93726 research outputs found
Sort by
A novel multicriteria methodology to assess the renovation of social buildings
No full textThe current Energy Performance Certificates of buildings are mostly based on the expected energy consumption. Recently, some certification schemes are starting to incorporate Indoor Air Quality aspects, but it is still not usual. Another topic of interest, such as energy poverty, is not considered in current certification schemes either. This paper presents a novel multicriteria methodology to assess the renovation of social housing buildings: the ARCAS methodology. The novelty lies in its transversal approach, which considers three axes: Energy Efficiency, Energy Poverty, and Indoor Environment Quality. The indicators set for assessment are calculated based on actual on-site measurements. The methodology has been tested in six demonstrator buildings selected throughout the SUDOE territory (southwestern Europe). The results show that the ARCAS methodology allows a comprehensive assessment of the current situation of the building to be achieved and helps to propose the renovation measures in a specific direction. Advice and guidance for professionals who want to test the methodology are also given. Although the methodology has been developed for its application in social housing, according to the EU recommendations, it can be applied to any type of housing.EHU - Euskal Herriko Unibertsitatea(undefined
New generation antigen delivery system for Candida albicans recombinant proteins as a vaccination strategy against systemic fungal infections
No full textPrograma doutoral em Molecular and Environmental Biology (ramo do conhecimento em Cellular Biology and Health)Over the past five decades, a rise in numbers of immunocompromised individuals has led to a significant increase in invasive fungal infections, including those caused by Candida albicans. This has created a pressing need for innovative approaches to combat these fungal pathogens. While several immunotherapeutic strategies and vaccines are in development, none have reached clinical use yet. Previous work emphasised the potential of liposomal formulations containing C. albicans cell wall surface proteins (ADS1) as a vaccine candidate, highlighting chitinase 3 (Cht3) as a promising antigen. A monovalent formulation incorporating DODAB:MO liposomes with purified, recombinant, P. pastoris-produced, Cht3 was prepared. Cht3 displayed stability and activity across a wide pH range (pH 3-9) and at elevated temperatures (up to 50-60 ºC), making it suitable for liposomal inclusion. Cht3 hydrolysed chitin and chito-oligomers into smaller chito-oligomers, with chitobiose being the principal end-product but with chitotriose and the monomer N-acetylglucosamide being also produced. Cht3 liposomes (DMC) showed enhanced antigen delivery and cellular internalisation compared to ADS1, eliciting a Th1/Th17 immune response and generating substantial levels of opsonising antigen-specific IgG and IgG1 antibodies. A divalent formulation (DMCS) including Cht3 liposomes and a new antigen, the C. albicans protease Sap2, was also prepared. DMCS maintained liposome characteristics and induced cytokine production similar to DMC but surprisingly provided greater protection against systemic infection, likely due to a stronger Th1/Th2 immune response, with a higher IFN-γ/IL-4 ratio, trained immunity, and higher antigen-specific IgG response. The role of Cht3 in C. albicans virulence was explored using knockout mutants, revealing a decrease in virulence with CHT3 deletion. In vitro studies showed no significant differences in physiological characteristics, but the release of small chito oligomers by Cht3 with potential immunomodulatory effects could play a role in virulence. The results obtained in this thesis signify a significant advancement in the field by introducing novel specific antigen formulations and confirming DODAB:MO as an effective delivery system and adjuvant. DMCS emerges as a promising candidate, demonstrating enhanced protection and immune response. Additionally, it highlights chitinase activity as a virulence factor in C. albicans infections and the complex interaction between chitinases and chito-oligomers. Further studies will delve into these interactions and their implications for fungal infection therapies.Nas últimas cinco décadas, o aumento no número de indivíduos imunocomprometidos levou a um aumento significativo de infeções fúngicas sistémicas, incluindo as causadas por Candida albicans. Este aumento criou uma necessidade urgente para abordagens inovadoras para combater estes fungos patogénicos. Embora várias estratégias imunoterapêuticas e vacinas estejam em desenvolvimento, nenhuma alcançou o uso na clínica. Trabalhos anteriores enfatizaram o potencial de formulações lipossomais contendo proteínas da superfície da parede de C. albicans (ADS1) como candidatas a vacinas, destacando a quitinase 3 (Cht3) como um antigénio promissor. Uma formulação monovalente foi produzida, incorporando lipossomas DODAB:MO com Cht3 recombinante purificada produzida em P. pastoris. Cht3 apresentou estabilidade e atividade numa ampla gama de pH (pH 3-9) e em temperaturas elevadas (até 50-60 ºC), tornando-a adequada para incorporação nos lipossomas. Cht3 hidrolisou quitina e quito-oligómeros em quito-oligómeros menores, sendo a quitobiose o principal produto final, mas com a quitotriose e o monómero N-acetilglucosamida a também serem produzidos. Os lipossomas com Cht3 (DMC) mostraram melhor entrega de antigénio e internalização celular em comparação com ADS1, induzindo uma resposta imune Th1/Th17 e gerando níveis substanciais de anticorpos IgG e IgG1 específicos e opsonizantes. Uma formulação divalente (DMCS) incluindo lipossomas com Cht3 e um novo antigénio, a protease Sap2 de C. albicans, também foi preparada. DMCS manteve as características dos lipossomas e induziu uma produção de citocinas semelhante a DMC, mas surpreendentemente induziu maior proteção contra uma infeção sistémica, provavelmente devido a uma resposta imune Th1/Th2 mais forte, com um rácio IFN-γ/IL-4 mais alto, imunidade treinada e maiores níveis de IgG específicos. O papel da Cht3 na virulência de C. albicans foi explorado usando mutantes knockout, revelando uma diminuição na virulência com a deleção de CHT3. Estudos in vitro não mostraram diferenças significativas nas características fisiológicas, mas a libertação de pequenos quito-oligómeros pela Cht3, com potenciais efeitos imunomoduladores, poderia desempenhar um papel na virulência. Os resultados obtidos nesta tese significam um avanço significativo nesta área ao introduzir novas formulações de antigénios específicos e a confirmar o DODAB:MO como um sistema de entrega e adjuvante eficaz. DMCS surge como um candidato promissor, demonstrando maior proteção e resposta imunológica. Além disso, destaca a atividade da quitinase como fator de virulência em infeções por C. albicans e a complexa interação entre quitinases e quito-oligómeros. Mais estudos irão aprofundar essas interações e suas implicações para terapias de infeções fúngicas.Agradeço à Fundação para a Ciência e Tecnologia pelo financiamento através das bolsas SFRH/BD/133513/2017 e COVID/BD/152169/2021
Investigar o processamento auditivo em esquizofrenia: uma abordagem de deep learning com base em sinais de EEG
No full textTese de doutoramento em Psicologia BásicaO diagnóstico da esquizofrenia apresenta desafios devido à sua natureza complexa e heterogénea,
ao relato subjetivo de sintomas, à possível sobreposição com outras condições psiquiátricas, e ao início
dos sintomas durante o período desenvolvimental de elevada vulnerabilidade para diferentes tipos de
psicopatologia (final da adolescência). A compreensão da percepção auditiva na esquizofrenia é crucial
para obter pistas sobre a fisiopatologia, uma vez que alucinações auditivas são sintomas proeminentes.
A electroencefalografia (EEG) fornece uma medida direta da atividade elétrica do cérebro, permitindo a
investigação de alterações neurais associadas à percepção auditiva em indivíduos com esquizofrenia.
A utilização de machine learning, especialmente métodos de deep learning, com dados de EEG tem o
potencial de aprender padrões neurais e relações complexas, predição e identificação de biomarcadores
em esquizofrenia.
Neste estudo, analisamos sinais de EEG de indivíduos saudáveis e de pacientes com esquizofrenia
conduzindo uma análise combinada de potenciais evocados (ERP), trial-by-trial e tempo-frequência para
investigar alterações no processamento auditivo. Com um tamanho de amostra maior do que em es tudos prévios (n>60), observamos reduções consistentes no componente ERP N100 em esquizofrenia,
especialmente em elétrodos fronto-centrais, em consonância com a literatura existente. A análise trial-by trial mostrou respostas N100 e P200 consistentes entre ensaios, mas maior variabilidade nos sinais de
pacientes com esquizofrenia. Diferenças significativas nas janelas temporais N100 e P200 enfatizaram
ainda mais a robustez desses componentes. A análise de tempo-frequência revelou padrões distintos de
sincronia da atividade neural nas oscilações theta e alfa entre os grupos de indivíduos saudáveis e com
esquizofrenia, contribuindo para diferentes assinaturas do sinal EEG em resposta a sons simples.
Tirando partido do sucesso dos métodos de deep learning e das diferenças robustas observadas nos
sinais EEG, desenvolvemos um modelo de classificação binária baseado em single-trials usando uma
arquitetura baseada em redes neuronais convolucionais (CNN). O modelo é proposto para a classificação
de segmentos de EEG, registados durante uma tarefa auditiva passiva usando a repetição de sons simples.
Um estudo ablativo permitiu analisar o efeito do número de elétrodos utilizados, bem como a lateralidade
e distribuição da atividade elétrica no escalpe. Os resultados mostram que o modelo proposto é capaz
de classificar indivíduos com esquizofrenia e sujeitos saudáveis com uma precisão média de 80% usando
apenas 5 elétrodos (Fz, FCz, Cz, CPz e Pz).
Este trabalho destaca a relevância dos métodos de deep learning na exploração de possíveis biomar cadores eletrofisiológicos para a esquizofrenia, bem como as potenciais aplicações clínicas dos sistemas
de classificação. Além disso, demonstra o potencial transformador da colaboração interdisciplinar entre
Psicologia, Psiquiatria e Engenharia Biomédica na revolução do diagnóstico, tratamento e prevenção de
doenças mentais como a esquizofrenia.The diagnosis of schizophrenia poses challenges due to its complex and heterogeneous nature, subjective symptom reporting, possible overlap with other psychiatric conditions, and typical symptom onset
in a developmental stage characterized by high vulnerability to distinct types of psychopathology (end of
adolescence). Understanding auditory perception in schizophrenia is crucial to get clues about the disorder’s pathophysiology, as auditory hallucinations are prominent symptoms. Electroencephalography (EEG)
provides a direct measure of brain electrical activity, allowing investigation of neural abnormalities associated with auditory perception in individuals with schizophrenia. The use of machine learning, especially
deep learning methods, with EEG data in schizophrenia research has the potential to unveil intricate neural
patterns and relationships, enabling more accurate diagnosis, prediction, and biomarker identification.
In this study, we analyzed EEG signals of healthy subjects and schizophrenia patients using combined
event-related potentials (ERP), trial-by-trial, and time-frequency analysis to investigate auditory processing
alterations. With a sample size larger than in previous studies (n>60), we observed consistent reductions in
the N100 ERP component in schizophrenia, particularly over frontal-central electrode sites, aligning with
existing literature. Trial-by-trial analysis showed consistent N100 and P200 responses across trials but
greater variability in schizophrenia signals. Significant differences in N100 and P200 temporal windows
further emphasized the robustness of these event-related components. Time-frequency analysis revealed
distinct neural activity synchrony patterns in theta and alpha oscillations between healthy and schizophrenia
groups, contributing to different EEG signatures in response to simple sounds.
Building upon the remarkable achievements of deep learning-based approaches and the observed
robust EEG differences, a single-trial-based binary classification model was developed using a convolutional
neural network-based architecture. The model is proposed to perform the classification based on multichannel auditory-related EEG single-trials recorded during a passive listening task of simple sounds. An
ablative study allowed us to analyze the effect of the number of electrodes used and the laterality and
distribution of the electrical activity over the scalp. Results show that the proposed model is able to classify
schizophrenia and healthy subjects with an average accuracy of 80% using only five midline channels (Fz,
FCz, Cz, CPz, and Pz).
This work highlights the relevance of deep learning methods in exploring potential electrophysiological biomarkers for schizophrenia and the potential clinical applications of such classification systems.
Furthermore, it showcases the transformative potential of interdisciplinary collaboration between Psychology, Psychiatry, and Biomedical Engineering in revolutionizing the diagnosis, treatment, and prevention of
psychiatric disorders such as schizophrenia.This work was supported by the Grant SFRH/BD/111083/2015, funded by Fundação para a Ciência
e Tecnologia (FCT) under the Programa Operacional Capital Humano (PO CH) co-funded by Portugal 2020
and European Social Fund
Relações para-sociais no âmbito dos entusiastas de celebridades tradicionais, ficcionais e micro: explorar os seus efeitos ao nível do comportamento do consumidor e emoções
No full textTese de doutoramento em Business AdministrationWith the proliferation of social media, there is an increasing presence of media figures and celebrities online, ranging from 1) real human celebrities, termed traditional celebrities (e.g. Adele); 2) microcelebrities or influencers (e.g. someone with a large follower base on social media platforms) to 3) fictional celebrities (e.g. Harry Potter). This increasingly drives the development of PSRs. Yet while scholarship on PSRs continually expanded, its effectiveness on consumer behaviors in new media contexts and among different celebrity types has received little attention. To address this gap three complementary studies are conducted. The first two studies are quantitative, focusing on the impact of PSRs with micro and traditional celebrities on online source credibility, brand credibility, and purchase intention, specifically on Instagram. The third study is qualitative, focusing on the hedonic aspects of PSR with fictional celebrities, specifically in virtual environments. This research contributes to theory by providing a comprehensive understanding of PSRs with three celebrity types. Moreover, it validates Djafarova and Trofimenko’s (2019) online source credibility model and highlights the limitations of using the traditional Ohanian (1990) source credibility model in online contexts. Results show increasing similarities between micro-celebrities and traditional celebrities affecting the development of PSRs, perceived credibility, and purchase intentions. Furthermore, findings show how hedonic aspects of PSR with fictional celebrities in virtual contexts affect consumers’ well-being. These findings offer valuable marketing insights for practitioners and highlight the importance of ethical considerations.A proliferação do uso de redes sociais no domínio do social media aumentou a presença de
personagens e celebridades online, variando entre 1) celebridades humanas (pessoas reais) (e.g.,
Adele); 2) micro-celebridades ou influencers (e.g., alguém que dispõe de uma base de seguidores
relevante nas diversas plataformas) e 3) celebridades ficcionais (e.g., Harry Potter). Estes diferentes
tipos de celebridades têm um papel relevante no desenvolvimento de relacionamentos de natureza
parassocial (RP). Todavia, apesar do conhecimento académico desta realidade estar em expansão, a
sua compreensão no domínio do comportamento do consumidor em novos contextos de media social,
designadamente os diferentes tipos de celebridades ainda não têm sido alvo de grande estudo e
interesse. Com a finalidade de preencher este gap foram conduzidos três estudos distintos mas
complementares. Os primeiros dois estudos de natureza mais quantitativa, focam-se no impacto dos
relacionamentos parassociais. Mais especificamente, numa primeira etapa nas micro-celebridades e,
numa segunda etapa, na comparação entre celebridades do tipo micro vs tradicionais. Para o efeito,
desenvolveu-se um modelo que incorporou três grandes conceitos, designadamente: a credibilidade da
fonte, a credibilidade da marca e a intenção de compra, mais especificamente no âmbito do Instagram.
O terceiro estudo, por seu turno, tem uma natureza qualitativa, e centra-se numa vertente mais
hedónica dos relacionamentos parassociais, em particular no que se refere às celebridades ficcionais
em ambientes virtuais. Neste sentido, esta pesquisa contribui para o conhecimento (ou teoria)
proporcionando uma melhor compreensão da natureza dos relacionamentos parassociais tendo em
conta três tipos de celebridades. Para além disso, o estudo valida o modelo de credibilidade da fonte
proposto por Djafarova e Tronfimenko (2019) sublinhando algumas limitações do modelo tradicional
proposto por Ohanian (1990) sobre a temática relativa à credibilidade da fonte em contextos online. Os
resultados evidenciam inúmeras semelhanças entre o grupo das micro-celebridades e das celebridades
tradicionais, ambos afetando o desenvolvimento de relacionamentos parassociais, de credibilidade
percebida e inteção de compra. Para além disto, os resultados revelaram como é que os aspetos
hedónicos dos relacionamentos parassociais com as celebridades ficcionais em ambientes virtuais
influenciam o bem estar dos consumidores. Estes resultados afiguram-se particularmente relevantes
para os académicos e marketeers, reforçando a importância da vertente ética
Dispositivos viscoelásticos para reduzir a vulnerabilidade sísmica de edifícios de alvenaria existentes
No full textTese de doutoramento em Civil EngineeringA preservação de edifícios históricos e a revitalização de centros históricos são cruciais para o
desenvolvimento de cidades sustentáveis, exigindo o cumprimento de normas modernas de conforto e
integridade estrutural. Os estudos pós-sismo demonstram que os colapsos para fora-do-plano são os
mecanismos de falha mais comuns em edifícios de alvenaria não reforçada, representando um risco
significativo para a vida humana. Esta tese tem por principal objetivo desenvolver um novo dispositivo
passivo de dissipação de energia para reduzir a vulnerabilidade sísmica destes edifícios, nomeadamente
melhorar a resposta das fachadas para fora do seu plano.
O novo dispositivo, conhecido como dispositivo viscoelástico geometricamente não linear (GNL VED), foi desenvolvido e testado através de simulações numéricas utilizando métodos explícitos. O GNL VED tira partido das propriedades de amortecimento dos materiais viscoelásticos (elastómeros) ao
amplificar os deslocamentos relativos entre os elementos estruturais durante um sismo. Os
deslocamentos induzem esforços de corte nas camadas do material, dissipando assim energia. O
dispositivo foi analisado num sistema de um grau de liberdade e, posteriormente, aplicado em modelos
parciais e completos de estruturas baseados no método de elementos finitos. O estudo incluiu dois casos
de estudo: edifícios do vice-reinado do México e edifícios Gaioleiros de Portugal. Estas tipologias foram
selecionadas devido às suas diferentes características arquitetónicas e estruturais, permitindo uma
análise aprofundada do desempenho do dispositivo. O GNL-VED mostrou-se eficaz no controlo dos
deslocamentos para fora do plano e na redução do dano, sobretudo nos edifícios do vice-reinado do
México, onde o mecanismo de colapso para fora do plano é dominante. No edifício Gaioleiro, com um
comportamento dinâmico mais complexo, a eficácia do GNL-VED é limitada para intensidades sísmicas
baixas, mas com uma melhoria para amplitudes mais elevadas, sugerindo que pode ser necessária uma
otimização dos dispositivos para edifícios com diferentes tipos de modos de falha.
Este estudo demonstrou que o GNL-VED é uma solução promissora para mitigar os danos
causados pela ação sísmica em edifícios existentes de alvenaria. No entanto, são necessárias mais
estudos para aperfeiçoar o dispositivo e otimizar a sua aplicação em edifícios com comportamentos
dinâmicos complexos.The preservation of historical buildings and revitalisation of historic city centres are crucial for
developing sustainable cities, while requiring to meet modern standards of comfort and structural integrity
without compromising their historical value. Post-earthquake surveys have shown that out-of-plane
collapse mechanisms are the most common failure modes in unreinforced masonry (URM) buildings,
representing a significant risk to human life. This thesis focuses on developing a new passive energy
dissipation (PED) device to reduce the seismic vulnerability of these buildings, namely, to improve the
out-of-plane response of façades.
The new PED device, denoted geometrically nonlinear viscoelastic device (GNL-VED), was
developed and tested through extensive numerical simulations using explicit solvers. The GNL-VED
exploits the damping properties of viscoelastic materials (elastomers) by amplifying relative displacements
between structural elements during an earthquake. The amplified displacements induce shear to
elastomeric layers, thereby dissipating energy. The device was initially analysed in a single-degree-of freedom system and then applied to partial and complete models of structures based on finite element
method. The study included two case studies: Viceregal dwellings in Mexico and Gaioleiro buildings in
Portugal. These typologies were selected due to their different architectural and structural characteristics,
allowing for an in-depth analysis of the device’s performance under various conditions. The GNL-VED
proved to be effective in controlling the out-of-plane displacements and minimising damage, especially in
the Viceregal dwelling, where the out-of-plane collapse mechanism is dominant. In this case study, the
device performed better compared to traditional strengthening techniques such as steel tie rods and rigid
beam-to-wall connections. In the Gaioleiro building, which exhibits more complex dynamic behaviour, the
GNL-VED's effectiveness was limited at lower seismic intensities but with an improvement at higher
amplitudes, suggesting that a device optimisation can be needed for buildings with different types of
failure modes.
This study demonstrates that the GNL-VED is a promising solution for mitigating the damage
caused by seismic actions in existing URM buildings. However, additional studies are recommended to
refine the device and optimise its application in buildings with complex dynamic behaviours.I acknowledge the Consejo Nacional de Humanidades, Ciencias y Tecnologías (Mexico), within the scope of the call CONACyT-Alianza FiiDEM 2018, for their funding, which made this work possible (2018-000013-01EXTF-00137). This dissertation was developed within the VISCO-D research project, 'Innovative viscoelastic devices for the reduction of the seismic vulnerability of existing masonry buildings' (EXPL/ECI-EGC/0940/2021), funded by the Fundação para a Ciência e a Tecnologia (Portugal)
Effect of hyaluronan molecular weight on the stability and biofunctionality of microfibers assembled by interfacial polyelectrolyte complexation
No full textNervous system disorders are characterized by a progressive loss of function and structure of neurons that ultimately leads to a decline in cognitive and motor functions. In this study, we used interfacial polyelectrolyte complexation (IPC) to produce fibers for neural tissue regeneration. IPC is a processing method that allows spinning of sensitive biopolymers. The rate of spinning and the properties of the used biopolymers (charge and molecular weight) influence different characteristics of the fibers such as size and stability, among others. We used two major components of the neuronal stem cell niche, the polycationic collagen (Col) and the polyanionic hyaluronic acid (HA), to obtain bioactive fibers. We tested HA with different molecular weights and found that HA with medium and high molecular weights (350 and 1200 kDa, respectively) enabled drawing of microfibers with a homogeneous distribution of Col and HA, whereas low-molecular-weight HA (40 kDa) did not allow spinning. The obtained microfibers showed high swelling ability in a physiological buffer: their diameters increased more than 5-fold from their dry state. At these conditions, the tensile storage moduli of the fibers were similar to nervous tissues. Collagenase and hyaluronidase did not change the morphology of the fibers for up to 3 days but reduced their moduli 2- to 3-fold. Assays with PC12 neuronal-like cells showed that IPC microfibers support cell adhesion and viability regardless of the molecular weight of the used HA.This work was supported by the Fundação para a Ciência e Tecnologia (project OncoNeoTreat, grant number PTDC/CTM-REF/0022/2020), cofinanced by FCT─OE component (DOI: 10.54499/PTDC/CTM-REF/0022/2020), Funding “Base” (DOI: 10.54499/UIDB/50026/2020), Funding “Programático” (DOI: 10.54499/UIDP/50026/2020), and Funding “Complementar─LA” (DOI: 10.54499/LA/P/0050/2020). R.R.C. acknowledges FCT for support through grant 2022.00764.CEECIND/CP1718/CT0020 (DOI: 10.54499/2022.00764.CEECIND/CP1718/CT0020). D.S.C. acknowledges FCT for support through grant DL57/2016/CP1377/CT0092 (DOI: 10.54499/DL57/2016/CP1377/CT0092). C.C. acknowledges FCT for support through grant SFRH/BD/143209/2019. The authors would like to thank the contributions to this research from the project “TERM RES Hub–Scientific Infrastructure for Tissue Engineering and Regenerative Medicine,” reference PINFRA/22190/2016 (Norte-01-0145-FEDER-022190), funded by the Portuguese National Science Foundation (FCT) in cooperation with the Northern Portugal Regional Coordination and Development Commission (CCDR-N), for providing relevant lab facilities, state-of-the art equipment, and highly qualified human resources
Climate activist groups’ discourses on science and knowledge: merging rhetorical strategies with political visions
No full textClimate activist groups such as Fridays for Future and Extinction Rebellion have recurrently demanded for science to be taken seriously in political decision-making. Scholars point out that science has been “consecrated” as a source of guidance for a climate-dependent future, with a focus on techno-scientific reasoning. This can potentially suppress values of inclusivity and pluralism, and power dynamics inherent to climate politics from public discourse. This study examines the discursive uses of science by three Portuguese youth-led climate groups—School Strike for Climate Portugal, Climáximo, and LIDERA. We found that they have invoked science in three distinctive communicative moves: constructing science into a strategic ally; turning climate science into a political and ethical referee; and debating and pluralizing science. We argue that while climate groups can bolster mainstream science, some also foster critical perspectives and try to push science away from the dominant symbolic territory of “universalism” and “neutrality”.This work received funding from the Fundação para a Ciência e a Tecnologia for the research project JUSTFUTURES Climate Futures and Just Transformations: Young People’s Narratives and Political Imaginaries
via grant PTDC/COMOUT/7669/2020
Implementation of quality tools to improve the production process of injected parts with high gloss finish
No full textDissertação de mestrado integrado em Engenharia de PolímerosA indústria automóvel é marcada pela elevada competitividade e exigência na qualidade dos produtos. A aplicação de métodos de melhoria contínua torna-se um trunfo imprescindível para que as empresas produtoras de peças para esta indústria alcancem a liderança no seu setor. Estas empresas procuram métodos que lhes permitam reduzir custos, melhorar a qualidade dos produtos e a satisfação do cliente. As empresas especializadas na produção de peças decorativas do interior do automóvel permanecem sempre a par das últimas tendências do mercado, sendo que a estética dos componentes assume um papel preponderante na valorização dos mesmos. As técnicas de acabamento superficial são um forte aliado deste setor, nomeadamente a pintura.
O presente estudo centra-se na análise do processo de produção de componentes decorativos com acabamento de alto brilho existente na Fehst Componentes, Lda. Estes componentes, após serem injetados, são pintados com tintas monocura por radiação ultravioleta, de modo a valorizar o seu aspeto estético. No fim do processo de pintura ocorre uma inspeção visual à qualidade superficial dos componentes, onde peças com defeitos, como inclusões, covas, riscos, casca de laranja, entre outros, são rejeitadas. A elevada percentagem de rejeição que ocorre em alguns componentes conduz à aplicação de metodologias e ferramentas da qualidade que visem a melhoria do processo produtivo. Através da aplicação da metodologia DMAIC (Definir, Medir, Analisar, Melhorar e Controlar), e de ferramentas da qualidade como brainstorming, diagramas de Ishikawa, 5Why’s, 5W2H, entre outros, pretende-se implementar medidas que reduzam e estabilizem o valor da percentagem de peças não conformes dos componentes decorativos com acabamento de alto brilho.
O estudo realizado permitiu concluir que o defeito inclusões é o defeito que apresenta maior incidência, tendo como origem operações secundárias na cadeia produtiva dos componentes. A alteração de alguns procedimentos e operações, permitiu melhorar significativamente a percentagem de defeitos ocorridos. A implementação da metodologia DMAIC foi uma mais valia na melhoria de todo o processo produtivo e na obtenção das metas definidas anualmente pela empresa, relativamente à percentagem de rejeição de peças com acabamento de alto brilho.The automotive industry is marked by competitiveness and demand for product quality. The application of continuous improvement methods becomes an essential asset for companies that produce parts for this industry to reach the leadership in their sector. These companies are looking for methods that will allow them to reduce costs, improve product quality and customer satisfaction. Companies specializing in the production of decorative parts for the interior of the car always keep up with the latest market trends, being that, the aesthetics of the components play a major role in their improvement. Surface finishing techniques are a strong ally of this sector, namely painting.
The present study focuses on the analysis of the production process of decorative components with high gloss finish existing at Fehst Componentes, Lda. These components, after being injected, are painted with monocure paints by ultraviolet radiation, in order to improve their aesthetic appearance. At the end of the painting process, there is a visual inspection of the surface quality of the components, where parts with defects, such as inclusions, pits, scratches, orange peel, among others, are rejected. The high rejection rate that occurs in some components has led to the application of quality methodologies and tools aimed at improving the production process. Through the application of the DMAIC methodology (Define, Measure, Analyse, Improve and Control), and with quality tools such as brainstorming, Ishikawa diagrams, 5Why's, 5W2H, among others, we intend to implement measures that stabilize and reduce the percentage value of non-compliant parts of the decorative components with high gloss finish.
The study made it possible to conclude that the inclusions defect is the defect that has the highest incidence, originating from secondary operations in the component production chain. The alteration of some procedures and operations, allowed to significantly improve the percentage of defects that occurred. The implementation of the DMAIC methodology was an asset in improving the entire production process and in achieving the goals set annually by the company regarding its percentage of rejection of parts with high gloss finish
A novel sustainable PHA downstream method
No full textThe pernicious impact of plastic originating from petroleum-based sources has been an increasing threat to our environment. Amongst several strategies tested and implied to manage these plastic wastes, one promising alternative is switching to biodegradable polymers. Polyhydroxyalkanoates (PHAs) are a class of microbial polyesters that are biodegradable, biocompatible and an attractive alternative for future packaging applications. They offer an approach to carbon neutrality and support a more sustainable industrialization. However, as these microbial polymers are stored intracellularly, development of downstream processes for extraction and purification are essential and are the most cost-intensive steps. A major drawback in most of the existing state-of-the-art is the use of harsh organic solvents which contributes to the non-eco-friendliness of the processes. This work proposes an alternative, sustainable PHA downstream process comprising the use of a green solvent. Biomass with an intracellular PHA content of ca. 49% (gPHA/gVSS) produced by mixed cultures was used in this study. A new hydrophobic natural deep eutectic solvent (hydrophobic NADES), Thymol : Vanillin (8 : 2 molar ratio) was synthesised and used as the extraction solvent while 1-heptanol was used as the purification solvent. Also, different intensities of sonication were applied to assist the cell wall rupture for the efficient release of PHAs. The optimised process yielded ∼99% polymer purity and recovery of ∼42% and did not alter the polymer composition. Furthermore, after successful extraction and purification of PHA, the organic solvents (hydrophobic NADES and 1-heptanol) used can be recycled or reused as per end-user requirements.The authors gratefully acknowledge Fundação para a Ciência e
a Tecnologia (FCT) for the financial support through the
project PTDC/BTA-BTA/31746/2017, project PTDC/BII-BIO/
50026/2020, grant CEECINST/00156/2018 and the PhD fellow
grant SFRH/BD/146967/2019. This work is financed by national
funds from FCT – Fundação para a Ciência e a Tecnologia, I.P.,
in the scope of the project UIDP/04378/2020 and UIDB/04378/
2020 of the Research Unit on Applied Molecular Biosciences –
UCIBIO, the project LA/P/0140/2020 of the Associate
Laboratory Institute for Health and Bioeconomy – i4HB, and
projects UIDP/CTM/05256/2020 and UIDB/CTM/05256/2020 of
the Institute for Polymers and Composites. This research was
also supported by the Associate Laboratory for Green
Chemistry – LAQV which is financed by national funds from
FCT/MCTES (UIDB/50006/2020 and UIDP/50006/2020)
Surface pathology on the walls of limestone and mortar masonry
No full textThe principles of this work are the study of masonry materials with a view to the conservation and rehabilitation of buildings. It is important to preserve and care for the historical heritage. Development and associated innovation require knowledge of the physical and mechanical characteristics of materials.
Situational causes of pathologies found in stone masonry can be identified by the photographs presented, showing a horizontal cracking due to physical natural actions, temperature variations and the presence of water.
The main causes of pathological changes are air pollution, but the water and the temperature also play a very important role in the development of the deterioration.
The test results are essential to obtain the quality and durability of the materials of the limestone and mortar to prevent damage the water in the environment with the incorporation of paint with good quality.(undefined