A performance comparison of heat pumps using a scroll (Sanden) and a reciprocating (Bitzer) compressor was conducted experimentally. The refrigerant used was R600a (isobutane). The heat pump components were evaluated performance-wise through: volumetric and isentropic efficiency of the compressors; the UA-value of the condensers and evaporators; sensible enthalpy differences between the liquid and the suction line in the internal heat exchanger; and overall heat pump system comparison as the coefficient of performance. The Bitzer heat pump had existing measuring devices and equipment installed, and it was already filled with refrigerant. The Sanden heat pump required installation of equipment and measuring devices and a refrigerant refill. The refrigerant charge was decided according to the criteria of the lowest compressor speed, which had an effect of overcharge for higher speeds. The measurements included the temperature of water and refrigerant, pressure of the refrigerant, water volume flow, and compressor power. The heat pumps performances were then evaluated based on these parameters. The Sanden compressor showed higher volumetric efficiency than the Bitzer compressor, for the two lower (out of three) speeds of the compressor. The isentropic efficiency of the Bitzer compressor proved to be higher for all pressure ratios out of the three speeds respectively. The condenser in the Bitzer heat pump showed proper UA-values based on the temperature differences between refrigerant and heat sink. The UA-values of the Sanden heat pump condenser did not increase with compressor speed which then gave a larger temperature difference between refrigerant and heat sink, for the two higher compressor speeds. The evaporators had a similar issue with the temperature difference between refrigerant and heat source, which also showed on the UA-values. The internal heat exchanger in the Bitzer heat pump had a larger sensible enthalpy difference on the suction side compared to the liquid side, when condenser subcooling was low, indicating that some fraction of refrigerant was being condensed instead. The Sanden heat pump instead had higher condenser subcooling and the sensible enthalpy difference showed to be very low in the internal heat exchanger. Finally, the coefficient of performance showed to be slightly higher in the Bitzer heat pump for almost all evaluated condensation and evaporation temperatures. / En jämförelse av prestandan för två värmepumpar, en med scroll- (Sanden) och en med kolvkompressor (Bitzer) har gjorts. Köldmedlet som användes var R600a (isobutan). Komponenterna och hur deras prestanda blev utvärderad följer här: kompressorernas volymetriska och isentropisk verkningsgrad; kondensorns och förångarens UA-värden; den sensibla entalpiskillnaden mellan gas- och vätskeledning i den interna värmeväxlaren; en övergripande jämförelse av värmepumparna i form av deras värmefaktor. Värmepumpen med Bitzerkompressor utvärderades med befintliga komponenter och mätutrustning, och en redan fylld mängd köldmedium. Värmepumpen med Sandenkompressor installerades med mätutrustning och komponenter, och fylldes med köldmedium. Mängden köldmedium bestämdes utifrån kriterier för det lägsta varvtalet på kompressorn, vilket visade sig ge en för stor mängd köldmedium vid de högre varvtalen. Mätningarna inkluderade temperatur på vatten och köldmedium, köldmediets tryck, vattnets volymflöde samt kompressorns effektbehov. Prestandan för värmepumparna är sedan utvärderad utifrån dessa data. Sandenkompressorn visade en högre volymetrisk verkningsgrad för de två lägre (av tre) hastigheterna utvärderade i experimenten, jämfört med Bitzer. Den isentropiska verkningsgraden var högre i Bitzerkompressorn för samtliga tryckförhållanden för de tre respektive hastigheterna. Kondensorn i Bitzervärmepumpen uppvisade goda UA-värden, baserat på temperaturskillnaderna mellan köldmedium och värmesänka. Kondensorn i Sandenvärmepumpen visade ingen förbättring av UA-värden när kompressorns hastighet ökade, vilket i sin tur gav upphov till stora temperaturskillnader mellan köldmediet och värmesänkan. Förångarna i båda värmepumparna uppvisade liknande problem med höga temperaturskillnader mellan värmekälla och köldmedium, vilket även deras UA-värden visade. Den interna värmeväxlaren i Bitzervärmepumpen visade större skillnad i den sensibla entalpin på sug- jämfört med vätskesidan, när underkylningen i kondensorn var låg, vilket indikerade på att en del av köldmediet istället kondenserade. I Sandenvärmepumpen var kondensorns underkylning högre vilket då uppvisade en liten skillnad i den sensibla entalpin mellan de båda sidorna. Slutligen så visade Bitzervärmepumpen en något högre värmefaktor än Sandenvärmepumpen för nästan alla utvärderade kondenserings- och förångningstemperaturer.

Submitted by Maicon Juliano Schmidt (maicons) on 2015-05-25T14:45:18Z No. of bitstreams: 1 Bruno de Sá Beckerle.pdf: 3991624 bytes, checksum: 500b993c0f3a1c3a8c3abae05ed51c5e (MD5) / Made available in DSpace on 2015-05-25T14:45:18Z (GMT). No. of bitstreams: 1 Bruno de Sá Beckerle.pdf: 3991624 bytes, checksum: 500b993c0f3a1c3a8c3abae05ed51c5e (MD5) Previous issue date: 2015-01-05 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Este trabalho apresenta um estudo experimental da transferência de calor e queda de pressão na ebulição do isobutano, R-600a, em um tubo composto por 7 mini canais paralelos, cujo diâmetro hidráulico é de 1,47 mm. Os testes em ebulição foram realizados com uma temperatura de saturação de 22 ºC e pressão de saturação de 302 kPa, com velocidade mássica entre 50 e 200 kg/(m²s) e fluxos de calor na seção de testes entre 7 e 40 kW/m². Com os testes realizados verificou-se que o coeficiente de transferência de calor aumenta conforme o incremento do fluxo de calor e velocidade mássica, sendo que esta última tem maior influência para baixos títulos de vapor. O coeficiente de transferência de calor atingiu valores máximos próximos a 3.200 W/(m²K) para a condição de maior vazão e fluxo de calor. A queda de pressão aumentou com o incremento da velocidade mássica e título de vapor em todos os testes, enquanto que o fluxo de calor apresentou influência na queda de pressão apenas nas maiores velocidades mássicas. A perda de pressão por atrito representou até 93,7% da perda total. Também foram analisados os padrões de escoamento, sendo observados os padrões de bolhas isoladas, bolhas alongadas, intermitente e anular, sendo que o padrão de bolhas isoladas foi observado para baixos fluxos e títulos de vapor, e o padrão anular mostrou-se presente para títulos superiores a 0,13. / This work presents an experimental study of heat transfer and drop pressure in flow boiling of the isobutane, R-600a, in a 7 mini channel of 1,47 mm hydraulic diameter. The tests were performed a boiling with a saturation temperature of 22 °C and saturation pressure of 302 kPa, with a mass velocity between 50 and 200 kg/(m²s) and heat fluxes in the test section between 7 and 40 kW/m². In the tests, it was found that the heat transfer coefficient increases with increasing heat flux and mass velocity has more influence at low quality. The heat transfer coefficient achieved values around 3.200 W/(m²K) for the condition of greater flow and heat flux. The drop pressure was increase by increasing the mass velocity and quality in all tests, while the heat flow have any influence on the drop pressure to the greater mass velocity. The friction drop pressure represented to 93.7% of the total loss. As well analyzed the flow patterns, and observing the patterns of isolated bubbles, plugs/slugs, intermittent and annular, and that the pattern of isolated bubbles were observed for low quality and heat flux and the annular pattern was present for quality from 0.13.

Microdosímetros gasosos geralmente empregam uma mistura gasosa equivalente ao tecido humano mole (Tissue-equivalent Gas TEG), composta de um hidrocarboneto, dióxido de carbono e nitrogênio, de forma que o poder de freamento na mistura e no tecido sejam semelhantes. Entretanto, independentemente do hidrocarboneto adotado, dados tanto teóricos como experimentais do primeiro coeficiente de Townsend de ionização (α) nestas misturas são raros, ainda que a primeira TEG, cujo metano é o gás majoritário, tenha sido proposta em 1956 por Rossi e Failla e continue sendo amplamente utilizada. Neste trabalho, dados do parâmetro α em TEGs baseadas no metano (CH4 64,4%, CO2 32,4% e N2 3,2%) e nos isômeros do butano (C4H10 51,4%, CO2 42,3% e N2 6,3%) são apresentados pela primeira vez em geometria planar para a faixa de campo elétrico normalizado pela densidade do gás (E/N) entre 100 290 Td (1 Td = 10-21 V.m2). O método de medidas adotado baseia-se na técnica de Townsend pulsada, onde o primeiro coeficiente de Townsend pode ser determinando comparando a corrente elétrica no regime de avalanche e a corrente de ionização primária gerada pela incidência de um feixe de laser de nitrogênio em um eletrodo metálico (catodo) de uma câmara de geometria planar, sendo o anodo um eletrodo de alta resistividade (ρ=2×1010Ω.m). O aparato experimental, até então operado apenas em pressão atmosférica, foi modificado para também trabalhar em baixa pressão (120 hPa), de modo a aumentar a faixa de E/N investigada. A validação do método e das alterações do sistema de detecção foi realizada utilizando os três gases componentes das TEGs cujos parâmetros de transporte são amplamente estudados: o nitrogênio, o dióxido de carbono e o metano. Observou-se que o parâmetro na TEG com metano assemelha-se com os valores determinados para o metano puro. Na TEG baseada no isobutano, ele é compatível com o primeiro coeficiente de Townsend do dióxido de carbono para campos acima de 170 Td. Já o parâmetro na mistura com n-butano é intermediário entre os valores obtidos para o dióxido de carbono e o nitrogênio. Os resultados experimentais, disponíveis em forma tabular, foram comparados com os simulados utilizando o programa Magboltz 2, evidenciando boa concordância dentro da incerteza experimental. / Gaseous microdosimeters usually employ a Tissue-equivalent Gas (TEG), made of a hydrocarbon, carbon dioxide and nitrogen, in order to make similar the stopping power in this mixture and in the human soft tissue. Notwithstanding, regardless the chosen hydrocarbon, both theoretical and experimental data on the first Townsend ionization coefficient (α) in this mixtures are rare, even though the first TEG, which methane is the most abundant gas, was proposed in 1956 by Rossi and Failla and it has been widely employed since then. In this work, data on the parameter α in TEGs based on methane (CH4 64.4%, CO2 32.4%, and N2 3.2%) and butanes isomers (C4H10 51.4%, CO2 42.3% e N2 6.3%) are presented for the first time in planar geometry in the gas density-normalized electric field (E/N) range between 100 290 Td (1 Td = 10-21 V.m2). The adopted method is based on the Pulsed Townsend Technique, where the first Townsend coefficient can be determined by comparing the electric current in the avalanche mode and the primary ionization current, produced by an nitrogen laser beam incidence in a metallic electrode (cathode) of a parallel plate chamber, which the anode is a resistive electrode (ρ=2×1010Ω.m). The experimental setup, previously operated only at atmospheric pressure, was adapted to work also at low pressure (120 hPa), in order to increase the investigated E/N range. The validation of both method and detection system modifications was made by employing three TEGs components, nitrogen, carbon dioxide and methane, whose transport parameters are extensively studied. The parameter in the methane-based TEG follows the behavior observed in pure methane. In the isobutane-based TEG, it is compatible with the first Townsend coefficient in carbon dioxide for E/N above 170 Td. The parameter in the n-butane-based TEG lies between the obtained values of in carbon dioxide and nitrogen. The experimental results, included in tabular form, agree with those from Magboltz 2 simulations within the experimental uncertainties.

Microdosímetros gasosos geralmente empregam uma mistura gasosa equivalente ao tecido humano mole (Tissue-equivalent Gas TEG), composta de um hidrocarboneto, dióxido de carbono e nitrogênio, de forma que o poder de freamento na mistura e no tecido sejam semelhantes. Entretanto, independentemente do hidrocarboneto adotado, dados tanto teóricos como experimentais do primeiro coeficiente de Townsend de ionização (α) nestas misturas são raros, ainda que a primeira TEG, cujo metano é o gás majoritário, tenha sido proposta em 1956 por Rossi e Failla e continue sendo amplamente utilizada. Neste trabalho, dados do parâmetro α em TEGs baseadas no metano (CH4 64,4%, CO2 32,4% e N2 3,2%) e nos isômeros do butano (C4H10 51,4%, CO2 42,3% e N2 6,3%) são apresentados pela primeira vez em geometria planar para a faixa de campo elétrico normalizado pela densidade do gás (E/N) entre 100 290 Td (1 Td = 10-21 V.m2). O método de medidas adotado baseia-se na técnica de Townsend pulsada, onde o primeiro coeficiente de Townsend pode ser determinando comparando a corrente elétrica no regime de avalanche e a corrente de ionização primária gerada pela incidência de um feixe de laser de nitrogênio em um eletrodo metálico (catodo) de uma câmara de geometria planar, sendo o anodo um eletrodo de alta resistividade (ρ=2×1010Ω.m). O aparato experimental, até então operado apenas em pressão atmosférica, foi modificado para também trabalhar em baixa pressão (120 hPa), de modo a aumentar a faixa de E/N investigada. A validação do método e das alterações do sistema de detecção foi realizada utilizando os três gases componentes das TEGs cujos parâmetros de transporte são amplamente estudados: o nitrogênio, o dióxido de carbono e o metano. Observou-se que o parâmetro na TEG com metano assemelha-se com os valores determinados para o metano puro. Na TEG baseada no isobutano, ele é compatível com o primeiro coeficiente de Townsend do dióxido de carbono para campos acima de 170 Td. Já o parâmetro na mistura com n-butano é intermediário entre os valores obtidos para o dióxido de carbono e o nitrogênio. Os resultados experimentais, disponíveis em forma tabular, foram comparados com os simulados utilizando o programa Magboltz 2, evidenciando boa concordância dentro da incerteza experimental. / Gaseous microdosimeters usually employ a Tissue-equivalent Gas (TEG), made of a hydrocarbon, carbon dioxide and nitrogen, in order to make similar the stopping power in this mixture and in the human soft tissue. Notwithstanding, regardless the chosen hydrocarbon, both theoretical and experimental data on the first Townsend ionization coefficient (α) in this mixtures are rare, even though the first TEG, which methane is the most abundant gas, was proposed in 1956 by Rossi and Failla and it has been widely employed since then. In this work, data on the parameter α in TEGs based on methane (CH4 64.4%, CO2 32.4%, and N2 3.2%) and butanes isomers (C4H10 51.4%, CO2 42.3% e N2 6.3%) are presented for the first time in planar geometry in the gas density-normalized electric field (E/N) range between 100 290 Td (1 Td = 10-21 V.m2). The adopted method is based on the Pulsed Townsend Technique, where the first Townsend coefficient can be determined by comparing the electric current in the avalanche mode and the primary ionization current, produced by an nitrogen laser beam incidence in a metallic electrode (cathode) of a parallel plate chamber, which the anode is a resistive electrode (ρ=2×1010Ω.m). The experimental setup, previously operated only at atmospheric pressure, was adapted to work also at low pressure (120 hPa), in order to increase the investigated E/N range. The validation of both method and detection system modifications was made by employing three TEGs components, nitrogen, carbon dioxide and methane, whose transport parameters are extensively studied. The parameter in the methane-based TEG follows the behavior observed in pure methane. In the isobutane-based TEG, it is compatible with the first Townsend coefficient in carbon dioxide for E/N above 170 Td. The parameter in the n-butane-based TEG lies between the obtained values of in carbon dioxide and nitrogen. The experimental results, included in tabular form, agree with those from Magboltz 2 simulations within the experimental uncertainties.

This thesis is based on eight articles all related to the characteristics of the cooling system and plate evaporator of a household refrigerator. Through these articles, knowledge is provided that can be used to increase the operational efficiency in household refrigeration. Papers A, B and C focus on heat transfer and pressure drop in a commonly used free convection evaporator – the plate evaporator. Applicable correlations are suggested on how to estimate the air side heat transfer, the refrigerant side pressure drop and the refrigerant side heat transfer. Papers D, E and F hold a unique experimental study of the refrigerant charge distribution in the cooling system at transient and steady state conditions. From this cyclic losses are identified and estimated and ways to overcome them are suggested. In paper G the topic “charging and throttling” is investigated in an unparalleled experimental study based on more than 600 data points at different quantities of charge and expansions device capacities. It results in recommendations on how to optimize the capillary tube length and the quantity of refrigerant charge. Finally, Paper H holds a thermographic study of the overall cooling system operating at transient conditions. Overall, a potential to lower the energy use by as much as 25 % was identified in the refrigerator studied. About 10 % was found on the evaporator’s air side. 1-2 % was identified as losses related to the edge effect of the evaporator plate. About 8 % was estimated to be cyclic losses. About 5 % was found in cycle length optimization.  It is believed that most of these findings are of general interest for the whole field of household refrigeration even though the results come from one type of refrigerator. Suggestions of simple means to reduce the losses without increasing the unit price are provided within the thesis / <p>QC 20120411</p>

Submitted by JOSIANE SANTOS DE OLIVEIRA (josianeso) on 2017-12-13T12:59:28Z No. of bitstreams: 1 Priscila Forgiarini da Silva_.pdf: 2273220 bytes, checksum: 4943272627a06de991d941e1c6bfd457 (MD5) / Made available in DSpace on 2017-12-13T12:59:28Z (GMT). No. of bitstreams: 1 Priscila Forgiarini da Silva_.pdf: 2273220 bytes, checksum: 4943272627a06de991d941e1c6bfd457 (MD5) Previous issue date: 2017-11-06 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / PROSUP - Programa de Suporte à Pós-Gradução de Instituições de Ensino Particulares / Este trabalho apresenta um estudo experimental da transferência de calor e queda de pressão na ebulição do isobutano, R600a, e do propano, R290, em um tubo composto por sete mini canais paralelos, cujo diâmetro hidráulico é de 1,47 mm. Os testes em ebulição foram realizados com uma temperatura de saturação de 20 ºC, para ambos os fluidos refrigerantes e pressão de saturação de 300 kPa, para o R600a e de 840 kPa para o R290, com velocidades mássicas entre 35 e 170 kg/(m²s) e fluxos de calor na seção de testes entre 5,3 e 21 kW/m². De acordo com os testes realizados verificou-se que o coeficiente de transferência de calor, para ambos os fluidos refrigerantes, aumenta conforme o incremento do fluxo de calor e velocidade mássica. O coeficiente de transferência de calor atingiu valores entre 1 a 18 kW/(m²K) para o R290 e de 1 a 9 kW/(m²K) para o R600a. A queda de pressão aumentou com o incremento da velocidade mássica e título de vapor em todos os testes, enquanto que o fluxo de calor apresentou influência na queda de pressão apenas nas maiores velocidades mássicas. Observou-se que a queda de pressão por aceleração apresenta a menor parcela, enquanto que, a queda de pressão por atrito apresenta a maior parcela. Na comparação entre o R290 e o R600a, verificou-se que o isobutano apresenta maior queda de pressão. Também foram analisados os padrões de escoamento, sendo observados os padrões de bolhas isoladas, pistonado, agitado, anular ondulado e anular, sendo que o padrão de bolhas isoladas foi observado somente para o R290, e o padrão anular mostrou-se presente para títulos superiores a 0,4. / This work presents an experimental study of heat transfer and pressure drop in boiling of isobutane, R600a, and propane, R290, in a tube composed of seven parallel mini channels, whose hydraulic diameter is 1.47 mm. Boiling tests were performed with a saturation temperature of 20 ºC for both refrigerants and saturation pressure of 300 kPa for R600a and 840 kPa for R290, with mass velocities between 35 and 170 kg/(m²s) and heat flux in the test section between 5.3 and 21 kW/m². According to the tests performed it was verified that the heat transfer coefficient for both refrigerant fluids increases as the heat flux and mass velocity increase. The heat transfer coefficient reached values between 1-18 kW/(m²K) for the R290 and 1-9 kW/(m²K) for the R600a. The pressure drop increased with increasing mass velocity and vapor quality in all tests, while the heat flux showed influence on the pressure drop only at higher mass velocities. It was observed that the pressure drop by acceleration presents the smallest portion, while the friction presents the largest portion. In the comparison between R290 and R600a, it was found that isobutane showed a higher pressure drop. Flow patterns were also analyzed, with isolated bubble, piston, agitated, annular and annular bubble patterns being observed, and the isolated bubble pattern was observed only for R290, and the annular pattern was present for quality higher than 0.4.

A presente tese trata da análise experimental do efeito da geometria da seção transversal do canal e do desempenho de refrigerantes de reduzido GWP (Global Warming Potential) durante a ebulição convectiva em canais de reduzidas dimensões. A tese inclui ainda um estudo extenso e crítico da literatura sobre métodos de previsão da perda de pressão e do coeficiente de transferência de calor, e sobre estudos experimentais em canais não-circulares e de refrigerantes com reduzido GWP na ebulição convectiva em canais de dimensões reduzidas. Resultados para o coeficiente de transferência de calor e perda de pressão durante a ebulição convectiva foram obtidos para canais com geometrias de seção circular, quadrada e triangular para o refrigerante R134a. Nos testes utilizou-se canais com perímetros internos similares obtidos a partir da conformação de um tubo com diâmetro interno igual a 1,1 mm. No caso do canal circular, dados foram também levantados para os HFOs R1234ze(E) e R1234yf e o hidrocarboneto R600a, fluidos com reduzido GWP. Ensaios foram executados para amplas faixas de fluxos de calor e velocidades mássicas, temperaturas de saturação de 31 e 41°C e títulos de vapor entre 0 e 0,95. Aspectos relacionados aos efeitos da geometria e do fluido refrigerante foram minuciosamente investigados através da análise paramétrica dos resultados. Com base na comparação do banco de dados coletado com os métodos de previsão disponíveis na literatura, constatou-se que estes proporcionam previsões satisfatórias apenas para condições experimentais especificas. Portanto, novos métodos de previsão da perda de pressão e do coeficiente de transferência de calor foram desenvolvidos com base nos dados levantados no presente estudo. Os métodos propostos preveem satisfatoriamente o banco de dados do presente estudo e resultados independentes disponíveis na literatura. Adicionalmente, com base nos resultados levantados, verificou-se que dissipadores de calor baseados em multi-microcanais com canais de seção triangular apresentam desempenho superior comparados a dissipadores com canais quadrados e circulares. / The present thesis concerns an experimental study on the effects of cross-sectional geometry and low GWP refrigerants on the thermal-hydraulic performance for convective boiling inside micro-scale channels. Experimental results for heat transfer coefficient and pressure drop gradient during convective boiling were obtained for circular, square and triangular channels for the fluid R134a. The evaluated channels present the same external perimeter and equivalent diameters of 1.1, 0.977 and 0.835 mm, respectively. In the case of the circular geometry, experimental results were also acquired for the HFOs R1234ze(E) and R1234yf and the hydrocarbon R600a (isobutane), which are fluids with low GWP and null ODP. Experiments were performed for a wide range of heat fluxes and mass velocities, saturation temperatures of 31 and 41°C and vapor qualities up to 0.95. The experimental data were carefully analyzed and discussed based on a parametrical analysis focusing on the effect of the cross-sectional geometry and the working fluid. Subsequently, the experimental data were compared against the most quoted predictive methods from literature. In general, it was verified that none of the predictive methods were able to accurately capture the experimental trends of the overall database. So, new predictive methods for the pressure drop and heat transfer coefficient were developed based on the broad database obtained in the present study. The proposed methods provided satisfactory results not only for the experimental database used for its development, but also for independent databases collected in the literature. Additionally, based on the data obtained in the present study and a performance analysis taken into account pressure drop, heat transfer coefficient and the channel packing factor, triangular cross sectional geometry is recommended for heat sinks.

A presente tese trata da análise experimental do efeito da geometria da seção transversal do canal e do desempenho de refrigerantes de reduzido GWP (Global Warming Potential) durante a ebulição convectiva em canais de reduzidas dimensões. A tese inclui ainda um estudo extenso e crítico da literatura sobre métodos de previsão da perda de pressão e do coeficiente de transferência de calor, e sobre estudos experimentais em canais não-circulares e de refrigerantes com reduzido GWP na ebulição convectiva em canais de dimensões reduzidas. Resultados para o coeficiente de transferência de calor e perda de pressão durante a ebulição convectiva foram obtidos para canais com geometrias de seção circular, quadrada e triangular para o refrigerante R134a. Nos testes utilizou-se canais com perímetros internos similares obtidos a partir da conformação de um tubo com diâmetro interno igual a 1,1 mm. No caso do canal circular, dados foram também levantados para os HFOs R1234ze(E) e R1234yf e o hidrocarboneto R600a, fluidos com reduzido GWP. Ensaios foram executados para amplas faixas de fluxos de calor e velocidades mássicas, temperaturas de saturação de 31 e 41°C e títulos de vapor entre 0 e 0,95. Aspectos relacionados aos efeitos da geometria e do fluido refrigerante foram minuciosamente investigados através da análise paramétrica dos resultados. Com base na comparação do banco de dados coletado com os métodos de previsão disponíveis na literatura, constatou-se que estes proporcionam previsões satisfatórias apenas para condições experimentais especificas. Portanto, novos métodos de previsão da perda de pressão e do coeficiente de transferência de calor foram desenvolvidos com base nos dados levantados no presente estudo. Os métodos propostos preveem satisfatoriamente o banco de dados do presente estudo e resultados independentes disponíveis na literatura. Adicionalmente, com base nos resultados levantados, verificou-se que dissipadores de calor baseados em multi-microcanais com canais de seção triangular apresentam desempenho superior comparados a dissipadores com canais quadrados e circulares. / The present thesis concerns an experimental study on the effects of cross-sectional geometry and low GWP refrigerants on the thermal-hydraulic performance for convective boiling inside micro-scale channels. Experimental results for heat transfer coefficient and pressure drop gradient during convective boiling were obtained for circular, square and triangular channels for the fluid R134a. The evaluated channels present the same external perimeter and equivalent diameters of 1.1, 0.977 and 0.835 mm, respectively. In the case of the circular geometry, experimental results were also acquired for the HFOs R1234ze(E) and R1234yf and the hydrocarbon R600a (isobutane), which are fluids with low GWP and null ODP. Experiments were performed for a wide range of heat fluxes and mass velocities, saturation temperatures of 31 and 41°C and vapor qualities up to 0.95. The experimental data were carefully analyzed and discussed based on a parametrical analysis focusing on the effect of the cross-sectional geometry and the working fluid. Subsequently, the experimental data were compared against the most quoted predictive methods from literature. In general, it was verified that none of the predictive methods were able to accurately capture the experimental trends of the overall database. So, new predictive methods for the pressure drop and heat transfer coefficient were developed based on the broad database obtained in the present study. The proposed methods provided satisfactory results not only for the experimental database used for its development, but also for independent databases collected in the literature. Additionally, based on the data obtained in the present study and a performance analysis taken into account pressure drop, heat transfer coefficient and the channel packing factor, triangular cross sectional geometry is recommended for heat sinks.

Les dévellopements récents des faisceaux exotiques, et en particulier avec le dispositif SPIRAL du GANIL, nous ont permis d'initier l'étude, dans l'9He, de l'inversion de couches déjà observée dans les isotone N=7, pour le 11Be et le 10Li. Une partie du travail est consacré à l'étude et au développement d'un détecteur servant aussi de cible (cible active) appelé MAYA, ainsi qu'au développement des outils informatiques nécessaires à son exploitation. l'9He a été étudier par la détermination des propriétés de ses états isobariques analogues (IAS) dans le 9Li. Les caractéristiques de ces états IAS sont obtenues par l'analyse des résonances dans les sections efficaces de diffusion élastique, du système 8He+p entre 2 et 3.9 MeV/n. Une étude rapide des réaction (p,d) et (p,t) est aussi présenté. Cette étude a permis, par la comparaison des résultats expérimentaux et des calculs de diffusion potentielle avec résonances, une assignation de spin et parité pour deux états de l'9He.

In der vorliegenden Arbeit wird erstmals das QQDS-Magnetspektrometer für die höchstauflösende Ionenstrahlanalytik leichter Elemente am Helmholtz-Zentrum Dresden-Rossendorf umfassend vorgestellt. Zusätzlich werden sowohl alle auf die Analytik Einfluss nehmenden Parameter untersucht als auch Methoden und Modelle vorgestellt, wie deren Einfluss vermieden oder rechnerisch kompensiert werden kann. Die Schwerpunkte dieser Arbeit gliedern sich in fünf Bereiche. Der Erste ist der Aufbau und die Inbetriebnahme des QQDS-Magnetspektrometers, der zugehörige Streukammer mit allen Peripheriegeräten und des eigens für die höchstauflösende elastische Rückstoßanalyse entwickelten Detektors. Sowohl das umgebaute Spektrometer als auch der im Rahmen dieser Arbeit gebaute Detektor wurden speziell an experimentelle Bedingungen für die höchstauflösende Ionenstrahlanalytik leichter Elemente angepasst und erstmalig auf einen routinemäßigen Einsatz hin getestet. Der Detektor besteht aus zwei Komponenten. Zum einen befindet sich am hinteren Ende des Detektors eine Bragg-Ionisationskammer, die zur Teilchenidentifikation genutzt wird. Zum anderen dient ein Proportionalzähler, der eine Hochwiderstandsanode besitzt und direkt hinter dem Eintrittsfenster montiert ist, zur Teilchenpositionsbestimmung im Detektor. Die folgenden zwei Schwerpunkte beinhalten grundlegende Untersuchungen zur Ionen-Festkörper-Wechselwirkung. Durch die Verwendung eines Magnetspektrometers ist die Messung der Ladungszustandsverteilung der herausgestreuten Teilchen direkt nach einem binären Stoß sowohl möglich als auch für die Analyse notwendig. Aus diesem Grund werden zum einen die Ladungszustände gemessen und zum anderen mit existierenden Modellen verglichen. Außerdem wird ein eigens entwickeltes Modell vorgestellt und erstmals im Rahmen dieser Arbeit angewendet, welches den ladungszustandsabhängigen Energieverlust bei der Tiefenprofilierung berücksichtigt. Es wird gezeigt, dass ohne die Anwendung dieses Modells die Tiefenprofile nicht mit den quantitativen Messungen mittels konventioneller Ionenstrahlanalytikmethoden und mit der Dickenmessung mittels Transmissionselektronenmikroskopie übereinstimmen, und damit falsche Werte liefern würden. Der zweite für die Thematik wesentliche Aspekt der Ionen-Festkörper-Wechselwirkung, sind die Probenschäden und -modifikationen, die während einer Schwerionen-bestrahlung auftreten. Dabei wird gezeigt, dass bei den hier verwendeten Energien sowohl elektronisches Sputtern als auch elektronisch verursachtes Grenzflächendurchmischen eintreten. Das elektronische Sputtern kann durch geeignete Strahlparameter für die meisten Proben ausreichend minimiert werden. Dagegen ist der Einfluss der Grenzflächendurchmischung meist signifikant, so dass dieser analysiert und in der Auswertung berücksichtigt werden muss. Schlussfolgernd aus diesen Untersuchungen ergibt sich für die höchstauflösende Ionenstrahlanalytik leichter Elemente am Rossendorfer 5-MV Tandembeschleuniger, dass die geeignetsten Primärionen Chlor mit einer Energie von 20 MeV sind. In Einzelfällen, wie zum Beispiel der Analyse von Bor, muss die Energie jedoch auf 6,5 MeV reduziert werden, um das elektronische Sputtern bei der notwendigen Fluenz unterhalb der Nachweisgrenze zu halten. Der vierte Schwerpunkt ist die Untersuchung von sowohl qualitativen als auch quantitativen Einflüssen bestimmter Probeneigenschaften, wie beispielsweise Oberflächenrauheit, auf die Form des gemessenen Energiespektrums beziehungsweise auf das analysierte Tiefenprofil. Die Kenntnis der Rauheit einer Probe an der Oberfläche und an den Grenzflächen ist für die Analytik unabdingbar. Als Resultat der genannten Betrachtungen werden die Einflüsse von Probeneigenschaften und Ionen-Festkörper-Wechselwirkungen auf die Energie- beziehungsweise Tiefenauflösung des Gesamtsystems beschrieben, berechnet und mit der konventionellen Ionenstrahlanalytik verglichen. Die Möglichkeiten der höchstauflösenden Ionenstrahlanalytik werden zudem mit den von anderen Gruppen veröffentlichten Komplementärmethoden gegenübergestellt. Der fünfte und letzte Schwerpunkt ist die Analytik leichter Elemente in ultradünnen Schichten unter Berücksichtigung aller in dieser Arbeit vorgestellten Modelle, wie die Reduzierung des Einflusses von Strahlschäden oder die Quantifizierung der Elemente im dynamischen Ladungszustandsnichtgleichgewicht. Es wird die Tiefenprofilierung von Mehrschichtsystemen, bestehend aus SiO2-Si3N4Ox-SiO2 auf Silizium, von Ultra-Shallow-Junction Bor-Implantationsprofilen und von ultradünnen Oxidschichten, wie zum Beispiel High-k-Materialien, demonstriert. / In this thesis the QQDS magnetic spectrometer that is used for high resolution ion beam analysis (IBA) of light elements at the Helmholtz-Zentrum Dresden-Rossendorf is presented for the first time. In addition all parameters are investigated that influence the analysis. Methods and models are presented with which the effects can be minimised or calculated. There are five focal points of this thesis. The first point is the construction and commissioning of the QQDS magnetic spectrometer, the corresponding scattering chamber with all the peripherals and the detector, which is specially developed for high resolution elastic recoil detection. Both the reconstructed spectrometer and the detector were adapted to the specific experimental conditions needed for high-resolution Ion beam analysis of light elements and tested for routine practice. The detector consists of two compo-nents. At the back end of the detector a Bragg ionization chamber is mounted, which is used for the particle identification. At the front end, directly behind the entrance window a proportional counter is mounted. This proportional counter includes a high-resistance anode. Thus, the position of the particles is determined in the detector. The following two points concern fundamental studies of ion-solid interaction. By using a magnetic spectrometer the charge state distribution of the particles scattered from the sample after a binary collision is both possible and necessary for the analysis. For this reason the charge states are measured and compared with existing models. In addition, a model is developed that takes into account the charge state dependent energy loss. It is shown that without the application of this model the depth profiles do not correspond with the quantitative measurements by conventional IBA methods and with the thickness obtained by transmission electron microscopy. The second fundamental ion-solid interaction is the damage and the modification of the sample that occurs during heavy ion irradiation. It is shown that the used energies occur both electronic sputtering and electronically induced interface mixing. Electronic sputtering is minimised by using optimised beam parameters. For most samples the effect is below the detection limit for a fluence sufficient for the analysis. However, the influence of interface mixing is so strong that it has to be included in the analysis of the layers of the depth profiles. It is concluded from these studies that at the Rossendorf 5 MV tandem accelerator chlorine ions with an energy of 20 MeV deliver the best results. In some cases, such as the analysis of boron, the energy must be reduced to 6.5 MeV in order to retain the electronic sputtering below the detection limit. The fourth focus is the study of the influence of specific sample properties, such as surface roughness, on the shape of a measured energy spectra and respectively on the analysed depth profile. It is shown that knowledge of the roughness of a sample at the surface and at the interfaces for the analysis is needed. In addition, the contribution parameters limiting the depth resolution are calculated and compared with the conventional ion beam analysis. Finally, a comparison is made between the high-resolution ion beam analysis and complementary methods published by other research groups. The fifth and last focus is the analysis of light elements in ultra thin layers. All models presented in this thesis to reduce the influence of beam damage are taken into account. The dynamic non-equilibrium charge state is also included for the quantification of elements. Depth profiling of multilayer systems is demonstrated for systems consisting of SiO2-Si3N4Ox-SiO2 on silicon, boron implantation profiles for ultra shallow junctions and ultra thin oxide layers, such as used as high-k materials.