<p>Vertically aligned nanocomposite (VAN) configuration has been recognized as the state-of-the-art architecture in the complex oxide epitaxial thin films, which are constructed by two immiscible phases simultaneously and vertically growing on a given substrate and forming various columnar microstructures, such as nanopillars embedded in matrix, nanomaze, and nanocheckboard. Due to its architectural features, VAN structure enables a powerful control on the multifunctionalities via vertical strain engineering, microstructural variations, and interfacial coupling. It provides flexibility in complex oxide designs with various functionalities (e.g., electrical, magnetic, optical, etc.), as well as a platform to explore the correlations between strain, microstructure, and multifunctionalities of the nanocomposite thin films.</p> <p>In this dissertation, integrated VAN systems with multilayer configuration have been constructed as a new three-dimensional (3D) framework, e.g., inserting 1-3 layers of CeO₂ (or LSMO) interlayers into the La_0.7Sr_0.3MnO₃ (LSMO)-CeO₂ VAN system and forming 3D interconnected CeO₂ (or LSMO) skeleton embedded in LSMO matrix. This new VAN 3D framework enables both lateral and vertical strain engineering simultaneously within the films and obtains highly enhanced magnetotransport properties, such as the record high magnetoresistance (MR) value of ~51-66%, compared with its VAN single layer counterpart. In order to demonstrate the flexibility of this design, other systems such as 3D ZnO framework embedded in LSMO matrix have been constructed to explore the thickness effects of the ZnO interlayers on the magnetotransport properties of the LSMO-ZnO system. The maximum MR value is obtained at the ZnO interlayer thickness of ~2 nm, which enables the optimal magnetoresistance tunneling effect. Meanwhile, the significance of the interlayer selection in the microstructure and magnetoresistance properties of the LSMO-ZnO system has been investigated by varying the interlayer materials yttria-stabilized zirconia (YSZ), CeO₂, SrTiO₃, BaTiO₃, and MgO. The formed 3D heterogeneous framework provides a new dimension to tailor the microstructure, strain and functionalities within the films.</p> <p>Moreover, a new strain engineering approach with engineered tilted interfaces has been demonstrated by multilayering different VAN layers with various two phase ratio and creating a hybrid nanodumbbell structure within the LSMO-CeO₂ VAN thin films. The nanodumbbell structure accomplishes a more efficient strain engineering and exhibits highly enhanced magnetic and magnetoresistance properties, compared with its VAN single layer and interlayer counterparts. </p> <p>These examples presented in the thesis demonstrate the flexibility and potential of 3D strain engineering in complex VAN systems and a higher level of property control, coupled with unique microstructures and interfaces. Beyond perovskites, these 3D designs can be extended to other material systems for a broader range of applications, such as energy conversion and storage related applications.</p>

<p>Electromigration in thin films is a well known failure mode for scaled microelectronics. While our understanding of electromigration physics has improved immensely in the last few decades, there are still some gaps in literature. In particular, the influence of interfaces on the mass transport rate is not well understood. Through reliability studies conducted on passivated metals films, marked improvement in electromigration lifetimes was observed. Specifically, some choices of materials for passivation appear to perform better than others. Qualitatively this improvement in electromigration performance is attributed to surface adhesion. However, a theoretical connection is largely missing in the literature. Lane et al. through in-situ electromigration experiments and separate interfacial debond experiments on sandwich specimens showed that a correlation exists between the void growth rate and the debond energy. However, a fundamental understanding of the relation between the two is missing. In this study we explore the connection between interfacial adhesion and void growth in a current driven system. Several experiments with varying test conditions are carried out on Blech-like test structures with different capping layers. The influence of these capping layers is captured through direct void growth measurements. Comparison of activation energy associated with electromigration was made against existing literature. It was found to be consistent with values reported for surface/interface dominated diffusion mechanisms. Further, an extension is proposed to the phase growth relations derived in existing literature to include the effect of surface adhesion. Interfacial adhesion energy ratios are extracted from the electromigration experiments for two of the test structures (Cu-Ta and Cu-SiNx) tested in this study. This ratio is compared to values reported in literature for the two interfaces and they show good agreement with experimental data.<br></p>

<div>Because of the limited reserve of fossil fuels and issues brought up by their combustion, the demand on renewable energy is considerably increasing. Solar energy is one of the most promising renewable energy sources considering the large amount of solar irradiation received by Earth and solar cell is such a device that allows us to directly convert sunlight directly into electricity. In this thesis, kesterite (I2-II-IV-IV4) system is the main focus as the light absorber material in thin film solar cells.</div><div><br></div><div>Cu2ZnSn(S,Se)4 (CZTSSe) has been first studied intensively. However, due to the band tailing resulting from Cu-Zn anti-site defects, further improvement on power conversion efficiency of this material has been hindered. Substituting Cu with Ag is expected to solve this problem by decreasing this defect density as a result of the high formation energy of Ag-Zn antisite defects. Herein, different concentrations of Ag are used to substitute Cu in the kesterite system through a nanoparticle-ink route for the fabrication of light absorber thin films. For Ag-alloying concentration less than 50%, it suggests that the Ag can induce inhomogeneity as well as secondary phase formation during nanoparticle formation. Moreover, Ag alloying is shown to enlarge the grain size and reduce film roughness after selenization, which are beneficial for the optoelectronic properties and device performance.</div><div><br></div><div>Additionally, the synthesis process for kesterite Ag2ZnSnS4 nanoparticles is explored. AZTS nanoparticles are achieved by solvent-thermal reaction. The reaction pathway during reaction is investigated by different material characterization methods to shed light on the Ag-based nanoparticle synthesis. The final nanoparticles obtained have high crystallinity and homogeneous composition, demonstrating great potential as light absorber materials. Also, the sulfide nanoparticles are converted into selenide thin films in Se vapor at elevated temperature (selenization). The selenization conditions, including temperature, heating ramp and selenization time, are optimized for the pure phase kesterite AZTSe thin films with large and dense grains. The optoelectronic properties are explored on these films and an initial research already demonstrates a 0.35% power conversion efficiency as the first solution processed AZTSe device.</div><div><br></div><div>In summary, multiple material characterization techniques are utilized to understand the microstructure evolution, phase transformation, and composition change for solution-processed nanoparticles and their resulting thin films. The material characteristics, process methods and film optoelectronic properties are associated for the future analysis and development of kesterite thin films for photovoltaic applications.</div><div><br></div>

Two-phase heat transfer techniques such a boiling make use of the high latent heat of fluids to enable dissipation of higher heat fluxes from surfaces compared to conventional single-phase cooling methods. To meet the increasing heat flux dissipation requirements of high-power electronic devices, modifications to the surface properties and roughness are often considered as a means to enhance two-phase heat transfer processes. Although surface roughness of varying length scales has been observed experimentally to enhance boiling heat transfer performance, the physical mechanisms that govern this improvement are not widely accepted. Correlations can be developed to map the behavior of specific surface structure geometries, but a broader investigation of the fundamental forces affecting evaporation at the three-phase contact line, which is critically important to the two-phase heat transfer process, may provide more widely applicable insights. In this thesis, an experimental setup was developed to investigate the effect of small scale surface roughness, with feature sizes below 1 micron, on the liquid film profile of a meniscus formed on a surface. This physical film profile can provide insight into how the surface roughness affects disjoining pressure, an important force that affects the phase change heat transfer process at the contact line. Using an interferometry technique to measure the liquid film profile for a model system of octane on silicon substrate with varying roughness, the change in disjoining pressure in the liquid film was observed. We found that the strength of disjoining pressure in the liquid film increases with increasing surface roughness feature depth.<br>

The primary aim of the research attempted to investigate whether ethnic differences in self-image existed amongst White and African females. It further investigated whether specific factors such as family, peer relations and psychosocial factors (specifically perceptions regarding the media) have had a positive or negative influence on weight. Additionally it explored whether black females may still perceive a fuller figure as being more acceptable or whether changes have occurred within past and/or present ideologies. The study used both qualitative and quantitative analysis. In order to establish whether ethnicity was significantly different from BMI, BSQ, SATAQ-3, t-tests (non-parametric one-way ANOVA comparisons) were performed. Ethnicity was not significantly different in relation to the dependent variables BSQ; overall SATAQ-3 scores and its dimensions; and affirmation, belonging and commitment of the MEIM. However significant differences were found between BMI, the overall MEIM and its subscale ethnic identity. Low and high scores of the MEIM revealed no differences across all measurement instruments used within the study, demonstrating no effect upon BSQ or the SATAQ-3 and its subscales. However there was a significant difference found between ethnic groups with regard to BMI. Fisher’s –z was used to conduct comparisons between the correlations established using Pearson’s Rank Coefficient Correlations. The results indicated that significant relationships did exist between the BSQ, specific subscales of the SATAQ-3, and BMI. Results obtained from Fisher’s-z revealed significant differences on the BSQ and SATAQ-3 (including its subscales) correlations. Multiple Regression was conducted to establish whether BMI, MEIM and SATAQ-3 have had an impact upon the outcome of BSQ. The comparisons revealed that BMI and sociocultural factors may have an impact upon the perceptions of body shape and size. Major themes identified within the qualitative analysis were culture, inherited concepts, media, family, friends and parental (mother) influences amongst others. The results revealed that White females may experience greater body dissatisfaction. African females had stated that their body concept is influenced by ethnicity, while White females communicated media and stereotyped images of white females as being the source of pressure to loose weight.

Master of Science - Science / The theoretical investigation of the Optical Kerr Effect (OKE) and Third- Harmonic Generation (THG) of Au-VO2 nanocomposite thin-films undertaken in this project is motivated by the potential use of the Au-VO2 nanocomposite in nonlinear optical (NLO) devices. NLO devices are a driving force in today fs high technology industry. In this thesis the OKE is estimated and computed and will be compared later to the measured OKE using the Z-scan interferometric laser technique [47]. This is a simple yet highly sensitive single-beam experimental technique used to determine the real part of the dielectric susceptibility. The study of the OKE in these composites has shown that it has a high value, comparable to that found in Au-SiO2, Au-TiO2 and Au-Al2O3 thin-films [29, 33]. This is despite the fact that the Au volumetric concentration in the Au-VO2 composites considered here is 10 percent at most, whereas in the other above mentioned composite thin-films the volumetric concentration range was between 5 − 60 percent. Moreover, it has been demonstrated that the OKE is thermally tunable in the Au-VO2 thin-films, owing to the thermally tunable optical properties of the VO2 [8, 38]. It is found that the magnitude of the OKE is of the order of 10−6 esu when the composite is below 68 0C and it is of the order of 10−7 esu when the nanocomposite thin-film is above 68 0C. The large enhancement of the OKE is due to the surface plasmon resonance (SPR) of the nanogold particles. Its fast response, which is of the order of few picoseconds [4, 6, 7], makes the Au-VO2 nanocomposite a good candidate as a fast thermally tunable optical switch or modulator. The modelling here of high-order harmonic generation in strongly absorbing media, as regards the amplitude of the primary beam, takes into account pump attenuation only, due to the absorption of light by the media. It is not concerned with pump depletion which is a consequence of the transfer of energy to the harmonics and which is small by comparison with attenuation in absorbing media. The modelling is applied to the so-called transmission and reflection configurations. The former refers to the case in which the high-order harmonic wave is monitored in the same direction as the input fundamental wave, whereas the latter describes the situation in which the fundamental wave is in the opposite direction to the high-order harmonic one. To analyse high-order harmonic generation one has to relate the high-order harmonic intensity to the fundamental intensity [1, 9, 10, 34-36]. In so doing, a general formula for analysing high-order harmonic generation is obtained and reduced to the particular case pertaining to THG in strongly absorbing media. The ratio of the third-harmonic intensity to the fundamental intensity is termed the conversion coefficient or the conversion efficiency, and it is denoted by ƒÅ [23, 34]. It is useful in the sense that it expresses quantitatively the amount of input light of frequency ƒÖ converted into light of frequency nƒÖ, where n is the order of nonlinear polarisation [1]. It is found that the THG conversion coefficient is higher the higher the laser intensity. It is thus advantageous to use a pulsed laser, which achieved very high intensities for short periods, separated by long off periods. The net harmonic output in this case is much higher than one would obtain with a continous laser of the same average output. It is found here that ƒÅ is greater in the reflection configuration compared to the transmission one above and below Tt in the photon energy range 1.0 − 3.0 eV, see Fig. 5.1 to 5.8. However, the conversion efficiency for THG in the Au-VO2 nanocomposites for the picoseconds laser illumination we have considered, is still extremely low, and it is difficult to see a potential use for this system as a tunable frequency converter. The situation would become more favorable with the use of femtosecond laser pulses, where for the same pulse energy the intensity is much greater. The laser pulse energy must be limited to avoid excessive heating of the thin-film. The heat generated as a consequence of the illumination of the thin-film by the laser [23] may be controlled by using a simple cooling device which consists of a substrate on which the thin-film is deposited. The choice of such a substrate depends on whether THG is monitored in the transmission or reflection configuration. In the former a transparent substrate must be used (for example diamond) whereas in the latter an opaque substrate may be used (for example Ag). Calculations pertaining to the removal of heat from the illuminated film are reported, and show that thermal control is manageable but only within limits. To avoid a temperature rise of more than 5 0K the peak laser intensity we found must not exceed 7.4 ~105Wcm−2 with a pulse duration of 5 ps.

A dissertation submitted to the Faculty of Engineering and the Built Environment, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Master of Science in Engineering, 2017 / This research has considered an idealistic non-interleaved planar transformer wherein only the electromagnetic parasitic capacitive and inductive elements arising out of the transformer geometry are taken into account, without considering material limitations. A suitable model for the planar transformer is used to analyse its frequency and power transfer characteristics; this model was validated by three dimensional electromagnetic simulations of various planar transformer structures in FEKO simulation software. The capacitive and inductive parasitics in this model have been found to be functions of the dielectric thickness. The theoretical bandwidth for the planar transformer is defined in this research as a function of dielectric thickness. The effect of dielectric thickness of the transformer windings on the bandwidth of the transformer is analysed, based on the premise that the inherent parasitic capacitive and inductive elements would affect the transfer characteristics of the transformer. Upon conclusion of this analysis, it is found that the dielectric thickness of a planar transformer can be optimised such as to present an optimised bandwidth. A closed form analytic expression for the optimum dielectric thickness value is derived and presented in this research. In a design example of a 4:1 50W transformer presented in this research, it has been shown that the bandwidth can be improved by 384%, along with a power density improvement of 45%, upon choosing of an optimum dielectric thickness of 0.156mm to replace a standard 0.4mm thick dielectric. It should be noted that the results derived in this research are purely theoretical, justified by many idealisations and assumptions that are argued throughout the research. It is thus expected that practical results should at best approach the theoretical results, due to the known non-ideal nature of reality. / CK2018

Orientador: Marystela Ferreira / Banca: Thiago Regis Longo Cesar da Paixão / Banca: Laura Oliveira Peres Philadelphi / Banca: Nilson Cristino da Cruz / Banca: Margarida Juri Saeki / Resumo: Neste trabalho, foram desenvolvidos biossensores eletroquímicos enzimáticos fabricados com óxido de grafeno reduzido (rGO) e funcionalizado depositados pelas técnicas LbL (Layer-by-Layer) e Langmuir-Blodgett (LB) para a detecção de glicose. Para isso, primeiramente foi sintetizado quimicamente o óxido de grafeno (GO) pela oxidação do grafite, em seguida, o GO foi reduzido para aumentar a sua condutividade elétrica e foi funcionalizado em meio estabilizante contendo cloridrato de poli(dialildimetilamônio), formando o GPDDA, ou poli(4-estireno sulfônico) PSS, formando GPSS. Com isso, foi possível formar dispersões aquosas estáveis das nanofolhas de grafeno, essencial para a fabricação de filmes LbL. Em um primeiro momento, foram fabricados biossensores com filmes LbL contendo GPDDA, GPSS e a enzima glucose oxidase (Gox). O desempenho desses biossensores na detecção de glicose foi avaliado em função do número de bicamadas contendo Gox. O melhor desempenho na detecção de glicose foi apresentado pelo biossensor com o filme LbL com a arquitetura (GPDDA/GPSS)/(GPDDA/GPSS)2. Este biossensor apresentou limite de detecção de 13,4 µmol.L-1, sensibilidade 2,47 μA.cm-2.mmol-1.L e faixa analítica entre 0,04 e 0,95 mmol.L-1. Este biossensor foi eficiente na detecção de glicose na presença de interferentes comumente encontrados em fluidos corporais, alimentos e fármacos. Quando avaliado na detecção de glicose em amostras reais, recuperou 100,8% para uma solução eletrolítica comercial e 88,8%... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: In this work, enzymatic electrochemical biosensors fabricated with reduced graphene oxide (rGO) and functionalized deposited by the LbL (Layer-by-Layer) and LangmuirBlodgett (LB) techniques were developed for the detection of glucose. Firstly, to get this, graphene oxide (GO) was chemically synthesized by the oxidation of graphite and after GO was reduced to increase its electrical conductivity and it was functionalized in stabilizing medium containing the poly (diallyl dimethyl ammonium chloride) PDDA, forming the GPDDA, or poly (4-styrene sulphonic) PSS, forming GPSS. Hence, stable aqueous dispersions of graphene nanosheets were formed, essential for the fabrication of LbL films. In a first moment, biosensors with LbL films containing GPDDA, GPSS and the enzyme glucose oxidase (Gox) were fabricated. The performance of these biosensors in the detection of glucose was evaluated as a function of the number of bilayers containing Gox. The best performance in the detection of glucose was presented by the biosensor with the LbL film with the architecture (GPDDA/GPSS)/(GPDDA/GPSS)2. It presented a detection limit of 13.4 μmol.L-1, sensitivity 2.47 μA.cm-2 .mmol-1 .L and analytical range between 0.04 and 0.95 mmol.L-1 . This biosensor was efficient in detecting glucose in the presence of interferents commonly found in body fluids, foods and drugs. When evaluated in the detection of glucose in real samples, it recovered 100.8% for a commercial electrolytic solution and 88.8% for lactose-free milk. In a second moment, biosensors were fabricated from LB films containing the same materials used to fabricate the biosensors with LbL films (GPDDA, GPSS and Gox), in order to compare the two methods of films deposition in the performance of the biosensors. The performance of biosensors fabricated with LB films was evaluated in relation to the amount of deposited... (Complete abstract electronic access below) / Doutor

Neste trabalho foram produzidos filmes finos de nitreto de carbono pelo método de deposição de íons e vapor (IVD) que consiste em evaporar um material (carbono) sobre um substrato e, simultaneamente, realizar sobre ele um bombardeamento por íons (N POT.+). Foram produzidas amostras com energia dos íons de 0,5, 0,8, 2,0, 5,0 e 10,0 keV e razão de transporte TR(C/N) = 04, 1,0, 1,5, 2,0 e 3,0, onde TR(C/N) é o quociente do número de átomos de carbono evaporados pelo canhão eletrônico e o número de íons acelerados que chegam simultaneamente ao mesmo substrato. Os filmes foram produzidos com espessuras de 200, 500 e 1000 nm sobre substratos de Si(100), sílica fundida e carbeto de tungstênio. As propriedades dos filmes analisadas por meio de espectroscopia de fotoeléctrons induzidos por Raios X (XPS), espectrofotometria do infravermelho por transformadas de Fourier (FT-IR), espectrofotometria do visível e ultravioleta, difração de Raios X (XRD) e ensaios de dureza Knoop. Os espectros de XPS foram utilizados para a análise do estado das ligações entre carbono e nitrogênio e também para se estimar a razão de composição CR(C/N) entre átomos de carbono e nitrogênio presentes nos filmes. Foram analisados os picos de C-1s POT. ½ e N-1s POT. ½ e a relação entre suas áreas forneceu a CR(C/N). Para amostras ricas em nitrogênio, o espectro XPS apresentou um pico de C-1s POT. ½ bem proeminente a 286.3 eV. De acordo com nossos resultados este pico pode ser atribuído à presença de ligação tripla CN. Os espectros de FT-IR apresentaram uma banda bem larga de 800 a 1700 CM POT.-1 e um pico proeminente centrado em 2190 CM POT.-1. Este último é atribuído à presença de ligação tripla CN e confirmou a formação de nitreto de carbono. Os estudos de difração de raios-X indicaram que os filmes formados apresentam uma estrutura amorfa ou que contenha pequenos cristalitos não detectáveis por XRD. A dureza e o band gap ótico das amostras mostraram uma forte dependência em relação à energia do feixe de íons e a CR(C/N). Filmes formados a baixa energia de implante (0,5 e 0,8 keV) e alta incorporação de nitrogênio (CR(C/N)=0,6~0,7) apresentaram alta dureza Knoop de até 63 GPa e band gap ótico de até 2,7 eV. Simulações computadorizadas foram utilizadas para ajudar no entendimento dos processos envolvidos na formação de filmes. / In this work thin carbon nitride films were produced by the ion and vapor deposition method (IVD). This method consists of the evaporation of a material (carbon) over a substrate and simultaneously to bombard this same substrate with an ion beam (N POT.+). Samples were produced with ion implantation energies of 0.5, 0.8, 2.0, 5.0 and 10.0 keV and transport ratio TR(C/N) = 0.4, 1.0, 1.5, 2.0 and 3.0, TR(C/N) is the ratio between the number of evaporated carbon atoms by the electron gun and the number of accelerated ions that simultaneously arrive at the substrate. The thicknesses of the produced films were 200, 500 and 1000 nm on Si(100) wafers, fused silica and tungsten carbide substrates. The properties of the films were analyzed by X-ray induced photoelectron spectroscopy (XPS), Fourier transform infrared (FT-IR) spectrophotometry, UV-VIS spectrophotometry, X-ray diffraction (XRD) and Knoop hardness test. The XPS spectra were used for the analyzes of the carbon and nitrogen bonding states and also for the estimation of the composition ratio CR(C/N) between carbon and nitrogen atoms in the films. The analyzes were carried specifically for C-1s POT. ½ and N-1s POT. ½ peaks and the CR(C/N) was estimated from the area ratio of the respective peaks. For nitrogen rich samples, the XPS spectra showed a prominent C-1s POT. ½ peak at 286.3 eV. This peak can be attributed to the presence of the triple bonding CN from our experimental results. The FT-IR spectra showed a wide band between 800 and 1700 CM POT.-1 and a prominent peak centered at 2190 CM POT.-1. This last peak is attributed to the CN triple bonding and confirmed the carbon nitride formation. X-ray diffraction patterns showed that all the films, that were formed has an amorphous structure or contained small crystallites not detectable by XRD. Hardness and optical band gap showed a strong dependence on the ion beam energy and the CR(C/N). Films deposited at low ion energy (0.5 and 0.8 keV) and high nitrogen incorporation (CR(C/N)=0.6~0.7) showed high Knoop hardness of up to 63 GPa and optical band gap of up to 2.7 eV. Computer simulations were used to understand the processes involved in the formation of the films.

Neste trabalho foram estudadas as propriedades magnéticas e características estruturais de filmes finos bicamadas e multicamadas formados com Ni{x}Cu{1-x} (ferromagneto, 50 x 90 e níquel puro), NiO, FeMn e IrMn (antiferromagnetos). Ligas de níquel-cobre podem ser classificadas como fracamente ferromagnéticas. Ferromagnetismo fraco gera efeitos únicos em propriedades de transporte eletrônico, junções com supercondutores e sistemas de exchange bias. Sendo uma solução sólida extremamente simples, diversas propriedades do NiCu dependem linearmente da estequiometria da liga. Os filmes finos foram depositados via magnetron sputtering em substratos monocristalinos de Si (100) a temperatura ambiente. A estrutura cristalina foi caracterizada por difração de raios-X e a morfologia foi analisada por retroespalhamento Rutherford. Propriedades magnéticas foram medidas com um SQUID a temperaturas entre 5K e 300K (curvas ZFC/FC e de magnetização a diferentes temperaturas) e com um VSM a temperatura ambiente (curvas de magnetização em diferentes ângulos no plano do filme). As amostras com Ni{x}Cu{1-x} mais ricas em cobre apresentam os maiores campos coercivo e de exchange bias a baixas temperaturas, mas os menores em altas temperaturas, geralmente respeitando as temperaturas de bloqueio indicadas nas curvas ZFC/FC. O acoplamento na interface das bicamadas NiO/Ni{x}Cu{1-x} foi muito reduzido em temperaturas mais altas, não sendo observada anisotropia unidirecional nas medidas a temperatura ambiente. As constantes de interação J{int} foram calculadas para as bicamadas Ni{x}Cu{1-x}/FeMn e Ni{x}Cu{1-x}/IrMn a 5K e temperatura ambiente. O sistema Ni/IrMn apresentou o maior valor dessa constante em ambas as temperaturas, diminuindo muito com o acréscimo de cobre na camada Ni{x}Cu{1-x}. O sistema Ni{90}Cu{10}/FeMn apresentou os maiores valores em ambas as temperaturas (excluindo o de níquel puro), comparável até com Ni/IrMn no caso de 5K. Esse comportamento pode ser explicado por uma maior afinidade entre as estruturas do FeMn e Ni{x}Cu{1-x} na direção [111], o que favorece a formação da face (111) no FeMn, mais consistentemente do que no IrMn. Nenhuma das multicamadas estudadas se comportou como uma válvula de spin, mas foi observado um aumento na coercividade e supressão do exchange bias, provavelmente por causa de acoplamento entre as camadas ferromagnéticas. / In this work, a study of magnetic properties and structural characteristics of bilayered and multilayered thin films was carried out. These films were made with Ni{x}Cu{1-x} (ferromagnet, 50 x 90 and pure nickel) and NiO, FeMn or IrMn as antiferromagnets. Nickel-copper alloys may be classified as weakly ferromagnetic. Weak ferromagnetism generates unique effects in electron transport, junctions with superconductors and exchange bias systems. Being an extremely simple solid solution, many properties of the NiCu alloy have a linear dependence with its stoichiometry. The thin films were deposited via magnetron sputtering on Si (100) monocrystalline substracts at room temperature. The crystalline structure was characterized by X-ray diffraction, while the morphology was analyzed by Rutherford back-scattering. Magnetic properties were measured by SQUID at temperatures between 5K and 300K (ZFC/FC curves and magnetization curves at different temperatures) and by VSM at room temperature (magnetization curves at varying angles at the film plan). The samples with copper-richer Ni{x}Cu{1-x} showed the largest coercive and exchange bias fields in lower temperatures, but the lowest ones in higher temperatures, usually following the blocking temperatures denoted by the ZFC/FC curves. The interfacial coupling in NiO/Ni{x}Cu{1-x} was very weak in higher temperatures, not showing unidirectional anisotropy at room temperature. The J{int} exchange interaction constants were calculated for the Ni{x}Cu{1-x}/FeMn and Ni{x}Cu{1-x}/IrMn bilayers at 5K and room temperature. Ni/IrMn had the largest values at both temperatures, heavily decreasing with the copper content in the Ni{x}Cu{1-x} layer. Ni{90}Cu{10}/FeMn showed the largest values at both temperatures (except for the pure Ni sample), even reaching the values of Ni/IrMn at 5K. This behavior may be explaned by a greater matching between the FeMn and Ni{x}Cu{1-x} crystalline structures on the [111] direction, which favors the formation of the (111) FeMn face more consistently than the corresponding IrMn face. All the studied multilayers showed no spin valve behavior, but an increased coercivity and supressed exchange bias were observed, probably because of coupling between the ferromagnetic layers.