Spelling suggestions: "subject:"anda figure off merit"" "subject:"anda figure oof merit""
21 |
Nanolaminated Thin Films for ThermoelectricsKedsongpanya, Sit January 2010 (has links)
Energy harvesting is an interesting topic for today since we face running out of energy source, a serious problem in the world. Thermoelectric devices are a good candidate. They can convert heat (i.e. temperature gradient) to electricity. This result leads us to use them to harvest waste heat from engines or in power plants to generate electricity. Moreover, thermoelectric devices also perform cooling by applied voltage to device. This process is clean, which means that no greenhouse gases are emitted during the process. However, the converting efficiency of thermoelectrics are very low compare to a home refrigerator. The thermoelectric figure of merit (ZTm) is a number which defines the converting efficiency of thermoelectric materials and devices. ZTm is defined by Seebeck coefficient, electrical conductivity and thermal conductivity. To improve the converting efficiency, nanolaminated materials are good candidate. This thesis studies TiN/ScN artificial nanolaminates, or superlattices were grown by reactive dc magnetron sputtering from Ti and Sc targets. For TiN/ScN superlattice, X-ray diffraction (XRD) and reciprocal space map (RSM) show that we can obtain single crystal TiN/ScN superlattice. X-ray reflectivity (XRR) shows the superlattice films have a rough surface, supported by transmission electron microscopy (TEM). Also, TiN/ScN superlattices grew by TiN as starting layer has better crystalline quality than ScN as starting layer. The electrical measurement shows that our superlattice films are conductive films. Ca-Co-O system for inherently nanolaminated materials were grown by reactive rf magnetron sputtering from Ca/Co alloy target. The XRD shows we maybe get the [Ca2CoO3]xCoO2 phase, so far. The energy dispersive X-ray spectroscopy (EDX) reported that our films have Al conmination. We also discovered unexpected behavior when the film grown at high temperature showed larger thickness instead of thinner, which would have been expected due to possible Ca evaporation. The Ca-Co-O system requires further studies.
|
22 |
Thin films for thermoeletric applicationsLin, Keng-Yu January 2014 (has links)
Global warming and developments of alternative energy technologies have become important issues nowadays. Subsequently, the concept of energy harvesting is rising because of its ability of transferring waste energy into usable energy. Thermoelectric devices play a role in this field since there is tremendous waste heat existing in our lives, such as heat from engines, generators, stoves, computers, etc. Thermoelectric devices can extract the waste heat and turn them into electricity. Moreover, the reverse thermoelectric phenomenon has the function of cooling which can be applied to refrigerator or heat dissipation for electronic devices. However, the energy conversion efficiency is still low comparing to other energy technologies. The efficiency is judged by thermoelectric figure of merit (ZT), defined by Seebeck coefficient, electrical conductivity and thermal conductivity. In order to improve ZT, thin film materials are good candidates because of their structural effects on altering ZT. Ca3Co4O9 thin films grown by reactive radio frequency magnetron sputtering followed by post-annealing process is studied in this thesis. Structural properties of the films with the evolution of elemental ratio (Ca/Co) of calcium and cobalt have been investigated. For the investigations, three samples having elemental ratio 0.82, 0.72, and 0.66 for sample CCO1, CCO2 and COO3, respectively, have been prepared. Structural properties of the films have been investigated by X-ray diffraction (XRD) θ-2θ and pole figure analyses. Surface morphology of the films has been investigated by scanning electron microscopic (SEM) analyses. The highly oriented and phase pure epitaxial Ca3Co4O9 thin films were obtained in the end. Mixing of ScN and CrN to obtain ScxCr1-xN solid solution thin films by DC magnetron sputtering is the other task in this thesis. Growth of ScN and CrN thin films were studied first in order to get the best mixed growth conditions. The phase shifts between ScN (111) and CrN (111) peaks were observed in mixed growth films by XRD θ-2θ measurements, indicating the formation of ScxCr1-xN. Surface morphology of the films were investigated by SEM. The (111)-oriented ScxCr1-xN thin films with decent surface smoothness grown by DC magnetron sputtering at 600 °C in pure nitrogen with bias were developed.
|
23 |
Implementação computacional de um novo método matricial para a determinação de fases em cristalografia / Computational implementation of a new matricial method for phase determination in crystallographyGabriela Castellano 25 March 1994 (has links)
Um novo critério, proposto por Jorge Navaza a partir de considerações teóricas para resolver o problema das fases, é avaliado numericamente. Este critério se baseia na propriedade de atomicidade da função densidade eletrônica, generalizando resultados obtidos por Goedkoop. O problema das fases é resolvido teoricamente pela minimização de uma função, R, que é formada pela soma dos menores autovalores de uma matriz, Q, construída a partir de todos os fatores de estrutura observados. O conjunto de fases procurado é aquele que minimiza R. Como a matriz Q depende em forma relativamente complexa do grupo de simetria espacial do cristal, teoria dos grupos é utilizada para reduzir a ordem desta matriz. O algoritmo e a implantação computacional do cálculo da função R, juntamente com testes numéricos que demonstram a utilidade do critério de Navaza, são descritos em detalhe. Como corolário, que pode talvez resultar de grande importância prática, é mostrado que a função R pode ser utilizada como uma nova figura de mérito nos métodos diretos por multissolução clássicos. Finalmente, é desenvolvida a álgebra correspondente ao cálculo do gradiente da função R, indicando a direção de trabalhos futuros / A new criterion, proposed by Jorge Navaza from theoretical considerations to solve the Phase Problem, is numerically tested. The criterion is based on the atomicity property of the electron density function, generalizing previous results by Goedkoop. The Phase Problem is theoretically solved by the minimization of a function, R, which is formed from the sum of the smallest eigenvalues of a matrix Q, constructed from the set of all observed structure factors. The sought set of phases is that which minimizes R. Because the matrix Q depends in a relatively complex fashion on the space group of the crystal, group theory is employed to reduce the order of Q. The algorithm and computational implementation for the calculation of R, together with numerical tests which demonstrate the usefulness of Navaza´s criterion, are described in detail. As a corollary, that might turn out to be of a practical importance, it is shown that the minimum value of the function R can be used as a novel Figure of Merit in the classical Multisolution Direct Methods. Finally, the rather complex algebra necessary for the calculation of the gradient of the function R is developed, indicating also the possible trends for future work
|
24 |
Utilização do sinal analitico liquido para avaliação de modelos de calibração multivariada atraves do calculo de figuras de merito e de cartas de controle / Utilization of net analyte signal for validation of models of multivariate calibration by using figure of merit and chart controlRocha, Werickson Fortunato de Carvalho 18 July 2007 (has links)
Orientador: Ronei Jesus Poppi / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Quimica / Made available in DSpace on 2018-08-08T23:09:23Z (GMT). No. of bitstreams: 1
Rocha_WericksonFortunatodeCarvalho_M.pdf: 1597999 bytes, checksum: b9442a7ba3c5fb6665bc87f6c0abf34f (MD5)
Previous issue date: 2007 / Resumo: Este trabalho teve como objetivo realizar a validação de modelos de calibração multivariada através do cálculo de figuras de mérito e de cartas de controle multivariadas do medicamento Nimesulida por meio do infravermelho próximo. Foram preparadas 69 amostras sintéticas contendo o princípio ativo (Nimesulida) na faixa 10,38-39,47% (m/m) em excipiente (lactose, povidona-KV29- 32, celulose 200, lauril sulfato de sódio, croscarmelose sódica e estereato de magnésio). Destas, 49 amostras foram utilizadas para a calibração e 20 para a validação as quais foram separadas através do algoritmo de Kennard-Stone.O tratamento utilizado nos espectros foi a correção do espalhamento multiplicativo. Na sequência, foram obtidos os seguintes valores para as figuras de mérito: limite de detecção (0,61), limite de quantificação (2,03), exatidão (RMESC 0,66 %(m/m), RMESCV 0,92 %(m/m), RMESP 1,05 %(m/m) de Nimesulida), seletividade média (0,0056), sensibilidade (0,0036), inverso da sensibilidade analítica (0,20 %(m/m)-1 de Nimesulida) e razão sinal ruído (181,11). Na segunda parte do trabalho foram utilizadas 113 amostras sintéticas para construção das cartas de controle multivariadas. Foram desenvolvidas três cartas de controle e calculado os limites de controle para cada carta. Os valores encontrados foram: - carta NAS (Superior = 6,54x10; Inferior = 5,25x10), carta interferente (7,45) e carta resíduo (2,38x10). Através dessas cartas de controle multivariadas foi possível identificar as amostras que estavam dentro e fora de controle. Obteve-se 64 amostras fora de controle e 20 amostras dentro de controle de acordo com o planejamento experimental realizado. Sendo assim, foi possível identificar, de forma qualitativa, as amostras de Nimesulida que estavam dentro e fora de controle. Portanto, a dissertação desenvolvida sugere um novo método analítico que pode ser usado para controle de qualidade em fármacos e para validação de modelos de calibração multivariada, pois os resultados obtidos, indicam que o modelo desenvolvido pode ser utilizado na indústria farmacêutica como uma alternativa ao método-padrão / Abstract: This study was mainly intended to elaborate the validation of multivariate calibration models based on the determination of figures of merit and multivariate control charts constructed with data from Nimesulide tablets upon near-infrared spectroscopy use. A total of 69 synthetic samples were prepared containing the active principle (Nimesulide) in the range of 10,38-39, 47% (m/m) in excipients (Cellulose,Sodium Lauryl Sulphate, Magnesium Stearate, Carmellose Sodium,Povidone and Lactose). It was used 49 samples for the calibration and 20 for the validation that were separated by the Kennard-Stone algorithm. It was used the multiplicative scatter correction on the spectra set. Then, the following values for figures of merit were calculated: limit of detection (0,61) , limit quantification (2,03) , accuracy (RMESC 0,66 %(m/m), RMESCV 0,92%(m/m), RMESP 1,05%(m/m) of Nimesulide), mean selectivity (0,0056), sensitivity (0,0036), inverse analytical sensitivity (0,20 %(m/m)- 1 of the Nimesulide) and signal-to-noise ration (181, 11). In the second application, 113 synthetic samples were used for the construction of multivariate control charts. Three control charts were designed and the control limits for each chart were calculated. The values found were: - chart NAS (upper = 6,54x10; below = 5,25x10), chart interferent (7,45) and chart residual (2,38x10). From these charts it was possible to identify the samples that were in and out of control. It was obtained 64 samples out of control and 20 samples in control, according to the experimental design. Therefore, it was possible to identify, in a qualitative way, the Nimisulide samples that were in control and the ones that were out of control. Hence it follows that, this research project suggests a new analytical method that can be used for quality control in drugs and also for the validation of multivariate calibration models. Thus, the results obtained indicate that this model can be used in pharmaceutical industries as an alternative to the standard procedure / Mestrado / Quimica Analitica / Mestre em Química
|
25 |
Estudo experimental da otimização em sistemas de mamografia digital CR e DR / Experimental study of optimization in CR and DR digital mammography systemsAlessandra Maia Marques Martinez Perez 29 January 2015 (has links)
A recente inserção e forte avanço da mamografia digital no Brasil como ferramenta de rastreamento do câncer mamário e as evidências de outras condições de otimização, quando comparadas à mamografia convencional (tela filme), requerem que novos parâmetros de qualidade sejam incluídos e estudados, bem como que as condições de otimização sejam revistas. O objetivo deste trabalho foi determinar a técnica radiográfica otimizada para dois sistemas de detecção (CR e DR) em uso em três unidades de mamografia: Mammomat 3000 Nova (Siemens), Senographe DMR (GE) e Senographe 2000D (GE). A otimização foi conduzida para uma variedade de combinações de fatores técnicos e configurações de simuladores de mama, tais como valores de kilovoltagem (26 a 32 kV), combinações anodo/filtro (Mo/Mo, Mo/Rh e Rh/Rh), material simulador de mama de várias espessuras (2 a 8 cm) e lesões simuladas como massas e calcificações, usando uma figura de mérito (FOM) como parâmetro. Verificou-se que o uso da combinação anodo/filtro que gera os espectros mais energéticos em cada equipamento proporcionou os maiores valores de FOM para todas as espessuras de simulador de mama e voltagens, devido a redução da dose. As combinações anodo/filtro que deram esses resultados foram Mo/Rh para o equipamento da marca Siemens e Rh/Rh para ambos os equipamentos da marca GE, correspondentes aos espectros mais energéticos de cada unidade. Foi observada ainda uma tendência de aumento do kV que maximiza FOM com o aumento da espessura. / The recent introduction and intense advance of digital mammography in Brazil as a tool in breast cancer screening and the evidences of new optimization conditions when compared to conventional mammography (screen-film) require adding and studying novel quality parameters, as well as revisiting optimization conditions. The objective of this work was to determine optimized radiographic technique for two detection systems (CR and DR) in use in three mammography units: Mammomat 3000 Nova (Siemens), Senographe DMR (GE) and Senographe 2000D (GE). Optimization was conducted for various combinations of technique factors and breast phantom configurations, such as kilovoltage settings (26 to 32 kV), target/filter combinations (Mo/Mo, Mo/Rh and Rh/Rh), breast equivalent material in various thicknesses (2 to 8 cm) and simulated mass and calcification lesions, using a figure of merit (FOM) as a parameter. When using anode/filter combination which generates higher energy spectra in each equipment, it was verified that higher FOM values were achieved for all voltages and phantom thicknesses, due to dose reduction. Anode/filter combinations which led to those results were Mo/Rh for Siemens equipment and Rh/Rh for both GE equipments, corresponding to the higher energy spectra in each unity. It was also observed an increasing tendency of kV which maximizes FOM with the increase of thickness.
|
26 |
Numerical Simulation of Thermoelectric Transport in Bulk and Nanostructured SiSn AlloysDusetty, Venkatakrishna 15 July 2020 (has links)
The current high demand for sustainable and renewable energy sources to solve world energy crisis has enormously increased interest in looking at alternative sources of energy. All the machines used in manufacturing process, electricity generation, residential applications, transportation etc., rejects energy in the form of heat into environment. Thermoelectric materials can convert thermal-to-electrical and electrical-to-thermal energy and can be utilized in waste-heat harvesting, more efficient cooling to reduce energy usage and CO2 emissions. Significant research efforts have been devoted over the past decade to thermoelectric materials, with particular emphasis being placed on combining materials selection with nanostructuring. The overarching goal was to reduce thermal conductivity through selective phonon scattering and thus boost the thermoelectric figure-of-merit (ZT). SiGe alloys, as well as superlattices and nanocomposites made from them, showed significant improvements upon nanostructuring and ZT exceeding one at high temperatures. Other group IV alloys were not studied in the context of thermoelectrics. However, SiSn alloys are widely studied for their optoelectronic properties because they were predicted to become direct-gap materials when Sn composition increased beyond about 50%. To address this gap, we study the thermoelectric properties of SiSn alloys. Furthermore, we develop an iterative full-band solver for the electron Boltzmann transport equation and use it to compute the electron and hole mobility and Seebeck coeffcient in SiSn alloys. The electronic structure of SiSn alloys was computed in the virtual crystal approximation from non-local empirical pseudopotentials, while the application of strain allowed us to extract the electron-phonon coupling deformation potentials for each alloy composition. We benchmark our code against available mobility data for Si and SiGe alloys and find that it accurately reproduces the measured values. Full phonon dispersion was computed from the adiabatic bond charge model, which was shown to accurately reproduce measured dispersion, and used in our phonon BTE solver to compute lattice thermal conductivities. Scattering rates include anharmonic phonon-phonon, impurity, isotope, alloy, and boundary mechanisms. The lowest thermal conductivity was obtained in SiSn alloys, which have been experimentally demonstrated with up to 18% Sn composition. This carries through when combined with calculations of electronic power factor, where mobilities and Seebeck coeffcients of SiSn alloys are comparable to those of SiGe. Furthermore, ZT is optimized through doping for every composition. The ZT improves dramatically at higher temperatures, reaching ZT of 1.9, 2.36 is obtained for Sn composition of 10% and 50% in a n-doped bulk SiSn alloys at a temperature of 1480 K. However, such high Sn composition of 50% is unlikely to be synthesized due to low solid solubility of Sn in Si. Lastly, we study the impact of nanostructuring in thin films on the ZT. We also establish the limits on how much the ZT can be improved through nanostructuring by studying thin films of SiSn alloys across temperature from room temperature up to 1500 K. We conclude that in bulk SiSn alloys, even at modest Sn concentration of 10%, ZT can reach 1.9, while in 20 nm thin films of n-type SiSn alloys, it can reach the long-sought target of ZT>3 and ZT of 2.16 is obtained in p-type nanostructured SiSn alloys.
|
27 |
Thermoelectric Transport and Energy Conversion Using Novel 2D MaterialsWirth, Luke J. January 2016 (has links)
No description available.
|
28 |
Next Generation Wideband Antenna Arrays for Communications and Radio AstrophysicsKolitsidas, Christos January 2017 (has links)
Wideband, wide-scan antenna arrays are a promising candidate for the future wireless networks and as well as an essential part of experimental radio astrophysics. Understanding the underline physics of the element performance in the array environment is paramount to develop and improve the performance of array systems. The focus of this thesis is to develop novel wideband antenna array technologies and develop new theoretical insights of the fundamental limits of antenna arrays. The developed methodologies have also been extended to include a radio astrophysics application for the global 21cm experiment. Investigating the fundamental antenna array limits and extracting general performance measures can provide a priori estimates for any application of arrays. In this thesis, a general measure for antenna arrays, the array figure of merit is proposed. This measure couples bandwidth, height from the ground plane and reflection coefficient in a bounded quantity. An extension of the array figure of merit that is able to provide matching, bandwidth and directivity/gain limits is also introduced. The soft Vivaldi array is introduced as a novel wideband, wide-scan angle array technology. Periodic structure loading has been utilized to improve the array's performance and mold the electromagnetic wave behavior to our benefit. The soft condition has been utilized in the same manner as the conventional soft-horn antenna at the Vivaldi element. An integrated matching layer in the form of periodic strip loading is introduced. A single polarized soft Vivaldi array prototype has been developed fabricated and measured. The developed finite array has been loaded with a soft condition in the periphery to mitigate edge effects. The results indicated improved cross-polarization and side-lobe levels. A new class of wideband antenna arrays, the Strongly Coupled Asymmetric Dipole Array (SCADA) was also proposed in this thesis. Exploiting asymmetry in the array element introduces an additional degree of freedom that improves bandwidth and scanning performance. A novel methodology for terminating finite arrays is also proposed. The theory and an experimental antenna array is presented with good agreement between measured and simulated results. An effort to integrate a vertical wide angle matching layer was also addressed and a prototype array with this concept is presented. In the last part of this thesis, a methodology for the detection of the global cosmological 21cm signal from the Epoch of Reionization (EoR) is developed. The main sources of errors in this experiment, the foregrounds and the antenna chromaticity are evaluated. A new algorithmic methodology for extracting the global EoR signal is proposed. The method is based on piecewise polynomial fitting and has successfully been applied and evaluated. An antenna array that is based on the methodologies described in this thesis has been developed and evaluated with the proposed algorithm. / Bredbandiga gruppantenner med stor utstyrningsvinkel är en av de lovande kandidaterna för nästa generations trådlösa kommunikationsnätverk samt en väsentlig del av experimentell radioastrofysik. Att förstå de bakomliggande fysikaliska principerna hos gruppantennens element är avgörande för att kunna utveckla och förbättra prestandan hos ett gruppantennsystem. Denna avhandling är fokuserad på att utveckla nya bredbandstekniker samt nya teoretiska insikter om de grundläggande gränserna för gruppantenner. De här utvecklade metoderna har förutom kommunikationstillämpningar också tillämpats på en radioastrofysik tillämpning i det globala 21cm experimentet. Att undersöka de fundamentala gränserna för gruppantenner och att utröna allmängiltiga mått på deras prestandaegenskaper kan möjliggöra a priori uppskattningar om gruppantenns tillämpbarhet för dess planerade användning. I den här avhandlingen föreslås ett allmänt kvalitetsmått på gruppantenner: gruppantennkvaliten. Detta mått kopplar samman främst bandbredd, reflektionskoefficienten med antennens tjocklek över ett jordplanet. En utvidgning av begreppet gruppantennkvaliten, presenters också i avhandlingen det kopplar samman bandbredd, matchning med antennens direktivitet/förstärkningsfaktor. En Vivaldi-gruppantenn med mjuka ytor introduceras här som en ny sorts bredbandig gruppantenn med stor utstyrningsvinkel. I antennen har en periodisk belastning inkluderats för att förbättra dess egenskaper, och för att forma antennens elektromagnetiska utstrålning till vår fördel. Den mjuka ytan på elementet har används på ett liknande sätt som det välkända korrigerade Vivaldihornets design, och har integrerats direkt i elementets design. Den här utvecklade ändliga gruppantennen har också en mjuk yta på dess yttre delar för att minska kanteffekternas påverkan av antennprestandan. Resultaten indikerade både förbättrad korspolarisations och lägre sidlobsnivåer hos antennen. En ny klass av bredbandiga gruppantenner har utvecklas i denna avhandling, den kallas en Starkt Kopplad Asymmetrisk Dipol-gruppAntennen - SCADA. Genom att utnyttja geometrisk asymmetri i antennelementet introduceras ytterligare en frihetsgrad som möjliggör förbättrad bandbredd och utstyrning. Vidare presenteras här en ny metod för impedansterminering av ändliga gruppantenner. Både SCADA-teorin samt dess verifiering i forma av en experimentell gruppantenn presenteras här. Teori, simulering och experiment visar god överenskommelse, vilket validerar idéerna. En prototyp av ett matchande skikt som stöder stor utstyrbarhet har integrerats med gruppantennprototypen och presenteras i avhandlingen. I den sista delen av avhandling utvecklas också en metod för detektering av den globala kosmologiska 21 cm-signalen från universums rejoniseringsepok - EoR. Huvudkällorna för mätfel i detta experiment utvärderas, de är antennens kromaticiteten och förgrundsstrålningen. En ny algoritmbaserad metod för att extrahera den globala EoR-signalen föreslås. Metoden är baserad på anpassning med multipla polynom och har med framgång tillämpats och utvärderats. En gruppantenn som baseras på de metoder som beskrivs i avhandling har också föreslagits och dess prestanda har utvärderats med den föreslagna metoden. / <p>QC 20171121</p>
|
29 |
Microwave oscillator with phase noise reduction using nanoscale technology for wireless systemsAqeeli, Mohammed Ali M. January 2015 (has links)
This thesis introduces, for the first time, a novel 4-bit, metal-oxide-metal (MOM) digital capacitor switching array (MOMDCSA) which has been implemented into a wideband CMOS voltage controlled oscillator (VCO) for 5 GHz WiMAX/WLAN applications. The proposed MOMDCSA is added both in series and parallel to nMOS varactors. For further gain linearity, a wider tuning range and minor phase noise variations, this varactor bank is connected in parallel to four nMOS varactor pairs, each of which is biased at a different voltage. Thus, VCO tuning gain reduces and optimal phase noise variation is obtained across a wide range of frequencies. Based on this premise, a wideband VCO is achieved with low phase noise variation of less than 4.7 dBc/Hz. The proposed VCO has been designed using UMC 130 nm CMOS technology. It operates from 3.45 GHz to 6.23 GHz, with a phase noise of -133.80 dBc/Hz at a 1 MHz offset, a figure of merit (FoM) of -203.5 dBc/Hz. A novel microstrip low-phase noise oscillator is based on a left-handed (LH) metamaterial bandpass filter which is embedded in the feedback loop of the oscillator. The oscillator is designed at a complex quality factor Qsc peak frequency, to achieve excellent phase noise performance. At a centre frequency of 2.05 GHz, the reported oscillator demonstrates, experimentally, a phase noise of -126.7 dBc/Hz at a 100 kHz frequency offset and a FoM of -207.2 dBc/Hz at a 1 MHz frequency offset. The increasing demands have been placed on the electromagnetic compatibility performance of VCO devices is crucial. Therefore, this thesis extends the potential of highly flexible and conductive graphene laminate to the application of electromagnetic interference (EMI) shielding. Graphene nanoflake-based conductive ink is printed on paper, and then it is compressed to form graphene laminate with a conductivity of 0.43×105 S/m. Shielding effectiveness is experimentally measured at above 32 dB as being between 12GHz and 18GHz, even though the thickness of the graphene laminate is only 7.7µm. This result demonstrates that graphene has great potential for offering lightweight, low-cost, flexible and environmentally friendly shielding materials which can be extended to offering required shielding from electromagnetic interference (EMI), not only for VCO phase noise optimisation, but also for sensitive electronic devices.
|
30 |
Avaliação do potencial da espectroscopia no infravermelho próximo como método de rotina para a determinação de carbono orgânico do solo / Evaluation of the potential of near infrared spectroscopy as routine method for soil organic carbon analysisSouza, André Marcelo de, 1977- 26 August 2018 (has links)
Orientador: Ronei Jesus Poppi / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Química / Made available in DSpace on 2018-08-26T13:27:08Z (GMT). No. of bitstreams: 1
Souza_AndreMarcelode_D.pdf: 34076641 bytes, checksum: 9b864adff6bf9857ffe47d6b6afee68a (MD5)
Previous issue date: 2015 / Resumo: As pesquisas atuais apontam que a espectroscopia no infravermelho próximo (NIR) é a técnica alternativa mais promissora para a determinação de carbono orgânico do solo (SOC) nos laboratórios de todo o mundo em substituição total ou parcial aos métodos tradicionais de via úmida. Considerado este fato, foi desenvolvido e validado um método para a determinação de SOC por espectroscopia NIR, visando seu empregado como método de rotina em laboratórios de análise de solos do Brasil. Para este fim, foram construídos modelos de calibração multivariada a partir de um número expressivo de amostras de solos (1490 amostras, 2.980 espectros) que englobam a variabilidade de solos brasileiros. Estes modelos foram validados através da submissão dos valores previstos das concentrações de matéria orgânica do solo (SOM) ao Programa de Análise de Qualidade de Laboratórios de Fertilidade (PAQLF). As questões envolvendo a transferência de calibração entre múltiplos instrumentos também foram abordadas e a regressão por vetores de suporte (SVR) foi avaliada como alternativa à regressão em mínimos quadrados parciais (PLS). Os resultados alcançados comprovaram de maneira contundente a robustez do método proposto e indicaram que o mesmo pode substitui o método de via úmida, superando seu desempenho em alguns casos. No estudo de transferência de calibração, foi demonstrado que quando dois ou mais espectrofotômetros NIR são empregados na aquisição dos dados, recomenda-se que ambos sejam de mesma configuração. Porém, quando instrumentos diferentes foram envolvidos, o método de atualização do modelo através da matriz aumentada apresentou resultados satisfatórios em relação aos demais métodos avaliados. Existem, no entanto, pelo menos dois gargalos da implementação da espectroscopia NIR em análises de rotina: (1) o elevado custo dos instrumentos em relação ao orçamento dos laboratórios de análise de solos no Brasil; e (2) a necessidade do emprego da quimiometria na etapa de modelagem dos dados. Ambas as questões podem ser solucionadas com políticas de subsídios para compra de instrumentos e intensivos treinamentos em quimiometria e espectroscopia NIR, que podem ser oferecidos pela Empresa Brasileira de Pesquisa Agropecuária e por instituições de ensino superior do Brasil / Abstract: Current research indicates that near infrared spectroscopy (NIR) is the most promising alternative technique for the determination of soil organic carbon (SOC) in laboratories around the world in total or partial replacement traditional wet chemistry methods. Considering this fact, in this research it was developed and validated a method for determination of SOC by NIR spectroscopy, aiming its use in soils laboratories in Brazil as a routine analysis method. To this end, multivariate calibration models were constructed from a large number of soil samples (samples 1490, 2980 spectra) covering the variability of brazilian soils. These models were validated by the submission of the predicted concentrations of soil organic matter (SOM) in the Quality Program of Analysis of Fertility Laboratories (PAQLF). Issues involving the calibration transfer among several instruments were discussed and the support vector regression (SVR) was evaluated as an alternative to PLS. The results proved in a conclusive way the robustness of the proposed method and indicated that it can replace the wet chemistry method, outperforming it in some cases. Some general recommendations that can be drawn from this work are: when two or more NIR spectrophotometers are involved, it is recommended purchase or apply those of same configuration because the transfer methods evaluated generate better models when compared to those of different configuration. However, when different instruments are involved, the method of model updating through the augmented matrix showed satisfactory results when compared to other methods evaluated. There are at least two bottlenecks of the implementation of the NIR spectroscopy method in routine analysis of soil laboratories: (1) the cost of NIR spectroscopy instruments and their maintenance are considered high in relation to the budget of the laboratories of soil analysis in Brazil and (2) the spectral data treatment, which requires the use of chemometrics. Both aspects that hinder the implantation of NIR spectroscopy as a routine method can be solved, with a policy of subsidies to purchase equipment and intensive training in chemometrics that can be offered by Brazilian Corporation of Agricultural Research (Embrapa) and also by brazilian teaching institutions / Doutorado / Quimica Analitica / Doutor em Ciências
|
Page generated in 0.1074 seconds