Development of infrared spectroscopy for assessing bacterial quality in foods

Rapid and specific detection of foodborne bacteria that can cause food spoilage or illness associated to its consumption is an increasingly important task in food industry. Bacterial detection, identification, and classification are generally performed using traditional methods based on biochemical or serological tests and the molecular methods based on DNA or RNA fingerprints. However, these methodologies are expensive, time consuming and laborious. Infrared spectroscopy is a reliable, rapid, and economic technique which could be explored as a tool for bacterial analysis in the food industry. In this thesis it was evaluated the potential of IR spectroscopy to study the bacterial quality of foods. In Chapter 2, it was developed a calibration model that successfully allowed to predict the bacterial concentration of naturally contaminated cooked ham samples kept at refrigeration temperature during 8 days. In this part, it was developed the methodology that allowed the best reproducibility of spectra from bacteria colonies with minimal sample preparation, which was used in the subsequent work. Several attempts trying different resolutions and number of scans in the IR were made. A spectral resolution of 4 cm-1, with 32 scans were the settings that allowed the best results. Subsequently, in Chapter 3, it was made an attempt to identify 22 different foodborne bacterial genera/species using IR spectroscopy coupled with multivariate analysis. The principal component analysis, used as an exploratory technique, allowed to form distinct groups, each one corresponding to a different genus, in most of the cases. Then, a hierarchical cluster analysis was performed to further analyse the group formation and the possibility of distinction between species of the same bacterial genus. It was observed that IR spectroscopy not only is suitable to the distinction of the different genera, but also to differentiate species of the same genus, with the simultaneous use of principal component analysis and cluster analysis techniques. The utilization of IR spectroscopy and multivariate statistical analysis were also investigated in Chapter 4, in order to confirm the presence of Listeria monocytogenes and Salmonella spp. isolated from contaminated foods, after growth in selective medium. This would allow to substitute the traditional biochemical and serological methods that are used to confirm these pathogens and that delay the obtainment of the results up to 2 days. The obtained results allowed the distinction of 3 different Listeria species and the distinction of Salmonella spp. from other bacteria that can be mistaken with them. Finally, in chapter 5, high pressure processing, an emerging methodology that permits to produce microbiologically safe foods and extend their shelf-life, was applied to 12 foodborne bacteria to determine their resistance and the effects of pressure in cells. A treatment of 300 MPa, during 15 minutes at room temperature was applied. Gram-negative bacteria were inactivated to undetectable levels and Gram-positive showed different resistances. Bacillus cereus and Staphylococcus aureus decreased only 2 logs and Listeria innocua decreased about 5 logs. IR spectroscopy was performed in bacterial colonies before and after HPP in order to investigate the alterations of the cellular compounds. It was found that high pressure alters bands assigned to some cellular components as proteins, lipids, oligopolysaccharides, phosphate groups from the cell wall and nucleic acids, suggesting disruption of the cell envelopes. In this work, bacterial quantification and classification, as well as assessment of cellular compounds modification with high pressure processing were successfully performed. Taking this into account, it was showed that IR spectroscopy is a very promising technique to analyse bacteria in a simple and inexpensive manner.

Efeito do revestimento nas propriedades das rolhas de cortiça

A utilização de revestimentos poliméricos em rolhas de cortiça natural é uma prática recorrente na indústria de produção de rolhas. A sua aplicação tem como principal objetivo a melhoria da qualidade visual das rolhas. Contudo, este tratamento altera também as suas caraterísticas de molhabilidade, pelo que é importante otimizar os processos de revestimento de superfície. O presente trabalho teve como objetivo principal a averiguação da influência de um revestimento colorido e de diferentes tratamentos de superfície no comportamento da rolha em garrafa. Foram utilizadas amostras, cedidas pela Cork Supply Portugal, provenientes de quatro branqueamentos distintos (LVL, CSP, CPI e SIL). Foram estudadas amostras com e sem revestimento colorido e, todas elas, tratadas superficialmente com quatro tipos de tratamento (TS1, TS2, TS3 e TS4), sendo as quantidades de parafina e silicone administradas todas diferentes. Para observar o efeito do revestimento colorido e dos tratamentos de superfície nas rolhas de cortiça, procedeu-se a testes para determinação de capilaridade, inserções e forças de extração, simulando as condições de engarrafamento/extração do cliente, de forma a verificar a conformidade dos resultados. Todos estes testes procederam-se nas instalações da CSP. Foram também efetuados testes de molhabilidade pela determinação de ângulos de contacto para três diferentes solventes, nomeadamente água, formamida e diiodometano, e obtidos os Wetting Envelopes, pelo método de OWRK, de forma a prever a molhabilidade das várias amostras. Os procedimentos experimentais descritos efetuaram-se nos laboratórios da Universidade de Aveiro. Aliando as análises aos vários testes efetuados, concluiu-se que o revestimento colorido tem um possível efeito favorecedor nas propriedades mecânicas das rolhas de cortiça, possui maior recetividade aos agentes lubrificantes e maior molhabilidade. Relativamente aos tratamentos de superfície, o melhor, tendo em conta os fatores de molhabilidade, foi o TS4, proporcionando melhor vedação ao vinho. Já a respeito da lubrificação, verificou-se melhores resultados para os tratamentos TS1 e TS2. A lavação SIL apresenta menor afinidade aos agentes vedantes e a CSP possui maior recetibilidade aos agentes lubrificantes.