented in this thesis, are good candidates for both FWA as well as backhaul communications. The two proposed arrays at 60 GHz (57-66 GHz) are i) a stacked patches array and ii) a connected slots array feeding a high gain lens antenna. The 2×16 stacked patches antenna array shows more than 20 dBi realized gain. The array is integrated with the front-end RFIC and the resulting module shows > 40 dBm measured effective isotropic radiated power (EIRP). The other 60 GHz antenna array is designed as linear connected slots with sixteen equidistant feeding points. The latest is then used as a feeder of a high gain dielectric lens. Peak measured gain of 25.4 dBi is achieved with this antenna.  Moreover, instead of experiencing scan loss, the lens is designed to get higher gain when the beam is steered away from the broadside direction.

Furthermore, two compact antenna arrays are designed at 28 GHz (24.25 - 29.50 GHz). A linear polarized (LP) and a circular polarized (CP) array are realized in the fan-out embedded wafer level ball-grid-array (eWLB) package. In comparison with the PCB arrays, this antenna in package (AiP) solution is not only cost-effective but it also reduces the integration losses because of shorter feed lines and no geometrical discontinuity.  The LP array is realized as a dipole antenna array feeding a novel horn-shaped heatsink.  The RF module gives 34 dBm peak EIRP with beam-steering in ±35°. Besides, the CP antenna array is realized with the help of crossed dipoles and the RF module provides 31 dBm peak EIRP with beam-steering in ±50°.

The data demands are not limited to the telecom industry as the upgradation of accelerators and experiments at the large hadron collider (LHC) at CERN will result in increased event rate thus demanding higher data rate front-end readout systems. This work thus investigates the feasibility of 60 GHz wireless links for the data readout at CERN. For this purpose, the 60 GHz wireless chips are irradiated with 17 MeV protons [dose 7.4 Mrad (RX) & 4.2 Mrad (TX)] and 200 MeV electrons [dose 270 Mrad (RX) & 314 Mrad (TX)] in different episodes. The chips have been found operational in the post-irradiation investigations with some performance degradation. The encouraging results motivate to move forward and investigate the realization of wireless links in such a complex environment.

Abstract

Nanopores are of great interest in study of DNA sequencing, protein profiling and power generation. Among them, solid-state nanopores show obvious advantages over their biological counterparts in terms of high chemical stability and reusability as well as compatibility with the existing CMOS fabrication techniques. Nanopore sensing is most frequently based on measuring ionic current through a nanopore while applying a voltage across it. When an analyte passes through the pore, the ionic current temporarily changes, providing information of the analyte such as its size, shape and surface charge. Although many magnificent reports on using solid-state nanopores have appeared in the literature, several challenges still remain for their wider applications, which include improvement of fabrication reproducibility for mass production of ultra-small nanopores and minimization of measurement instability as well as control of translocation speed and reduction of background noise. This thesis work explores different techniques to achieve robust and high throughput fabrication of sub-10 nm nanopores for different applications.

The thesis starts with presenting various fabrication techniques explored during my PhD studies. Focused ion beam method was firstly employed to drill nanopores in free-standing SiN_x membranes. Sub-10 nm nanopores could be obtained with a focused helium ion beam. But the fabrication throughput was limited with this technique. A new fabrication process combing electron beam lithography (EBL) with reactive ion etching/ion beam etching, which is compatible with the existing CMOS fabrication technology, was developed to realize a high throughput, mass production of nanopores in free-standing SiN_x membranes. However, the smallest size that could be controllably achieved with this process was around 40 nm, which is still far from sub-10 nm in size required for, e.g., DNA sequencing. Finally, by using anisotropic etching of single-crystal silicon in KOH solution, sub-5 nm truncated pyramidal nanopores were mass produced with good process controllability in a silicon-on-insulator (SOI) substrate. In addition, nanopore arrays were also successfully fabricated using a modified EBL based fabrication process.

Then, several sensing application examples using either single nanopores or nanopore arrays were investigated. Translocation of nanoparticles, DNA and proteins were demonstrated using the fabricated single nanopores or nanopore arrays in a single freestanding membrane. Moreover, the kinetics and mechanism of the lipid bilayer formation in nanopore array, aiming to prevent non-specific adsorption, were studied using ionic current measurements. In addition, individual addressability of a solid-state nanopore array on separated freestanding membranes was realized by integrating microfluidics and a customized multiplexer.

Abstract

Gold based label-free electrochemical DNA sensors have been widely studied for biomarker diagnostics. The sensitivity and reproducibility of these sensors are determined by the sensing interface: the DNA modified gold surfaces. This thesis systematically studies the preparation processes of the DNA sensor interfaces as well as their effects on the sensor performance. First, three pretreatment methods to clean the gold electrode surface and their influence on the subsequent binding of thiolated molecules were carefully investigated. As we found that the surface pretreatment method involving cyclic voltammetry (CV) in H₂SO₄ may induce structural changes to the gold surface, thus greatly impacting the thiolated molecule binding, the factors influencing this pretreatment method were studied. Practical guidelines were summarized for preparing a clean and reproducible gold surface prior to functionalization. Afterwards, the effects of the surface coverage density of probe DNA and the salt concentration on the probe-target DNA hybridization on a gold sensing surface were systematically investigated using surface plasmon resonance (SPR) analysis. Based on the SPR results, the maximum potentiometric signal that could be generated by the DNA hybridization on the surface, and the detection limits, were estimated for different experimental conditions. These estimations were further compared with experimental results obtained using silicon nanowire field effect transistors (SiNW FET) with DNA modified gold on the gate oxide. Practical limitations for the potentiometric DNA sensor were analysed and discussed. Finally, the stability and reproducibility issues on the electrochemical impedance spectroscopy (EIS) analyses of DNA hybridization were also studied on the aptamer/mercaptohexanol (MCH)-modified gold surface. The root cause for the drift problems in this type of sensor and the temperature effects on the aptamer/MCH modified surface were identified. This thesis could serve as a practical reference for the preparation and understanding of the sensing interface of gold-based electrochemical DNA sensors.

Abstract

The rise of Big Science projects brings issues related to the energy consumption and the associated environmental impacts of such large-scale facilities. Therefore, environmentally-sustainable developments are undertaken towards the adoption of energy savings and improved energy-efficient approaches. The advent of the superconducting (SC) radio frequency (RF) accelerating cavity is bringing answers to these issues. Such superconducting RF (SRF) cavity is made of niobium that allows much higher accelerating gradients with a minimization of the energy consumption. The SC RF technology is increasingly used in many modern particle accelerators, including: the European Spallation Source (ESS), the X-ray Free Electron Laser (XFEL), the Linac Coherent Light Source (LCLS)-II and the proposed International Linear Collider (ILC).

The innovation of solid state PA technology pushes limits regarding packaging, efficiency, frequency capability, thermal stability, making them more attractive than other well-established alternative technologies, such as vacuum tube technology in mid-range power applications. Through the investigations of designs and techniques, this research goal of the thesis allows to improve solid-state based power generation systems from module to the overall system design. This thesis introduces the single-ended PA design approach in planar technology and at kilowatt level. The design solution unlocks different possibilities including: improved integration, layout flexibility for tuning, and suitably for mass productions that are demanded in future high peak power generation systems. The novel amplifier design is followed by time domain characterization to fully evaluate the pulse profiles of such amplifiers when delivering kilowatt output power level for operation in conjunction with SRF accelerating cavities. Amplitude and phase stability of those amplifiers are also investigated in time-domain. The extracted data can then be used as measurement-based model for predicting factors which could degrade the overall stability of the associated PA.

Future RF power generation systems built around solid state PAs need also efficient combining strategies. Two engineering design solutions are investigated in this thesis aiming for mid- and high- range power combination. One solution is based on a combination of the Gysel structure using suspended strip-line technology for improved power handling capability. Another solution is implementing a radial combiner, which uses re-entrant cavity resonator at 352 MHz and door-nob geometry for coupling at inputs and at the output. These solutions facilitate the scaling up 400 kW for powering ESS spoke cavities while maintaining a high degree of efficiency in RF power generation. This thesis gives insights of system integration and tuning procedures with a demonstration of combining 8 modules, delivering a total of 10 kW output power. Along with the proposed combining solutions at higher power levels, the nominal power block of 10 kW is used as an elementary block to propose scaling up in power till the 400 kW nominal power required by ESS.

Finally, this thesis focuses on implementing an optimal charging scheme for SRF cavities, which helps reducing the wasted energy and improves the overall efficiency operation at future accelerating facilities. Therefore, these results contribute further to the larger adoption of solid state technologies in the future power generation systems for particle accelerators.

Abstract

In the last two decades, a significant development in the field of medical technology occurred worldwide. This development is characterized by the materialization of various body implants and worn devices, that is devices attached to the body. These devices assist doctors and paramedical staff in effectively monitoring the patient’s health and helping increase patients’ average life expectancy. Furthermore, the various implants inside the human body serve different purposes according to the humans’ needs. As this situation became more prominent, the development of protocols and of reliable transmission media is becomes essential to improve the efficiency of inter-device communications. Positive prospects of the use of human tissue for intra-body communication were proven in recent studies. Fat tissues, for example, which also work as energy banks for human beings, can be potentially used in intra-body communications as transmission media. In this thesis, the fat (adipose) tissue’s function as an intra-body communication channel was investigated. Therefore, various simulations and experimentations were performed in order to characterize the reliability of the fat tissue in terms of communication, considering, for example, the effect that the variability in the thickness of adipose and muscular tissues could have on the communication performance, and the possible effect that the variability in the transmitted signal power could have on the data packet reception. Fat tissue displays superior performance in comparison to muscle tissue in the context of a low loss communication channel. For example, at 2.45 GHz, the path losses of ~0.7 dB/cm and ~1.9 dB/cm were observed for phantom and ex-vivo measurements, respectively. At a higher frequency of 5.8 GHz, the ex-vivo path loss was around 1.4 dB/cm. It was concluded from the results that the adipose tissue could function as a reliable medium supporting intra-body communication even under low power transmitted signals. Moreover, although the presence of thick blood vessels could degrade the signal strength, the results show that communication is possible even under the presence of perturbant tissues. Overall, the results of this thesis would provide a foundation in this area and assist researchers in developing innovative and solutions for intra-body communication.

Abstract

In recent years, the use of microwave techniques for medical diagnostics has experienced impressive developments. It has demonstrated excellent competencies in various modalities such as using non-ionizing electromagnetic waves, providing non-invasive diagnoses, and having the ability to penetrate human tissues within the GHz range. However, due to anatomical, physiological, and biological variations in the human body, certain obstacles are present. Moreover, there are accuracy problems such as the absence of numerical models and experimental data, difficulty in conducting tests due to safety issues with human subjects, and also practical restrictions in clinical implementation. With the presence of these issues, a better understanding of the microwave technique is essential to further improve its medical application and to introduce alternative diagnostic methods that can detect and monitor various medical conditions in real time.

The first part of this thesis focuses on measurement systems for the microwave technique in terms of sensor design and development, numerical analysis, permittivity measurement, and phantom fabrication. The aim is to investigate the feasibility of flexible systems with different fields of application including a microwave sensor system for measuring the healing progression of bone defects present in lower extremity trauma, bone regeneration in craniotomy for craniosynostosis treatments, and dielectric variation for burn injuries. The microwave sensor which utilizes the contrast in dielectric constant between various tissues was used as the primary sensor for the proposed application. This involved detailed optimization of the sensor for greater sensitivity. The experimental work carried out in the lab environment showed that the microwave sensor was able to detect the contrast in dielectric properties so that it can give an indication of the healing status for actual clinical scenarios.

The second part of the thesis is making a significant step towards its practical implementation by establishing a system that can detect and monitor the rate of healing progression with fast data acquisition speed of microseconds, and developing an efficient user interface to convert raw microwave data into legible clinical information in terms of bone healing and burn injuries. As an extension to this thesis, clinical studies were conducted and ethical approval for conducting tests on human subjects was obtained for the development of a microwave medical system. The results showed a clear difference in healing progressions due to high detection capability in terms of dielectric properties of different human tissues. All of these contributions enable a portable system to complement existing medical applications with the aim of providing more advanced healthcare systems.

Abstract

The material supply to build renewable energy conversion systems needs to be considered from both a cost and an energy security perspective. For Cu(In,Ga)Se₂ (CIGS) thin film solar cells the use of indium in the absorber layer is most problematic. The material input per service unit can be reduced, if the absorber layers are thinned down without a loss in power conversion efficiency.

Thinning down absorber layers can increase the conversion efficiency. However, for real CIGS solar cells absorption losses and recombination rates at the rear surface between the CIGS absorber and the Mo rear contact as well as shunt-like behavior increase. Thus, both rear surface passivation and optical management are essential for maintaining high power conversion efficiencies.

In this work, thin oxide layers, so-called passivation layers, are introduced between the CIGS absorber layer and the Mo contact. They can passivate the CIGS surface, if the CIGS-oxide interface has a lower defect density than the CIGS-Mo interface and/or if they contain a negative fixed oxide charge, which increases the hole concentration and reduces the electron concentration in the CIGS in the vicinity of the oxide.

As these oxides are insulators, electrical conduction through the passivation layer has to be ensured. In this work, nanopoint contacts were etched into ALD-Al₂O₃ passivation layers in CIGS solar cells. These solar cells had 0.5 -1.5 µm thin absorber layers with a low In content and a high band gap. Ga grading was not used. Although absorber layers with a high Ga content have a short minority carrier diffusion length, a passivation effect could be discerned with the help of external quantum efficiency measurements and current-voltage measurements under varying temperatures in combination with optical and electrical modeling with a two-diode model. Moreover, the possibility of leaving out the additional fabrication step has been explored for ALD-Al₂O₃ and HfO₂ as passivation layers. The results suggest that the passivation layer does not necessarily need to be opened for electrical conduction in an additional fabrication step, if sodium fluoride (NaF) is deposited onto Al₂O₃ layers prior to CIGS evaporation. In this case solar cells with 215 nm absorber layers and 6 nm thin passivation layers have a power conversion efficiency of 8.6 %, which is 3 % (absolute) higher than the conversion efficiency on a reference. Shunt-like behavior is additionally reduced. For the HfO₂ layers photoluminescence data indicate a good passivation effect, but the layers need to be opened up to ensure conduction.

Abstract

Three-dimensional integrated circuits have great potential for further increasing the number of transistors per area by stacking several device tiers on top of each other and without the need to continue the evermore complicated and expensive down-scaling of transistor dimensions. Among the different approaches towards the realization of such circuits, the monolithic approach, i.e. the tier-by-tier fabrication on a single substrate, is the most promising one in terms of integration density. Germanium is chosen as a substrate material instead of silicon in order to take advantage of its low fabrication temperatures as well as its high carrier mobilities. In this thesis, the work on two key components for the realization of such germanium-based three-dimensional integrated circuits is presented:the source/drain contacts to germanium the interconnects.

As a potential source/drain contact material, nickel germanide is investigated.In particular, the process temperature windows for the fabrication of morphologically stable nickel germanide layers formed from initial nickel layers below 10 nm are identified and the reaction between nickel and germanium is further studied by means of in-situ x-ray diffraction. The agglomeration temperature of nickel germanide is increased by 100 °C by the addition of tantalum and tungsten interlayers and capping layers. In an effort to more thoroughly characterize the contacts, a method to reliably extract the specific contact resistivity is implemented on germanium.

As a potential interconnect material cobalt is investigated. In a first step, highly conductive cobalt thin films are demonstrated by means of high-power impulse magnetron sputtering. The high conductivity of the cobalt films is owing to big grains, high density, high purity, and smooth interfaces. In a second step, the potential of high-power impulse magnetron sputtering for the metallization of nanostructures is further explored.

Abstract

Nanopore based sensing technology has been widely studied for a broad range of applications including DNA sequencing, protein profiling, metabolite molecules, and ions detection. The nanopore technology offers an unprecedented technological solution to meeting the demands of precision medicine on rapid, in-field, and low-cost biomolecule analysis. In general, nanopores are categorized in two families: solid-state nanopore (SSNP) and biological nanopore. The former is formed in a solid-state membrane made of SiN_x, SiO₂, silicon, graphene, MoS₂, etc., while the latter represents natural protein ion-channels in cell membranes. Compared to biological pores, SSNPs are mechanically robust and their fabrication is compatible with traditional semiconductor processes, which may pave the way to their large-scale fabrication and high-density integration with standard control electronics. However, challenges remain for SSNPs, including poor stability, low repeatability, and relatively high background noise level. This thesis explores SSNPs from basic physical mechanisms to versatile applications, by entailing a balance between theory and experiment.

The thesis starts with theoretical models of nanopores. First, resistance of the open pore state is studied based on the distribution of electric field. An important concept, effective transport length, is introduced to quantify the extent of the high field region. Based on this conductance model, the nanopores size of various geometrical shapes can be extracted from a simple resistance measurement. Second, the physical causality of ionic current rectification of geometrically asymmetrical nanopores is unveiled. Third, the origin of low-frequency noise is identified. The contribution of each noise component at different conditions is compared. Forth, a simple nano-disk model is used to describe the blockage of ionic current caused by DNA translocation. The signal and noise properties are analyzed at system level.

Then, nanopore sensing experiments are implemented on cylinder SiN_x nanopores and truncated-pyramid silicon nanopores (TPP). Prior to a systematic study, a low noise electrical characterization platform for nanopore devices is established. Signal acquisition guidelines and data processing flow are standardized. The effects of electroosmotic vortex in TPP on protein translocation dynamics are excavated. The autogenic translocation of DNA and proteins driven by the pW-level power generated by an electrolyte concentration gradient is demonstrated. Furthermore, by extending to a multiple pore system, the group translocation behavior of nanoparticles is studied. Various application scenarios, different analyte categories and divergent device structures accompanying with flexible configurations clearly point to the tremendous potential of SSNPs as a versatile sensor.

Abstract

This thesis work explores electrical conductors based on few-layer graphene flakes as an enabler for low-cost, mechanically flexible, and high-conductivity conductors in large area flexible and printed electronic devices. The flakes are deposited from aqueous solutions and processed at low temperature.

Graphene is selected for its excellent properties in mechanical, optical, electronic, and electrical aspects. However, thin films of pristine few-layer graphene flakes deposited from dispersions normally exhibit inferior electrical conductivity. One cause responsible for this problem is the loose stacking and random orientation of graphene flakes in a graphene deposition. We have solved this problem by implementing a simple post-deposition treatment leading to dramatically densified and planarized thin films. Significantly increased electrical conductivity by ~20 times is obtained. The 1-pyrenebutyric acid tetrabutylammonium salt as an exfoliation enhancer and dispersant in water yields ~110 S/m in conductivity when the graphene based thin films are processed at 90 °C. In order to achieve higher conductivity, a room-temperature method for site-selective copper electroless deposition has been developed. This method is of particular interest for the self-aligned copper deposition to the predefined graphene films. The resultant two-layer graphene/copper structure is characterized by an overall conductivity of ~7.9 × 10⁵ S/m, an increase by ~7000 times from the template graphene films. Several electronic circuits based on the graphene/copper bilayer interconnect have been subsequently fabricated on plastic foils as proof-of-concept demonstrators. Alternatively, highly conductive composites featuring graphene flakes coated with silver nanoparticles with electrical conductivity beyond 10⁶ S/m can be readily obtained at 100 ^oC. Moreover, a highly conductive reduced-graphene-oxide/copper hybrid hydrogel has been achieved by mixing aqueous graphene oxide solution and copper-containing Fehling's solution. The corresponding aerogel of high porosity exhibits an apparent electrical conductivity of ~430 S/m and delivers a specific capacity of ~453 mAh g⁻¹ at current density of 1 A/g. The experimental results presented in this thesis show that the solution-phase, low-temperature fabrication of highly conductive graphene-based materials holds promises for flexible electronics and energy storage applications. 

Abstract

In the past decades, silicon nanowire field-effect transistors (SiNWFETs) have been explored for label-free, highly sensitive, and real-time detections of chemical and biological species. The SiNWFETs are anticipated for sensing analyte at ultralow concentrations, even at single-molecule level, owing to their significantly improved charge sensitivity over large-area FETs. In a SiNWFET sensor, a change in electrical potential associated with biomolecular interactions in close proximity to the SiNW gate terminal can effectively control the underlying channel and modulate the drain-to-source current (I_DS) of the SiNWFET. A readout signal is therefore generated. This signal is primarily determined by the surface properties of the sensing layer on the gate terminal, with sensitivity close up to the Nernstian limit widely demonstrated. To achieve a high signal-to-noise ratio (SNR), it is essential for the SiNWFETs to possess low noise of which intrinsic device noise is one of the major components. In metal-oxide-semiconductor (MOS)-type FETs, the intrinsic noise mainly results from carrier trapping/detrapping at the gate oxide/semiconductor interface and it is inversely proportional to the device area.

This thesis presents a comprehensive study on design, fabrication, and noise reduction of SiNWFET-based sensors on silicon-on-oxide (SOI) substrate. A novel Schottky junction gated SiNWFET (SJGFET) is designed and experimentally demonstrated for low noise applications. Firstly, a robust process employing photo- and electron-beam mixed-lithography was developed to reliably produce sub-10 nm SiNW structures for SiNWFET fabrication. For a proof-of-concept demonstration, MOS-type SiNWFET sensors were fabricated and applied for multiplexed ion detection using ionophore-doped mixed-matrix membranes as sensing layers. To address the fundamental noise issue of the MOS-type SiNWFETs, SJGFETs were fabricated with a Schottky (PtSi/silicon) junction gate on the top surface of the SiNW channel, replacing the noisy gate oxide/silicon interface in the MOS-type SiNWFETs. The resultant SJGFETs exhibited a close-to-ideal gate coupling efficiency (60 mV/dec) and significantly reduced device noise compared to reference MOS-type SiNWFETs. Further optimization was performed by implementing a three-dimensional Schottky junction gate wrapping both top surface and two sidewalls of the SiNW channel. The tri-gate SJGFETs with optimized geometry exhibited significantly enhanced electrostatic control over the channel, thereby confined I_DS in the SiNW bulk, which greatly improved the device noise immunity to the traps at bottom buried oxide/silicon interface. Finally, a lateral bipolar junction transistor (LBJT) was also designed and fabricated on a SOI substrate aiming for immediate sensor current amplification. Integrating SJGFETs with LBJTs is expected to significantly suppress environmental interference and improve the overall SNR especially under low sensor current situations.

Abstract

The outstanding properties of graphene make it attractive ink filler for conductive inks which plays an important role in printed electronics. This thesis focuses on the ink formulation based on graphene and graphene oxide (GO).

Liquid phase exfoliation of graphite is employed to prepare graphene dispersions, i.e., shear- and electrochemical exfoliation. High concentration graphene dispersions with small size, few-layer graphene platelets are obtained by both methods. With the addition of ethyl cellulose stabilizer, shear-exfoliated graphene platelets in NMP were successfully inkjet printed on different substrates. The printed graphene film with electrical conductivity of ~3^10⁴ S/m was obtained after annealing at 350 °C for one hour. Alternatively, the electrochemically exfoliated graphene nano-platelets were collected and redispersed in DMF to form inks. The printed film of conductivity ~2.5^10³ S/m was obtained after annealing at 300 °C for one hour.

Water based GO/Ag hybrid inks were developed for screen printing. When high concentration GO aqueous dispersion was mixed with reactive silver ink, the viscosity of the mixture increased instantly to above 1000 cP as a result of reactions between oxygen functional groups (OFGs) on GO sheets and ingredients in the reactive silver ink. When the screen printed lines with different GO:Ag ratios were annealed in air, the conductivity of the resultant reduced graphene oxide/silver nanoparticles (RGO/AgNP_X) composites decreased as silver content increased. As oxygen enriched compounds in RGO/AgNP_X composites were detected, we proposed that AgOx compounds were generated on the AgNPs surface, which raised the contact resistance between AgNPs and RGO flakes. To solve this problem, the printed patterns were instead annealed in reducing gas (Ar/H₂ 5%). The electrical conductivity ~2.0^10⁴ S/m was then achieved.

Furthermore, the reduction of GO using ammonium formate as reducing reagent was investigated. When applying a hydrothermal method, ammonium formate shows excellent reduction ability, surpassing the widely used reducing agent, L-ascorbic acid, under same condition. Elemental analysis shows the C/O ratio of RGO as high as ~11 and most OFGs were removed in the reduction process. Meanwhile, incorporated nitrogen atoms introduced active sites in resultant RGO, making it a promising electrocatalyst for oxygen reduction reaction.

Abstract

The aim of this thesis is to understand the influence of the deposition process and resulting film properties on Cu₂ZnSnS₄ (CZTS) thin film solar cells. Two main aspects are studied, namely formation of absorber precursors by sputtering, and alternative Cd-free buffer materials with improved band alignment.

Reactive sputtering is used to grow dense and homogeneous precursor films containing all elements needed for CZTS absorbers. The addition of H₂S gas to the inert Ar sputter atmosphere leads to a drastic decrease of Zn-deposition rate due to the sulfurization of the target. Sulfurization also leads to instabilities for targets made of CuSn, Cu and Cu₂S, while sputtering from CuS gave acceptable process stability.

The H₂S/Ar-ratio also affects film morphology and composition. Precursors with sulfur content close to stoichiometric CZTS have a columnar, crystalline structure. Materials analysis suggests a non-equilibrium phase with a cubic structure, where each S atom is randomly surrounded by 2:1:1 Cu:Zn:Sn-atoms, respectively. Substrate heating during sputtering is shown to be important to avoid cracks in the annealed films while stress in the precursor films is not observed to affect the absorber or solar cell quality.

Sputtering from compound targets in Ar-atmosphere yields precursor properties similar to those from reactive sputtering at high H₂S/Ar-ratios and both types can be processed into well-performing solar cells.

Additionally, a low temperature treatment of CZTS absorbers in inert atmosphere prior to buffer layer growth is shown to affect the device properties, which indicates that the thermal history of the CZTS absorber is important.

The alternative buffer system ZnO_1-xS_x is found to yield lower efficiencies than expected, possibly due to inferior interface or buffer quality. The Zn_1-xSn_xO_y (ZTO) buffers instead give better performance than their CdS references. For optimized parameters, the activation energy for recombination coincides with the energy of the photoluminescence peak of the absorber. This can be interpreted as a shift of dominant recombination path from the interface to the CZTS bulk. A well-performing CZTS-ZTO device with antireflective coating yielded an efficiency of 9.0 %, which at the time of publication was the highest value published for a Cd-free pure-sulfide CZTS solar cell.

Abstract

The CIS solar cell family features both a high stability and world-class performances. They can be deposited on a wide variety of substrates and absorb the entire solar spectrum only using a thickness of a few micrometers. These particularities allow them to feature the most positive Energy returned on energy invested (EROI) values and the shortest Energy payback times (EPBT) of all the main photovoltaic solar cells. Using mainly electron microscopy characterization techniques, this thesis has explored the questions related to the interface control in thin-film photovoltaic solar cells based on CuInSe₂ (CIS) absorber materials. Indeed, a better understanding of the interfaces is essential to further improve the solar cell conversion efficiency (currently around 23%), but also to introduce alternative substrates, to implement various alloying (Ga-CIS (CIGS), Ag-CIGS (ACIGS)…) or even to assess alternative buffer layers.

The thread of this work is the understanding and the improvement of the interface control. To do so, the passivation potential of Al₂O₃ interlayers has been studied in one part of the thesis. While positive changes were generally measured, a subsequent analysis has revealed that a detrimental interaction could occur between the NaF precursor layer and the rear Al₂O₃ passivation layer. Still within the passivation research field, incorporation of various alkali-metals to the CIS absorber layer has been developed and analyzed. Large beneficial effects were ordinarily reported. However, similar KF-post deposition treatments were shown to be potentially detrimental for the silver-alloyed CIGS absorber layer. Finally, part of this work dealt with the limitations of the thin-barrier layers usually employed when using steel substrates instead of soda-lime glass ones. The defects and their origin could have been related to the steel manufacturing process, which offered solutions to erase them.

Electron microscopy, especially Transmission electron microscopy (TEM), was essential to scrutinize the local changes occurring at the different interfaces within a few nanometers. The composition variation was measured with both Electron energy loss spectroscopy (EELS) and Energy dispersive X-ray spectroscopy (EDS) techniques. Finally, efforts have been invested in controlling and improving the FIB sample preparation, which was required for the TEM observations in our case.

Abstract

The sun provides us with a surplus of energy convertible to electricity using solar cells. This thesis focuses on solar cells based on chalcopyrite (CIGSe) as well as kesterite (CZTS(e)) absorber layers. These materials yield record efficiencies of 20.4 % and 11.1 %, respectively. Especially for CZTS(e), the absorber layers often do not consist of one single desired phase but can exhibit areas with deviating material properties, referred to as secondary phases. Furthermore, several material layers are required for a working solar cell, each exhibiting interfaces. Even though secondary phases and interfaces represent a very small fraction of the solar cell they can have a profound influence on the over-all electrical solar cell characteristics. As such, it is crucial to understand how secondary phases and interfaces influence the local electrical characteristics.

Characterising secondary phases and interfaces is challenging due to their small sample volume and relatively small differences in composition amongst others. This is where electronmicroscopy, especially transmission electron microscopy, offers valuable insight to material properties on the microscopic scale. The main challenge is, however, to link these material properties to the corresponding electrical characteristics of a solar cell.

This thesis uses electron beam induced current imaging and introduces a new method for JV characterisation of solar cells on the micron scale. Combining microscopic structural and electrical characterisation techniques allowed identifying and characterising local defects found in the absorber layer of CIGS solar cells after thermal treatment. Furthermore, CZTSe solar cells in this thesis exhibited a low photo-current density which is traced to the formation of a current blocking ZnSe secondary phase at the front contact interface. The electron microscopy work has contributed to an understanding of the chemical stability of CZTS and has shown the need for an optimised back contact interface in order to avoid chemical decomposition reactions and formation of detrimental secondary phases. With this additional knowledge, a comprehensive picture of the material properties from the macroscopic down to the microscopic level can be attained throughout all required material layers.

Abstract

Emerging macro- and flexible electronic applications such as foldable displays, artificial skins, and smart textiles grow rapidly into the market. Solution-processed thin-film transistors (TFTs) based on single-walled carbon nanotubes (SWCNTs) as the semiconductor channel can offer high performance, low cost and versatility for macro- and flexible electronics. Major challenges to the development of SWCNT-based TFTs include: (i) hysteresis in their transfer characteristics (TCs), (ii) difficulties in simultaneous achievements of high on-state current I_on and large on/off current ratio I_on/I_off, and (iii) poor uniformity and scalability resulting from the poor solution processability. This thesis aims at developing reliable and simple process techniques for fabrication of the SWCNT-based TFTs that possess the afore-stated characteristics. It presents a systematic investigation to not only explore the fundamental device physics, but also develop novel fabrication methods for enhancement of device performance.

First, issues related to the measurement of gate capacitance (C_g), the determination of current scalability, and the hysteresis in randomly networked SWCNTs are properly addressed. This leads to the establishment of a comprehensive methodology for extraction of carrier mobility (μ) for the SWCNT-based TFTs. In detail, the large hysteresis is effectively suppressed by adopting a pulsed drain current-gate voltage (I_d-V_g) method in which the polarity of the gate pulse was alternating during the measurement. Different from most reported methods in the literature, C_g is accurately determined in our case by performing direct capacitance-voltage measurement on the TFTs.

Second, with the employment of functional composites comprising SWCNTs embedded in a semiconducting polymer, poly-9,9 dioctyl-fluorene-cobithiophene (F8T2), as the semiconducting channel via facile solution processes under ambient conditions, the fabricated TFTs exhibit outstanding electrical performance with: (i) negligible hysteresis, (ii) high μ, (iii) high I_on and large I_on/I_off, (iv) excellent uniformity and dimensional scalability, and (v) good stability. These highly desired performance parameters are achieved owing to an ideal composite structure with metallic SWCNTs being selectively removed and the remaining semiconducting SWCNTs being well wrapped by the polymer matrix.

Finally, the developed TFTs basing on the SWCNT/F8T2 composite are used as the building block to construct some logic circuits. The resultant inverters, NANDs, and NORs are found to retain the small-hysteresis characteristics, with a cut-off frequency reaching 100 kHz. The results presented in this thesis advance the state-of-art SWCNT-based macroelectronics.

Abstract

Point-of-care (POC) diagnostics presents a giant market opportunity with profound societal impact. In particular, specific detection of DNA and protein markers can be essential for early diagnosis of e.g. cancer, cardiovascular disease, infections or allergies. Today, identification of these markers often requires extensive laboratory work and hence is expensive and time consuming. Current methods for recognition and detection of specific biomolecules are mostly optics based and thus impose severe limitations as to convenience, specificity, sensitivity, parallel processing and cost reduction.

Electronic sensors based on silicon nanowire field-effect transistors have been reported to be able to detect biomolecules with concentrations down to femtomolar (fM) level with high specificity. Although the reported capability needs further confirmation, the CMOS-compatible fabrication process of such sensors allows for low cost production and high density integration, which are favorable for POC applications. This thesis mainly focuses on the development of a multiplex detection platform based on silicon nanowire field-effect sensors integrated with a microfluidic system for liquid sample delivery. Extensive work was dedicated to developing a top-down fabrication process of the sensors as well as an effective passivation scheme. The operation mechanism and coupling efficiencies of different gate configurations were studied experimentally with the assistance of numerical simulation and equivalent circuits. Using pH sensing as a model system, large effort was devoted to identifying sources for false responses resulting from the instability of the inert-metal gate electrode. In addition, the drift mechanism of the sensor operating in electrolyte was addressed and a calibration model was proposed. Furthermore, protein detection experiments were performed using small-sized Affibody molecules as receptors on the gate insulator to tackle the Debye screening issue. Preliminary results showed that the directionality of the current changes in the sensors was in good agreement with the charge polarities of the proteins. Finally, a graphene-based capacitor was examined as an alternative to the nanowire device for field-effect ion sensing. Our initial attempts showed some attractive features of the capacitor sensor.

Abstract

The recent development of the commercially viable thin film electro-acoustic technology has triggered a growing interest in the research of plate guided wave or Lamb wave components owing to their unique characteristics. In the present thesis i) an experimental study of the thin film plate resonators (FPAR) performance operating on the lowest symmetrical Lamb wave (S0) propagating in highly textured AlN membranes versus a variety of design parameters has been performed. The S0 mode is excited through an Interdigital Transducer and confined within the structure by means of reflection from metal strip gratings. Devices operating in the vicinity of the stop-band center exhibiting a Q-value of up to 3000 at a frequency around 900MHz have been demonstrated. Temperature compensation of this type of devices has been studied theoretically and successfully realized experimentally for the first time. Further, integrated circuit-compatible S0 Lamb based two-port FPAR stabilized oscillators exhibiting phase noise of -92 dBc/Hz at 1 kHz frequency offset with feasible thermal noise floor below -180 dBc/Hz have been tested under high power for a couple of weeks. More specifically, the FPARs under test have been running without any performance degradation at up to 27 dBm loop power. Further, the S0 mode was experimentally demonstrated to be highly mass and pressure sensitive as well as suitable for in-liquid operation, which together with low phase noise and high Q makes it very suitable for sensor applications; ii) research in view of FPARs operating on other types of Lamb waves as well as novel operation principles has been initiated. In this work, first results on the design, fabrication and characterization of two novel type resonators: The Zero Group Velocity Resonators (ZGVR) and The Intermode-Coupled Thin Film Plate Acoustic Resonators (IC-FPAR), exploiting new principles of operation have been successfully demonstrated. The former exploits the intrinsic zero group velocity feature of the S1 Lamb mode for certain combination of design parameters while the latter takes advantage of the intermode interaction (involving scattering) between S0 and A1 Lamb modes through specially designed metal strip gratings (couplers). Thus both type of resonators operate on principles of confining energy under IDT other than reflection.

Abstract

Thin film technology is of great significance for a variety of products, such as electronics, anti-reflective or hard coatings, sensors, solar cells, etc. This thesis concerns the synthesis of thin functional films, reactive magnetron sputter deposition process as such and the physical and functional characterization of the thin films synthesized. Characteristic for reactive sputtering processes is the hysteresis due to the target poisoning. One particular finding in this work is the elimination of the hysteresis by means of a mixed nitrogen/oxygen processing environment for dual sputtering of Alumina-Zirconia thin films. For a constant moderate flow of nitrogen, the hysteresis could be eliminated without significant incorporation of nitrogen in the films. It is concluded that optimum processing conditions for films of a desired composition can readily be estimated by modeling. The work on reactively sputtered SiO₂–TiO₂ thin films provides guidelines as to the choice of process parameters in view of the application in mind, by demonstrating that it is possible to tune the refractive index by using single composite Si_x/TiO₂ targets with the right composition and operating in a suitable oxygen flow range. The influence of the target composition on the sputter yield is studied for reactively sputtered titanium oxide films. It is shown that by using sub-stoichiometric targets with the right composition and operating in the proper oxygen flow range, it is possible to increase the sputter rate and still obtain stoichiometric coatings. Wurtzite aluminum nitride (w-AlN) thin films are of great interest for electro-acoustic applications and their properties have in recent years been extensively studied. One way to tailor material properties is to vary the composition by adding other elements. Within this thesis (Al,B)N films of the wurtzite structure and a strong c-axis texture have been grown by reactive sputter deposition. Nanoindentation experiments show that the films have nanoindentation hardness in excess of 30 GPa, which is as hard as commercially available hard coatings such as TiN. Electrical properties of w-(Al,B)N thin films were investigated. W-(Al,B)N thin films are found to have a dielectric strength of ~3×10⁶ V/cm, a relatively high k-value around 12 and conduction mechanisms similar to those of AlN. These results serve as basis for further research and applications of w-(Al,B)N thin films. An AlN thin film bulk acoustic resonator (FBAR) and a solidly mounted resonator (SMR) together with a microfluidic transport system have been fabricated. The fabrication process is IC compatible and uses reactive sputtering to deposit piezoelectric AlN thin films with a non-zero mean inclination of the c-axis, which allows in-liquid operation through the excitation of the shear mode. The results on IC-compatibility, Q-values, operation frequency and resolution illustrate the potential of this technology for highly sensitive low-cost micro-biosensor systems for applications in, e.g. point-of-care testing.

Abstract

Today’s MEMS technology allows manufacturing of miniaturized, low power sensors that sometimes exceeds the performance of conventional sensors. The pressure sensor market today is dominated by MEMS pressure sensors.

In this thesis two different pressure sensor techniques are studied. The first concerns ways to improve the sensitivity in the most commonly occurring pressure sensor, namely such based on the piezoresistive technique. Since the giant piezoresistive effect was observed in silicon nanowires, it was assumed that a similar effect could be expected in nano-thin silicon films. However, it turned out that the conductivity was extremely sensitive to substrate bias and could therefore be controlled by varying the backside potential. Another important parameter was the resistivity time drift. Long time measurements showed a drastic variation in the resistance. Not even after several hours of measurement was steady state reached. The drift is explained by hole injection into the buried oxide as well as existence of mobile charges. The piezoresistive effect was studied and shown to be of the same magnitude as in bulk silicon. Later research has shown the existence of such an effect where the film thickness has to be less than around 20 nm. 

The second area that has been studied is the pressure sensitivity of in acoustic resonators. Aluminium nitride thin film plate acoustic resonators (FPAR) operating at the lowest-order symmetric (S0), the first-order asymmetric (A1) as well as the first-order symmetric (S1) Lamb modes have been theoretically and experimentally studied in a comparative manner. The S0 Lamb mode is identified as the most pressure sensitive FPAR mode. The theoretical predictions were found to be in good agreement with the experiments. Additionally, the Lamb modes have been tested for their sensitivities to mass loading and their ability to operate in liquids, where the S0 mode showed good results.

Finally, the pressure sensitivity in aluminium nitride thin film bulk wave resonators employing c- and tilted c-axis texture has been studied. The c-axis tilted FBAR demonstrates a substantially higher pressure sensitivity compared to its c-axis oriented counterpart. 

Abstract

The use of solar cells for energy production has indeed a bright future. Reduction of cost for fabrication along with increased efficiency are key features for a market boom, both achieved as a result of increased knowledge of the technology. Especially the thin film solar cell technology with absorbers made of Cu(In,Ga)Se₂ (CIGS) is promising since it has proven high power conversion efficiency in combination with a true potential for low cost fabrication.

In this thesis different recipes for fabrication of the Cu(In,Ga)Se₂ absorber layer have been studied. The deposition technique used has been co-evaporation from elemental sources. For all depositions the substrate has been heated to a constant temperature of 500 ºC in order for the growing absorber to form a chalcopyrite phase, necessary for the photovoltaic functionality. The selenium has been evaporated such to always be in excess during depositions whereas the metal ratio Cu/(In+Ga) has been varied according to different recipes but always to be less than one at the end of the process. In the work emphasis has been on the radiative properties of the CIGS film during growth.

The substrate heater has been temperature controlled to maintain the constant set temperature of the substrate, regardless of varying emitted power caused by changing surface emissivity. Depending on the growth conditions the emissivity of the growing film is changing, leading to a readable variation in the electrical power to the substrate heater.

Since the thermal radiation from the substrate during growth has been of central focus, this has been studied in detail. For this reason the substrate has been treated as an optical stack composed of glass/Mo/Cu(In,Ga)Se₂/Cu_xSe which determine the thermally radiated power by its emissivity. An optical model has been adopted to simulate the emissivity of the stack. In order to use the model, the optical constants for Cu(In,Ga)Se₂ and Cu_xSe have been derived for the wavelength interval 2 μm to 20 μm. The simulation of the emissivity of the stack during CIGS growth agreed well with what has been seen for actual growth. Features of the OP-signal could hereby be explained as a result of film thickness of Cu(In,Ga)Se₂ and Cu_xSe respectively. This is an important knowledge for an efficient fabrication in large scale.

Abstract

A deeper understanding of Cu(In,Ga)Se₂ (CIGS) solar cells is important for the further improvement of these devices. This thesis is focused on the use of electrical modelling as a tool for pursuing this aim. Finished devices and individual layers are characterized and the acquired data are used as input in the simulations. Band gap gradients are accounted for when modelling the devices. The thesis is divided into two main parts. One part that treats the influence of cadmium free buffer layers, mainly atomic layer deposited (Zn,Mg)O, on devices and another part in which the result of CIGS absorber layer modifications is studied. Recombination analysis indicates that interface recombination is limitting the open circuit voltage (V_oc) in cells with ZnO buffer layers. This recombination path becomes less important when magnesium is introduced into the ZnO giving a positive conduction band offset (CBO) towards the CIGS absorber layer. Light induced persistent photoconductivity (PPC) is demonstrated in (Zn,Mg)O thin films. Device modelling shows that the measured PPC, coupled with a high density of acceptors in the buffer-absorber interface region, can explain light induced metastable efficiency improvement in CIGS solar cells with (Zn,Mg)O buffer layers. It is shown that a thin indium rich layer closest to the buffer does not give any significant impact on the performance of devices dominated by recombination in the CIGS layer. In our cells with CdS buffer the diffusion length in the CIGS layer is the main limitting factor. A thinner CIGS layer improves V_oc by reducing recombination. However, for thin enough absorber layers V_oc deteriorates due to recombination at the back contact. Interface recombination is a problem in thin devices with Zn(O,S) buffer layers. This recombination path is overshadowed in cells of standard thickness by recombination in the CIGS bulk. Thin cells with Zn(O,S) buffer layers have a higher efficiency than CdS cells with the same absorber thickness.

Abstract

Solar cells constitute the most direct way of converting solar energy to electricity, and thin-film solar-cell technologies have lately been growing in importance, allowing the fabrication of less expensive modules that nonetheless have good power-conversion efficiencies. This thesis focuses on solar cells based on Cu(In,Ga)Se₂, which is the thin-film technology that has shown the highest conversion efficiency to date, reaching 20.3 % on the laboratory scale. Solar modules still have some way to go to become entirely competitive with existing energy technologies, and there are two possible paths to this goal: Firstly, reducing their manufacturing costs, for instance by minimizing the material usage per module and/or by increasing the throughput of a given factory; and secondly, increasing the power output per module in other words, the module efficiency. The subject matters of this thesis are related to those two approaches.

The first issue investigated is the possibility for reducing the thickness of the Cu(In,Ga)Se₂ layer and compensating for lost absorption by using a ZrN back reflector. ZrN layers are fabricated by reactive sputtering and I present a method for tuning the sputtering parameters so as to obtain a back reflector with good optical, electrical and mechanical properties. The reflector layer cannot be used directly in CIGS devices, but relatively good devices can be achieved with a precursor providing a homogeneous supply of Na, the addition of a very thin sacrificial Mo layer that allows the formation of a film of MoSe₂ passivating the back contact, and optionally a Ga gradient that further keeps electrons away from the back contact.

The second field of study concerns the three-stage CIGS coevaporation process, which is widely used in research labs around the world and has yielded small-area cells with highest efficiencies, but has not yet made it to large scale production. My focus lies on the development and the effect of gradients in the [Ga]/[In+Ga] ratio. On the one hand, I investigate 'intrinsic' gradients (ones that form autonomously during the evaporation), and present a formation model based on the differing diffusivity of Ga and In atoms in CIGS and on the development along the quasi-binary tie line between (In,Ga)₂Se₃ and Cu₂Se. On the other hand, I determine how the process should be designed in order to preserve 'extrinsic' gradients due to interdiffusion. Lastly, I examine the electrical effects of Ga-enhancement at the back and at the front of the absorber and of In-enhancement at the front. Over a wide range, In-rich top layers prove to have no or a weakly beneficial effect, while Ga-rich top regions pose a high risk to have a devastating effect on device performance.

Abstract

Solar irradiation is a vast and plentiful source of energy. The use of photovoltaic (PV) devices to convert solar energy directly to electrical energy is an elegant way of sustainable power generation which can be distributed or in large PV plants based on the need. Solar cells are the small building blocks of photovoltaics and when connected together they form PV modules. Thin film solar cells require significantly less energy and raw materials to be produced, as compared to the dominant Si wafer technologies. CIGS thin film solar cells are considered to be the most promising thin film alternative due to its proven high efficiency.

Most thin film PV modules utilise monolithic integration, whereby thin film patterning steps are included between film deposition steps, to create interconnection of individual cells within the layered structure. The state of the art is that CIGS thin film modules are made using one laser patterning step (P1) and two mechanical patterning steps (P2 and P3). Here we present work which successfully demonstrates the replacement of mechanical patterning by laser patterning methods. The use of laser ablation promises such advantages as increased active cell area and reduced maintenance and downtime required for regular replacement of mechanical tools.

The laser tool can also be used to transform CIGS into a conducting compound along a patterned line. We have shown that this process can be performed after all semiconductor layers are deposited using a technique we call laser micro-welding. By performing patterning at the end of the process flow P2 and P3 patterning could be performed simultaneously. Such solutions will further reduce manufacturing times and may offer increased control of semiconductor interfaces.

While showing promising performance on par with reference processes there are still open questions of importance for these novel techniques, particularly that of long term stability. Thin film modules are inherently sensitive to moisture and require reliable encapsulation. Before the techniques introduced here can be seen industrially they must have achieved proven stability. In this work we present a proof of existence of stable micro-welded interconnections.

Abstract

CdS is conventionally used as a buffer layer in Cu(In,Ga)Se₂, CIGS, solar cells. The aim of this thesis is to substitute CdS with cadmium-free, more transparent and environmentally benign alternative buffer layers and to analyze how the material properties of alternative layers affect the solar cell performance. The alternative buffer layers have been deposited using Atomic Layer Deposition, ALD. A theoretical explanation for the success of CdS is that its conduction band, E_c, forms a small positive offset with that of CIGS.

In one of the studies in this thesis the theory is tested experimentally by changing both the E_c position of the CIGS and of Zn(O,S) buffer layers through changing their gallium and sulfur contents respectively. Surprisingly, the top performing solar cells for all gallium contents have Zn(O,S) buffer layers with the same sulfur content and properties in spite of predicted unfavorable E_c offsets. An explanation is proposed based on observed non-homogenous composition in the buffer layer.

This thesis also shows that the solar cell performance is strongly related to the resistivity of alternative buffer layers made of (Zn,Mg)O. A tentative explanation is that a high resistivity reduces the influence of shunt paths at the buffer layer/absorber interface. For devices in operation however, it seems beneficial to induce persistent photoconductivity, by light soaking, which can reduce the effective E_c barrier at the interface and thereby improve the fill factor of the solar cells.

Zn-Sn-O is introduced as a new buffer layer in this thesis. The initial studies show that solar cells with Zn-Sn-O buffer layers have comparable performance to the CdS reference devices.

While an intrinsic ZnO layer is required for a high reproducibility and performance of solar cells with CdS buffer layers it is shown in this thesis that it can be thinned if Zn(O,S) or omitted if (Zn,Mg)O buffer layers are used instead. As a result, a top conversion efficiency of 18.1 % was achieved with an (Zn,Mg)O buffer layer, a record for a cadmium and sulfur free CIGS solar cell.

Abstract

This thesis presents various integrated antenna solutions for different types of systems and applications, e.g. wireless sensors, broadband handsets, advanced base stations, MEMS-based reconfigurable front-ends, automotive anti-collision radars, and large area electronics.

For wireless sensor applications, a T-matched dipole is proposed and integrated in an electrically small body-worn sensor node. Measurement techniques are developed to characterize the port impedance and radiation properties. Possibilities and limitations of the planar inverted cone antenna (PICA) for small handsets are studied experimentally. Printed slot-type and folded PICAs are demonstrated for UWB handheld terminals.

Both monolithic and hybrid integration are applied for electrically steerable array antennas. Compact phase shifters within a traveling wave array antenna architecture, on single layer substrate, is investigated for the first time. Radio frequency MEMS switches are utilized to improve the performance of reconfigurable antennas at higher frequencies. Using monolithic integration, a 20 GHz switched beam antenna based on MEMS switches is implemented and evaluated. Compared to similar work published previously, complete experimental results are here for the first time reported. Moreover, a hybrid approach is used for a 24 GHz switched beam traveling wave array antenna. A MEMS router is fabricated on silicon substrate for switching two array antennas on a LTCC chip.

A concept of nano-wire based substrate integrated waveguides (SIW) is proposed for millimeter-wave applications. Antenna prototypes based on this concept are successfully demonstrated for automotive radar applications.

W-band body-worn nonlinear harmonic radar reflectors are proposed as a means to improve automotive radar functionality. Passive, semi-passive and active nonlinear reflectors consisting of array antennas and nonlinear circuitry on flex foils are investigated.

A new stretchable RF electronics concept for large area electronics is demonstrated. It incorporates liquid metal into microstructured elastic channels. The prototypes exhibit high stretchability, foldability, and twistability, with maintained electrical properties.

Abstract

Biomolecular interaction analysis to determine the kinetics and affinity between interacting partners is important for the fundamental understanding of biology, as well as for the development of new pharmaceutical substances. A quartz crystal microbalance instrument suitable for kinetics and affinity analyses of interaction events was developed. The functionality of the sensor system was demonstrated by development of an assay for relative affinity determination of lectin-carbohydrate interactions.

Sensor surfaces allowing for effective immobilization of one interacting partner is a key functionality of a biosensor. Here, three different surfaces and immobilization methods were studied. First, optimized preparation conditions for sensor surfaces based on carboxyl-terminated self assembled monolayers were developed and were demonstrated to provide highly functional biosensor surfaces with low non-specific binding. Second, a method allowing for immobilization of very acidic biomolecules based on the use of an electric field was developed and evaluated. The electric field made it possible to immobilize the highly acidic C-peptide on a carboxylated surface. Third, a method for antibody immobilization on a carboxyl surface was optimized and the influence of immobilization pH on the immobilization level and antigen binding capacity was thoroughly assessed. The method showed high reproducibility for a set of antibodies and allowed for antibody immobilization also at low pH.

Three broadly different strategies to increase the sensitivity of electroacoustic sensors were explored. A QCM sensor with small resonator electrodes and reduced flow cell dimensions was demonstrated to improve the mass transport rate to the sensor surface. The use of polymers on QCM sensor surfaces to enhance the sensor response was shown to increase the response of an antibody-antigen model system more than ten-fold. Moreover, the application of high frequency thin film bulk acoustic resonators for biosensing was evaluated with respect to sensing range from the surface. The linear detection range of the thin film resonator was determined to be more than sufficient for biosensor applications involving, for instance, antibody-antigen interactions. Finally, a setup for combined frequency and resistance measurements was developed and was found to provide time resolved data suitable for kinetics determination.

Abstract

In recent years there has been a huge increase in the growth of communication systems such as mobile phones, wireless local area networks (WLAN), satellite navigation and various other forms of wireless data communication that have made analogue frequency control a key issue. The increase in frequency spectrum crowding and the increase of frequency into microwave region, along with the need for minimisation and capacity improvement, has shown the need for the development of high performance, miniature, on-chip filters operating in the low to medium GHz frequency range. This has hastened the need for alternatives to ceramic resonators due to their limits in device size and performance, which in turn, has led to development of the thin film electro-acoustics industry with surface acoustic wave (SAW) and bulk acoustic wave (BAW) filters now fabricated in their millions. Further, this new technology opens the way for integrating the traditionally incompatible integrated circuit (IC) and electro-acoustic (EA) technologies, bringing about substantial economic and performance benefits.

In this thesis the compatibility of aluminium nitride (AlN) to IC fabrication is explored as a means for furthering integration issues. Various issues have been explored where either tailoring thin film bulk acoustic resonator (FBAR) design, such as development of an improved solidly mounted resonator (SMR) technology, and use of IC technology, such as chemical mechanical polishing (CMP) or nickel silicide (NiSi), has made improvements beneficial for resonator fabrication or enabled IC integration. The former has resulted in major improvements to Quality factor, power handling and encapsulation respectively. The later has provided alternative methods to reduce electro- or acoustomigration, reduced device size, for plate waves, supplied novel low acoustic impedance material for high power applications and alternative electrodes for use in high temperature sensors.

Another method to enhance integration by using the piezoelectric material, AlN, in the IC side has also been explored. Here methods for analysing AlN film contamination and stoichiometry have been used for analysis of AlN as a high-k dielectric material. This has even brought benefits in knowledge of film composition for use as a passivation material with SiC substrates, investigated in high power high frequency applications. Lastly AlN has been used as a buried insulator material for new silicon-on-insulator substrates (SOI) for increased heat conduction. These new substrates have been analysed with further development for improved performance indicated.

Abstract

Electroacoustic technology has in many ways revolutionised the wireless telecommunication industry. The IC compatible fabrication technique of thin film electroacoustic devices has so far provided a considerable increase in device performance and reduction in size. At the moment, new areas where this technology can be of use is under investigation. In particular, thin film bulk acoustic wave resonators are promising candidates for biochemical and gravimetric sensor applications.

For bulk acoustic waves, the thesis addresses a number of aspects in the design, fabrication, characterisation, and integration of thin film electroacoustic devices. The object of the studies conducted in the thesis has been to improve on design and thereby optimise the performance of the device to fit a particular application of interest. For high frequency and high power applications, a conceptually new design of the solidly mounted resonator has been investigated. A 1 GHz plate wave resonator with a much higher Q factor than its surface acoustic counterpart have also been fabricated. A multi-chip-module 2 GHz microwave oscillator featuring a monolithically integrated solidly mounted resonator and a flip chip transistor have been fabricated and characterised with a phase noise of -125 dBc/Hz at 100 kHz. For sensor applications, the fabrication of shear mode solidly mounted resonators featuring c-axis inclined AlN films has been studied. A process for the bonding of a microfluidic system on top of the resonator has been realised. Further, the effect of conductive liquids on the resonator performance has been investigated.For surface acoustic wave devices, acoustic manipulation of particles in microfluidic channels has been studied. Two functional devices have been fabricated by bonding piezoelectric substrates to glass or fused silica superstrates. By generating an interface acoustic wave, that propagates along the bonded interface, manipulation of sub-micrometer particles was realised.

Abstract

Biosensors are today important devices within various application areas.

In this thesis a new type of label-free biosensor device is studied, which is fabricated using the same processes used for the fabrication of integrated circuits. This enables tighter integration and further sensors/biosensor miniaturization. The device is a so-called Thin Film Bulk Acoustic Resonator (FBAR). Within this thesis a low temperature reactive sputtering process for growing AlN thin films with a c-axis inclination of 20-30^o has been developed. This enables shear mode FBAR fabrication suitable for in-liquid operation, essential for biosensor applications. Shear mode FBARs were fabricated operating at frequencies above 1GHz exhibiting Q values of 100-200 in water and electromechanical coupling factors k_t² of about 1.8%. This made it possible to move the thickness excited shear mode sensing of biological layers into a new sensing regime using substantially higher operation frequencies than the conventionally used quartz crystal microbalance (QCM) operating at 5-20MHz. Measured noise levels of shear mode FBARs in contact with water showed the resolution to be in the range 0.3ng/cm² to 7.5ng/cm². This demonstrated the FBAR resolution without any averaging or additional stabilization measures already to be in the same range as the conventional QCM (5ng/cm²), suggesting that FBARs may be a competitive and low cost alternative to QCM. The linear thickness limit for sensing of biomolecular layers was concluded to be larger than the thickness of the majority of the molecular systems envisaged for FBAR biosensor applications. A temperature compensated shear mode FBAR composite structure was demonstrated with retained coupling factor and Q-value by utilizing the second mode of operation. Understanding has been gained on the sensor operation as well as on how the design parameters influence its performance. Specifically, sensitivity amplification utilizing low acoustic impedance layers in the FBAR structure has been demonstrated and explained. Further, temperature compensated Lamb mode (FPAR) devices were also studied and demonstrated with optimized electromechanical couplings.

Abstract

Thin-film solar cells based around the absorber material CuIn_1-xGa_xSe₂ (CIGS) are studied with respect to their stability characteristics, and different ways of modelling device operation are investigated. Two ways of modelling spatial inhomogeneities are detailed, one fully numerical and one hybrid model. In the numerical model, thin-film solar cells with randomized parameter variations are simulated showing how the voltage decreases with increasing material inhomogeneities.

With the hybrid model, an analytical model for the p-n junction action is used as a boundary condition to a numerical model of the steady state electrical conduction in the front contact layers. This also allows for input of inhomogeneous material parameters, but on a macroscopic scale. The simpler approach, compared to the numerical model, enables simulations of complete cells. Effects of material inhomogeneities, shunt defects and grid geometry are simulated.

The stability of CIGS solar cells with varying absorber thickness, varying buffer layer material and CIGS from two different deposition systems are subjected to damp heat treatment. During this accelerated ageing test the cells are monitored using characterization methods including J-V, QE, C-V and J(V)_T. The degradation studies show that the typical V_OC decrease experienced by CIGS cells subjected to damp heat is most likely an effect in the bulk of the absorber material.

When cells encapsulated with EVA are subjected to the same damp heat treatment, the effect on the voltage is considerably reduced. In this situation the EVA is saturated with moisture, representing a worst case scenario for a module in operation. Consequently, real-life modules will not suffer extensively from the V_OC degradation effect, common in unprotected CIGS devices.

Abstract

In mobile communication new applications like wireless internet and mobile video have increased the demand of data-rates. Therefore, new more wideband systems are being implemented. Power amplifiers in the base-stations that simultaneously handle these wideband signals for many terminals (handhelds) need to be highly linear with a considerable band-width.

In the past decade LDMOS has been the dominating technology for use in these RF-power amplifiers. In this work LDMOS transistors possible to fabricate in a normal CMOS process have been optimized and analyzed for RF-power applications. Their non-linear behavior has been explored using load-pull measurements. The mechanisms of the non-linear input capacitance have been analyzed using 2D TCAD simulations. The investigation shows that the input capacitance is a large contributor to phase distortion in the transistor.

Computational load-pull TCAD methods have been developed for analysis of RF-power devices in high-efficiency operation. Methods have been developed for class-F with harmonic loading and for bias-modulation. Load-pull measurements with drain-bias modulation in a novel measurement setup have also been conducted. The investigation shows that the combination of computational load-pull of physical transistor structures and direct measurement evaluation with modified load-pull is a viable alternative for future design of RF-power devices. Simulations and measurements on the designed LDMOS shows a 10 to 15 % increase in drain efficiency in mid-power range both in simulations and measurements. The computational load-pull method has also been used to investigate the power capability of LDMOS transistors on SOI. This study indicates that either a low-resistivity or high-resistivity substrate should be used in manufacturing of RF-power LDMOS transistors on SOI to achieve optimum efficiency. Based on a proper substrate selection these devices exhibit a 10 % higher drain-efficiency mainly due to lower dissipated power in the devices.

Abstract

The explosive development of the telecom industry and in particular wireless and mobile communications in recent years has lead to a rapid development of new component and fabrication technologies to continually satisfy the mutually exclusive requirements for better performance and miniaturization on the one hand and low cost on the other. A fundamental element in radio communications is time and frequency control, which in turn is achieved by high performance electro-acoustic components made on piezoelectric single crystalline substrates. The latter, however, reach their practical limits in terms of performance and cost as the frequency of operation reaches the microwave range. Thus, the thin film electro-acoustic technology, which uses thin piezoelectric films instead, has been recently developed to alleviate these deficiencies.

This thesis explores and addresses a number of issues related to thin film synthesis on the one hand as well as component design and fabrication on other. Specifically, the growth of highly c-axis textured AlN thin films has been studied and optimized for achieving high device performance. Perhaps, one of the biggest achievements of the work is the development of a unique process for the deposition of AlN films with a mean c-axis tilt, which is of vital importance for the fabrication of resonators operating in contact with liquids, i.e. biochemical sensors. This opens the way for the development of a whole range of sensors and bio-analytical tools. Further, high frequency Lamb wave resonators have been designed, fabricated and evaluated. Performance enhancement of FBAR devices is also addressed, e.g. spurious mode suppression, temperature compensation, etc. It has been demonstrated, that even without temperature compensation, shear mode resonators operating in a liquid still exhibit an excellent performance in terms of Q (200) and coupling (~1.8%) at 1.2 GHz, resulting in a mass resolution better than 2 ng cm^-2 in water, which excels that of today’s quartz sensors.

Abstract

The development and implementation of a metal gate technology (alloy, compound, or silicide) into metal-oxide-semiconductor field effect transistors (MOSFETs) is necessary to extend the life of planar CMOS devices and enable further downscaling. This thesis examines possible metal gate materials for improving the performance of the gate stack and discusses process integration as well as improved electrical and physical measurement methodologies, tested on capacitor structures and transistors.

By using reactive PVD and gradually increasing the N₂/Ar flow ratio, it was found that the work function (on SiO₂) of the TiN_x and ZrN_x metal systems could be modulated ~0.7 eV from low near nMOS work functions to high pMOS work functions. After high-temperature anneals corresponding to junction activation, both metals systems reached mid-gap work function values. The mechanisms behind the work function changes are explained with XPS data and discussed in terms of metal gradients and Fermi level pinning due to extrinsic interface states.

A modified scheme for improved Fowler-Nordheim tunnelling is also shown, using degenerately doped silicon substrates. In that case, the work functions of ALD/PVD TaN were accurately determined on both SiO₂ and HfO₂ and benchmarked against IPE (Internal Photoemission) results. KFM (Kelvin Force Microscopy) was also used to physically measure the work functions of PVD TiN and Mo deposited on SiO₂; the results agreed well with C-V and I-V data.

Finally, an appealing combination of novel materials is demonstrated with ALD TiN/Al₂O₃/HfAlO_x/Al₂O₃/strained-SiGe surface channel pMOS devices. The drive current and transconductance were measured to be 30% higher than the Si reference, clearly demonstrating increased mobility and the absence of polydepletion. Finally, using similarly processed transistors with Al₂O₃ dielectric instead, low-temperature water vapour annealing was shown to improve the device characteristics by reducing the negative charge within the ALD Al₂O₃.

Abstract

Thin films of metals and compounds have a very wide range of applications today. Many of the deposition methods used for the production of such films utilize plasma to support the growth the film, e.g. by the supply of energy and the enhancement of reactivity. This thesis focuses on the physical vapor deposition (PVD) of thin films by high density plasma sources based on hollow cathodes and aims to increase the understanding of the deposition process and its influence on the film properties.

Titanium nitride films reactively deposited by the low-pressure hybrid plasma (HYP LP) source exhibited excellent properties and was deposited at considerable higher rates than films deposited by conventional methods.

An original finding in this work is the influence of substrate material on the deposition process and consequently on the properties of the deposited film. In the deposition of TiN films by the HYP LP source it was found that the substrate temperature was higher for Si substrates than for steel substrates due to a more efficient absorption of microwave power in Si than in steel. Further, it was found that ferromagnetic substrates influence the film growth in magnetized plasma systems. An effect of the ferromagnetic substrates is the enhancement of ion bombardment that increases the growth temperature and affects the texture and morphology of the growing films. It was also found that a DC bias can change the TiN film properties considerably and compensate the effect of ferromagnetic substrates.

High rate depositions of chromium and chromium nitride films by the RF hollow cathode plasma jet (RHCPJ) source were studied. The performance of the reactive diffuse arc process and the CrN film properties indicates that the process can be transferred from small cylindrical cathodes to linear magnetized hollow cathodes which allow deposition on considerable larger areas and this is important for industrial applications.

Abstract

Today, micro- and nano-electronic devices are becoming more complex and advanced as the dimensions are shrinking. It is therefore a very challenging task to develop new device technologies with performance that can be predicted. This thesis focuses on advanced measurement techniques and TCAD simulations in order to characterize and understand the device physics of advanced semiconductor devices.

TCAD simulations were made on a novel MOSFET device with asymmetric source and drain structures. The results showed that there exists an optimum range of implantation doses where the device has a significantly higher figure-of-merit regarding speed and voltage capability, compared to a symmetric MOSFET. Furthermore, both 2D and 3D simulations were used to develop a resistive model of the substrate noise coupling.

Of particular interest to this thesis is the random dopant fluctuation (RDF). The result of RDF can be characterized using very advance and reliable measurement techniques. In the thesis an ultra-high precision parametric mismatch measurement system was designed and implemented. The best ever reported performance on short-term repeatability of the measurements was demonstrated. A new bipolar parametric mismatch phenomenon was also revealed using the measurement system.

A complete simulation platform, called SiSPET (Simulated Statistical Parameter Extraction Tool), was developed and integrated into the framework of a commercial TCAD environment. A special program for randomization of the doping was developed and proven to provide RDF effects in agreement measurement. The SiSPET system was used to investigate how different device models were able to take RDF effects into account. The RDF effects were translated in to parameter fluctuations using the developed extraction routines. It was shown that the basic MOSFET fluctuation model could be improved by including the field dependenent mobility. However, if a precise description of the fluctuations is required an advanced compact-model, such as MOS Model 11 should be used.

Abstract

Electro-acoustic devices such as surface acoustic wave (SAW) and bulk acoustic wave (BAW) devices have been in commercial use for over 60 years and can be found in applications ranging from specialised scientific and military equipment to consumer products, such as mobile telephones, TV and radio receivers, etc. Today by far the largest market for electro-acoustic devices is the telecommunication industry which annually consumes approximately three billion acoustic wave filters for frequency control alone.

The development of new materials and technologies for electro-acoustic devices has gained a substantial and growing interest from both academic and industrial research communities in recent years due to the enormous growth in the telecommunication industry and other forms of wireless data communication. One of the bigger issues has been to replace the single crystalline substrates with thin film piezoelectric materials deposited by reactive sputtering. This would not only reduce the manufacturing costs but will also enable high frequency of operation and a wider choice of substrate materials. However, in order to obtain the material properties required for the intended application a detailed theoretical description of the reactive sputtering process is necessary since the texture and other functional properties of the piezoelectric material are extremely sensitive to the process parameters in addition to the structure of the underlying material.

This thesis studies the reactive sputtering process and its application for the fabrication of thin film electro-acoustic devices. The aim has been to gain a further insight into the process and make use of this knowledge to improve the fabrication of electro-acoustic devices. In this work modelling of the reactive sputtering process has been improved by studying certain fundamental aspects of the process and in particular the dynamics of the processes taking place during sputtering both at the target and the substrate surfaces. Consequently, highly textured thin piezoelectric aluminium nitride films have been synthesized and thin film bulk acoustic resonators (FBAR) operating in the GHz range have been fabricated and studied.

Abstract

Thin-film solar cells based on Cu(In,Ga)Se₂ contain a thin buffer layer of CdS in their standard configuration. In order to avoid cadmium in the device for environmental reasons, Cd-free alternatives are investigated. In this thesis, ZnO-based films, containing Mg or S, grown by atomic layer deposition (ALD), are shown to be viable alternatives to CdS.

The CdS is an n-type semiconductor, which together with the n-type ZnO top-contact layers form the pn-junction with the p-type Cu(In,Ga)Se₂. From device modeling it is known that a buffer layer conduction band (CB) position of 0-0.4 eV above that of the Cu(In,Ga)Se₂ layer is consistent with high photovoltaic performance. For the Cu(In,Ga)Se₂/ZnO interface this position is measured by photoelectron spectroscopy and optical methods to –0.2 eV, resulting in increased interface recombination. By including sulfur into ZnO, a favorable CB position to Cu(In,Ga)Se₂ can be obtained for appropriate sulfur contents, and device efficiencies of up to 16.4% are demonstrated in this work. From theoretical calculations and photoelectron spectroscopy measurements, the shift in the valence and conduction bands of Zn(O,S) are shown to be non-linear with respect to the sulfur content, resulting in a large band gap bowing.

ALD is a suitable technique for buffer layer deposition since conformal coverage can be obtained even for very thin films and at low deposition temperatures. However, deposition of Zn(O,S) is shown to deviate from an ideal ALD process with much larger sulfur content in the films than expected from the precursor pulsing ratios and with a clear increase of sulfur towards the Cu(In,Ga)Se₂ layer.

For (Zn,Mg)O, single-phase ZnO-type films are obtained for Mg/(Zn+Mg) < 0.2. In this region, the band gap increases almost linearly with the Mg content resulting in an improved CB alignment at the heterojunction interface with Cu(In,Ga)Se₂ and high device efficiencies of up to 14.1%.

Abstract

Today, the silicon technology is well established for RF-applications (f~1-100 GHz), with emphasis on the lower frequencies (f < 5 GHz). The field of RF power devices is extensive concerning materials and devices. One of the important RF-devices is the silicon LDMOS transistor. A large extent of the research presented in the thesis concerns studies of this device, which have resulted in increased understanding of the device behavior and improved performance. The thesis starts with a brief survey of the RF-field, including the LDMOS transistor, followed by a description of the methods used in the investigations; simulations, modeling and measurements. Specific results presented in the appended papers are also briefly summarized.

A new concept for LDMOS transistors, which allows for both high frequency and high voltage operation, has been developed and characterized. World-record performance in terms of output power density was obtained: over 1 W/mm at 50 V and 3.2 GHz. Further understanding and improvements of the device are achieved using simulations and modeling. For determination of model parameters a new general parameter extraction technique was developed. The method has been successfully used for a large variety of high-frequency devices, and has been frequently used in the modeling work in this thesis.

Important properties of RF-power devices are the device linearity and power efficiency. Extensive studies regarding the efficiency were conducted using numerical simulations and modeling of the off-state output resistance, which is correlated to the efficiency. The results show that significant improvements can be obtained for devices on both bulk- and SOI-substrates, using thin high-resistivity substrates and very low-resistivity SOI-substrates, respectively.

Finally a new approach to drastically reduce substrate crosstalk by using very low-resistivity SOI substrate is proposed. Experimentally, a reduction of 20-40 dB was demonstrated in the GHz range compared to high-resistivity SOI substrate.

Abstract

Microengineering has developed a broad range of production techniques to reduce size, increase throughput, and reduce cost of electrical and mechanical devices. The miniaturisation has also entailed entirely new opportunities.

In this work, a piezoresistive silicon sensor measuring mechanical deformation has been designed and fabricated with the help of microengineering. Due to the large variety of used processes, this device can serve as a survey of techniques in this field. Four basic process categories are recognised: additive, subtractive, modifying, and joining methods.

The last category, joining methods, has previously been the least investigated, especially when it comes to compatibility with the other categories. The adaptability of wet chemical etching to established silicon wafer bonding technique has been investigated. Further, phenomena related to oxygen plasma pre-treatment for direct bonding has been investigated by blister bond adhesion tests, X-ray photoelectron spectroscopy, and atomic force microscopy.

Wafer bonding has been adapted to monocrystalline quartz. For wet chemical pre-treatment, characteristics specific for quartz raise obstacles. Problems with limited allowable annealing temperature, low permeability of water released in the bond at annealing, and electrostatic bonding of particles to the quartz surface, have been studied and overcome. The influence of internal bond interfaces on resonators has been investigated.

Chemical polishing of quartz by ammonium bifluoride has been experimentally investigated at high temperatures and concentrations. Chemometrical methods were used to search for optimum conditions giving the lowest surface roughness. These extreme conditions showed no extra advantages.

Adhesion quantification methods for wafer bonding have been comprehensively reviewed, and augmentations have been suggested. The improved techniques’ usefulness for three areas of use has been forecasted: general understanding, bonding scheme optimisation, and quality control. It was shown that the quality of measurements of all commonly used methods could be dramatically improved by small means.

Abstract

The solar cell technology based on Cu(In,Ga)Se₂ (CIGS) thin-films provides a promising route to cost competitive solar electricity. The standard device structure is ZnO:Al/ZnO/CdS/CIGS/Mo films on a glass substrate, where the first three layers are n-type semiconductors with wide bandgaps, forming a pn-junction with the p-type CIGS absorber layer; the Mo layer serves as a back contact. This thesis deals with analysis of the generation and recombination of electron-hole pairs throughout the device. These processes determine the maximum output power: generation limits the current; recombination limits the voltage.

The generation is calculated with an optical model based on complex refractive indices determined for the individual layers. The main features of the optical response of the solar cell can be reproduced with a specular model neglecting scattering. A model including ideally Lambertian scattering at the front and back surface of the CIGS absorber layer is introduced to investigate the possibility to maintain a high current generation with thin absorber layers. The result highlights the relatively poor optical performance of the Mo back contact. TiN and ZrN are explored as alternatives, and improved optical performance is experimentally demonstrated for both materials.

The recombination analysis emphasizes that, in general, more than one recombination path of comparable magnitude are operative in parallel. For cells with absorber bandgap increasing from 1.0 eV (CuInSe₂) to 1.7 eV (CuGaSe₂), a relative increase of interface recombination is found. When these cells are subject to accelerated ageing, degradation is smallest for intermediate bandgaps; an explanation involving different sensitivity to decreased absorber band bending and activation of grain boundaries is suggested. The optical gain with ZrN back contacts is counteracted by increased back contact recombination and contact resistance, but an intermediate layer of MoSe₂ is shown to alleviate these problems, allowing for an overall improved efficiency.

Abstract

The demands for faster, smaller, and less expensive electronic equipments are basically the driving forces for improving the speed and increasing the packing density of microelectronic components. Down-scaling of the devices is the principal method to realize these requests. For future CMOS devices, new materials are required in the transistor structure to enable further scaling and improve the transistor performance.

This thesis focuses on novel metal gate and high-κ dielectric materials for future CMOS technologies. Specifically, TiN and ZrN gate electrode materials were studied with respect to work function and thermal stability. High work function, suitable for pMOS transistors, was extracted from both C-V and I-V measurements for PVD and ALD TiN in TiN/SiO₂/Si MOS capacitor structures. ZrN_x/SiO₂/Si MOS capacitors exhibited n-type work function when the low-resistivity ZrN_x was deposited at low nitrogen gas flow. Further, variable work function by 0.6 eV was achieved by reactive sputter depositing TiN_x or ZrN_x at various nitrogen gas flow. Both metal-nitride systems demonstrate a shift in work function after RTP annealing, which is discussed in terms of Fermi level pinning due to extrinsic interface states. Still, the materials are promising in a gate last process as well as show potential as complementary gate electrodes.

The dielectric constant of as-deposited (Ta₂O₅)_1-x(TiO₂)_x thin films is around 22, whereas that of AlN is about 10. The latter is not dependent on the degree of crystallinity or on the measurement frequency up to 10 GHz. Both dielectrics exhibit characteristics appropriate for integrated capacitors. Finally, utilization of novel materials were demonstrated in strained SiGe surface-channel pMOSFETs with an ALD TiN/Al₂O₃ gate stack. The transistors were characterized with standard I-V, charge pumping, and low-frequency noise measurements. Correlation between the mobility and the oxide charge was found. Improved transistor performance was achieved by conducting low-temperature water vapor annealing, which reduced the negative charge in the Al₂O₃.

Abstract

Chemical mechanical polishing (CMP) has been used for a long time in the manufacturing of prime silicon wafers for the IC industry. Lately, other substrates, such as silicon-on-insulator has become in use which requires a greater control of the silicon CMP process. CMP is used to planarize oxide interlevel dielectric and to remove excessive tungsten after plug filling in the Al interconnection technology. In Cu interconnection technology, the plugs and wiring are filled in one step and excessive Cu is removed by CMP. In front end processing, CMP is used to realize shallow trench isolation (STI), to planarize trench capacitors in dynamic random access memories (DRAM) and in novel gate concepts.

This thesis is focused on CMP for front end processing, which is the processing on the device level and the starting material. The effects of dopants, crystal orientation and process parameters on silicon removal rate are investigated. CMP and silicon wafer bonding is investigated. Also, plasma assisted wafer bonding to form InP MOS structures is investigated.

A complexity of using STI in bipolar and BiCMOS processes is the integration of STI with deep trench isolation (DTI). A process module to realize STI/DTI, which introduces a poly CMP step to planarize the deep trench filling, is presented.

Another investigated front end application is to remove the overgrowth in selectively epitaxially grown collector for a SiGe heterojunction bipolar transistor.

CMP is also investigated for rounding, which could be beneficial for stress reduction or to create microoptical devices, using a pad softer than pads used for planarization.

An issue in CMP for planarization is glazing of the pad, which results in a decrease in removal rate. To retain a stable removal rate, the pad needs to be conditioned. This thesis introduces a geometrically defined abrasive surface for pad conditioning.

Abstract

Cu(In,Ga)Se₂ (CIGS) thin film solar cells contain a thin layer of CdS. To avoid toxic heavy-metal-containing waste in the module production the development of a cadmium-free buffer layer is desirable. This thesis considers alternative Cd-free buffer materials deposited by Atomic Layer Deposition (ALD). Conditions of the CIGS surface necessary for ALD growth are investigated and the heterojunction interface is characterized by band alignment studies of ZnO/CIGS and In₂S₃/CIGS interfaces. The thesis also includes investigations on the surface modification of the CIGS absorber by sulfurization.

According to ALD theory the growth process is limited by surface saturated reactions. The ALD growth on CIGS substrates shows nucleation failure and generally suffers from surface contaminations of the CIGS layer. The grade of growth disturbance varies for different ALD precursors. The presence of surface contaminants is related to the substrate age and sodium content. Improved growth behavior is demonstrated by different pretreatment procedures.

The alignment of the energy bands in the buffer/absorber interface is an important parameter for minimization of the losses in a solar cell. The valence band and conduction band offsets was determined by in situ X-ray and UV photoelectron spectroscopy during layer by layer formation of buffer material. The conduction band offset (ΔE_c) should be small but positive for optimal solar cell electrical performance according to theory. The conduction band offset was determined for the ALD ZnO/CIGS interface (ΔE_c = -0.2 eV) and the ALD In₂S₃/CIGS interface (ΔE_c = -0.25 eV).

A high temperature process for bandgap grading and a low temperature process for surface passivation by post deposition sulfurization in H₂S were investigated. It is concluded that the high temperature sulfurization of CuIn_(1-x)Ga_xSe₂ leads to phase separation when x>0. The low temperature process did not result in enhanced device performance.

Abstract

The Cu(In,Ga)Se₂-based thin film solar cell is a promising candidate for becoming one of the more important solar cell technologies in the near future. In order to realize such a development a significant reduced production cost of the Cu(In,Ga)Se₂ (CIGS) layer is needed. This work shows a possible way towards such a reduction by increasing the deposition rate and decreasing the CIGS thickness with almost maintained device efficiency.

Obtaining an improved device performance in CIGS-based solar cells by using an in-depth variation of the band gap has earlier been investigated without any clear conclusions. In this work an extensive experimental study of the beneficial effect of band gap profiling has been performed and firmly based conclusions have been made. For standard CIGS devices the band gap profiling can result in an improved efficiency of around 0.4 % units. This gain is related to improved field-assisted carrier collection. For reduced CIGS thicknesses the importance of a band gap profiling is enhanced, and at a CIGS thickness of 0.5 μm an efficiency gain of 2.5 % units is obtained, resulting in a 13.4 % efficient device. The main reason for the gain is passivation of the back contact, which becomes increasingly detrimental for the device performance as the CIGS thickness is reduced. With an optimized band gap profile the CIGS thickness can be reduced 3-4 times, with almost solely absorption related losses.

The potential for increasing the deposition rate of co-evaporated CIGS layers is shown to be large. An increase of up to 10 times compared to commonly used deposition rates is possible with only minor losses in efficiency. By using band gap profiled thin CIGS layers deposited at high rates, the production from a single evaporation system can be increased up 30 times. Such an increase will lead to the needed reduction of the production cost of the complete solar cell module.

Abstract

Recently, the enormous growth in personal communications systems (PCS), satellite communication and various other forms of wireless data communication has made analogue frequency control a key issue as the operation frequency increases to the low/medium GHz range. Surface acoustic wave (SAW) and bulk acoustic wave (BAW) electroacoustic devices are widely used today in a variety of applications both in consumer electronics as well as in specialized scientific and military equipment where frequency control is required. Conventional piezoelectric materials such as quartz, LiNbO3 and LiTaO3 suffer from a variety of limitations and in particular medium to low SAW/BAW velocity as well as being incompatible with the IC technology. Thin piezoelectric films offer the great flexibility of choosing at will the substrate/film combination, thus making use of the electroacoustic properties of non-piezoelectric substrates, which widens greatly the choice of fabrication materials and opens the way for integration of the traditionally incompatible electroacoustic and IC technologies.

This thesis focuses on the synthesis and characterization of novel thin film materials for electroacoustic applications. A prime choice of material is thin piezoelectric AlN films which have been grown using both RF and pulsed-DC reactive sputter deposition on a variety of substrate materials. A unique synthesis process has been developed allowing the deposition of high quality AlN films at room temperature, which increases greatly the process versatility. The films are fully c-axis oriented with a 1.6° FWHM value of the rocking curve of the AlN-(002) peak. Complete process flows for the fabrication of both SAW and BAW devices have been developed. Electroacoustic characterization of 2 GHz BAW resonators yielded an electromechanical coupling coefficient (kt²) of 6.5%, Q-value of 600 and a longitudinal velocity of 11350 m/s. AlN thin films based SAW resonators on SiO2/Si yielded a SAW velocity of around 5000 m/s and a piezoelectric coupling coefficient (K²) of around 0.3%. Finally, AlN on polycrystalline diamond 1 GHz SAW resonators exhibited an extremely high SAW velocity of 11800 m/s, a piezoelectric coupling coefficient (K²) of 1% and a Q-value of 500.

Abstract

More and more electronic products, like battery chargers and power supplies, as well as applications in telecommunications and automotive electronics are based on System-on-Chip solutions, where signal processing and power devices are integrated on the same chip. The integration of different functional units offers many advantages in terms of reliability, reduced power consumption, weight and space reduction, leading to products with better performance at a hopefully lower price.

This thesis focuses on the integration of vertical high-voltage double-diffused MOS transistors (DMOSFETs) on Silicon-on-Insulator (SOI) substrates. MOSFETs possess a number of features which makes them indispensable for Power Integrated Circuits (PICs): high switching speed, high efficiency, and simple drive circuits. SOI substrates combined with trench technology is superior to traditional Junction Isolation (JI) techniques in terms of cross-talk and leakage currents.

Vertical DMOS transistors on SOI have been manufactured and characterized, and an analytical model for their on-resistance is presented. A description of self-heating and operation at elevated temperatures is included. Furthermore, the switching dynamics of these components is investigated by means of device simulations with the result that the dissipated power during unclamped inductive switching tests is reduced substantially compared to bulk vertical DMOSFETs.

A large number of defects is created in the device layer if the trenches are exposed to high temperatures during processing. A new fabrication process with back-end trench formation is introduced in order to minimize defect generation. In addition, a model for the capacitive coupling between trench-isolated structures is developed.

Abstract

Further, a highly selective dry etch process for etching Al on AlN has been developed for the fabrication of MIM, MIS, SAW and BAW test structures for electrical and electroacoustic characterization of the films. A dielectric constant of 10 for AlN and 25 for Ti doped Ta₂O₅ have been measured. With respect to electroacoustic characterization, BAW measurements gave a longitudinal velocity of 11350 m/s and a TCD of -25ppm/K. AlN thin film test structures on SiO₂/Si yielded a SAW velocity of around 5000 m/s, while those on polycrystalline diamond exhibited a SAW velocity of 11800 m/s. The latter results illustrate one of the biggest advantages of thin film SAW technology, namely one can exploit both the piezoelectric properties of the film and the acoustic properties of the substrate and hence devise components with superior performance.

The explosive development of personal communications systems, navigation, satellite communications as well as personal computer and data processing systems together with the constant demand for higher speeds and larger bandwidths has driven the fabrication technology to its limits. This in turn necessitates the development of novel functional materials for the fabrication of devices with superior performance and higher capacity at reduced manufacturing costs. This thesis focuses on the synthesis and characterization of such materials for IC and electroacoustic applications. Specifically, AlN thin films as well as Ti doped Ta₂O₅ thin films have been grown using both RF and pulsed-DC reactive sputter deposition on a variety of substrate materials. AlN is a piezoelectric material and hence its crystallographic structure and film texture are of prime interest, while Ta₂O₅ is a material with a relatively high dielectric constant. A significant part of the work deals with the optimization of the deposition processes. The latter have been optimized both empirically and theoretically by modeling the reactive sputter process. Subsequently, highly textured AlN thin films have been synthesized and characterized. The films were fully c-axis oriented with a typical value for the FWHM of the (002) rocking curve of 1.6°. In addition, epitaxial AlN films have been grown on sapphire at 500^oC with relatively low defect density.

Abstract

Cu(In,Ga)Se₂ (CIGS) is one of the most promising semiconductor compounds for large-scale production of efficient, low-cost thin film solar cells, and several research institutes have announced their plans for CIGS production lines. But for the CIGS technology to become a commercial success, a number of issues concerning manufacturability, product definition, and long-term stability require further attention.

Several studies indicate that CIGS-based modules are stable over many years in field operation. At the same time, it is shown in the present work that they may have difficulties in passing standard accelerated lifetime test procedures like the IEC 1646 damp heat test. In particular, CIGS modules are sensitive to humidity penetrating through the module encapsulation, which will increase the resistive losses in the front contact and cause severe corrosion of the back contact. It is also shown that cells experience degradation in both voltage and fill factor, and the causes of these effects are addressed.

By concentrating the light falling onto a solar cell, the device will deliver a higher power output per illuminated absorber area, which can lower the electricity production costs. For CIGS-based solar cells, low-concentrated illumination could be an economically viable approach. In this work it is shown that the yearly performance of a photovoltaic system with CIGS modules can be significantly improved at a moderate cost by using parabolic aluminum mirrors as concentrating elements. However, in order to avoid detrimental power losses due to high temperatures and current densities, the modules need to be designed for the higher light intensity and to be sufficiently cooled during operation. A design where the front contact of the module is assisted by a metal grid has shown promising results, not only for concentrated illumination but also for normal operation. The benefits are enhanced window processing tolerance and throughput, as well as improved degrees of freedom of the module geometry.

Abstract

The lateral double-diffused MOS (LDMOS) transistor has traditionally been a high-voltage device used in switching applications. The use as a high-frequency device has become more important lately since the LDMOS offers an low cost solution for telecommunication applications. An important property of the LDMOS concept is that it can be manufactured in virtually the same process used in standard CMOS production. It only requires one extra process step, which is easily implemented. The other important aspect that gives the LDMOS the good high-frequency performance is that the channel length is a process parameter and not a lithography parameter.

This thesis investigates the LDMOS transistor primarily from two aspects. The first is the high-voltage performance. For a high-voltage device the most important parameter is the breakdown voltage. The second most important parameter is the on-resistance that has the property of being in contradiction of the breakdown voltage and usually trade-offs are made to achieve acceptable performance. In the thesis several methods to improve the breakdown voltage/on-resistance relation are presented.

The other part covers the high-frequency behavior of the LDMOS transistor. High-frequency characterization has been made to gain valuable information for the fundamental understanding of the physical mechanisms inside the transistor. A large part of the thesis covers modeling and parameter extraction of the devices. A new general method for parameter extraction of small-signal equivalent circuit models is presented, which has the appealing properties of not needing any approximation during the extraction which is common with other techniques.

Abstract

Polysilicon is used, among other materials, in today's integrated circuits. In this thesis, the structural and electrical properties' of polysilicon thin film resistors have been studied. The aim has been to improve the long-term stability and increase the knowledge concerning grain-boundary related phenomena. Several methods, which improve the long-term stability at least by a factor of two, have been proposed and investigated.

A new model for the time dependence of the resistivity in the presence of electrical and thermal stress has been developed. From this, valuable information on the trap density and f-factor have been obtained. It was found that a small group of hydrogen atoms at the grain-boundary, with a concentration of about 10¹⁰ cm^-2 and activation energies not exceeding 0.3 - 0.7 eV, was responsible for the resistivity drift. Heat treatments in the interval 150 °C to 350 °C also established the presence of weakly bound hydrogen.

Blocking some of the dangling bonds with atoms having higher bond strength to silicon than hydrogen, such as phosphorus and fluorine, reduces the amount of weakly bound hydrogen and thereby improves the long-term stability. When using fluorine, the annealing temperature must not exceed 750 °C. A formation of BF complexes are shown to decrease the active carrier concentration. In the case of phosphorus doping, boron has to be added to obtain the desired resistivity. In such compensation doped polysilicon, the presence of BP complexes leads to an increase in the number of hole traps.

Trace amounts of Ti and W have been found to improve the long-term stability. This was due to a retardation of the diffusion of hydrogen through the oxide, caused by an ability of Ti and W to change the relative amounts of atomic and molecular hydrogen.

Semi-insulating polysilicon (SIPOS) has a lower temperature dependence (TCR) than polysilicon. A sub-oxide at the grain boundaries was shown to explain the experimental data. It not only influences the grain size, it also causes a shift from thermionic emission towards tunneling, with a reduced TCR as a result.

Abstract

This thesis consider aspects concerning the transport, distribution and incorporation of Na in Cu(In,Ga)Se₂ absorber layers for thin film solar cells. The influence of Na on the growth of Cu(In,Ga)Se₂ layers has been studied, as well as the influence on the device performance. In particular, the relationship between the oxygen content in the Mo back contact and the Na transport properties has been investigated.

Na is supplied by out diffusion from the soda lime glass substrate through the Mo back contact, or by deposition of an Na precursor onto the Mo film prior to the Cu(In,Ga)Se₂ growth. All investigated Na precursor compounds, Na₂Se, Na₂S and NaF, have resulted in high quality devices. The effect of Na on the microstructure and crystal quality of Cu(In,Ga)Se₂ layers has been studied by Scanning Electron Microscopy and X-Ray Diffraction. The Na transport properties have been investigated by Secondary Ion Mass Spectrometry.

It has been found that the presence of MO oxide phases in the MO back contact is necessary for the Na transport through the Mo film if Na is supplied from the soda lime glass. Na has beneficial effects on both the structural properties of the Cu(In,Ga)Se₂ layer, yielding large grains and a flat surface, and on the electrical properties. The main influence of Na is a substantial increase in device efficiency compared to devices fabricated without Na. In order to obtain these beneficial effects a minimal quantity of Na is necessary, but the system is tolerant to excess Na. Both out-diffusion from the soda lime glass as well as Na precursors can supply the required Na concentrations.

Abstract

In this thesis, aspects of the Silicon Germanium (SiGe) Heterojunction Bipolar Transistor (HBT) are addressed. A physics-based electrical large-signal model including thermal dependence has been developed and is implemented using a commercially available simulator package. Good agreement-is found between calculated data using the model and measured data. Equations for the electrical parameters based on physical data and a fitting procedure for finding parameter values concerning parasitic effects are presented. In addition, a technique for extracting very short thermal time constants using small signal measurements is presented.

Using the large-signal model, a frequency multiplier employing a single SiGe HBT as the non-linear device has been designed and fabricated. The doubler operates with an output frequency of 55 GHz and performance can be well explained using the model.

Low-frequency noise in the SiGe HBT has been studied, primarily using transimpedance amplifiers. Problems related to the measurement of low-frequency noise are discussed. The dominant noise source in a SiGe HBT is discriminated using direct two-channel noise measurements for a sweep of base resistance terminations of the device. By employing a device temperature variation the temperature dependence of the dominant source is further studied. A method for improved coherence measurements during a sweep of base resistance terminations is presented. A method for modeling low-frequency noise in a SPICE based simulator and aspects of the noisecorner frequency are discussed.

Abstract

Microstructure technology can be defined as a set of fabrication processes enabling three-dimensional structures featuring dimensions in the micrometer to millimetre range, with micrometer accuracy. Microstructures can be assembled or manufactured integrated with other structures and/or microelectronics forming complex microsystems.

The thesis includes articles presenting different fabrication processes and microstructures in three completely different materials: silicon, quartz, and diamond. Only single crystalline silicon and quartz, and polycrystalline chemical vapour deposited diamond are considered.

An introduction is given to the field of microstructure technology in relation to crystal structure and wafer fabrication. Material properties are listed and commented, followed by an overview of relevant fabrication processes.

A major part of the thesis is related to silicon. A system based on a capacitive pressure sensor, followed by an electrostatic actuator, both made in single crystalline silicon, are presented. The use of electrodeposited photoresist as a non-planar lithography technique is also demonstrated on silicon aiming at an optoelectronicmodule.

Two papers treat fundamental fabrication processes on single crystalline quartz. Experiments and simulation of wet etching in quartz as well as a novel quartz-to-quartz direct wafer bonding technique are presented. The process compatibility of wet etching and quartz direct bonding is demonstrated by a hermetically sealed cavity.

A microstructure technology for thick film diamond replicas, using hot filament chemical vapour deposition on microstructured silicon has been developed. Demonstrator structures for different application areas, especially microfluidics, are shown.

Finally, the concept of a miniature X-ray source based on field emission is presented. A tentative medical application is discussed and candidate field emission structures in metal, silicon, and diamond are investigated. X-ray spectra from different anode materials have been collected using diamond cold cathodes, thereby providingexperimental verification of the device concept.

Silicon is without competition the most mature material for microstructures. The properties of quartz are well known but this material is lagging far behind in the number of fabrication techniques. Diamond is the most recent material to be considered for microstructure technology. Although the possibilities to sculpture diamond are more limited, the extreme material properties are very attractive in many applications.

Abstract

This thesis brings together multiply faceted work within the field of microstructure technology. The different facets contribute to the field in various ways; thereby making an example of the structure of the technology brought forward in the summary. First, the field of microstructure technology is contextually identifled: "By 'microstructure technology' I mean the processing and application of structures that look man-made and that have a shape that ought to be imaged using a scanning electron microscope." The actions taken to perform microstructure technology is then discussed as a structure considered in five parts: the learning about the material, the working of the material, the giving of a shape, the relating of cause and action, and finally the attaching of names to the things.

Among the papers included in this thesis are studies on thedevelopment of a process to establish the precise crystallographicdirections of silicon and indium phosphide which is of importance for high precision micromachining. The nature of the phenomenological etching behaviour of near (111) faces of silicon in KOH solutions is investigated in detail in order to help even higher precision of wet chemical etching. The shortcomings of existing simulators to model the etch behaviour is exemplified by the demonstration of a new class of vertical corner compensations. Work has also been directed towards microsystems. A micromechanical snap-in structure is both analyticallymodelled in detail and also demonstrated as a fabricated device. Finally, another light steering device, a scanning mirror, is demonstrated for the use in head mounted display systems.

Abstract

The microsystems technology (MST) is strongly evolving, constantly finding new application areas. Here, the area of microoptics is presented with examples of different solutions that MST has made possible within optical sensors, actuators, and optoelectronic systems. The most commonly used processes for microfabrication are presented together with a description of single crystalline silicon as a material.

In this thesis, bulk silicon micromachining techniques have been used to make optical mirrors and assembly systems for low cost assembly of optical devices. The optical mirrors were formed by revealing crystallographic planes by wet anisotropic etching of silicon. The assembly systems consist of thin flexible holding structures which press optical fibres and quadratic chips into well-defined grooves and pits in a silicon carrier. These systems were produced in one etch step by using a photovoltaic etch stop technique.

Further, a fibre-optic pressure sensor for measurements in the coronary arteries of the heart is presented. It is a light intensity modulating pressure sensor which is integrated with the guide wire used during balloon dilatation of constricted coronary arteries. The sensor element was made by wet anisotropic etching of silicon. The improved fabrication process is described in detail.

The last part of the thesis treats fundamental etch studies on single crystalline silicon. First, a method for investigation of the angular dependence during dry etching (RIE) is presented. Differently oriented surfaces, formed in silicon by wet anisotropic etching, were covered by a material for subsequent dry etching and examination.

Finally, an investigation on the influence from bond interfaces on subsequent wet anisotropic etching of silicon is presented. This investigation was made on hydrophobically and hydrophilically bonded silicon wafers etched in KOH and tetramethyl ammonium hydroxide (TMAH).