ENGINEERING FOR RURAL DEVELOPMENT Jelgava, 24.-26.05.2017. 88 KINETIC MODELING OF ETHANOL FERMENTATION BY YEAST KLUYVEROMYCES MARXIANUS FROM LACTOSE- AND INULIN- CONTAINING SUBSTRATES Jekaterina Martynova, Martins Mednis, Armands Vigants, Peteris Zikmanis University of Latvia jek.martinova@gmail.com Abstract. An unstructured kinetic model was developed in this study for the batch production of bioethanol by the yeast Kluyveromyces marxianus DSM 5422 from the renewable sources of agricultural and food processing origin, such as whey permeate or inulin, which include the terms of both substrate and product inhibitions. Experimental data collected from multiple fermentations in bioreactors with three different initial concentrations for each substrate were used to estimate the parameters and to validate the proposed model. The growth of K.marxianus can be expressed by the Haldane-type extended Monod model in combination with the Jerusalimsky term for the non-competitive product inhibition and the Luedeking–Piret equation was adequate to describe the growth-associated formation of ethanol as the target product. The parameters in the models were estimated by minimizing mean-squared errors between the predictions of the models and the experimental data using the differential evolution (DE) algorithm and the L-BFGS-B nonlinear optimization code. In all cases, the model simulation matched well with the fermentation data being confirmed by the high R-squared values (0.984, 0.992 and 0.965 for WP, lactose and inulin, respectively). The kinetic models proposed here can be employed for the development and optimization of the bioethanol production processes from renewable resources. Keywords: bioethanol, kinetic model, whey permeate, inulin, Klyuveromyces marxianus. Introduction In recent years the yeast Kluyveromyces marxianus has attracted an increasing attention due to versatile biotechnological applications. It can be used as an efficient producer for valuable microbial products including a number of enzymes, flavor and fragrance compounds as well as bioethanol, particularly from renewable resources [1-3]. The ability to utilize a variety of carbon sources, an enhanced thermotolerance, a rapid growth and a strong Crabtree-negative character of cells are the advantageous traits, which promote the use of K. marxianus for industrial bioprocesses [1; 3]. Although, these non-conventional food-grade yeasts have been subjected to still insufficient investigation efforts and quantitative studies of technologically important processes are rarely reported [4; 5]. Such in-depth studies are particularly needed because significantly different growth parameters have been reported not only for different strains of K. marxianus but also for the same strain when investigated in different laboratories [2; 6]. Such a metabolic diversity makes it difficult to generalize the knowledge about these yeasts and therefore encourages researchers to focus at least initially, on the reduced number of strains chosen from key culture collections [2]. A substantial phenotypic variation in the growth parameters can be observed in the production of bioethanol from lactose- or inulin-containing substrates, which are provided by the operation of β- galactosidase or endo- and exo-inulinases in K. marxianus cells [2; 3; 7-9]. Within the foregoing, particularly the proposed need [2] to study a limited number of strains, a certain attention should be given to the K.marxianus DSM 5422. This strain has been proposed as an efficient producer of ethyl acetate and appears relatively well-studied in this context [10, 11]. Although, it has been also used for the production of bioethanol from the renewable lactose-containing substrate such as cheese whey [12-14]. There are several reports on the ethanol production from another technologically promising substrate such as inulin and inulin - containing raw materials by a variety of K. marxianus strains [15-21] although DSM 5422 remains unrepresented among them. Therefore, a comparative analysis for the above two substrates could give a fuller insight into the potential of K. marxianus DSM 5422 for production of bioethanol from the renewable, inexpensive and abundant raw materials [8; 22]. It is well known that the behavior of microbial systems can be evaluated by the growth kinetic parameters, which constitute appropriate mathematical models [23-25]. Even relatively simple kinetic models could be indispensable for the design and successful operation of industrial bioprocesses and for obtaining quantitative information about the function of microbial cells [23]. Thus, relevant parameters of Monod kinetics such as the maximum specific growth rate (μ max ), the saturation constant (K s ) and the yield of biomass (Y x/s ) can be considered as DOI: 10.22616/ERDev2017.16.N016