The growing demand for energy in the world, the implications of climate change, the increasing damages to our environment and the diminishing fossil fuel reserves have created the appropriate conditions for renewable energy development. Biofuels such as bioethanol can be produced by breaking down the lignocellulosic structure of plant materials to release fermentable sugars. Sweet sorghum bagasse has been shown to be an important lignocellulosic crop residue and is potentially a significant feedstock for bioethanol production. The aim of this study was to investigate suitable microwave assisted pretreatment conditions of sweet sorghum bagasse for bioethanol production. A chemical pretreatment process of sweet sorghum bagasse, using different concentrations (1 to 7 wt%) of sulphuric acid (H2SO4) and calcium hydroxide (Ca (OH)2) was applied to break up the lignocellulosic matrix of sweet sorghum bagasse. The pretreated broth, which contained pentose and hexose sugars, was fermented using a combination of Zymomonas mobilis ATCC31821 and Saccharomyces cerevisiae to produce bioethanol at pH 4.8 and 32oC for 24 hours. The highest reducing sugar yield of 0.82 g/g substrate was obtained with microwave irradiation at 180 W for 20 minutes in a 5 wt% sulphuric acid solution. The highest ethanol yield obtained was 0.5 g/g from 5 wt% H2SO4 pretreated bagasse at 180 W using a 10:5% v/v of Saccharomyces cerevisiae to Zymomonas mobilis ratio, whereas for 3 wt% Ca (OH)2 microwave pretreatment, a sugar yield of 0.27 g/g substrate was obtained at 300 W for 10 minutes. Thereafter, an ethanol yield of 0.13 g/g substrate was obtained after 24 hours of fermentation when using a 10:5% v/v of Saccharomyces cerevisiae to Zymomonas mobilis ratio. The effect of microwave pretreatment on the bagasse was evaluated using Scanning Electron Microscopy (SEM) and Fourier Transform Infrared Spectroscopy (FTIR) analysis. The reducing sugars formed were quantified using High Performance Liquid Chromatography (HPLC). The results showed that microwave pretreatment using 5 wt% H2SO4 is a very effective pretreatment that can be used to obtain sugars from sweet sorghum bagasse. The analytic results also showed physical and functional group changes after microwave pretreatment. This confirms that microwave irradiation is very effective in terms of breaking up the lignocellulose structure and improving fermentable sugar yield for fermentation. Bioethanol yields obtained from microwave pretreatment using different solvents also show that Saccharomyces cerevisiae and Zymomonas mobilis ATCC31821 is a good combination for producing ethanol from sweet sorghum bagasse. Sweet sorghum bagasse is clearly a very effective and cheap biomass that can be used to produce bioethanol, since very high yields of fermentable sugars were obtained from the feedstock. / Thesis (MSc (Engineering Sciences in Chemical Engineering))--North-West University, Potchefstroom Campus, 2013.

Zymomonas mobilis is one of the most promising microorganisms for bioethanol production. However, its practical use on industrial scale is impeded by its high sensitivity to acetate, which is present in high concentration in pretreated biomass. This research develops an adaptive mutation method for generating acetate-tolerant strains for bioethanol production. The goal is to obtain Zymomonas mobilis strain capable of growing and producing ethanol in the presence of acetate at a concentration typical of a pretreated biomass (2-3%). The interplay between the ability of fermentative production of ethanol and acetate tolerance will be investigated through careful fermentation studies. The potential cross-tolerance to other inhibitors, commonly present in pretreated biomass will be evaluated. A preliminary study on the mechanism of acetate tolerance at the cell membrane level will be conducted. The strain developed through this research will be useful in bioethanol production from biomass. The insights into tolerance mechanisms gained through this study will allow a more rational approach to further engineer a better producing strain.

Thesis (M.S. (Hons.))-- University of Western Sydney, 2008. / A thesis sumitted to the University of Western Sydney in fulfilment of the requirements for the degree of Masters of Science (Honours), School of Natural Sciences, College of Health and Science. Includes bibliography.

Orientador: Crispin Humberto Garcia-Cruz / Banca: Rodrigo Simões Ribeiro Leite / Banca: Marcos de Lucca Junior / Banca: Vanildo Luiz Del Bianchi / Banca: Maurício Boscolo / Resumo: A geração de álcool combustível a partir de resíduos lignocelulósicos pode ser uma fonte alternativa e renovável de energia. O Brasil é um dos maiores produtores de frutas tropicais do mundo, com destaque para a laranja e banana. Em consequência disto, é capaz de gerar grandes quantidades de resíduos que podem ser utilizados como biomassa que, uma vez hidrolisada, libera pentoses e hexoses. A fim de se buscar uma maior produção de etanol, a utilização de co-culturas que metabolizem tanto pentoses quanto hexoses torna-se bastante interessante. Dentro desse contexto, o objetivo do presente trabalho foi estudar o efeito da hidrólise de cascas de banana e laranja para a produção de etanol por co-culturas de Zymomonas mobilis e Pichia stipitis. A hidrólise ácida foi feita com ácido sulfúrico 5% e 15 min em autoclave, para a hidrólise enzimática utilizou-se kit fornecido pela Novozymes em reação a 50°C durante 36 h. Após as hidrólises o meio foi desintoxicado com carvão ativado. Foram avaliados os efeitos da fermentação somente por Z. mobilis e por co-culturas de Z. mobilis e P. stipitis, do pH, da agitação e da quantidade de açúcares no meio. As hidrólises ácida e enzimática liberaram 64,97 e 134,75 g/L de açúcares totais das cascas de banana e 101,30 e 176,70 g/L de açúcares totais das cascas de laranja, respectivamente. A desintoxicação do hidrolisado resultou em remoção de 58% a 93% dos fenóis e 1,7% a 4% dos açúcares totais. Nas fermentações descontínuas com meio sintético o maior crescimento celular e máxima produção de etanol foram 1,52 e 11,29 g/L para a monocultura de Z. mobilis, e 8,00 e 77,02 g/L para a co-cultura, respectivamente. Para o meio com cascas de banana essas respostas foram 1,87 e 4,16 g/L para a monocultura, e 15,19 e 23,92 g/L para a co-cultura, respectivamente. Para o meio com cascas de laranja, as mesmas respostas foram 0,39 e 1,85 g/L para a monocultura, e 6,73 e... / Abstract: The generation of ethanol from lignocellulosic waste, such as fruit crops, can be an alternative and renewable energy source. Brazil is one of the largest producers of tropical fruits in the world, especially orange and banana. As a result, it is able to generate large amounts of waste that may be used as biomass, which once hydrolyzed, releases pentoses and hexoses. In order to get a higher ethanol yield, the use of co-cultures which metabolize both pentose and hexose becomes quite interesting. In this context, the objective of this work was to study the hydrolysis effect of banana and orange peels for ethanol production by co-cultures of Zymomonas mobilis and Pichia stipitis. The acid hydrolysis was performed with 5% sulfuric acid and 15 min in autoclave; for the enzymatic hydrolysis was used a commercial kit supplied by Novozymes, in reaction at 50°C for 36 h. After the hydrolysis the medium was detoxified with activated carbon. The effects of fermentation by Z. mobilis monoculture and Z. mobilis and P. stipitis co-cultures, pH, agitation and the initial amount of sugars in the middle were evaluated. The sequential acid and enzymatic hydrolysis released 64.97 and 134.75 g/L of total sugars from banana peels and 101.30 and 176.70 g/L of total sugars from orange peels, respectively. The detoxification of the hydrolysate resulted in a removal of 58% to 93% of phenol and 1.7% to 4% of the total sugars. In Erlenmeyer fermentations with synthetic medium, the largest cell growth and maximum ethanol production were 1.52 and 11.29 g/L for Z. mobilis monoculture, and 8.00 and 77.02 g/L for the co-culture, respectively. In the medium with banana peels, these responses were 1.87 and 4.16 g/L for monoculture, and 15.19 and 23.92 g/L for the co-culture, respectively. To the medium with orange peels, the same responses were 0.39 and 1.85 g/L for monoculture and 6.73 and 11.36 g/L for the co-culture, respectively. For all evaluated media, the Z. ... / Doutor

Made available in DSpace on 2016-04-01T17:54:49Z (GMT). No. of bitstreams: 0 Previous issue date: 2015-09-11. Added 1 bitstream(s) on 2016-04-01T18:00:30Z : No. of bitstreams: 1 000860213_20200911.pdf: 451488 bytes, checksum: 42a973a544500493638c49349249b14c (MD5) / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / A geração de álcool combustível a partir de resíduos lignocelulósicos pode ser uma fonte alternativa e renovável de energia. O Brasil é um dos maiores produtores de frutas tropicais do mundo, com destaque para a laranja e banana. Em consequência disto, é capaz de gerar grandes quantidades de resíduos que podem ser utilizados como biomassa que, uma vez hidrolisada, libera pentoses e hexoses. A fim de se buscar uma maior produção de etanol, a utilização de co-culturas que metabolizem tanto pentoses quanto hexoses torna-se bastante interessante. Dentro desse contexto, o objetivo do presente trabalho foi estudar o efeito da hidrólise de cascas de banana e laranja para a produção de etanol por co-culturas de Zymomonas mobilis e Pichia stipitis. A hidrólise ácida foi feita com ácido sulfúrico 5% e 15 min em autoclave, para a hidrólise enzimática utilizou-se kit fornecido pela Novozymes em reação a 50°C durante 36 h. Após as hidrólises o meio foi desintoxicado com carvão ativado. Foram avaliados os efeitos da fermentação somente por Z. mobilis e por co-culturas de Z. mobilis e P. stipitis, do pH, da agitação e da quantidade de açúcares no meio. As hidrólises ácida e enzimática liberaram 64,97 e 134,75 g/L de açúcares totais das cascas de banana e 101,30 e 176,70 g/L de açúcares totais das cascas de laranja, respectivamente. A desintoxicação do hidrolisado resultou em remoção de 58% a 93% dos fenóis e 1,7% a 4% dos açúcares totais. Nas fermentações descontínuas com meio sintético o maior crescimento celular e máxima produção de etanol foram 1,52 e 11,29 g/L para a monocultura de Z. mobilis, e 8,00 e 77,02 g/L para a co-cultura, respectivamente. Para o meio com cascas de banana essas respostas foram 1,87 e 4,16 g/L para a monocultura, e 15,19 e 23,92 g/L para a co-cultura, respectivamente. Para o meio com cascas de laranja, as mesmas respostas foram 0,39 e 1,85 g/L para a monocultura, e 6,73 e... / The generation of ethanol from lignocellulosic waste, such as fruit crops, can be an alternative and renewable energy source. Brazil is one of the largest producers of tropical fruits in the world, especially orange and banana. As a result, it is able to generate large amounts of waste that may be used as biomass, which once hydrolyzed, releases pentoses and hexoses. In order to get a higher ethanol yield, the use of co-cultures which metabolize both pentose and hexose becomes quite interesting. In this context, the objective of this work was to study the hydrolysis effect of banana and orange peels for ethanol production by co-cultures of Zymomonas mobilis and Pichia stipitis. The acid hydrolysis was performed with 5% sulfuric acid and 15 min in autoclave; for the enzymatic hydrolysis was used a commercial kit supplied by Novozymes, in reaction at 50°C for 36 h. After the hydrolysis the medium was detoxified with activated carbon. The effects of fermentation by Z. mobilis monoculture and Z. mobilis and P. stipitis co-cultures, pH, agitation and the initial amount of sugars in the middle were evaluated. The sequential acid and enzymatic hydrolysis released 64.97 and 134.75 g/L of total sugars from banana peels and 101.30 and 176.70 g/L of total sugars from orange peels, respectively. The detoxification of the hydrolysate resulted in a removal of 58% to 93% of phenol and 1.7% to 4% of the total sugars. In Erlenmeyer fermentations with synthetic medium, the largest cell growth and maximum ethanol production were 1.52 and 11.29 g/L for Z. mobilis monoculture, and 8.00 and 77.02 g/L for the co-culture, respectively. In the medium with banana peels, these responses were 1.87 and 4.16 g/L for monoculture, and 15.19 and 23.92 g/L for the co-culture, respectively. To the medium with orange peels, the same responses were 0.39 and 1.85 g/L for monoculture and 6.73 and 11.36 g/L for the co-culture, respectively. For all evaluated media, the Z. ...

Submitted by Mario BC (mario@bc.ufrpe.br) on 2017-07-31T14:52:51Z No. of bitstreams: 1 Edna Barboza de Lima.pdf: 1887608 bytes, checksum: 310d47dc918d66209fc26060683ce16f (MD5) / Made available in DSpace on 2017-07-31T14:52:51Z (GMT). No. of bitstreams: 1 Edna Barboza de Lima.pdf: 1887608 bytes, checksum: 310d47dc918d66209fc26060683ce16f (MD5) Previous issue date: 2016-07-29 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / This work utilized a flow-batch based system aiming to study the growth kinetic profile and resistance of microorganisms tested with inorganic drug cisplatin (CDDP), commonly used in cancer treatment. Aiming to reduce costs and waste production during the tests, Solenoid mini-pumps were used as fluid propulsion system. Such a device was connected to an integrated circuit ULN 2803 which was coupled to an electronic serial interface (USB 4000) to connect to computer's USB output that had the function to control the pulses of solenoid mini-pumps sent to reaction chamber. A multichannel UV-Vis spectrophotometer with a linear array of photodetectors was used as detection system. The data acquisition and control devices were obtained from program written in LABVIEW 8.5 language. Initially, the tests with Saccharomyces cerevisiae yeast were carried out to obtain their growth kinetic profile; however, the micro-organism produced a high amount of biomass and this caused interference in turbidimetric analysis, causing what we call backscattering or backscatter effect. In order to reduce interference, subsequent tests were performed with the bacterium Zymomonas mobilis, which not only produces a smaller amount of biomass but is also a non-pathogen used in cell biology. For such bacteria, it was was possible to obtain the kinetic growth profile and verify its behavior with the use of inorganic drugs cisplatin for a period of 24 hours. After this time the system presented interference and reading errors caused by the increase in cell concentration. To evaluate the CDDP-resistance profile of the Z. mobilis turbidimetric readings were performed at 600 nm in UV-Vis without the flow-batch system. The cell count of the microbial were carried out in a Neubauer chamber and Gram stain was checked on glass slides. The result from such analyzes, reaffirm the anti-mitotic action of cisplatin as well allowed the chronological elucidation of the biochemical mechanisms of the resistance against anticancer tested (CDDP). / Neste trabalho foi empregado um sistema baseado em fluxo-batelada para estudo do perfil cinético de crescimento e resistência de microrganismos na presença do fármaco cisplatina, usada comumente no tratamento de câncer. Com o intuito de minimizar a geração de resíduos, foram utilizadas mini-bombas solenoide como sistema de propulsão de fluidos. Inicialmente foram realizados testes com a levedura Saccharomyces cerevisiae para verificar o seu perfil cinético de crescimento, porém tal microrganismo produziu uma elevada quantidade de biomassa aglomerada gerando assim interferência na análise turbidimétrica, chamada de efeito backscaterring ou retroespalhamento. A fim de reduzir esse tipo de interferência, os testes posteriores, foram realizados com a bactéria Zymomonas mobilis, pois esta, mesmo produzindo uma quantidade considerável de biomassa, possui mobilidade (1 a 4 flagelos polares) o que garante melhor sua distribuição, dificultando a formação de aglomerados. Desde modo, foi possível obter o perfil cinético de crescimento e verificar seu comportamento frente ao uso da droga inorgânica cisplatina, no período de 24 h. Após esse tempo, o sistema apresentou interferências causadas pelo aumento da concentração celular. Para verificar a resistência da Z. mobilis ao agente antineoplásico (cisplatina) foram realizadas leituras turbidimétricas a 600 nm, empregando espectrofotômetro UV-Vis sem utilização do sistema fluxo-batelada. Também foram realizadas contagens microbianas em câmara de Neubauer e coloração de gram em lâminas de vidro. Os resultados obtidos com tais análises, além de reafirmar a ação anti-mitótica da cisplatina, também possibilitaram a elucidação temporal para os mecanismos bioquímicos de resistência do microrganismo testado frente ao antineoplásico em questão (CDDP).

Current energy and environmental challenges are driving the use of cellulosic materials for biofuel production. A major obstacle in this pursuit is poor ethanol tolerance among cellulolytic Clostridium species. The first objective of this work was to establish a potential upper boundary of ethanol tolerance for the cellulosome itself. The hydrolytic function of crude cellulosome extracts from C. cellulolyticum on carboxymethyl cellulose (CMC) with 0, 5, 10, 15, 20 and 25% (v/v) ethanol was determined. Results indicated that the endoglucanase activity of the cellulosome incubated in 5% and 10% ethanol was significantly different from a control without ethanol addition. Furthermore a significant difference was observed in endoglucanase activity for cellulosome incubated in 5%, 10%, 15%, 20% and 25% ethanol in a standalone experiment. Endoglucanase activity continued to be observed for up to 25% ethanol, indicating that cellulosome function in ethanol will not be an impediment to future efforts towards engineering increasing production titers to levels at least as high as the current physiological limits of the most tolerant ethanologenic microbes. The second objective of this work was to study bioethanol production by a microbial co-culture involving Clostridium cellulolyticum and a recombinant Zymomonas mobilis engineered for the utilization of oligodextrans. The recombinant Z. mobilis ZM4 pAA1 and wild type ZM4 were first tested on RM medium (ATCC 1341) containing 2% cellobiose as the carbon source. Ethanol production from the recombinant Z. mobilis was three times that observed from the wild type Z. mobilis. Concomitant with ethanol production was the reduction in OD from 2.00 to 1.580, indicating the consumption of cellobiose. No such change in OD was observed from the wild type. The recombinant ZM4 was then co-cultured with C. cellulolyticum using cellobiose and microcrystalline cellulose respectively as carbon sources. Results indicate that the recombinant ZM4 acted synergistically with C. cellulolyticum to utilize 2.0 g L-1 cellobiose, producing as much as 0.40 mM concentration of ethanol whereas only 0.20 mM ethanol was detected for the wild type ZM4 co-cultured with C. cellulolyticum under the same conditions. A co-culture of the recombinant ZM4 and C. cellulolyticum using 7.5 g L-1 microcrystalline cellulose gave lower ethanol yield than when using cellobiose. In the latter case, the recombinant began producing ethanol in 5 days whereas the wild type required 10 days to produce detectable ethanol. Future efforts will concentrate on identifying the correct concentration of cellulosic substrate at which synergy will be observed using the recombinant ZM4 and other cellulose degrading microorganisms, as well as optimizing medium formulations to better support both organisms.

Thesis (MEng) -- Stellenbosch University, 2014. / ENGLISH ABSTRACT: See PDF for abstract. / AFRIKKANSE OPSOMMING: Sien PDF vir die opsomming.

A bactéria Zymomonas mobilis atraiu considerável interesse nas últimas décadas devido ao seu metabolismo único e eficientes características fermentativas na produção de etanol a partir de açúcares simples. No entanto, apesar das aparentes vantagens na conversão e taxas específicas quando comparada com as leveduras, estas ainda dominam o mercado produtivo de etanol. Dentre os diversos modelos encontrados na literatura para representar o processo fermentativo de glicose através de Z. mobilis, o modelo proposto por Jöbses et al. (1986) aparenta ser bem estruturado e ideal para a criação de estratégias de controle e otimização de processos fermentativos com este microrganismo, viabilizando sua utilização em escala industrial. Fermentações de Z. mobilis em regime contínuo apresentam comportamento oscilatório para baixas taxas de diluição, fenômeno este comprovado experimentalmente. Experimentos laboratoriais foram utilizados para a estimação de parâmetros do modelo de Jöbses, através da técnica de minimização da derivada do erro da função objetivo. O novo modelo ajustado apresenta dinâmica equivalente ao modelo original de Jöbses, fato este comprovado através da construção dos diagramas de bifurcação. Os diagramas de fase dos modelos apresentam algumas diferenças estruturais entre si, entretanto, a ocorrência de multiplicidade de estados estacionários em baixos valores de taxa de diluição está presente em todos. Em bioprocessos, a obtenção de informações sobre o sistema é um tanto complexa, devido a estes meios serem muito suscetíveis a distúrbios, contaminações, além de que medidas de concentrações são relativamente dispendiosas, associadas com um alto tempo morto, o que impossibilita, muitas vezes, a implementação de sistemas de controle eficientes. Para contribuir com a resolução deste problema foi desenvolvida uma nova inferência das variáveis de estado através da informação de consumo de hidróxido de sódio, utilizado no controle de pH do meio. Inferidores de variáveis de estado que utilizam a variação no consumo de hidróxido de sódio foram ajustados e apresentaram resultados promissores, mostrando a viabilidade do desenvolvimento de metodologias para este tipo de análise, tornando a caracterização de sistemas fermentativos mais rápida e acessível, não apenas em nível acadêmico, que a desenvolvida atualmente, principalmente devido ao baixo custo associado, e a dinâmica rápida deste tipo de sensor. / The bacterium Zymomonas mobilis has attracted considerable interest in recent decades due to their unique metabolism and efficient characteristics in the production of ethanol by simple sugars. However, despite the apparent advantages in the conversion and specific rates, when compared with yeast, there is no industrial-scale fermentations with these bacteria. Among the various models found in literature to represent glucose fermentation with Z. mobilis, the model proposed by Jöbses et al. (1986) appears to be well structured and ideal for the creation of control strategies and optimization methods for increase productions. Fermentation in continuous mode with this organism exhibit oscillatory behavior at low dilution rates, phenomenon showed or demonstrated by experiments performed on laboratory scale. A laboratorial experiment also was used for estimation of new parameters group of Jöbses models, using the technique of minimization of error derivative for objective function. The new adjusted model set presents dynamics similar to the original model, fact confirmed by bifurcation analyses of both models. The diagram for these models show structural differences, but both presents steady states multiplicity and Hopf bifurcations. The obtained of fermentation characteristics is very complex, due to these reactive means being very susceptible to disturbances, contamination, among others, and concentrations measures are relatively expensive and whit a high dead time associated, which often prevents the implementation of control tools. The proposed inference of state variables by consumption rate of hydroxide sodium, used for maintenance pH medium, was tested and confirmed. Virtual sensors using the information of consumption rate of hydroxide were adjusted and have shown promising results, demonstrating the feasibility of developing methods of analysis based on this methodology, making the characterization of fermentative systems faster and cheaper them currently developed, mainly due to low cost associated, and the dynamics of this fast type of sensor.

Orientador: Crispin Humberto Garcia-Cruz / Banca: Eleni Gomes / Banca: Ranulfo Monte Alegre / Banca: Vanildo Luiz Del Bianchi / Banca: Hamilton Cabral / Resumo: O principal produto da fermentação de açúcares por Zymomonas mobilis é o etanol quando glicose e frutose são utilizadas como fontes de carbono. Entretanto, quando sacarose é empregada na fermentação, o rendimento do etanol diminui devido à formação de subprodutos como levana, sorbitol, acetaldeído, ácido acético, pequenas quantidades de alguns álcoois superiores e fenol. A utilização de produtos agroindustriais, como o caldo e melaço de cana-deaçúcar, é uma alternativa para reduzir o custo final dos produtos de fermentação devido à disponibilidade de aquisição e composição química desses substratos. Este trabalho teve como objetivo utilizar substratos alternativos e otimizar as condições de fermentação para a produção de etanol, levana e sorbitol por Zymomonas mobilis CCT 4494. Também foi considerado o efeito da variação do substrato e de sais minerais adicionados nos meios de produção (sintético, caldo e melaços de cana-de-açúcar). Para a obtenção dos produtos de fermentação, foi aplicada a metodologia de superfície de resposta, seguindo um planejamento fatorial do tipo 27-2, de acordo com o modelo proposto por Box e Hunter, onde as variáveis independentes estudadas foram: pH inicial do meio de cultivo, temperatura de incubação, concentração do substrato e efeito da adição de KCl, K2SO4, MgSO4, CaCl2. Durante a realização das fermentações foi observado que a bactéria Zymomonas mobilis CCT 4494 se adaptou nos meios de fermentação contendo altas concentrações de sacarose e suportou a variação do pH e da temperatura de fermentação. O aumento da concentração da fonte de carbono favoreceu a formação dos produtos levana e etanol, entretanto, não houve produção de sorbitol. O meio sintético proporcionou maior rendimento de levana e etanol, enquanto que, os meios alternativos caldo e melaços de cana-de-açúcar... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: The main product from fermentation of sugars by Zymomonas mobilis is ethanol when glucose and fructose are used as carbon sources. However, when sucrose is used in the fermentation medium, ethanol yield decreases due to the formation of by-products such as levan, sorbitol, acetaldehyde, acetic acid, small amounts of some superior alcohols and phenol. The use of agro industrial by products, such as sugarcane juice and molasses, is an alternative to reduce the final cost of fermentation products due to the constant availability and to the chemical composition of these by substrates. This study had the aim of using alternative substrates and of optimizing fermentation conditions for the production of ethanol, levan and sorbitol by Zymomonas mobilis CCT 4494. The effect of variation of substrate and mineral salts added to the production media (synthetic, sugarcane juice and molasses) was also considered. To obtain the fermentation products, response surface methodology was employed, following a 27-2 factorial planning, according to the model proposed by Box and Hunter, where the independent variables studied were: initial medium pH, incubation temperature, substrate concentration and effect of the addition of KCl, K2SO4, MgSO4, CaCl2. During the fermentations, it was noted that the bacteria Zymomonas mobilis CCT 4494 well adapted in the media containing high concentrations of sucrose and tolerated pH and temperature variations. The increase of carbon source concentration favored the formation of levan and ethanol, however, there was no sorbitol production. The synthetic medium offered higher levan and ethanol yield, whereas alternative media sugarcane juice and molasses, favored cellular growth. Among the independent variables analyzed with the best medium (synthetic) for biosynthesis of the biopolymer and ethanol, the ones that significantly (p<0.05) affected were KCl, K2SO4, CaCl2... (Complete abstract click electronic access below) / Doutor