BRPI1105142B1 - extractive fermentation process and use - Google Patents

Abstract

extractive fermentation process and use. the present invention relates to an extractive fermentation process. more specifically, the present invention relates to an extractive fermentation process that employs a single solvent capable of removing products and inhibitors simultaneously, in addition to cooling the medium. furthermore, the present invention relates to the use of the process. specifically, the present invention refers to the use of biodiesel, such as biodiesel from soybean oil, corn, castor oil, oil palm, macauba, macauba, among others, as a biocompatible extraction agent for in-situ product extraction and and / or fermentation inhibiting components, specifically the fermentation product and inhibitory compounds present in musts from hydrolytic processes of lingo-cellulosic raw material, starch, directly extractable carbohydrates, molasses strongly depleted with a high content of inhibitors such as, among others, biotin, or similar musts rich in carbohydrates, as well as for cooling the fermentation medium by the solvent itself.

Description

EXTRACTIVE FERMENTATION PROCESS AND USE OF THE SAME

Field of the Invention

The present invention relates to an extractive fermentation process. More specifically, the present invention relates to an extractive fermentation process that employs a single solvent capable of simultaneously removing products and inhibitors, and / or cooling the medium. In addition, the present invention relates to the use of the process, for example, to obtain ethanol.

Background of the Invention

It is a known fact that obtaining fermentation products, such as organic acids, solvents and alcohols, from carbohydrates depends a lot on raw material costs and capital and energy costs, making other factors of secondary importance. Development of technology for the production of second generation fuels, alcohols, solvents or organic acids has the principle of converting biomass, usually ligno-cellulosic, starch and the like, or directly extractable carbohydrates, transforming them into products of higher economic value. aggregate.

In conventional alcoholic fermentation and mainly in second generation alcoholic fermentation, its productivity is limited by inhibition effects,

2/91 either by the fermentation product or by inhibitors introduced in the extraction of carbohydrates from the raw material or inhibitors from the hydrolysis of lignocellulosic biomass. The same problem also arises in fermentation for the production of other products such as, for example, acetone, butanol, and products of the organic acid type such as acetic acid, propionic acid, lactic acid, succinic acid, butanoic acid, among others or any combination of these . Technologies developed and currently used in the industry aim to reduce this effect of inhibition by detoxification of the medium before fermentation and / or removal of the product after fermentation.

To avoid the introduction of inhibitors during the process of extracting carbohydrates from the raw material, for example sugar cane or beet, the temperature used in this step must be kept low, to the detriment of higher extraction yields of these carbohydrates that could be obtained for higher temperatures.

For hydrolyzed biomass fermentations, research has been carried out to develop means of removing inhibitory components from the hydrolytic broth before its introduction in the fermentation stage.

A conventionally applied method is the alkaline treatment of the broth, for example, the callus: increase of pH at

3/91

9-10 with calcium hydroxide. In addition to the neutralization of the broth and sedimentation of salts with sulfuric acid in the form of gypsum, compounds such as furfural and HMF are chemically altered, decreasing the toxicity of these inhibitors, which was described by Martinez and collaborators, in the work entitled Detoxification of dilute acid hydrolysates of

lignocellulose with file ' '' published at Biotechnology Progress 17; 287-293 in 2001. However, O method has the disadvantage of reduces will quality and The quantity of sugar. For treatments adsorption-based of inhibitors, like adsorption by active coal or ion exchange, a excellent adsorption From compounds can hinder

regeneration of coal or resin, so that its use becomes impracticable, which was proven by Sainio and collaborators in the work entitled Adsorptive removal of fermentation inhibitors from concentrated hydrolyzates of lignocellulosic biomass published in Bioresource Technology 102; 6048-6057 in 2011.

The application of all of these methods has positive effects on fermentation, but the installation and use of a separate must treatment stage before fermentation makes the process more expensive, except for in-situ treatments, such as in-situ microbiological detoxification or

4/91 liquid-liquid extraction using a biocompatible solvent.

The simplest in-situ microbiological treatment uses the yeast itself, possibly in high concentration, to reduce the toxicity of the medium, either by immobilization, cell recycling or controlled feeding in batch-fed fermentation, as described by Talebnia and Taherzadeh, in work entitled In situ detoxification and continuous cultivation of diluteacid hydrolyzate to ethanol by encapsulated S. cerevisiae published in the Journal of Biotechnology 125, 377-384, in 2006.

However, the disadvantage of these methods is the limited rate of detoxification by yeast, impairing growth and productivity and increasing the time of the fermentation process, in addition to the accumulation of inhibitors that cannot be converted as various anions of organic acids, as proved by Almeida et al. work entitled: Increased tolerance and conversion of inhibitors in lignocellulosic hydrolysates by Saccharomyces cerevisiae ”published in the Journal of Chemical Technology and Biotechnology, 82 (2007) 340-349 in 2007.

continuous vacuum extractive fermentation regime was studied by Atala in his doctoral thesis entitled

5/91

Assembly, instrumentation, control and experimental development of an extractive fermentative process for ethanol production, FEA / UNICAMP in 2004 to remove ethanol from the medium during fermentation with a flash-type vacuum evaporator. The system reduces the concentration of ethanol in the medium to around 5 ° GL, that is, below the inhibition level. As a result, fermentation, fed with a flow with a high sugar content (up to around 500g / L), showed less inhibition by the ethanol produced, high viability and volumetric productivity. In addition, the concentration of ethanol in the vapor stream from the evaporator can reduce costs in the recovery of the product as described by Junqueira and collaborators in the work entitled Simulation and optimization of the continuous vacuum extractive fermentation for bioethanol production and evaluation of the influence on distillation process published Computer Aided Chemical Engineering 26; 827832 in 2009. The high concentration of substrate results in less water used in the integral ethanol production process, implying less energy expenditure, and more concentrated phlegm and vinasse. The latter results in a lower cost of drying and transporting vinasse to the sugarcane fields to fertilize them. An efficient cooling of the vat was also foreseen, by the

6/91 vacuum system.

In this system, the use of hydrolytic broth as a substrate, the economics of using a membrane for recycling cells and the flash evaporator on an industrial scale in combination with the high rate of recycling by the system have not yet been studied. The vacuum evaporator can probably decrease the concentrations of volatile inhibitors like furfural and acetic acid, while other inhibitors like vanilla and 4-hydroxybenzaldehyde can accumulate if the cell mass is not high enough for their effective conversion. Also, the high stress of the yeast, generated by the high rate of yeast recycling can negatively influence its viability in an industrial scale system, although this effect has not been observed in the bench scale. However, it is possible that the low concentration of ethanol is of greater importance to maintain high viability. In this context, it is important to remember that the stress factor due to the presence of the product and inhibitors in the hydrolytic broth can be added to the stress that occurs in the cell membrane through the cell recycling system. For example, phenolic inhibitors precisely disturb the integrity of the cell membrane and other inhibitors cause ATP diversion for intracellular repairs, to the detriment of maintaining

7/91 integrity of the cell membrane.

Liquid-liquid extractive fermentation involves the use of a biocompatible solvent during alcoholic fermentation to remove the product from solvent fermentation and, above all, different from similar technologies previously proposed by other authors, the inhibitors in the medium from the carbohydrate extraction process the raw material and / or hydrolytic broth, in addition to extracting heat from this fermentation medium.

There are some examples in the state of the art where the solvent used is not biocompatible, requiring the separation of the yeast from the medium or wine before extraction or use of a membrane to avoid direct contact between the yeast and the solvent, such as decanol (Minier M., Goma G., 1998, Eckert and Schügerl, 1987, Kapucu and collaborators, 1999), that-octane, octane, gasoline, kerosene and diesel (Rahman and collaborators, 2007), supercritical fluids such as compressed gases CO2, ethane and propane (Bothun et al., 2003) to remove ethanol, or ethyl ether or ethyl acetate (Cruz et al., 1999). In these cases, it is also impossible to apply the solvent as an in-situ cooling agent.

On the other hand, there are still examples of the use of a biocompatible solvent. Kollerup and Daugulis, in

8/91 work entitled Ethanol production by extractive fermentation - solvent identification and prototype development published in The Canadian Journal of Chemical Engineering, 64, 598 - 606 in 1986, made a theoretical assessment of 1361 possible solvents for ethanol extraction through properties as coefficients partition, biocompatibility, availability and cost. For biocompatibility, it was observed that solvents with a carbon chain with a number of carbon atoms less than 12 have toxic or inhibitory effects on yeast growth (Minier and Goma, 1982). Examples of these and other solvents described in the prior art are the use of oleic oil (Daugulis et al., 1987, Roffler and et al., 1987, Weilnhammer and Blass, 1994), dodecanol (Gyamerah and Glover, 1996, Minier and Coma, 1981) and oleic acid for ethanol removal (Jassal and collaborators, 1994) or for simultaneous production of biodiesel (Oliveira and collaborators, 1998, Csányi and collaborators, 2004), ricinoleic acid or other fatty acids (Waibel and collaborators, 2010, Boudreau and Hill, 2006).

The use of biodiesel, methyl esters of fatty acids from vegetable oils, to remove ethanol, butanol or acetone were also evaluated and described in the state of the art (Waibel collaborators, 2010, Grobben collaborators,

9/91

1993, Ishizaki collaborators, 1999), but not for removing inhibitors from the process of extracting carbohydrates from the raw material, inhibitors from the hydrolytic broth and much less the use of the solvent for the refrigerant in the fermentation medium. Using oleic alcohol in a process of saccharification and extractive fermentation with cellulase, Moritz and Duff (1996) came close to performing the simultaneous extraction of product and inhibitors of the hydrolytic broth, but their literature does not show this understanding.

Other studies show positive results using polymers for ethanol extraction (Seiler and collaborators, 2003, Offeman and collaborators, 2008) and, for extraction of phenolic compounds from the hydrolytic broth, polymers whose separation of the medium is induced by an increase in temperature (Hasmann and employees, 2008). The use of variations in polyethylene glycol was reviewed by Banik et al. (2003). The removal of ethanol by adsorbents, in combination with immobilized yeast, was studied by Cartón and collaborators (1998).

Only for in-situ ethanol removal (without removing other inhibitors or cooling the medium), with biocompatible solvent, the process was tested on the bench scale and modeled, showing returns on investment costs

10/91 for existing plants and new plants (Daugulis et al., 1991, 1994). The process led to the use of high concentrations of sugar, less inhibition of ethanol during fermentation and less energy requirement.

US patent 4,865,973, filed on August 13, 1986 in the name of UNIV KINGSTON and entitled: Process for extractive fermentation describes a process for the production of ethanol by fermentation coupled to an extraction unit of the product formed, citing acid ricinoleic and other fatty acids as solvent options. It is also described that the extraction is in situ and listed as advantages: reduced costs due to the size of the fermenter, reduced costs of product recovery due to higher final concentrations and reduced pretreatment and cost of waste treatment, due to the lower flow aqueous. In addition, the extracting agent can be recirculated. The fermentation described in that document takes place at 20 to 80 ° C.

Using a solvent with the same or similar characteristics to biodiesel can have a similar effect for the removal of inhibitors since all other conditions of the process are also the same. However, the extraction of inhibitors is not just a side effect, but a

11/91 pre-treatment technology itself, recognizing the fact that the removal must be such that the final concentration of the inhibitors in the fermentation medium must be equal to or less than the maximum concentration that still allows satisfactory performance of the microorganism. Knowing that inhibitors show synergistic inhibition effects, aggravated by certain pH and temperature ranges, the level of desirable inhibitor removal becomes a complex calculation, involving partition coefficients for each inhibitor and the dependence of these coefficients on temperature, pH, concentration of ethanol and salts in the biphasic system. For this reason, the level of collateral removal of inhibitors using the solvent in the first place only to remove the product as suggested by the other authors of the prior art, is hardly satisfactory in the use of hydrolytic broth as raw material, even more so because the natural concentration of inhibitors depends a lot on how to prepare this raw material, either by acid hydrolysis or enzymatic hydrolysis. The latter also applies to inhibitors in musts obtained by extracting carbohydrates from the raw material, such as sugar cane or beet, whose concentrations depend mainly on the temperature used in this extraction process.

In these cases, it will be necessary to employ a

12/91 pre-treatment of the must in an earlier stage of fermentation, while the technology proposed by the present invention intends, precisely, among others, to make such stage obsolete.

However, the simultaneous and satisfactory removal of inhibitors employing a solvent for in situ removal of fermentation product is not obvious and has not been described and much less suggested in the prior art. The removal of inhibitors can happen, but the removal sufficient for a good performance of the fermentation process depends, among others, on the flow rate of the solvent, temperature, pH, concentration of polar compounds, as well as the fermentation product, concentrations of other components as salts, which can promote or limit the removal of inhibitors in addition to contributing to total inhibition, even if the removal of the product is satisfactory for the liquid-liquid extraction processes of the product as suggested in the prior art.

It is observed that the use of the solvent as a cooling agent for the medium is not described by other authors. Conventional cooling, by external recycling of the fermentation medium through heat exchangers or cooling coils in the vat, is disadvantageous due to the formation of incrustations from the medium itself, between diamonds. These

13/91 fouling makes heat transport difficult, increases the risk of contamination, promotes biofilm formation and complicates heat exchange due to low thermal conduction and limiting available temperature gradients. Other disadvantages of such refrigeration equipment are thermal shock to the yeast; physical stress for yeast caused by the high flow of the medium in the pipeline and the need for external cooling equipment.

In order for an agent to function as a cooling agent, the specific enthalpy of the solvent at the moment of its introduction in the middle of fermentation, the flow rate of the solvent, and the integral of the heat capacity of the solvent over the temperature range between the solvent temperature and the final temperature of the medium, are determining factors in the cooling capacity of the solvent. In fact, the solvent only works as a cooling agent when it is introduced at a temperature below the temperature of the fermentation medium.

What is disclosed in the present invention is the complete refrigeration of the medium by the solvent, making a conventional secondary refrigeration system obsolete. For this purpose, the temperature and flow rate of the solvent need to be accurately calculated, predicting the generation of heat by metabolic activity and agitation of the medium. After all,

14/91 the temperature of the medium cannot be too low or too high to guarantee a good fermentation performance. This requirement makes use of the solvent as a cooling agent, allowing cooling by direct contact, eliminating heat exchange surfaces. Therefore, the use of the solvent for the purpose of cooling the medium was not foreseen, described or much less suggested in any state of the art document.

The international patent application WO 2009/042950, filed on September 26, 2008, in the name of LS9 INC and others and entitled: Reduction of the toxic effect of impurities from raw materials by extractive fermentation describes the production of ethanol and butanol by fermentation followed by in situ extraction and the agent is also recycled. However, in that document the solvent used is not defined.

The international patent application WO 2010/100642, filed on March 2, 2010 in the name of EYAL RES CONSULTANTS LTD and EYAL AHARON and entitled: Fermentation processes describes a method of producing at least one fermentation product and a derivative thereof, comprising among the steps, that of providing a fermentable aqueous solution comprising a fermentable compound. In that document, there is removal of inhibitors with

15/91 octanol adsorption and treatment suggestions with active carbon treatment with heat, base, ion exchanger and separation using membrane.

A disadvantage of the aforementioned document is the removal of inhibitors outside the vat, which requires the inhibition of the generation of inhibitors in the process of extracting carbohydrates from the raw material using low temperatures and / or a step of pretreatment of the must based on ligno- cellulose, which requires greater investment in equipment and less efficiency of the process in general.

The article by Grobben N. et al., Entitled

Production of acetone, butanol and ethanol (ABE) from potato wastes: fermentation with integrated membrane extraction published in Applied Microbal Biotechnology 39

494-498, in 1993 describes the production of butanol, acetone and ethanol (ABE) fermentation with article mentions solvent mentions starting from potato residues by extracting only the extractant. In addition and does not even suggest the use of an article entitled Improved acids published in the integrated membrane. The said application of biodiesel as an in-situ addition

Boudreau T.

ethanol-water said article not from the solvent.

M.

and collaborators separation using fatty

Process Biochemistry 41

980-983 in

2006, describes an improved ethanol-water separation

16/91 using fatty acids. The article describes the extraction of only the product and without using the solvent as a cooling agent.

In this case, it will be necessary to employ a step of pretreatment of the broth in a previous stage of fermentation for fermentation of wort based on hydrolytic broth, while the technology proposed by the present invention intends, among others, precisely to make such step obsolete.

As can be seen, no document of the state of the art describes or much less suggests approaches to combining the pretreatment of the wort with the removal of inhibitory product from the fermentation medium by liquid-liquid extraction. Therefore, this gap is the main subject of the present invention. Furthermore, there are no examples in the state of the art of refrigeration of the fermentation medium by the solvent, much less in combination with simultaneous product extraction and / or inhibitors. Liquid-liquid extraction can make fermentation a more efficient process, causing less stress to the yeast and resulting in lower energy costs.

Summary of the Invention

To solve the problems mentioned above, the present invention, based on extractive fermentation,

17/91 will provide significant advantages in relation to conventional fermentation processes, enabling an increase in its performance and presenting a more favorable cost / benefit ratio.

The present invention if extractive fermentative for amylaceous processes strongly refers to hydrolytic musts of raw material depleted carbohydrates with, among others, biotin, carbohydrates with a process derived from ligno-cellulosic, directly extractable, molasses high content of inhibitors as or similar musts for the purpose of in fermentation.

Specifically, the present invention refers to the use of a biocompatible solvent, such as biodiesel from soybean oil, corn, African oil palm (oil palm), macaúba among others, as an extraction agent for the extraction in-situ and simultaneous fermentation inhibiting product and / or components, specifically the fermentation product and inhibiting compounds present in the wort, and / or as a cooling agent in the fermentation medium, the latter in combination or not for the purposes of extraction product and / or inhibitors.

Fermentation productivity in terms of products such as ethanol, butanol, propanediol, 2-3 butanediol, acetone, enzyme, amino acids, such as glutamic acid, lysine and

18/91 monosodium glutamate, organic acids such as acetic acid, propionic acid, lactic acid, succinic acid, butanoic acid, raw material for bioplastics and other products, is limited by the concentration of the product itself, that is, there is an inhibition effect by product. In addition, fermentation of ligno-cellulose biomass or another type of biomass is limited by the chemical inhibition of the process due to inhibitors present in this raw material. These inhibitors, being the main furfural, HMF, phenolic compounds and carboxylic acids, inhibit fermentation in terms of microbial growth and production rate. The same effect occurs caused by inhibitors present in other types of must such as musts prepared with sugars directly extractable from the raw material, such as sugar cane or beet, when the temperature of the process of extracting carbohydrates from this raw material is higher, often resulting in higher extraction yields.

The final wort can also contain organic components, such as biotin, which inhibit more specific fermentations such as amino acid biosynthesis, such as glutamic acid, lysine and monosodium glutamate.

The present invention allows the removal of all these compounds from the fermentation medium, accelerating the

19/91 process. Thus, the technology proposed by the present invention will effectively enable the production of biofuels from biomass, that is, second generation fuels, as well as other products from hydrolyzed biomass, allows the use of higher temperatures in the extraction processes of carbohydrates. raw materials such as sugar cane and beet, resulting in higher yields from this process, and will enable and promote the use of musts in general that contain compounds with solvent-extractable inhibitory effects.

In the present invention, the wort pretreatment step and the product removal step are integrated. In this way, the ϊη-situ removal of product and other inhibitors from the fermentation medium with biocompatible solvent, without or with the use of the solvent as a refrigerant, in fact eliminates the need for an extra medium treatment unit and can increase the return on the total investment of the production process, both by increasing productivity and yield, and by decreasing energy costs and water used in the entire process.

The solvent itself can also be used to cool the fermentation process. Fermentation is an exothermic process, that is, heat is being generated during the conversion of the substrate. This heat can be

20/91 effectively removed from the fermentation medium using the solvent itself in a continuous system, and the solvent can be introduced into the vat with low internal energy after previous cooling and eventual storage. In this way, the fermentation temperature can be maintained around the optimum, due to the large heat exchange surface, resulting from the liquidity property of the solvent and the speed of mixing with the medium, resulting in lower temperature gradients in the vat and, consequently, increasing efficiency in general, both in terms of energy costs and in terms of cooling and maintenance equipment costs.

Brief Description of the Figures

The structure and operation of the present invention, together with its additional advantages, can be better understood by reference to the attached drawings and the following description:

Figure 1 shows the process of the present invention in a simplified form, where (A) is the wort input (substrate); (B) is an optional entry for the recycled microorganism; (C) is the solvent input; (D) is the solvent and product output; (E) is the departure of wine with yeast; (F) is the product and water mixture output; (G) is the output of clarified wine; (H) is the output of the purified product and

21/91 (I) is the output of waste and by-products.

Figure 2 shows the block diagram of a process with the technology proposed by the present invention.

Figure 3 shows the biocompatibility of biodiesel (4A) and castor oil (4B) and carbon dioxide production during fermentation without an organic phase (dotted line), with biodiesel (black circle) and with castor oil (white triangle) .

Figure 4 shows a schematic drawing explaining how the solvent has the capacity to absorb a certain amount of water, sugar concentration occurs during the extraction, leading to fermentation at a higher yield.

Figure 5 shows a bar graph with the absorption of water and ethanol in castor oil and biodiesel for various concentrations of ethanol in the aqueous phase.

Figures 6, 7 and 8 show the results for the

fermentation THE, B and C, where (THE) and the fermentation without inhibitors, without solvent; (B) and the fermentation with inhibitors, without solvent; and (Ç) and the fermentation with

inhibitors, with solvent.

Figures 9 and 10 show the concentration profiles of furfural, furfuryl-alcohol, vanilla and vanilla alcohol during fermentation without solvent and with solvent, respectively.

22/91

Figure 11 shows a viscosity comparison chart between castor oil and biodiesel for various temperatures.

Figure 12 shows a flow chart of an extractive fermentation plant according to the present invention.

Figure 13 shows a flowchart of an extractive fermentation plant with a vat, according to a first embodiment of the present invention.

Figure 14 shows a flowchart of an extractive fermentation plant with three vats, only the first vat keeping an organic layer in accordance with a second embodiment of the present invention.

Figure 15 shows a flowchart of an extractive fermentation plant for fermentation in batch or fed batch mode, with a vat according to a third embodiment of the present invention.

Figure 16 shows a flowchart of an extractive fermentation plant for continuous fermentation with cooling and removal of inhibitors in all vats, according to a fourth modality of the present invention.

Figure 17 shows an embodiment of the present invention of a schematic process with fluids and processing units from an extractive fermentation system. The process stages on the dotted line have been modeled.

23/91

Detailed Description of the Invention

Although the present invention may be susceptible to different modalities, it is shown in the drawings and in the following detailed discussion, a preferred modality with the understanding that the present modality should be considered an example of the principles of the invention and is not intended to limit the present invention to what has been illustrated and described here.

The present invention relates to an extractive fermentation process. More specifically, the present invention relates to an extractive fermentation process that employs a single solvent capable of removing products and inhibitors simultaneously, in addition to cooling the medium. In addition, the present invention relates to the use of the process.

The present invention reveals an extractive fermentation process with high productivity, which will be able to make the conversion of fermentable carbohydrates obtained by extraction or chemical hydrolysis into products maximum and fast.

What happens is that some compounds contained in pure hydrolysates, the main ones being furfural, HMF, phenolic compounds and carboxylic acids, inhibit fermentation in terms of growth and production rate.

The technology proposed by the present invention allows the in-situ removal of these components by

24/91 of a biocompatible extracting agent. This procedure to eliminate the inhibitory effect of these components will result in an increase in the productivity of the microorganism, viability, efficiency of the fermentation, and high stability of the process, as well as favoring a fast fermentation with a high rate of conversion of the substrate. In the case of extracellular products, in addition to the inhibitors present in the must, the fermentation product itself, which inhibits the process from a certain concentration, will also be removed by the solvent. For intracellular products, such as, for example, lipids, polyhydroxyalkanoates and various enzymes, the present invention applies to the removal of inhibitors from the wort and / or to the cooling of the fermentation medium.

Raw materials and inputs

The preferred raw material envisioned for the application of the technology involves hydrolysates of solid ligno-cellulosic material, starches, such as sugarcane bagasse, corn or cassava starch hydrolyzate. They may also involve the use of directly extractable carbohydrates, such as carbohydrates conventionally used in fermentation: molasses, cane juice, beet extract and molasses heavily depleted with a high content of inhibitors, such as biotin, in addition to similar, rich musts

25/91 in carbohydrates and for fermentation purposes or any combination or derivative thereof, hereinafter referred to as 'must'. Carbohydrates being the general name for substrates for glucose, fructose, xylose, arabinose fermentation, among others, such as sucrose.

For certain fermentations, the choice of the raw material for the wort or the concentration of carbohydrates in the wort is limited by the presence of natural inhibitors such as biotin.

One of the main advantages of the technology proposed by the present invention is precisely the removal of inhibitors present in the must, from the extract broth of the raw material, from the hydrolyzed liquor or from another source such as carboxylic acids, such as acetic acid, furan derivatives, such as furfural and hydroxymethyl furfural, and phenolic compounds, such as vanilla, coniferyl alcohol, catechol, among others. Natural compounds in the wort that exhibit inhibitory properties like biotin can also be removed by the proposed technology.

The present invention is based on the use of a biocompatible extracting agent hereinafter referred to as solvent ', during fermentation. The solvent used as the organic phase, specifically part of the specificity of the present invention involves a biodiesel selected from the

26/91 group comprising vegetable oils such as soybean oil, corn oil, castor oil, palm oil biodiesel and macaúba biodiesel.

Other biocompatible liquids also like long chain alcohols (alcohol used can be

5 oleic, phytol, isofitol, stearyl alcohol, alcohol cetyloco, octyldodecanol), acids fatty (acid ricinoleic , oleic acid, acid linoleic, acid linolenic, lauric acid, acid palmitic, acid palmitoleic, stearic acid) , etanoates (dodecyl acetate, 10 butyl dodecanoate), or other chain substances it was long how dibutylsebacate, di (2- -ethylhexyl) sebacate,

di (2dibutyladipate di (2-ethylhexyl) adipate ethylhexyl) phthalate, di (3,5,5-trimethylhexyl) phthalate, glycerol tridecanoate,

2-dodecanone dodecanal.

Preferably, a methyl ester or ethyl ester of a fatty acid, such as ricinoleic acid, is employed.

Ricinoleic acid (cis-12-hydroxyoctadeca-9enoic acid) is a long chain molecule, having a hydroxyl group at the C-12 position.

This hydroxyl group increases the polarity of the molecule favoring partition coefficients of inhibitors and fermentation products with polar characteristics. The double connection in position

C-9 favors lower viscosity. The source of the fatty acid can be an agricultural product such as castor oil that has a

27/91 85 to 90% percentage of ricinoleic acid esters.

The solvent has a higher boiling point than the product and volatile inhibitors, and its size is such that it limits diffusion and integration with the microorganism's cell wall, which would impair its viability and growth.

More specifically, the solvent has the following essential and favorable properties:

• Partition coefficients favorable to the components to be extracted (inhibitors and / or product);

• High selectivity for the components of interest (product and inhibitors) compared to other components (substrate and minerals);

• Immiscible in water and with negligible solubility in the aqueous phase;

• Absorption capacity of a certain amount of water, elevated by the fraction of ethanol in the solvent, which effectively concentrates the substrate in the aqueous phase, reducing its loss and thus increasing the yield of the product;

• Low toxicity (positive fact for the human element - company operators, for the environment and for the organism used in fermentation (bio-compatibility);

• Possibility of recovering the product from the solvent and

28/91 its regeneration;

• No formation of (stable) emulsions during fermentation and down-stream processing under conditions of low turbulence of the medium;

• Rapid separation of organic and aqueous phases;

• High chemical stability, regardless of the use of high temperatures in the recovery or other components such as acids present in the process;

• Low melting point and viscosity in the used temperature range, and boiling point above the used temperature range;

• Low cost and possibility of producing it on site;

Possibility of being used as fuel, biodiesel, after several cycles of use as a process solvent.

The solvent can be injected at low temperatures, in order to act in the cooling of the fermented medium, to maintain the desired temperature. The fermentation product is extracted by the solvent, which is removed from the bioreactor, and then rectified with removal of inhibitor products and contaminants, and continuously recycled, free of products and inhibitors. The diagram shown in Figure

29/91 illustrates the process in a simplified way.

The process of the present invention allows the use of high concentrations of carbohydrates, less inhibition by the inhibitors from the must and by the inhibitor product (ethanol, butanol, acetone, propanediol, 2-3 butanediol, organic acids, solvents, enzymes, amino acids, such as glutamic acid, lysine and monosodium glutamate, terpenes, raw material for bioplastics, among others) during fermentation and less energy requirement.

Products

The present invention reveals an extractive fermentation process with high productivity, which will be able to make the conversion of fermentable carbohydrates obtained by extraction or chemical hydrolysis into products such as: butanol, acetone, propanediol, 2-3 butanediol, xylitol, acids organics such as acetic acid, propionic acid, lactic acid, succinic acid, butanoic acid, citric acid, enzymes, amino acids such as glutamic acid, lysine and monosodium glutamate, terpenes, raw material for bioplastics, lipids, polyhydroxyalkanoates, the yeast itself, or other organic products or any combination thereof, hereinafter referred to as 'product'.

30/91

The focus of the present invention is ethanol, but the characteristics of two-phase systems involving the distribution of a component with polarities less than the polarity of ethanol, which is the case for, for example, butanol, acetone and most amino acids and other products mentioned above, will inevitably show a good performance in terms of extraction, even better than ethanol itself. This is because, compared to ethanol, the partition coefficients of components with less polarity are higher, and as the selectivity for these components is higher, then the extraction for the organic phase will be favored.

Note that several products mentioned can be produced as intracellular products, extracellular products, or a combination thereof, depending on the microorganism (s) employed and their ability to transport the product through the cell wall.

For the production of biomass or intracellular products such as, for example, in many cases lipids, polyhydroxyalkanoates, various enzymes and amino acids, the present invention applies only to the removal of inhibitors from the wort and / or to the cooling of the fermentation medium.

What happens is that some compounds contained in

31/91 pure hydrolysates, the main ones being furfural, HMF phenolic compounds and fermentation in terms of

Similar effects of carboxylic acids inhibit growth and production rate.

inhibitions occur for musts prepared by extracting carbohydrates from the raw material, for example, sugar cane or beet, a process in which high temperatures promote extraction, but cause generation

Equipment

Extractive fermentation can be continuous, batched continuously.

high yields of such inhibitors.

be run in a batch or fed way and

The fermentation vat, the execution of the process for the solvent, with the same possibly fermentative, should prefer the integrated in the deflectors, counting on a pressure to overcome the weight of the column of the content phases of the vat.

aqueous volume volume of the vat should be greater and the organic phase of these phases can be between 1: 0.01 to 1:20.

Other equipment used to recover the product and with preference necessary to have an entrance to the bottom of the agitator or rather high for aqueous and organic than the sum of the phase, the reason being adjusted to necessary for the recycling of any reason the production and solvent can

32/91 be: medium mixer, absorption column, decanter, distillate cascade, distiller, rectifier, centrifuge, valve or equipment for bleeding a fraction of yeast cells, equipment for cell reactivation of eventual recycled yeast, heat exchangers , and others

Figure 2 represents the block diagram of a process with the technology proposed by the present invention. Biological process

The biological process for converting substrate to product involves fermentation using a microorganism such as, but not limited to, yeast, such as Saccharomyces cerevisia or bacteria, such as Zymomonas mobilis, genetically modified or not, among others or in any combinations thereof, hereinafter called 'yeast'. Yeast consumes carbon sources such as sucrose, glucose, fructose and / or xylose, arabinose or any combination of these, and converts them into products such as ethanol, butanol, acetone, propanediol, 2-3 butanediol, xylitol, organic acids such as acid acetic, propionic acid, lactic acid, succinic acid, butanoic acid, citric acid, enzymes, amino acids such as glutamic acid, lysine and monosodium glutamate, terpenes, raw material for bioplastics, lipids,

33/91 polyhydroxyalkanoates, yeast itself or other organic products or any combination thereof. The process can be carried out under anaerobic conditions or using aeration, depending on the microbiological needs of the yeast used to produce the desired product. Any inhibitors present in the fermentation medium are removed from the medium by the solvent, either completely or partially. Extracellular products, often an inhibitory factor from certain concentrations as in the case of ethanol for Saccharomyces cerevisiae, are also partially removed from the medium by the solvent. In this way, the microorganism's performance in terms of growth rate, viability, volumetric and specific productivity and yield can be optimized, as well as the energy use of the total production process including product recovery and purification and product recycling. solvent.

Fermentation is an exothermic process, that is, heat is being generated during the conversion of the substrate. This heat can be effectively removed from the fermentation medium using the solvent flow itself in a continuous system, and the solvent can be introduced into the vat with low internal energy after previous cooling and eventual storage. In this way, the temperature of the

34/91 fermentation can be maintained around the optimum, with lower temperature gradients in the vat and with more efficiency in general, both in terms of energy costs and in terms of equipment and maintenance costs.

The extractive fermentation process according to the present invention comprises the following steps:

a) Entrance of a must rich in carbohydrates in the bioreactor;

B) Input in an inoculum: in the bioreactor; ç) Injection of solvent at the bioreactor; d) Removal of solvent of bioreactor;

e) solvent recovery;

f) Purification of the vapor phase;

g) Removal of the aqueous phase from the bioreactor;

h) Recovery of the microorganism.

In the present invention, the wort detoxification step, provided for in cases as described above, and removal of the product from the fermentation medium, are integrated. In this way, the in-situ removal of product and other inhibitors from the fermentation medium with biocompatible solvent actually facilitates the concentration of the product, eliminates an extra must treatment unit and reduces the total cost of ethanol production, both in terms of increase in productivity and in terms of minimizing

35/91 energy and water costs used in the integral process.

Step (a) is to feed the Joioreactor with must from hydrolytic processes of ligno-cellulosic raw material, starch, directly extractable carbohydrates, molasses strongly depleted with a high content of inhibitors, such as biotin, among others or similar musts rich in carbohydrates , where the content of the fermenter is composed of an aqueous and an organic phase.

Step (b) involves the entry of an inoculum into the bioreactor, by recycling, where a small fraction of the microorganism is added, or by reactivation, using aerobic treatment, acid treatment and / or nutrient feeding. This step may or may not be optional, and this choice depends on the type of product, being indicated in the production of extracellular products and not applied for the production of intracellular products, where the yeast cell wall will be perforated or destroyed to obtain products such as lipids , enzymes, polyhydroxyalkanoates, among others.

Step (c) is injection of the solvent into the bioreactor, which occurs from the bottom of the bioreactor or through holes in the propellants and / or deflectors. The solvent is injected into the fermented at low temperatures, varying between -10 and 25

36/91 ° C. The solvent may contain sterile oxygen to assist in the maintenance mechanisms of the microorganism. The density of the solvent is relatively low, which makes it flow naturally to the upper part of the bioreactor, where an organic layer is formed, with another fraction partially dispersed in the aqueous phase. On the other hand, agitation of the medium can result in greater mixing of the solvent in the fermentation medium, increasing the heat exchange surface and the time this surface is exposed to the medium. The solvent therefore functions as an in-situ product extractor, inhibitors and / or heat from the fermentation medium.

In step (d), the solvent is removed from the bioreactor, with limited agitation of the medium, preferably from the upper part, where it is consequently richer in product, by decantation or centrifugation, after or during fermentation.

In the next step (e) the solvent is sent to a recovery unit. As the solvent has less volatility than water and product, the latter can then be recovered by heating and vacuum evaporation and can be reintroduced into the bioreactor.

Volatile inhibitors can also be removed at this stage. For products sensitive to high temperatures, such as

37/91 in the case of several enzymes, or products of low volatility, as in the case of amino acids, this step can be modified to prevent damage to its molecular structure or to allow its purification. In this case, other separation methods can be applied, such as the use of a second solvent with greater volatility, ultrafiltration, application of electromagnetic fields, reduction of the product's solubility, for example, by application of low temperature or modification of the solvent polarity, among others. .

The steam phase, composed of fermentation product and water, is sent to purification sectors, such as separation or distillation and rectification columns (step (f)).

The aqueous layer is pumped from the bioreactor like wine (step (g)) and, optionally, the microorganism can be recovered by centrifugation, ultrafiltration, membrane separation or any method of separation and concentration of cells, in step (h).

As mentioned earlier, for intracellular products, step (h) can be used for the concentration of the cells. For extracellular products, for example, in the case of ethanol produced by yeast, most of the inoculum can be reused, while the light phase is sent to the final treatment sector.

38/91 as clarified wine, which may be the distillation or separation and rectification columns. The final treatment sector generates the product and residues or by-products, such as vinasse, phlegm and oil-fusel, for example.

The following benefits of the technology proposed by the present invention can be mentioned:

• Reduce the energy cost of the entire fermentation and product recovery process;

• Control of inhibition caused by the fermentation product and / or by-products originating from the must, resulting in greater fermentation efficiency and greater volumetric productivity and process stability;

• Greater viability of microorganisms, high general microbial stability and lower risks of application of genetically modified microorganisms;

• Use of higher concentrations of sugar, implying less need for water, energy and less waste at the end of the process (for example, for the vinasse of the ethanol production process, which is currently 11 to 13 L / L EtOH, can become 3-5 L / L EtOH);

• Application in continuous or batch fed process. During continuous fermentation, an additional advantage is the reduction of substrate loss: the flow rate of

39/91 solvent can be used as an extra control factor to eliminate oscillatory behavior in the bioprocess;

• Implicit cooling of the fermentation medium by the solvent, introducing it into the vat with low internal energy, eliminating the need for expensive external refrigeration instrumentation, which introduces an unnecessary stress factor to the microorganism;

• Use of the solvent can also result in lower concentrations of impurities in the vinasse, promoting the option of recycling vinasse in the process, resulting in less amounts of water used in the process and less energy required to treat the vinasse;

• The solvent will act as a buffer for inhibitors, ensuring for a long time a limited amount of inhibitors also in the aqueous phase, which can act as a natural antibiotic, because contaminating organisms such as viruses and bacteria are exposed to inhibitory effects similar to the effects on the microorganism producer (eg yeast);

• Ensuring for a long time a limited amount of inhibitors in the aqueous phase can increase the fermentation yield because for certain concentrations of furfural, the yield of ethanol on glucose may be higher due to the balance in the metabolic pathway of NOTHING (P) and inhibition

40/91 repressed or induced expression of key enzymes.

• Use of a bio-based renewable solvent, such as the biodiesel suggested in this invention, can decrease the cost of the extractor, as it is produced on a large scale and the option of on-site production by the plant itself and, after several recycling in the fermentative process, can be reused as fuel.

As an example, the present invention can be applied in conventional alcoholic fermentation and, more preferably, in second generation alcoholic fermentation. However, the technology proposed by the present invention can also be applied in fermentation for the production of other products with properties suitable for solvent extraction, being, for example, acetone, butanol, propanediol, 2-3 butanediol, acetic acid, lactic acid, propionic acid, butanoic acid, succinic acid, enzymes, amino acids, such as glutamic acid, lysine and monosodium glutamate, terpenes, raw material for bioplastics, among others or any combination of these. The process of the present invention also makes it possible to recover by-products from hydrolyzed broth such as furfural and phenolic compounds, which have economic value. In hydrogen production specifically, the composition of about 35-50% ethanol-water in the

41/91 solvent from the fermenter is ideal for steam reforming from ethanol by catalysis, eliminating a distillation step in this process.

As can be seen, the differential of the process of the present invention is in the use of biodiesel, as a biocompatible extraction agent for the in-situ and simultaneous extraction of fermentation inhibiting product and components, specifically the fermentation product and inhibiting compounds present in the must, coming from the biomass hydrolyzate, from the process of extracting carbohydrates from the raw material, or directly from the raw material. Another differential is the use of the solvent as a refrigerant in the fermentation process, in combination or not with the use as an extraction agent for inhibitory components.

Other extracting agents, toxic or biocompatible, have been explored in the state of the art, in-situ, or outside the vat, but not the biodiesel proposed by the present invention, much less for the purposes of combining the extraction of inhibitory components and / or refrigeration.

The differential in the use of the preferred solvent, biodiesel, in the process of the present invention is great, since it is renewable and its production sustainable. In addition to being a suitable extracting agent in terms of biocompatibility,

42/91 extractive properties and physical properties, such as low viscosity.

The preferred solvent has implicit advantages such as on-site production from vegetable oil and reuse as a fuel after several recycles, as proposed in the process of the present invention.

EXAMPLE I

Comparison of vegetable oil and biodiesel based on vegetable oil as an organic solvent for the extraction of in-situ fermentation inhibitors in hydrolyzed bagasse.

Castor oil and its ethylester, hereafter indicated as biodiesel, are compared in terms of biocompatibility, partition coefficients for various inhibitors, substrates and ethanol, viscosity and water absorption. Water absorption can be significant by concentrating the substrate in the aqueous phase, on the one hand, but diluting the fermentation product in the organic phase, on the other hand. Ferments of synthetic hydrolytic liquor are also carried out to verify the expected increase in the performance of solvent-mediated ethanol.

Biocompatibility

The biocompatibility of the solvents (castor oil and biodiesel based on this oil) was determined using four Erlenmeyer flasks (125mL), equipped with

43/91 swan for carbon dioxide outlet. The swan neck contained approximately 2 mL of sulfuric acid for drying the exhaust gas. The fermentation medium was composed of 10 g / L peptone, 10 g / L yeast extract, 2.5 g / L K2HPO4 and 60 g / L glucose and was autoclaved at 121 ° C for 15 min. The inoculum medium was similar and an industrial strain of Saccharomyces cerevisiae (Santa Adélia), previously grown in petri dishes with agar-agar of the same composition, was used. After 12 hours of incubation, 50 ml of the inoculum was added to each flask. Two flasks were also added to 10 ml of one of the solvents. The weight of each bottle was monitored during fermentation, using an analytical balance (Scientech SA210), indicating the loss of carbon dioxide. The variation of the weights of the bottles with and without organic solvent were compared to verify the biocompatibility of the solvents. Figure 3 shows the weight profile for fermentation with and without solvent. It can be seen that there is practically no difference between the two processes, proving the biocompatibility of both solvents.

At the end of fermentation, all bottles with extractive solvents had a loss of 3% less carbon dioxide per liter than bottles without solvent, but as there was a consistent delay in the weight loss profile of

44/91 bottles with organic solvent, this may be an indicator that there was an accumulation of carbon dioxide in the organic layer, instead of the lack of biocompatibility.

Water and ethanol absorption

The solvent has the capacity to absorb a certain amount of water, which rises as the fraction of ethanol in the solvent rises, which effectively concentrates the substrate in the aqueous phase as shown in figure 4. Since the microorganism used in fermentation converts the substrate until it reaches a certain concentration at the end of the fermentation or at the end of a continuous fermentation, the total loss of the substrate, in terms of the absolute amount, decreases, that is, the fermentation yield increases.

The absorption of water in the solvent was determined for various concentrations of ethanol in the aqueous phase (0 to 200 g / L). For both solvents, biodiesel and castor oil, a two-phase system was built from 5 grams of Milli-Q water, 5 grams of organic solvent and a quantity of ethanol.

The system was kept in a water bath at 34 ° C without stirring for 48 hours to establish a complete balance. After this period samples of the organic phase were transferred to a rotary evaporator flask,

45/91 with a known weight, and kept under low pressure (0.5 bar) at 35 ° C in a rotary evaporator (Marconi MA120). The weight of the flask with an analytical balance was determined at 5-minute intervals until the weight stabilized, meaning the complete removal of ethanol and water previously present in the organic phase. A sample of the aqueous phase was taken, filtered with a 0.22 filtron sample (Millipore), and the ethanol concentration was analyzed with HPLC, equipped with an auto-sampler (Varian 9095), a UV detector (Varian 9095), a refractive index detector (Varian RI-4), a binary pump (Varian 9010) and using an Aminex HPX-87H separation column (Bio-Rad, 7.8 * 300 mm), with a Biorad micro-guard Cation-H125-0129 pre-column, at 25 ° C, and 5 mM H2SO4 as the mobile phase at 0.7ml / min. The concentrations of ethanol and water in the organic phase were calculated from the mass balance of the biphasic system, using the known total mass of water and ethanol present in the system and the concentration of ethanol in the aqueous phase.

As a result, it was observed that biodiesel absorbed significant amounts of water and on average twice as much water as ethanol. There is no constant relationship between absorbed water and absorbed ethanol, as suggested by Malinowski et al. (1993) for oleic alcohol. In Figure 5,

46/91 observe that with the increase in the concentrations of ethanol in the aqueous phase, both organic solvents absorbed more ethanol and water and the ratio between water and ethanol absorbed decreases. Biodiesel absorbed 2.4 to 6.8 times more water than 5 castor oil, while ethanol absorption was approximately the same for both organic solvents.

Although a clear relationship between the water-ethanol ratio in the organic phase and the concentration of ethanol in the aqueous phase could not be concluded from the experimental data, an increase in the ethanol / water absorption ratio can be seen as an increase in the concentration of ethanol, especially for biodiesel.

This means that with a higher concentration of ethanol in the aqueous phase, ethanol is relatively more absorbed. This is much more significant in the case of biodiesel, where both water and ethanol fractions absorbed are greater with increased ethanol concentration. While the difference in water absorption is considerable between biodiesel and castor oil, the difference in absorption of 20 ethanol lies more within the deviation patterns of the experiments.

It should be noted that, with another product, for example butanol, other quantities can be expected.

On the whole, biodiesel has an advantage over the

47/91 castor oil for greater water absorption and can thus effect the concentration of substrates in the aqueous phase during fermentation, which leads to a reduction in the loss of substrates in continuous fermentations or an increase in production and yield for batch fermentation. and batch fed.

Partition coefficients

The partition coefficients (see equation below) were determined for hydrolysis by-products: furfural, 5hydroxymethyl-furfural, vanilla, syringaldehyde, coniferyl aldehyde and acetic acid. The partition coefficients for glucose, sucrose, fructose, xylose and for glycerol and ethanol were also determined.

W aq m = —— p (partition coefficient) ^W org where: m _p is the partition coefficient (-), w _org : the mass fraction of the component in the organic phase (-), w _aq : the mass fraction of the component in the aqueous phase.

All coefficients were determined for six different concentrations of ethanol in the aqueous phase ranging from 0 to 200 g / L. For each of these components and for each concentration of ethanol, the biphasic system was established in an Eppendorf tube containing 1 gram of a solution of

48/91 lg / L of the component, except for ethanol, in Milli-Q water as the aqueous phase and 1 gram of biodiesel as the organic phase. The biphasic system was kept closed and not stirred for 48 hours in a water bath at 34 ° C before centrifuging for 5 minutes at 2000 RPM. A sample of about 500 mg of the organic phase was added to 7 00 mg Milli-Q water, forming a second biphasic system for the reverse extraction of the component. This reverse two-phase system was also maintained without stirring for 48 hours at 34 ° C. The aqueous phases of the original two-phase and reverse two-phase systems were analyzed on an HPLC equipment. HPLC measurements for glucose, xylose, glycerol and ethanol were made using an ion exchange column (BioRad, Aminex HPX-87H 7.8 * 300mm), at 25 ° C and 5 mMH ₂ SO ₄ as a mobile phase at 0.7ml / min; the detection was made by a refractive index detector (Varian RI-4). Acetic acid and lactic acid were separated with the same column, but quantified with a UV detector (Varian 9095) at 210nm.

Furfural, furfural 5-hydroxymethyl, vanilla, syringaldehyde and coniferylaldehyde, were retained using a C-18 column (pBondapak, 10 pm, 3.9 * 300nm) and quantified with detection

UV (Varian 9095) in wave lengths that were most appropriate for each compound individually. A solution of acetonitrile in water was used as an eluent (Iml / min).

49/91

The eluents were prepared with Milli-Q water, filtered with 0.45 pm filter (Millipore).

From the mass balance on the original and reverse biphasic system, the partition coefficients for each component and for each concentration of ethanol in the aqueous phase were determined. All measurements were performed in duplicates or triplicates and the results are summarized in Table 1.

The partition coefficients depend a lot on the amount of ethanol present in the two-phase system. Partition coefficients for HMF and furfural increase with increasing ethanol concentration. However, for vanilla, seringaldehyde and coniferyl aldehyde, the partition coefficients tend to decrease with a higher concentration of ethanol in the system.

It can also be seen that for these components there are maximum partition coefficients for concentrations of ethanol within the chosen range of 0 to 200 g / L and, more specifically, for concentrations below 80 g / L. Coincidentally, this value is around the concentration of ethanol which is seen as the concentration at which the inhibition by ethanol is significant, at standard fermentation temperatures according to the work of Rivera et al., 2006 and incorporated here by reference in

50/91 in its entirety. During ethanol fermentation, the highest biodiesel partition coefficients can then be exploited without prejudice to inhibition by ethanol.

For sucrose, glucose, fructose and xylose substrates, the partition coefficients are relatively small, even negligible, with values of 0.04 for glucose and fructose up to 1.2 · 10 ~ ³ for xylose. The selectivity for sugars is low and the partition coefficients for the studied substrates are negligible and still decrease for greater amounts of ethanol in the system.

Partition coefficients of glycerol, acetic acid and lactic acid are relatively low and no greater than 0.3. For the extraction of these components, biodiesel does not perform optimally. However, the optimal concentration of ethanol in the aqueous phase for maximum exploitation of partition coefficients of glycerol, acetic acid and lactic acid is below 80 g / L.

The same experiment was carried out, with biodiesel being replaced by castor oil. From the mass balance over the two-phase systems, the original system and the secondary system being described above for biodiesel, the partition coefficients were determined for each component and for each concentration of ethanol in the aqueous phase.

51/91

The measured partition coefficients are shown in Table 1. Until ethanol concentrations of 100g / L in the aqueous phase, biodiesel exceeds the performance of castor oil as an extracting agent for all measured fermentation inhibiting components, being HMF, furfural, vanilla, syringaldehyde and coniferyl aldehyde acetic acid and lactic acid. Apparently, all of these compounds have more affinity for biodiesel than castor oil. However, this affinity depends heavily on the amount of ethanol present in the two-phase system. The partition coefficients for HMF and furfural increase with the increasing concentration of ethanol, at least up to 100 g / L of ethanol in the aqueous phase.

This is also the case for vanilla, syringaldehyde and coniferylaldehyde partition coefficients in the case of castor oil as an organic solvent, but in the case of biodiesel, each partition coefficient tends to decrease faster with the increasing amount of ethanol in the system. It can also be seen that the maximum values of the partition coefficients of these compounds are reached for concentrations of ethanol within the chosen ethanol concentration range of 0 to 200g / L, but that the concentrations of ethanol to reach these maximum values are not the same for biodiesel and castor oil. For biodiesel as a solvent

52/91 organic, maximum partition coefficients can be found for lower ethanol concentrations than for castor oil. Coincidentally, for biodiesel, these maxims are found in ethanol below 5 of 80 g / L, a concentration at which ethanol becomes an inhibitor for fermentation at most standard fermentation temperatures (Rivera et al., 2006). During the fermentation of ethanol, the extractive properties of biodiesel can thus be exploited to the fullest while the inhibition by ethanol can still be kept low.

For sucrose, glucose, fructose and xylose substrates, no significant difference can be seen between the partition coefficients for biodiesel and castor oil, but in all cases the partition coefficients 15 are relatively small, with measured values ranging between 0.04 for glucose and below 1,210 ³ for xylose.

In addition, partition coefficients for the studied substrates decrease with an increased amount of ethanol in the system, for both biodiesel and castor oil.

Thus, as desired, the selectivity for sugar is negligible in both cases, biodiesel and castor oil as an organic solvent. The use of any of these solvents during fermentation will therefore not lead to any significant decrease in sugar concentrations

53/91 in the aqueous phase

In general, the optimal concentration of ethanol in the aqueous phase for maximum exploitation of the partition coefficients for glycerol, acetic acid and lactic acid can be found below 80 g / L

Table 1: Partition coefficient of various compounds for

biodiese (BD) and Oil in Castor bean (CO) at presence in concentrations many different of ethanol . at water phase 3 dehyde aldehyde Ethanol HMF furfural vanillin syring al coniferyl Sucrose glucose fructose xylose glycerol acetic ac lactic aci Ethanol 0 BD 0.33 1.8 3.1 1.6 18 0.018 0.034 0.035 0.0031 0.093 0.071 0.132 - CO <0.01 0.37 1.1 0.5 2.5 0.025 0.040 0.040 0.0026 0.044 0.006 0.0019 . 37 BD 0.26 2.1 3.7 1.8 28 0.018 0.033 0.034 0.0018 0.068 0.127 0.096 0.12 CO <0.01 0.45 1.0 0.4 2.6 0.0092 0.027 0.027 0.0032 0.023 0.048 0.0024 0.09 74 BD 0.17 2.3 2.8 1.4 29 0.0073 0.022 0.022 0.0012 0.084 0.258 0.070 0.12 CO 0.012 0.85 1.5 0.6 3.8 0.011 0.024 0.024 0.0038 0.021 0.068 0.0042 0.11 110 BD 0.40 2.4 1.8 0.97 22 0.0046 0.020 0.023 0.0014 0.041 0.304 0.065 0.12 CO <0.01 1.5 5.4 1.98 16 0.0061 0.019 0.019 0.0034 0.018 0.076 0.0084 0.12 149 BD 0.49 2.5 1.5 0.86 18 0.0093 0.020 0.021 0.0014 0.054 0.337 0.071 0.15 CO 0.15 1.4 4.5 1.84 13 0.0075 0.022 0.019 0.0022 0.018 0.104 0.013 0.14 183 BD 0.61 2.5 1.4 0.81 15 0.0022 0.014 0.014 0.0016 0.038 0.346 0.068 0.15 CO 0.19 1.4 4.1 1.72 9.7 0.0058 0.017 0.018 0.0018 0.016 0.102 0.013 0.17

Fermentations

Fermentations were performed to illustrate the technical feasibility of the present invention. Furfural, acetic acid and vanilla were chosen as inhibitors representing the main groups of inhibitors: furans carboxylic acids and phenolic compounds.

These inhibitors were introduced in the pre-inoculum in the

54/91 inoculum and in the final fermentations in the furfural: acetic acid: vanilla ratio of 1: 2.2: 0.6 g / L, which is a representative ratio for hydrolytic broth. In addition to these inhibitors, the media were prepared with 10 g / L of peptone, 10 g / L yeast extract, 2.5 g / L of K ₂ HPO ₄ and 110 g / L of glucose, sterilized in an autoclave at 121 ° C for 15 minutes. The incubation of the inoculum was done in Erlenmeyer, using an agitator (Tecnallab TE420) at 34 ° C and 150rpm and monitoring the optical density every 12 hours, using a spectrophotometer (Beckman Coulter DU 640) at 600nm.

The biomass of the final fermentations was determined by the dry mass of each sample, each of which was washed twice with milliQ water. These final fermentations, carried out using an industrial Saccharomyces cerevisiae strain (Usina Santa Adélia) in temperature ranges between 30 and 37 ° C and pH between 4.0 and 4.5, were distinguished as follows:

A. Fermentation without

B. Fermentationwith

C. Fermentationwith inhibitors, without solvent inhibitors, without solvent inhibitors, with solvent

The volumes of fermentations A and B were 1L of medium.

For fermentation C, a total of 400mL of the solvent was added to 800mL of medium. Before fermentation, the solvent was washed 4 times in batches of each 500mL each

55/91 with 2500mL of distilled water at 25 ° C, using compressed air bubbling with a flow of 0.2 L-min ¹ for 2 hours. With this method, air bubbles take a micro-layer of clean water to the organic phase, at the top of the biphasic system. On the surface of this phase the bubble breaks, spreading the water on the surface of the biodiesel layer and then submerges, taking impurities to the lower aqueous phase. Between each washing cycle, the solvent was left for two hours before draining the aqueous layer. Samples of this aqueous phase were analyzed with HPLC to ensure complete removal of traces of /

glycerol, ethanol and other impurities. After repeating the washing process, the solvent was emptied, dried and clarified on a rotary evaporator (Marconi MA 120) at 55 ° C and 0.5 bar.

Dry yeast mass and concentrations of substrate, inhibitors and product (ethanol) were determined using samples obtained during fermentations. Figures 6, 7 and 8 show the results for fermentation A, B and C.

Figures 9 and 10 show the concentration profiles of furfural, furfuryl-alcohol, vanilla and baunyl alcohol during fermentation without solvent (Figure 9) and with solvent (Figure 10). The result of these fermentations shows the feasibility of the process invention, which can be

56/91 summarized as follows: comparing the fermentation without inhibitors (A), which took 24 hours, with the fermentation with inhibitors and without solvent (B), which took 90 hours to complete, it appears that there was a almost 4-fold increase in fermentation time.

Thus, the introduction of solvent in the fermenter with inhibitors, reduced the total fermentation time, which returned to 24 hours, that is, four times less, approximately. Correcting the aqueous volume to 20% less, these amounts correspond to a reduction of fermentation time of 67%.

In addition to the detoxification of the medium by the solvent, natural detoxification occurs via the metabolic pathway of yeast. As for furfural and vanilla, the microbiological detoxification mechanism works by reducing furfural to furfuryl alcohol and vanilla to baunyl alcohol. At first, furfuraldehyde is reduced at a rate that decreases when furfuraldehyde concentration becomes lower. Ά vanilla reduction, on the contrary, is slower or slower at first, and after the complete conversion of furfural, its rates increase until the whole vanilla has been reduced.

Fermentation with inhibitors, but without solvent (Figure 9) remains in the lag phase until furfural and vanilla are

57/91 converted to less toxic levels, in this case after 60 hours. Thereafter, fermentation continues in a conventional manner for the same 24 hours, as seen for fermentation without inhibitors (A).

In the presence of the solvent (Figure 10), alcoholic fermentation begins right after the introduction of yeast to the medium. In this case, the inhibitors are distributed between the aqueous phase and the organic phase. The concentration of inhibitors in the aqueous phase is therefore lower than in fermentation B, and yeast which is only susceptible to inhibitors in this aqueous phase, has a lesser need to detoxify the medium to maintain a viable fermentative performance.

In the presence of biodiesel, the same pattern of reduction can be seen as was the case for fermentation without solvent, with the exception of two main observations. First, there is a delay phase of 5 hours before furfuraldehyde starts to be significantly reduced, while the reduction of inhibitors in fermentation B begins directly after the introduction of the inoculum into the medium. Although the inoculum grew under equal inhibitory conditions for both fermentations, the yeast had to again induce its metabolic capacity to reduce furfural, but it was triggered to do so later in

58/91 fermentation C and to a lesser extent. The lower concentrations of furfural, vanilla and acetic acid in the aqueous phase may have reduced the need for an immediate reaction to the presence of these toxic compounds. Second, the maximum specific rate of reduction of furfuraldehyde, calculated from the sum of the reduction in the aqueous phase and in the organic phase, is higher in the biphasic system (0.27 gg “ ¹ -h ' ¹ ) than in solvent-free fermentation ( 0.05 gg ' ¹ -h ¹ ). For vanilla, a similar difference is seen: 0.20 gg ^ -h ' ¹ specific reduction versus 0.03 g * g ^_1 ' h ^_1 ; . Apparently, the lowest level of synergistic inhibition in the presence of an organic phase allows a more effective reduction of furfural and vanilla. Thus, while in the presence of an organic phase the concentration of furfural to which the yeast is exposed (0.9 g ^-1 in the aqueous phase) is lower than in fermentation without the organic phase (2 g'l ^_1 ) the rate of specific reduction is five to six times higher.

Therefore, for hydrolyzed liquor fermentation, keeping the inhibitor concentration limited is important both for the yeast growth rate and for the faster reduction of the inhibitors and thus for the productivity of the fermentative process in general.

The high rate of furfural reduction resulted in a total reduction of furfural within 13 hours, well before the end of

59/91 fermentation. This has implications for ethanol yield compared to glycerol, which is higher while furfural is present in the medium at low concentrations. On the other hand, the specific growth rate and specific ethanol production rate suffer in the presence of inhibitors.

Concentrations of acetic acid are not shown, but have increased slightly during fermentations. Table 2 summarizes the different fermentation parameters for each fermentation batch.

Table 2: Initial concentrations of inhibitors and parameters

fermentation following order: maximum growth specific, reduction in furfural and vanilla and rates production of ethanol, rate maximum volumetric production of

ethanol, ethanol, glycerol and biomass production.

furfural gr ¹ vanillin gr ¹ acetic acid gr ¹ biodiesel% | lmax g-gh ' ¹ Qf max ggh ' ¹ Qv max gg-h ' ¹ dEtOH max ggh ' ¹ ^ EtOH max gT ¹ h * ¹ Yeioh gg ' ¹ Yglycerol gg ' ¹ Ybiomass gg ' ¹ 0 0 0 0 0.47 - - 1.7 3.4 0.46 0.040 0.043 2 1.2 4.4 0 0.09 0.05 0.03 1.2 3.0 0.41 0.014 0.018 2 1.2 4.4 33.3 0.13 0.27 0.20 1.3 4.1 0.45 > 0.026 0.045

In contrast, it has been shown in the prior art that the reduction in furfural rates increased with higher furfural concentrations up to 4 g'l ^-1 of furfural. For even higher concentrations, the reduction rates decreased

60/91 as presented in the work by Palmqvist et al., In 1999 and incorporated here in its entirety by reference.

Probably, the effect of the synergistic toxicity of vanilla and acetic acid in combination with furfural reduces the furfural concentration, for which the reduction rate is maximum. An important implication of this observation is that for fermentations carried out in batch fed regime, with must containing inhibitors such as furfural, regardless of the application of an organic phase, the conversion rates of furfural and vanilla may be more favorable than for fermented batch regime. , and the regulated feeding in the fed batch regime allows to keep the inhibitor concentrations constant, but low. However, for the viability of a continuous fermentation, growth inhibition may be the real limiting factor instead of the maximum conversion rate of the inhibitors, even though the conversion rate will be lower than for batch or batch fed fermentations as noted also by Horvath et al., 2001, incorporated here in its entirety by reference.

On top of that, before reaching the biomass 'washout' conditions in continuous fermentations, low substrate conversion levels will cause loss

61/91 of the same and low levels of the ethanol production rate. Thus, especially for continuous fermentations, the proposed two-phase system is a promising option for increased productivity and yield with less dependence on inhibition reduction rates by the yeast itself.

Therefore, the fermentation process of the present invention can be optimized to obtain maximum yield, optimizing the flow of fed substrate and flow of the organic phase through the system, and in the case of batch or batch fermentation, keeping in optimized level the concentration of inhibitors present in the medium, until the end of fermentation.

Viscosity

For both solvents, biodiesel and castor oil, viscosity measurements were made using a Physica MCR301 rheometer (Anton Paar GMbh, Graz, Austria), equipped with a parallel steel plate (75mm in diameter, 0.5mm opening). The measurements for mounting the flow curve were performed in triplicates with the shear rate ranging from 0 to 300s ^_1 . A program of up-down steps (up-down-up) was carried out in order to evaluate the thixotropy of the evaluated product.

The flow behavior of biodiesel and castor oil was modeled in order to obtain rheological parameters

62/91 (shear stress (σο), consistency index (K) and fluid behavior index (η)), according to ο Herschel-Bulkley model and incorporated here by reference in its entirety, of which the viscosity.

The temperature effect on the viscosity of biodiesel and castor oil was studied within the range of 5 to 40 ° C, using a temperature range of 5 ° C between each measurement. The results of this temperature effect on viscosity were evaluated according to the Arrhenius equation, and incorporated here by reference in its entirety.

^σ-σ ° ⁺ ^ · Ϋ (Herschel-Bulkley model) (E A 1 ln (77) = ln (77 ₀ ) + '(Arrhenius equation)

Where 11: viscosity (Pa's), E _a : activation energy for viscous fluid (Jmol ^_1 K ^_1 ), T: temperature (K).

For both biodiesel and castor oil, the shear stress showed a strong linear relationship with the shear rate with zero intercession on both axes and an average regression of 0.9999 for all temperatures. Consequently, there is no shear stress ( ₀ ) and the flow behavior index (n) is equal to 1, leading to Newtonian behavior with an index of

63/91 consistency (K) equal to viscosity (Π) at the same temperature.

Figure 11 shows the effect of temperature on the viscosity of biodiesel and castor oil. Graphing the natural logarithm of viscosity against the inverse of temperature in Kelvin shows a clearly linear relationship for both solvents, with regressions of 0.999. For both biodiesel and castor oil, the increase in temperatures resulted in decreasing viscosities. However, biodiesel has consistently had a lower viscosity than castor oil for all temperatures, which was expected since biodiesel was originally developed to be a low viscosity oil substitute fuel. For biodiesel and castor oil, n ₀ is found 3,126'10 ⁸ and 5,751 ^'11 respectively; Πο biodiesel is therefore higher than that of castor oil, but both are relatively small. And _a was found 33556 for biodiesel and 57431 for castor oil, which means that the activation energy needed to start the flow is considerably lower for biodiesel, and it can be understood that the temperature required for low viscosity is much higher for castor oil.

64/91

Choice of preferred extraction agent

Both biodiesel and castor oil demonstrate excellent biocompatibility with industrial yeast. However, as demonstrated in the items above, castor oil-based biodiesel, the solvent suggested in the present invention, has several characteristics that make it an interesting choice as an extraction agent for in-situ extractive fermentation of inhibitors as derived from ligno-biomass. hydrolyzed cellulosic or other musts and raw materials defined above for fermentation purposes. It is relatively inexpensive and a renewable bioproduct that can be produced locally. It is biodegradable and a product that, after several cycles of use as an extraction agent and / or refrigerant, can still be sold or used with biofuel for the plant's own fleet or subsidiaries.

More specifically, biodiesel has the lowest density compared to castor oil, favoring its separation from the fermentation broth.

Another advantage of biodiesel is that it absorbs more water and ethanol than castor oil, which, in fact, concentrates substrate in the aqueous phase. For a batch or batch fed fermentation, this means that the sugar is more concentrated at the end of the fermentation,

65/91 promoting maximum productivity at this stage of fermentation, and increases the yield of fermentation in continuous regime.

The partition coefficients for the main fermentation inhibitors are higher for biodiesel, resulting in the removal of these inhibitors from the fermentation broth in favor of the fermentation process as a whole.

For a batch or batch fed fermentation, ethanol production starts earlier and production rates will be higher than without an extractive solvent. In order to optimize the use of biodiesel extractive capacities, the concentration of ethanol in the aqueous phase should be slightly lower than 50-80 g / 1, coincidentally the concentration at which ethanol becomes an inhibitor.

The comparison of a conventional fermentation without inhibitors, a fermentation with inhibitors and a fermentation with inhibitors and biodiesel, confirm the positive effect of using biodiesel for a series of fermentation with furfural, vanilla and acetic acid in terms of total fermentation time, rates growth rates and ethanol production rates.

Unit Operations involved and Flowchart of modalities

66/91 of the Process of the Present Invention

Figure 12 shows an example of a possible flowchart of an ethanol production plant by extractive fermentation, based on the technology proposed by the present invention. According to the diagram, the medium is prepared in the mixer. After preparation and a heat treatment procedure, the medium is cooled by a heat exchanger and introduced into the fermenter, where the substrate will be converted into a product (be it ethanol, butanol, acetone, enzyme, amino acid, among others) by a micro- organism (whether Saccharomyces cerevisia, Zymomonas mobilis, among others, genetically modified or not). The mixture of wine and organic solvent at the outlet of the fermenter is sent to a decanter, where the solvent will be separated by difference in density, possible application of heat, another method of separation or any combination of these.

Then, as in the conventional production process, the wine is clarified in centrifugation units and, via another heat exchanger, introduced into the distillation system. In the case of extracellular product, the microorganism can be recycled and introduced in a unit where eventual bleeding of cells can occur, followed by a unit for reactivation of the cells by acidic, aerobic treatment, with nutrients or other methods for purposes

67/91 reactivation.

A volatile product such as ethanol, butanol or acetone partially present in the one recovered by the organic column, which, for example, exchangers was separated from the recovery using only the temperature exchanger introduced in the fermenter outlet gas, is absorption of the product. The solvent in the decanter absorbs heat via heat, before cascading distillers into a system for the purpose of solvent, product and water, of low pressure steam.

fermenter, the solvent lowers heat and pressure. Product and introduced for example

Others

Figures 13 and

Before preference is to be cooled to one of the fermentation temperature by a cooling system under low water directly in a rectifier.

settings

14.

recovered from the solvent are purification system, as possible are shown in the

Figure 13 shows the removal of inhibitors and product with simultaneous refrigeration, with recovery of the solvent product, water inhibitors in a system with only one bioreactor.

The Figure shows the use of more bioreactors in series in this case three fermentation vats. In this configuration, the first vat is responsible for

68/91 fermentation of most of the substrate fed, while the second and third vats are responsible for the fermentation of the rest of the substrate. Only in the first vat is it necessary to remove enough inhibitors and product by the solvent to obtain very low concentrations of at least most inhibitors at the exit of the first vat, thus guaranteeing minimal synergistic inhibition and effective fermentation both in this vat and in the vats. connected in series.

Figure 15 shows a flowchart of an extractive fermentation plant for batch or batch fermentation fed with a vat according to a third embodiment of the present invention. During fermentation, the solvent is being recycled by the system, increasing the productivity of the fermentation through continuous extraction of the inhibitors, the product and for keeping the fermentation medium at the optimum temperature. After fermentation, the aqueous phase removed from the vat is centrifuged and the rest of the product present in this phase is recovered and purified by the distillation and rectification step.

Figure 16 shows a flowchart of an extractive fermentation plant for continuous fermentation with cooling and removal of inhibitors in several vats, according to a fourth modality of the present invention, where the

69/91 solvent is sent across all vats countercurrent to the fermented medium. Thus, the concentrations of all inhibitors and the product are minimized in the last vat, ensuring maximum productivity and yield in the entire process.

Preferential Modality of the Present Invention

EXTRACTIVE ETHANOL FERMENTATION MODELING IN COMBINATION WITH IN-SITU _T PRODUCT RECOVERY AVERAGE DETOXICATION AND COOLING.

A continuous fermentation process was modeled with a series of three fermenters, representing a realistic industrial frame.

Figure 17 shows a schematic view of the process.

In a first stage, the wort, composed of clarified juice and hydrolyzed broth, is mixed, concentrated by evaporation, cooled and fed to the first fermenter as a flow (B). Evaporated water (A) can be used for cleaning the solvent after removing ethanol in the ethanol recovery unit, but this is not considered in the model. The first fermenter is the largest in volume, contains an organic phase and is responsible for the conversion of about 40-70% of the substrate. The aqueous phase contains the whole, the substrate and yeast. The organic phase is partially dispersed in the aqueous phase, but tends to migrate

70/91 to the top of the fermenter due to its low density. The fraction of the ethanol produced, inhibitors and a small amount of water from the process are extracted into the organic phase, thus improving the performance and productivity of the fermentation and concentrating the substrate. Optimized concentrations of ethanol in the aqueous volume also allow the maximum partition of most hydrolyzed inhibitors to the organic phase.

Other flows that enter the fermenter are the possible concentrated and reactivated microorganism from the cell reactivation unit (Q) and recycled and refrigerated solvent (I) that does not contain inhibitors and ethanol. The temperature of the solvent at the entrance of the vat is adjusted to allow cooling of the fermentation medium, while remaining around three degrees above the point of solidification or turbidity of this solvent. In the simulation, a variety of target fermentation temperatures are evaluated for their influence on the performance of the fermentation.

The mixture of wine and solvent is removed from the top of the bioreactor where the solvent concentration is higher due to the difference in density between the solvent and the average fermentation. This fluid (D) is taken to a decanter unit from which the solvent, rich in ethanol and water, is attracted from the top (E) while the wine is decanted or

71/91 aqueous phase (J) is recovered at the bottom of the decanter. Because it has a higher boiling point than both ethanol and water, the solvent enriched with ethanol (E) is heated and the ethanol and water are separated from the solvent by evaporation, a stage that is not included in the model. The fraction of water and ethanol recovered from the solvent (F) is taken directly to a rectifier, a unit from which ethanol can be recovered with a high degree of purity (R). The decanted wine fluid (J) is fed to a second and third fermenter, which are placed in series through the flow (K).

The two fermenters contain only an aqueous phase one-third the size of the aqueous phase of the first fermenter and serve to convert the remaining substrate. Because most of the ethanol produced in the first fermenter has been removed, inhibition by further accumulation of ethanol produced in the last two fermenters is limited. The additional removal of hydrolyzate inhibitors is also not necessary because the concentration of these inhibitors should already be at least low enough to allow fermentation in the first fermenter and low amounts of furfural have been shown to have a positive effect on ethanol yield. The cooling of the second and third fermenters is done conventionally, assuming

72/91 fermentation temperatures equal to the target temperatures of the first fermenter and considering that the generation of heat in these vats is limited due to the small amount of substrate to be converted in these volumes. After leaving the fluid from the third fermenter (L), the yeast in this fluid is separated by centrifugation and conducted to bleeding units and treatment by clarified wine (M) is pumped into flow (N), while the distillation and rectifying units , purifying the by-products, vinasse model stage.

aeration fraction ethanol and oils (R) of fusion fluids (S), one of the process that

A smaller portion remaining and nutrients in itself from the yeast treated by a yeast unit before re-introducing in the not included bled (O) in the aeration, acid, from treatment to fermenter through flow (Q).

The choice of solvent, a ricinoleic or biodiesel from castor oil ethyl ester, was elaborated according to the work described in the items above and based on the desired properties: favorable partition coefficients for product and inhibitors, high selectivity for these compounds compared to substrate, high biocompatibility, low solubility in the aqueous phase, high water absorption capacity to concentrate the substrate of the

73/91 fermentor, favorable thermodynamic characteristics for removing the product from the solvent and regenerating the solvent as a high boiling point, without forming stable emulsions, rapid separation of the aqueous phase, high chemical stability at high temperatures, low viscosity, low cost, possibly manufactured locally and reused for other purposes, among others.

The kinetics of the microbiological fermentation process applied in the present invention was modeled using kinetic functions of temperatures and concentrations of substrate, ethanol, biomass and hydrolyzed inhibitors, which were elaborated in the work of Zautsen et al., 2011 (submitted in Nov. 2011 for publication under the title 'Kinetics of ethanol fermentation and inhibition byhydrolyzed lignocellulosic biomass'), and incorporated herein in its entirety by reference. For most steps in the process, only simplified mass balances were modeled. Ethanol recovery and solvent recycling, distillation and rectification have not been modeled. The heat energy introduced in the bioreactor vats by metabolic activity related to ethanol production and the energy dissipated by the agitators in these vats were included in the model. The energy introduced by movement of

74/91 mixture was only considered for the fermenters and the heat introduced by exothermic dissolution of ethanol production was not considered.

Dimensions and assumptions The first fermenter was dimensioned with an aqueous phase volume of 600m ³ , while the fermenters both contained 200m ³ second and third broth. The organic phase in the first fermenter had a volume of 20% of the aqueous phase.

The dilution rate of the aqueous phase of the first fermenter was fixed at 0.125h ¹ , by adjusting the feed rate, recognizing the different densities of all fluids with their compositions and temperatures.

in

The raw material costs clarified cane juice hydrolytic liquor were reported as 3: 1.

The components considered in the model were substrate, biomass, ethanol, furfural, vanilla, acetic acid, water, biodiesel as a solvent and carbon dioxide. Sucrose, glucose, fructose and xylose were considered as equal substrates, without differentiation for kinetic conversion that would depend on the specific characteristics and capacities of the yeast strain applied in the real process. Clarified sugarcane juice and hydrolytic liquor are supposed not to evaporate during wort preparation, although in a more realistic scheme, the concentrations of

75/91 some more volatile inhibitors such as acetic and furfural acid may well be reduced at this stage.

Compounds present in molasses and syrup that inhibit fermentation such as sulfide, organic acids, high concentrations of salt, calcium and magnesium and insoluble solids are not considered. However, the biological conversion rates of inhibitors differ for each furan or phenolic compound, depending mainly on the distinct functional group. Inhibition by furfural and HMF is considered cumulative, and is represented in the model by furfural as its total sum of furans, as is vanilla, a phenolic aldehyde, chosen as a representative for the sum of all phenols. The acetic acid represents all organic acids. The densities of liquids are calculated as the average sum of the densities of all components and where possible with temperature dependence. Ethanol was recovered at 100% of the outflow solvent and the outflow solvent was reintroduced dry in the first fermenter and free of inhibitor compounds previously extracted. The lowest temperature limit for solvent entering the fermenter was determined to be three degrees above its freezing point, which is approximately -5 ° C. The kinetics of the transport of compounds from the aqueous to the organic phase was not

76/91 considered and mixing conditions in all the equipment involved were considered ideal. The Reynolds Re number was fixed at 10 ⁶ as was the N _p power number at 5.5 for the agglomerated mixer turbine. For calculating power and turbulence, three mixers were assumed per fermenter, each with a diameter equal to 30% diameter of the vat, and the ratio between the height and the diameter of the vat was two for all fermenters. In these conditions, the tip speed of the agitator was between 3.3 and 6.4 m-s' ¹ , depending on the viscosity and density of the juice and the organic phase. The values of the partition coefficients of the components over the organic and aqueous phase in the fermenter and in the outflow of the bioreactor were equal to the partition coefficients as measured at 34 ° C in equilibrium for different concentrations of ethanol in the aqueous phase.

Cell viability was modeled only temperature dependent, with cell death rates relative to the wort temperature deviation from the maximum temperature, set at 39 ° C, and only the viable fraction of yeast contributed to cell growth and ethanol production while the remaining fraction added to the loss of product yield. The carbon dioxide production rate was assumed to be equivalent to 95.7% of production

77/91 of ethanol by weight, and the capture of ethanol in the exhaustion of CO ₂ was not considered, nor of water, by any of the three bioreactors. Water generation due to metabolic chemical processes was not considered. Yeast density was assumed to be 1200 Kg m ' ³ . For every 1000 liters of ethanol produced, a 20 kg fraction of yeast cream was bled and the rest was reactivated and returned to the fermenter. No heat was introduced by the centrifugation process or at any other stage of the process, except in fermentation as detailed above. The transfer of heat from the fermentation medium to the solvent was assumed to be rapid in all cases, so that the temperatures of the organic phase and the fermentation broth were equal shortly after the introduction of the solvent into the vat.

Fermentation process calculation method was modeled using Python as a computing language. A numerical calculation method was chosen for which millions of combinations of input parameter values were evaluated in parallel. The calculation continued until the state variables were constant for all combinations of input parameter values.

Results

The calculation results showed the ranges for the

78/91 better fermentation conditions, optimizing volumetric productivity and minimizing raw material costs. As a reference point for evaluating the performance of extractive fermentation, a simulation was first carried out for conventional fermentation, that is, without solvent as an extracting agent.

Conventional fermentation

Solvent-free fermentations were also simulated using equal dimensions and assumptions to allow comparison with fermentation with the proposed applied extraction technology. In this case, three input parameters were varied: the proportion of hydrolytic liquor in the must, the final concentration of substrate in the feed must and the temperature of the fermentation.

The results showed that the optimization of the volumetric production rate, and the minimization of raw material costs, both depend on the ratio between clarified broth and hydrolytic liquor and on the substrate concentration. Because cooling was conventionally provided for this fermentation, the amount of inhibitors in the final wort was the only limiting factor for fermentation. For low fractions of hydrolytic liquor, the growth rate was high enough to prevent total loss of yeast by diluting the medium.

79/91

For wort containing only juice, the substrate volumetric production rates of substrates were maximized using a concentration around 24, consequently reduced the (volumetric). Although substrate, the yield merely 53% of the high biomass cost of maximum fees

Higher concentrations of growth rate of ethanol production at total concentration (non-theoretical data due to the detriment of ethanol.

ideal substrate by volume of 24% demonstrated) was a production plus ethanol produced was lower only for concentrations of sugars reaching less than 20 for both lower fermentation substrates used in the simulations of the calculated values but not less conventional and extractive fermentation. The higher concentrations of substrates caused incomplete conversion and decreased the overall yield, related to the production by cost of sugars, and, therefore, that is directly substrate.

For conventional fermentations, the proportions of hydrolytic broth could not be greater than 2% and ideal fermentation temperatures for ideal volumetric production rates were approximately 32 ° C, while for a lower substrate cost, the ideal temperatures were higher near 37 ° C.

80/91

Extractive fermentation

For solvent fermentations, the solvent flow rate was used as a fourth independent parameter and expressed as the solvent dilution rate over the aqueous volume.

Results showed the dependence on the volumetric production rate and the substrate costs per volume of ethanol produced, as the final must dilution rate was the proportion of the solvent broth. The amount is a limiting factor for hydrolytic and inhibitors the performance of the fermentation level for any given solvent dilution rate, causing yeast depletion for high concentrations of these inhibitors, but not as severe as in the case of conventional fermentation.

Another limiting factor was the cooling offered by the solvent. For low solvent dilution rates, both volumetric yield rates, and substrate cost production have collapsed due to insufficient cooling. Above a dilution rate of

0.25h ^_1 the solvent provided sufficient cooling to

every wort compositions. With taxes of dilution of solvent higher, the rates in production volumetric increased due to removal in ethanol and inhibitors

from the hydrolytic broth.

What's more, rates

81/91 increased dilution of the solvent allowed higher fractions of hydrolyzate to be fermented and decreased the minimum substrate cost, although even the dilution rate of ^1h was not sufficient to enable the fermentation of a pure hydrolyzate. However, the economic consequences of high solvent dilution rates, in terms of operating and energy costs, have not been calculated and are likely to increase total production costs, thus limiting the feasibility of using very high solvent dilution rates. solvent.

For substrate cost production, with the juice being three times more expensive than the hydrolyzate, an optimized wort composition was found, different for each solvent dilution rate.

For a wort composition with a proportion of hydrolytic broth fixed at 30%, substrate concentrations above 20% required considerable solvent dilution rates to obtain high volumetric production rates. For volumetric production rates, different optimized substrate concentrations were found for different solvent dilution rates. On the other hand, minimal substrate costs were only obtained for minimum concentrations of substrates and decreased slightly with the solvent dilution rate above the apex of the

82/91 provision of cooling.

For this same proportion of hydrolytic broth fixed to

30%, musts with higher substrate concentrations generated higher fermentation rates and increased heat production by metabolic activity.

Consequently, higher solvent dilution rates were necessary to provide sufficient cooling for higher concentrations of substrates.

For the rate of volumetric production, an optimized concentration of substrate was found at a lower percentage than for conventional fermentation an optimized concentration of substrate yield or minimum cost of substrate, which and differs from the maximum were both the lowest concentration 15% substrate.

As was also noted for fermentations without removal of inhibitors and in a continuous regime, in the work of Horvath et al. 2001 and incorporated here in its entirety by reference, total yeast loss due to the dilution rate of the medium already occurs at relatively low levels of inhibitor concentrations.

The synergistic toxic effect of vanilla, acetic and furfural acid decreased the furfural concentration for which the furfural reduction rate was at its maximum. Growth inhibition was a limiting factor for maximum rates of

83/91 conversion of specific inhibitors by metabolic reduction.

However, it was proved, through modeling and simulation, that the continuous extractive fermentation allowed a simultaneous in-situ extraction of inhibitors present in the hydrolyzate and ethanol as a fermentation product, as well as effective cooling of the fermentation process by the same extractive agent. .

For modeled extractive fermentation, furfural migrated to the organic phase as with vanilla and acetic acid, each with different partition coefficients depending on the ethanol concentrations in the system, improving inhibitor conversion rates, increased rates and rates ethanol production, especially at increased solvent rates.

Thus, the model combined removal of several inhibitors by extraction and biological conversion, both affected by temperatures and composition of the fermentation broth. Different values for the proportion of hydrolytic broth in the wort, substrate concentration, target fermentation temperatures and solvent dilution rates were evaluated, including the generation of mechanical and metabolic heat, inhibition effect by the ethanol product, partition coefficients of the different compounds and temperature effects on kinetic conversion

84/91 inhibitor. In this way, combinatorics effects on the performance of several keys of fermentation indicators were revealed.

The dilution rate of the solvent needed only to be high enough to ensure yeast growth as fast as the dilution rate of the aqueous phase, over the three fermenters, and providing sufficient cooling of the fermentation medium. Before reaching yeast washout conditions, due to high concentrations of inhibitor or substrate, low sugar conversion rates caused loss of substrate and low ethanol production rates.

The optimized solvent dilution rates reduced the inhibition better for both, inhibitors from hydrolytic broth and the product ethanol, improving fermentation rates.

In the context of different fermentation regimes, the inhibitor concentrations can be kept low in relation to the biomass concentration for hydrolytic liquor batch fermentation, by controlling the wort feeding rate for fermentation in batch-fed or continuous regimes, and dilution rate of the organic phase in the case of continuous fermentation as presented in this simulation. However, for fermentations

85/91 batched or continued without the application of an organic phase, this is not possible. Consequently, conversion rates of furfural and vanilla were higher in the extractive fermentation presented than in conventional continuous fermentation, allowing the fermentation of fractions of hydrolytic broth in the must, generating higher volumetric production rates and decreasing the totoal costs of substrate. .

It should also be noted that the concentrations of inhibitors from the hydrolytic broth were assumed not to decrease during the preparation of the wort, which represents a worse scenario considering the high volatility of inhibitors such as acetic acid. Depending on the source of hydrolytic liquor and the method and severity of the hydrolysis of the raw material, it can be expected that the final concentrations of inhibitors in the wort food are lower than the values assumed in this work, leading to even more favorable results. Altogether, it can be concluded that for continuous fermentation, the liquid-to-liquid two-phase extractive fermentation system is an efficient integration and a promising integration of in-situ reduction of inhibitor concentration, product recovery and fermentation cooling.

resume

86/91

Organic esters, better known as biodiesels, based on fatty acids such as, for example, ricinoleic acid or oleic acid, are the most suitable solvents to be used in the extractive fermentation process as revealed by the present invention, since they have characteristics fundamental principles such as:

> Biocompatibility - the solvent has a high biocompatibility, which is vital for the performance of the bioprocess. The size of the molecule being greater than 12 10 carbons prevents the miscibility of the molecule with the yeast cell wall and, thus, guarantees biocompatibility, since there are no special groups in the molecule that can turn it into a toxic component;

Viscosity, structure point, containing at least one methyl, ethyl or boiling and density ester - Due to a double bond because it is propyl, it makes the molecular interaction surface of the solvent in the third dimension, resulting in a limited one and both are size of of bridges molecules are limited to low density viscosity compared to favorable properties. Per molecule, the surface of the same hydrogen that guarantees an aqueous phase point, on the other hand, the and the quantity can be formed between those of high boiling and quite differentiated from the fermentation product and from the others

87/91 volatile components, allowing the recovery of the product and the solvent itself in an economically viable way;

> Polarity - The polarity of the molecule, mainly determined by the size of the main chain of the molecule and by the polar groups and bonds, promotes the extraction of polar and supportive substances (inhibitors and products such as ethanol, butanol, acetone, among others.), But just not substances that are as polar as sugars (substrate). For environmental reasons, the solvent cannot dissolve in the aqueous phase. On the other hand, water can be absorbed in the organic phase and thus concentrate the substrate. Many possible polar groups, such as OH, NH3, esters, among others and combinations of them can form an ideal polarity.

> Partition coefficients - absorption capacities and high partition coefficients for the product and inhibitors from the wort and low selectivity for substrate.

> Economic viability - the high availability in the local market and the low price of its raw material, in the case of biodiesels, a vegetable oil such as castor oil, are very important factors for this molecule to be economically viable or not, in addition to the option of solvent reuse for other purposes, such as, for example,

88/91 biofuel;

> Toxicity - low toxicity is important both for workers at the site where the solvent is used, in this case the plant, and for the eventual introduction into the environment, in the case that minimal fractions of the solvent mix with an outflow of the process that will be exposed to the environment, for example, vinasse or bled cell.

> Emulsion - do not form (stable) emulsions during fermentation and / or downstream processes;

> Chemical stability of the solvent - the solvent cannot degrade the product or cause its degradation, or even react with the product to form another one, regardless of the use of high temperatures in the recovery processes or the presence of reactive components, such as acids, in the fermentative medium;

In a two-phase system, it applies to both solvents, castor oil and biodiesel based on this oil, for higher concentrations of ethanol in the aqueous phase, the partition coefficients of the main inhibitors are higher and the ratio between ethanol and water absorption is also higher, especially for biodiesel. Biodiesel absorbs relatively more water, resulting in a greater capacity for substrate concentration in the middle of fermentation.

89/91

The fermentations in IL fermenters in batch mode showed the positive effect of using biodiesel as a solvent in the liquid-liquid extractive fermentation, in the presence of inhibitors in the must. The rates of products presence rate of no effect of solvent volumetric growth, conventional furfural inhibitors, vanilla and reduction of furfural inhibitor were more vanilla acetic acid in decreasing the lag phase and increasing the specific productivity and while the same concentrations of these often complete death

A series inhibits fermentation, decreasing the viability and even leading to the inoculum kinetic model before it leaves the lag phase.

experiments allowed the construction of the reduction of furfural presence of acetic acid and for concentrations of these others, the rates of growth rate and a various and vanilla, temperatures in the inhibitors. The model conversion describes, between furfural and vanilla, the ethanol production rate as a function of temperature, substrate and concentration of inhibitors and product, adjusted by parameter estimation and validated for several possible key combinations of the independent variables involved.

The modeling and simulation of an integrated liquid-liquid extractive fermentation system with

90/91 solvent and cooling of the fermentation medium by the solvent of the present invention, allowed the optimization of the solvent dilution rate, the substrate concentration, the target temperature of the fermentation and the hydrolytic broth fraction. Using a model of relative costs of raw materials, you obtain an important relationship between the costs of the mixture of the must and the productivity of the process, based on the total yield of the same and assuming the limit for the use of hydrolytic juice. Above all, the advantage of using biodiesel as an extracting agent of inhibitors from hydrolytic broth, ethanol and cooling agent for the fermentative process was shown, allowing the fermentation of larger fractions of hydrolytic broth and increasing the productivity and yield of the integrated process.

Thus, although only a few modalities and models of the present invention have been shown, it will be understood that several omissions, substitutions and alterations in the extractive fermentation process for musts originating from hydrolytic processes of ligno-cellulosic raw material, starch, directly extractable carbohydrates, molasses heavily depleted with a high content of natural inhibitors such as, among others, biotin, or similar musts, rich in carbohydrates and for fermentation purposes,

The present invention, which employs a solvent capable of removing products and inhibitors simultaneously, in addition to cooling the medium, can be made by a person skilled in the art, without departing from the spirit and scope of the present invention.

It is expressly provided that all combinations or omissions of the elements that perform the same function in substantially the same way, to achieve similar results, are within the scope of the invention. Substitutions for elements of a modality described by another 10 or omissions thereof are also fully intended and contemplated.

It is also necessary to understand that the drawings are not necessarily to scale, but that they are only of a conceptual nature. The intention is therefore to be limited, 15 as indicated by the scope of the appended claims.

Claims (2)

1. Extractive fermentation process characterized by being carried out continuously, batch or batch fed and comprising the following steps: a) Entrance of a must selected from hydrolysates of ligno-cellulosic material, starches, directly extractable carbohydrates and similar musts rich in carbohydrates in the bioreactor; b) Entry into the bioreactor of an inoculum by recycling or reactivation by aerobic treatment, acid treatment and / or nutrient feeding and be optional; c) Injection of castor oil biodiesel; d) Removal of biodiesel and removal of inhibitors and / or products by biodiesel by decantation or centrifugation carried out during or after fermentation; e) Recovery of biodiesel by heating or vacuum evaporation and re-introduction to the bioreactor; f) Purification of the vapor phase by separation or distillation and rectification columns; g) Removal of the aqueous phase from the bioreactor by pumping; h) Recovery of the microorganism is optional and by centrifugation, ultrafiltration, separation by Petition 870190101589, of 10/09/2019, p. 4/9 2/2 membrane or any method of separation and concentration of cells.