A method for predicting the reaction rate of substances in anaerobic ethanol fermentation

Abstract

The present invention relates to a method for predicting the reaction rates of substances during anaerobic ethanol fermentation. This method uses changes in the specific growth rate μ to predict changes in the specific consumption or specific production rate of each substance under different substrate and product concentration conditions. Compared with the existing technology, the present invention, based on a process model constructed based on thermodynamic principles, can predict the stoichiometric relationship between the rates of raw materials, products, heat production, and other factors during the ethanol production process of brewer's yeast under anaerobic culture at different specific growth rates, based on minimal understanding of process metabolic regulation mechanisms and experimental data requirements. Based on this, the thermodynamically based black box model constructed in this solution can provide excellent predictions of changes in the anaerobic ethanol fermentation process under steady-state and dynamic conditions.

Description

Technical Field

[0001] The present invention relates to the field of anaerobic ethanol fermentation engineering, in particular to a method for predicting the reaction rate of substances in the anaerobic ethanol fermentation process. Background Art

[0002] Driven by environmental concerns stemming from the use of fossil fuels, the world is urgently seeking alternative, clean, and renewable energy sources. Bioethanol is a renewable, environmentally friendly energy source, with production reaching approximately 110 billion liters in 2019. Blending a certain percentage of ethanol into gasoline can alleviate energy challenges. It is estimated that using bioethanol as a gasoline substitute can reduce automotive CO2 emissions by 90%, leading to its widespread use in the automotive industry as a fuel extender. Furthermore, anaerobic production offers significant advantages, such as higher yields, reduced fermentation heat, reduced biomass production, and lower mechanical energy input, significantly reducing production costs. Furthermore, the transition from non-renewable to renewable resources for fuels and chemicals is a key step in developing a circular economy. While challenges remain in industrial production, such as the toxicity of lignocellulosic feedstocks, the presence of mixed carbon sources in the matrix, batch variability, and bacterial contamination, the use of lignocellulosic feedstocks could offer environmental, economic, and strategic advantages for biofuel production in the near future.

[0003] For many years, yeast has been gaining increasing attention in systems biology and synthetic biology research. A key factor driving this is the high similarity between human and yeast cells in many key cellular metabolic processes, such as protein folding and chaperone function, heat shock, protein transport and secretion, and autophagy. The most prominent example is the Crabtree effect, which enables yeast cells to produce ethanol even under aerobic conditions. This phenomenon is similar to the Warburg effect observed in human cancer cells. Because yeast is an excellent strain for producing alcohol, the fermentation industry has already applied it to fuel and chemical production. Currently, ethanol can be produced from sugar and starch substrates using Saccharomyces cerevisiae at concentrations of 12%-15% (v / v). In recent decades, extensive research has focused on the kinetics and process design of ethanol production using Saccharomyces cerevisiae, as well as key metabolic engineering strategies for developing high-yield strain platforms. Undoubtedly, the introduction of kinetic theory and models has provided a foundation for process design, optimization, and plant operation. However, the gap between model complexity and data availability has hindered its industrial application, particularly in early process development, when historical data on the strain or process is limited. Therefore, most of the existing industrial fermentation kinetic models are qualitative rather than quantitative.

[0004] Kinetics describes the rate and pathways of a specific process, while thermodynamics describes the overall properties, behavior, and equilibrium components of a system. The two complement each other. In chemical engineering, the systematic application of chemical thermodynamics to process technology development facilitates the design of petrochemical plants and the development of petrochemical processes, sometimes requiring only minimal experimental effort. Similarly, in biochemistry, researchers face a significant challenge in bioprocess development, where extensive experimental work is required. Therefore, it can be hypothesized that the rational incorporation of thermodynamic principles in biochemical engineering could streamline bioprocess development while reducing, or even eliminating, tedious experimental work. In particular, from the perspective of quantitative modeling, general thermodynamic principles such as percolation limit, Gibbs free energy dissipation, approximate equilibrium, and thermodynamic driving force describe the characteristics of biological systems. For example, extensive data demonstrate that a thermodynamic model can be constructed to describe the maintenance of microorganisms that is independent of either electron donors or electron acceptors, with only temperature being the significant influence.

[0005] Recently, basic thermodynamic calculations have been used to assess the feasibility of anaerobic substrate-to-product conversion. By doing so, different substrates and a full range of process conditions can be taken into account, allowing for a rapid and simple feasibility assessment of anaerobic processes. The results show that many products can theoretically be produced under anaerobic conditions using several conventional and unconventional feedstocks. In recent years, researchers have hoped to use this to obtain quantitative relationships between fluxes and metabolite levels, and how they are affected by perturbations of substrates, products, and gene expression at the enzyme level. However, more research is currently focused on analyzing and modifying microbial cell factories, and few studies have reported on the analysis and design of bioprocesses based on thermodynamics, making it extremely difficult to predict the specific consumption or specific production rates of various substances in anaerobic ethanol fermentation processes along this direction. Summary of the Invention

[0006] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a method for predicting the reaction rate of substances in the anaerobic ethanol fermentation process. This technical solution is based on a process model constructed based on thermodynamic principles to predict the stoichiometric coefficients involved in the ethanol production process of brewer's yeast under anaerobic culture at different specific growth rates.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] The present invention provides a method for predicting the reaction rate of substances in an anaerobic ethanol fermentation process, characterized in that the change in the specific consumption or specific generation rate of each substance under different substrate concentrations and product concentrations is predicted by the change in the specific growth rate μ.

[0009] Furthermore, the specific consumption or specific generation rate of each substance is linearly related to the specific growth rate μ.

[0010] Furthermore, the specific growth rate μ is obtained by:

[0011]

[0012] Where: μ mas is the maximum specific growth rate, k S is the substrate affinity constant, K I Substrate inhibition constant, S is the substrate concentration.

[0013] Furthermore, during the anaerobic ethanol fermentation process, the growth of yeast cells is inhibited by ethanol in a non-competitive manner, that is, only the maximum specific growth rate μ max Affected by ethanol concentration.

[0014] Furthermore, when only the maximum specific growth rate μ max When affected by ethanol concentration, the specific growth rate μ is obtained as follows:

[0015]

[0016] Where: P is the ethanol concentration, K P is the maximum inhibition constant of ethanol on growth.

[0017] Furthermore, based on the dual inhibition of substrate and product, the specific growth rate μ is obtained as follows:

[0018]

[0019] Furthermore, when the dilution rate during the continuous culture process is lower than a preset threshold, the specific growth rate μ is obtained by summing and correcting the experimental parameter μ0;

[0020] The experimental parameter μ0 is the experimental parameter obtained when the dilution rate of the limiting substrate concentration is close to zero.

[0021] Furthermore, when the dilution rate is relatively low, the specific growth rate μ is obtained as follows:

[0022]

[0023] Furthermore, the maximum specific growth rate μ max , substrate affinity constant K S , substrate inhibition constant K I ,, ethanol inhibition constant K P They are 0.28 1 / h, 17.5g / L, 651.4g / L and 25.41g / L respectively.

[0024] Furthermore, when the dilution rate D during the continuous culture process is lower than a preset threshold, μ0=D.

[0025] Compared with the prior art, the present invention has the following technical advantages:

[0026] 1) This technical solution is based on a process model constructed based on thermodynamic principles. Compared with other existing process model construction technologies, it requires the least understanding of the cell metabolic mechanism and experimental data during the culture process.

[0027] 2) This technical solution is based on a process model based on thermodynamic principles to predict the stoichiometric coefficients of raw materials, products, heat production, etc. involved in the ethanol production process under different specific growth rates of Saccharomyces cerevisiae under anaerobic culture.

[0028] 3) By combining the necessary conservation of mass, this technical solution verifies the applicability of the thermodynamic black box model constructed in this solution in predicting process dynamics under steady-state and dynamic conditions. That is, the black box model constructed based on thermodynamic principles can provide good predictions of changes in the anaerobic ethanol fermentation process under steady-state and dynamic conditions, and is stable. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 The metabolic pathway for ethanol fermentation and glycerol production in Saccharomyces cerevisiae;

[0030] Figure 2 Schematic diagram of substrate distribution during anaerobic ethanol synthesis in Saccharomyces cerevisiae;

[0031] Figure 3 The relationship between the specific consumption (or specific production) rate of each substance in ethanol production under anaerobic conditions and the specific growth rate of the bacteria;

[0032] Figure 4 is the yield of each substance to the product under different specific growth rate conditions, q i The unit is (mol i / CmolX / h), where i = (S, P, NH4 + 、H + , H2O, CO2, heat);

[0033] Figure 5 is the cell growth (μ) and the specific consumption (or specific production) rate of each substance simulated by the yeast cell growth kinetics model under different substrate and ethanol concentration conditions;

[0034] Figure 6 Comparison of the predicted values ​​of the substrate and product dual inhibition model with the experimental values ​​under high-sugar fermentation culture conditions.

[0035] Figure 7 Middle: Sub-graph a is the q predicted by the model s The value is consistent with the experimentally measured q s Comparison of values; sub-graph b is the q predicted by the model p The value is consistent with the experimentally measured q p Comparison of values; sub-graph c is the model prediction The value and experimental measurement The light blue solid line represents the model prediction result, the blue dots represent the steady-state experimental values, the red dots represent the linear variation experimental values ​​of the dilution rate, and the black dots represent the linear variation experimental values ​​of the dilution rate in the literature. DETAILED DESCRIPTION

[0036] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. Any features such as structure / module name, control mode, algorithm, process or composition ratio not clearly described in this technical solution are regarded as common technical features disclosed in the prior art.

[0037] metabolic pathways

[0038] In the anaerobic fermentation of Saccharomyces cerevisiae, ethanol production mainly involves the glycolysis pathway. Through the glycolysis process, one molecule of glucose is converted into two molecules of pyruvate and further reduced to ethanol, while releasing two molecules of carbon dioxide and producing two ATP molecules, such as Figure 1 Under anaerobic conditions, ethanol replaces oxygen as an electron acceptor to produce ATP. Therefore, ethanol production is closely related to yeast cell growth.

[0039] Process Reaction Model

[0040] According to the linear equation of substrate consumption, namely the Herbert-Pirt (HP) relationship ( Figure 2 ), substrate consumption is used for cell growth, product synthesis, and cell maintenance. For products that produce energy under anaerobic conditions (such as ethanol), product formation is coupled to cell growth and maintenance. Therefore, there is no separate product production term in the HP equation.

[0041] Cell maintenance

[0042] Maintenance energy primarily occurs in living cells. This energy is used by cells to cope with numerous adverse conditions, such as metabolite leakage, DNA breakage, and protein denaturation, through repair mechanisms. The Gibbs free energy expenditure (KJ / mol-Biomass / h) for cellular maintenance has been shown to be solely temperature-dependent and unrelated to intracellular electron donors and acceptors. Maintenance energy expenditure follows the following Arrhenius relationship (Equation 1).

[0043]

[0044] The equation states that the energy required to maintain the process doubles for every 8°C increase in temperature. R and T represent the gas constant (8.314 J / mol / K) and absolute temperature (K), respectively.

[0045] Cell growth

[0046] Synthesizing bacteria requires a Gibbs free energy expenditure, and the Gibbs free energy required to generate 1 mol of bacteria depends on the carbon source used. For heterotrophic growth (e.g., Saccharomyces cerevisiae growing on glucose), the Gibbs free energy consumption required to generate 1 mol of bacteria can be estimated using the following formula (Equation 2).

[0047]

[0048] C and γ s where are the number of carbon atoms and the degree of reduction of the carbon source. It can be estimated that 240 kJ of Gibbs free energy is required to synthesize 1 mol of bacteria using glucose.

[0049] Under anaerobic conditions, Saccharomyces cerevisiae can only provide energy for growth and maintenance by producing ethanol. Therefore, the process reaction can be divided into two major parts, namely catabolism and anabolism.

[0050] catabolism

[0051] In this section, the Gibbs free energy required to produce 1 mol of bacteria can be generated by the process of forming ethanol from glucose. The following formula (3) shows that when 1 mol of glucose is converted into 2 mol of ethanol and 2 mol of carbon dioxide, 225 kJ of Gibbs free energy can be generated.

[0052] -1C6H 12 O6+2c2H6O2+2CO2+225KJ+95KJ heat (3)

[0053] Therefore, the following formula (4) is obtained based on the relationship between the 240 kJ of energy produced by catabolism required to synthesize 1 mol of bacteria.

[0054]

[0055] Similarly, the maintenance process consumes Gibbs free energy, which is also derived from the ethanol production process equation (3). For ethanol production at 30°C, the maintenance Gibbs free energy consumed can be estimated from equation (1) to be 7.13 kJ / mol-biomass / h, equivalent to 0.03 mol glucose / mol-biomass / h.

[0056] anabolism

[0057] The production of 1 mol of bacteria consumes 240 kJ of energy generated during the ethanol production process. According to the principle of conservation of elements and charge, the stoichiometric formula (5) corresponding to the production of 1 mol of bacteria can be obtained. The bacteria are in the form of 1 carbon (C1H 1.8 O 0.5 N 0.2 )express.

[0058]

[0059] Cell growth process reaction

[0060] Combining formula (4) and formula (5) we can get the following reaction formula (6), which gives the complete stoichiometric relationship for generating 1 mol of bacteria.

[0061]

[0062] Maintain the reaction process reaction formula

[0063] The energy consumption for cell maintenance can only come from the production of ethanol, namely reaction (3), and the reaction rate is 0.03 mol glucose / mol-biomass / h.

[0064] Herbert-Pirt substrate partition equation

[0065] After completing the stoichiometric model for cell growth and maintenance reactions, the Herbert-Pirt distribution equation for nitrogen source, product, water, carbon dioxide, and heat of reaction can be derived based on the distribution of each substance in the growth and maintenance terms (Equations 7-13). Furthermore, the reactants and products involved in the process reaction are all functions of μ.

[0066] q s =-1.242μ-1*0.03 (7)

[0067]

[0068] q P =2.133μ+2*0.03 (9)

[0069]

[0070]

[0071]

[0072] q heat =101μ+95*0.03 (13)

[0073] Yeast cell growth kinetics

[0074] In the absence of inhibition by toxic metabolites, the Monod model provides a good fit between the specific growth rate and the limiting substrate concentration. However, previous studies have found that high concentrations of glucose and ethanol can limit yeast cell growth. In such cases, the Logistic-Monod mixed growth model can be employed to account for the system's multiple growth-dependent factors.

[0075] Substrate inhibition can be described by incorporating the substrate inhibition constant into the following kinetic expression (Equation 14).

[0076]

[0077] Yeast cell growth is inhibited by ethanol in a noncompetitive manner, indicating that only the maximum specific growth rate is affected by ethanol concentration. A generalized nonlinear equation (Equation 15) can be used to describe the effect of ethanol concentration on yeast cell growth rate.

[0078]

[0079] Combining Equation 14 and Equation 15, a comprehensive model (Equation 16) can be established to describe the dual inhibition of substrate and product, as shown below.

[0080]

[0081] However, in continuous cultures, when the dilution rate is relatively low, the model's prediction of the relative growth rate is significantly lower than the experimentally measured result. Therefore, this deviation can be corrected by adding an experimental parameter (μ0) corresponding to the dilution rate at which the limiting substrate concentration is close to zero, as shown in the following formula.

[0082]

[0083] The kinetic parameters in the above formula are obtained by data fitting, among which the maximum specific growth rate μ max , substrate affinity constant K S , substrate inhibition constant K I ,, ethanol inhibition constant K P They are 0.28 1 / h, 17.5 g / L, 651.4 g / L, and 25.41 g / L respectively. When the dilution rate D is extremely low during continuous culture, μ0 = D. (When S≤K S ).

[0084] Thermodynamic Description of Ethanol Production Process under Anaerobic Culture

[0085] In large-scale fermentation processes, key performance indicators such as titer, yield, and productivity are relevant to process evaluation and design and therefore require significant consideration. However, obtaining these indicators requires extensive and tedious laboratory-scale experiments. To circumvent this tedious task, whole-cell thermodynamic modeling has the potential to advance bioprocess research and development by predicting cell culture performance, understanding cell growth parameters, and developing, monitoring, and controlling cell cultures. The thermodynamic theory of Gibbs free energy consumption is well-developed, enabling predictions of biomass and catabolic product yields. However, existing kinetic models describing bacterial growth and product formation rarely incorporate thermodynamic considerations. Consequently, developing models for process design and optimization remains challenging.

[0086] Based on the general thermodynamic principle of Gibbs energy dissipation, a reaction model for ethanol production in anaerobic culture of Saccharomyces cerevisiae was constructed. Based on elemental, charge, and reducing power balances, a black-box thermodynamic model was constructed to predict the stoichiometric coefficients for biomass growth, substrate consumption, product, and heat formation.

[0087] Figure 3 The relationship between the stoichiometric coefficient and the specific growth rate of the process reaction of Saccharomyces cerevisiae under anaerobic culture is shown. Under anaerobic conditions, the HP relationship of fermentation production of ethanol is linearly related to the specific growth rate.

[0088] Figure 4 The graph shows the change in the yield of each substance to the product as the specific growth rate increases, that is, the change in the consumption or production of substances required to produce 1 mol of product. From the graph, we can see that when μ = 0, the substrate required to produce 1 mol of product is the least, indicating that the yield of product to substrate is the highest at this time. As the specific growth rate increases, more and more substrate is required to produce 1 mol of product, and the yield of product to substrate gradually decreases. As the specific growth rate increases, the consumption of NH4 required to produce 1 mol of product increases. + And the generated H + The changing trends of substances such as H2O and CO2 are the same as those of the substrates. As the specific growth rate increases, the yield of the product gradually decreases.

[0089] Unlike the above, the heat generated per mol of product is greatest when μ = 0. Subsequently, as the specific growth rate increases, the heat generated per mol of product decreases. This is likely due to the increasing proportion of heat released as bacterial growth accompanies this. Overall, ethanol production is most favorable when μ = 0, the maintenance phase. At this time, the yield of ethanol from substrates and various intermediates is maximized, and the proportion of energy generated within the bacteria that goes toward growth and maintenance is minimized.

[0090] Model prediction results under double inhibition conditions

[0091] The model simulates the growth and production of bacteria under different substrate and product concentrations, where the maximum simulated substrate concentration is 300g / L and the maximum simulated product concentration is 200g / L. The simulation results are as follows: Figure 5 shown.

[0092] First, from the results of the bacterial growth simulation, it can be seen that the substrate begins to inhibit the bacteria only when it reaches a certain concentration, while the inhibitory effect of ethanol on the bacteria begins when it is generated. When the ethanol concentration is zero and the substrate concentration is about 80g / L, μ reaches a maximum value of 0.2078 1 / h, and then the specific growth rate of the bacteria begins to decrease with the increase of substrate concentration. Figure 5 It can be seen that ethanol has a great influence on the activity of bacteria. With the increase of ethanol concentration, the specific growth rate of bacteria decreases rapidly. When the substrate concentration is 80g / L and the ethanol concentration is 130g / L, μ≈0.03 1 / h.

[0093] For the HP model, under anaerobic fermentation conditions, the specific consumption or specific production rate of each substance is linearly related to μ. Therefore, the changes in the specific consumption or specific production rate of each substance under different substrate and product concentrations are the same as the changes in μ.

[0094] Model prediction performance under dual inhibition conditions

[0095] The predictive performance of the substrate and product dual inhibition model was tested using experimental data under high sugar fermentation conditions. Figure 6 As shown in the figure, the model has the following characteristics: the specific growth rate μ and substrate consumption rate q of the bacteria under the initial high sugar fermentation conditions. s and product ratio generation rate q p Has good predictive ability.

[0096] Predictive performance of the model under steady-state and linear dilution conditions

[0097] The prediction performance of the model in this scheme was verified under steady-state conditions with linear dilution rate changes. The data used were from experiments conducted by the authors of this paper and Beckers et al. (Bekers KM, Heijnen JJ, van Gulik WM. Determination of the in vivo NAD:NADH ratio in Saccharomyces cerevisiae under anaerobic conditions, using alcohol dehydrogenase as sensor reaction. Yeast (Chichester, England), 2015, 32(8): 541-557.). The model simulation results and the linear dilution rate change conditions were compared. s ,q P 、 The experimental data were compared with the results of Figure 7 As shown in the figure, we can see that when μ=0.05-0.2h -1 The model has good stability over time. Figure 7 It can be seen that the model is stable, where sub-graph a is the q predicted by the model s The value is consistent with the experimentally measured q s Comparison of values, sub-graph b is the q predicted by the model p The value is consistent with the experimentally measured q p Comparison of values, sub-graph c is the model prediction The value and experimental measurement The light blue solid line in the figure represents the model prediction result, the blue data points are the applicant's steady-state experimental values, the red dots are the applicant's dilution rate linear variation experimental values, and the black dots are the Beckers et al.'s dilution rate linear variation experimental values.

[0098] This study demonstrates that, based on thermodynamic principles, a process reaction model for ethanol production by Saccharomyces cerevisiae under anaerobic conditions has been established. This model can predict the relationship between the specific growth rate and the stoichiometric coefficients involved in the ethanol production process, including feedstock, product, and heat production, under different specific growth rates during anaerobic cultivation. Furthermore, the applicability of this thermodynamically based black-box model for predicting process dynamics under dynamic conditions has been demonstrated.

[0099] The above description of the embodiments is intended to facilitate understanding and use of the invention by those skilled in the art. It will be apparent that those skilled in the art can readily make various modifications to these embodiments and apply the general principles described herein to other embodiments without requiring inventive effort. Therefore, the present invention is not limited to the above-described embodiments. Improvements and modifications made by those skilled in the art based on the disclosure of the present invention, without departing from the scope of the present invention, should be within the scope of protection of the present invention.

Claims

1. A method for predicting the reaction rate of substances in an anaerobic ethanol fermentation process, characterized in that: The changes in the specific consumption or specific production rate of each substance under different substrate concentrations and product concentrations can be predicted by the changes in the specific growth rate μ. The specific consumption or specific generation rate of each substance is linearly related to the specific growth rate μ; The specific growth rate μ is obtained as follows: Where: μ max is the maximum specific growth rate, K S is the substrate affinity constant, K I Substrate inhibition constant, S is the substrate concentration; During the anaerobic ethanol fermentation process, the growth of yeast cells is inhibited by ethanol in a non-competitive manner, that is, only the maximum specific growth rate μ max Affected by ethanol concentration; When only the maximum specific growth rate μ max When affected by ethanol concentration, the specific growth rate μ is obtained as follows: Where: P is the ethanol concentration, K P is the maximum inhibition constant of ethanol on growth; Based on the dual inhibition of substrate and product, the specific growth rate μ is obtained as follows: When the dilution rate during continuous culture is lower than a preset threshold, the specific growth rate μ is obtained by summing and correcting the experimental parameters μ0; The experimental parameter μ0 is the experimental parameter obtained when the dilution rate of the limiting substrate concentration is close to 0; When the dilution rate is relatively low, the specific growth rate μ is obtained as follows:

2. The method for predicting the reaction rate of substances in an anaerobic ethanol fermentation process according to claim 1, wherein: The maximum specific growth rate μ max , substrate affinity constant K S , substrate inhibition constant K I , ethanol inhibition constant K P They are 0.281 / h, 17.5g / L, 651.4g / L and 25.41g / L respectively.

3. The method for predicting the reaction rate of substances in an anaerobic ethanol fermentation process according to claim 1, wherein: When the dilution rate D is lower than the preset threshold during the continuous culture process, μ0=D.

Images