Premature infant / low birth weight infant formula milk powder, low side reaction preparation method and evaluation analysis method

Abstract

The invention discloses premature infant / low birth weight infant formula milk powder, a low side reaction preparation method and an evaluation analysis method, and belongs to the field of emulsion preparation. The technical problems that in the prior art, premature infant / low-birth-weight infant formula milk powder has the characteristics of high protein and high fat, and the protein-lipid co-oxidation phenomenon is prone to occurring are solved, the low-side-reaction preparation method of the premature infant / low-birth-weight infant formula milk powder is provided, and by changing homogenizing process parameters, the low-side-reaction content of the premature infant / low-birth-weight infant formula milk powder is improved. Through cooperation of sterilization and vacuum spray drying technologies, the interface protein content and emulsifying activity of the formula milk powder are improved, protein-lipid co-oxidation is reduced to the greatest extent, and the stability of the formula milk powder is enhanced; based on a quantitative structure-activity relationship (QSAR) model, the invention also provides an evaluation analysis method, which can accurately predict the influence of homogeneous process parameters and interface characteristics on oxidation stability, and provides a new method for accurate manufacturing of premature infant / low birth weight infant formula food.

Description

Technical Field

[0001] This invention belongs to the field of food preparation, and in particular relates to an infant formula milk powder and a preparation method and evaluation and analysis method with low side effects. Background Technology

[0002] Infants born at a gestational age greater than 28 weeks but less than 37 weeks are defined as premature infants; furthermore, newborns with a birth weight of less than 2500 grams in the first hour after birth are classified as low birth weight infants. Compared to full-term infants, premature infants, unable to accumulate sufficient nutritional reserves, require higher levels of energy, protein, and fat to meet their nutritional needs and achieve a growth rate comparable to full-term infants. Therefore, infant formula specifically designed for premature and low birth weight infants is characterized by high energy density, high protein, and high fat. However, in the design of such high-fat / high-protein infant formula, the protein behavior at the oil-water interface is a core factor determining system stability; interfacial proteins anchor the interface through the hydrophobic domains of β-lactoglobulin, forming a viscoelastic adsorption layer to resist droplet aggregation. Notably, the interfacial properties induced by interfacial proteins (such as zeta potential and interfacial rheological parameters) not only affect short-term physical stability but also determine long-term chemical stability by regulating the initiation of lipid oxidation.

[0003] The oil-water interface in milk powder is a key structural domain that interacts with various components. During milk powder emulsification, proteins tend to adsorb onto the oil-water interface, thereby encapsulating lipids and forming a viscoelastic interfacial protein layer. Therefore, the structure and characteristics of this interfacial layer directly determine the stability of the milk powder. In milk powder, lipid and protein oxidation are intertwined and occur concurrently. Byproducts of lipid oxidation, including malondialdehyde (MDA) and 4-hydroxy-2-enal, are major factors promoting protein oxidation and altering protein structure and function; these changes affect the stability of the interfacial protein layer in milk powder. Simultaneously, byproducts of protein oxidation further accelerate lipid oxidation; this co-oxidation phenomenon not only reduces the nutritional value of milk powder but also leads to the loss of functional properties and may potentially result in the formation of toxic and harmful substances.

[0004] Currently, although homogenization is commonly used in milk powder production to improve its stability, its impact on protein-lipid co-oxidation in infant formula remains poorly understood, especially for formula intended for premature and low birth weight infants. Existing technologies lack predictive mathematical models for rapidly assessing the intrinsic relationship between processing parameters, interfacial properties, and oxidative stability in infant formula.

[0005] Premature / low birth weight infant formula is prone to protein-lipid co-oxidation due to its high protein and fat content. Therefore, those skilled in the art desire to develop a method for preparing premature / low birth weight infant formula with high protein, high fat content and strong oxidative stability, as well as a method suitable for evaluating the characteristics and oxidative stability of premature / low birth weight infant formula. Summary of the Invention

[0006] To solve the above-mentioned technical problems, this invention provides a formula milk powder for premature / low birth weight infants, a preparation method with low side effects, and an evaluation and analysis method.

[0007] One objective of this invention is to provide a method for preparing formula milk powder for premature / low birth weight infants with low side effects, the method comprising the following steps: S1: Dissolve whey protein powder and carbohydrates separately in deionized water at 60°C, stir at 600 rpm for 2 h, and store at 4°C overnight to obtain hydrated whey protein powder and carbohydrates respectively. S2: Mix the hydrated whey protein powder, carbohydrates and edible vegetable oil from S1 evenly, and process them using an IKA high shear homogenizer at 10,000 rpm for 1 min to obtain an emulsified mixture. S3: The emulsified mixture in S2 is homogenized in a high-pressure homogenizer to obtain a homogenized liquid; the homogenized liquid is sterilized in a 95°C oil bath for 15 seconds, and then dried into powder using a vacuum spray drying tower to obtain a premature / low birth weight infant formula.

[0008] In a preferred embodiment of the present invention, the whey protein powder in S1 has a protein content of 80%; the carbohydrate is obtained by mixing lactose and maltodextrin in a mass ratio of 7:3.

[0009] In a preferred embodiment of the present invention, the mass-volume ratio of the whey protein powder, carbohydrates and edible vegetable oil in S2 is 7:21:12.

[0010] In a preferred embodiment of the present invention, the conditions for homogenization in S3 are: the homogenization pressure is 20-60 MPa, and the number of homogenization cycles is 1-2.

[0011] In a preferred embodiment of the present invention, the process parameters of the vacuum spray drying tower in S3 are: vacuum pressure of -20 kPa, inlet air temperature of 115°C, outlet air temperature of 60°C, gas flow rate of 30 L / min, and peristaltic pump speed of 30 r / min.

[0012] The second objective of this invention is to provide a premature / low birth weight infant formula milk powder, which is obtained by the above-mentioned low-side-reaction preparation method.

[0013] The third objective of this invention is to provide an evaluation and analysis method, which is based on the QSAR model and uses homogenization parameters and interfacial properties as engineering descriptors for predicting oxidative stability, thereby evaluating and analyzing the oxidative stability of the aforementioned premature / low birth weight infant formula.

[0014] In a preferred embodiment of the present invention, the method specifically comprises: Step 1: Obtain the key parameters of the formula milk powder; Step 2: Construct a quartic polynomial function and fit the key parameters into the quartic polynomial function. At the same time, use the coefficient of determination and root mean square error to evaluate the goodness of fit. The fourth-degree polynomial function is:

[0015] Where x is the number of homogenization cycles and y is the homogenization pressure; Step 3: Construct a multivariate model based on QSAR and perform multiple regression analysis on homogenization times, homogenization pressure, interfacial proteins, and emulsifying activity.

[0016] In a preferred embodiment of the present invention, the key parameters include average particle size, zeta potential, emulsifying activity (EAI), emulsifying stability (ESI), adsorbed protein (AP), malondialdehyde (MDA), protein oxidation product (N'-formylkynurenine), and protein oxidation product (DT).

[0017] In a preferred embodiment of the present invention, the multiple regression analysis specifically includes: First, before the analysis, all predictor and response variables were transformed using monotonically ordered number spline curves to generate a completely monotonically nonlinear CATREG regression. Then, relevant analysis was performed to establish a quantitative relationship between homogenization parameters and oxidation characteristics; Finally, hierarchical cluster analysis was performed to highlight vertical clustering of homogenization parameters and horizontal clustering of physicochemical properties.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention provides a method for preparing formula milk powder for premature / low birth weight infants with low side effects. By homogenizing, the content of interfacial proteins and emulsifying activity of the formula milk powder are improved, thereby minimizing protein-lipid co-oxidation and enhancing the emulsification stability of the formula milk powder. The synergistic sterilization and vacuum spray drying technologies, by adjusting the homogenization pressure and the number of homogenization cycles, achieve the optimal homogenization parameters (homogenization pressure of 20-60 MPa, homogenization cycles of 1-2). The optimal homogenization parameters help to form a suitable droplet size distribution and zeta potential, promote the construction of a stable interfacial protein layer, and achieve the purpose of improving the content of adsorbed proteins, absolute zeta potential value, emulsification activity and emulsification stability in the interfacial film.

[0019] However, excessive homogenization can have multiple negative effects: it can cause droplet aggregation and increased particle size, reduce the absolute value of the zeta potential, and lead to desorption and reduced content of interfacial proteins; at the same time, it can induce protein oxidative damage, manifested as conformational changes (reduced intrinsic fluorescence intensity), accumulation of oxidation products (N'-formylkynurenine and dityrosine), and a significant increase in lipid oxidation indicators (peroxide value and malondialdehyde).

[0020] To accurately obtain the optimal formula for premature / low birth weight infants, this invention provides an evaluation and analysis method based on the quantitative structure-activity relationship (QSAR) model. It integrates a ternary descriptor-response framework comprising homogenization parameters (processing descriptor), interfacial properties (structural descriptor), and oxidative stability (response activity). A monotonic nonlinear regression model is constructed, including homogenization process parameters (homogenization cycle number X1 and homogenization pressure X2), interfacial properties (interfacial protein content Y1 and emulsifying activity Y2), and an oxidation index (malondialdehyde Z). This invention achieves, for the first time, quantitative prediction of the complete action chain of "processing parameters-interfacial properties-oxidative stability," and provides the first quantitative model for precisely regulating the homogenization sterilization process to inhibit protein-lipid co-oxidation.

[0021] Curve fitting, correlation analysis, and hierarchical cluster analysis showed that homogenization cycle number, homogenization pressure, emulsifying activity index, and emulsifying stability index were closely related to emulsifying properties and protein-lipid co-oxidation levels. Among these, sample S-2-40 was particularly prominent due to its high emulsifying activity and low content of protein-lipid co-oxidation products. A regression equation (R²) was successfully established for homogenization cycle number, homogenization pressure, interfacial protein content, emulsifying activity, and malondialdehyde content. 2 =0.959, p <0.05, which can accurately predict the impact of homogenization process parameters and interface characteristics on oxidation stability.

[0022] This invention provides a new method for the precise manufacturing of formula for premature / low birth weight infants, and provides important theoretical basis and processing strategies for the development of highly stable formula milk powder for premature / low birth weight infants. Attached Figure Description

[0023] Figure 1 A is a particle size distribution chart; B is a chart for the group with one homogenization treatment; C is a chart for the group with two homogenization treatments; D is a bar chart for particle size distribution. Figure 2 This is a graph showing the detection of ζ potentials. Figure 3 A is the emulsification index test chart; B is the emulsification activity index (EAI) test chart; C is the emulsification stability index (ESI) test chart. Figure 4 This is a graph showing the detection of adsorbed protein AP content; Figure 5 This is a graph showing the detection of POV (Polyoxymethylene) content. Figure 6 This is a graph showing the detection of malondialdehyde (MDA) content. Figure 7 This is a graph showing the carbonyl content detection. Figure 8 This is a graph showing the detection of free thiol-SH content; Figure 9 A represents the endogenous fluorescence spectrum detection; B represents the detection image of the group with 1 homogenization treatment; C represents the detection image of the group with 2 homogenization treatments; D represents the detection image of the group with 3 homogenization treatments. Figure 10 A is a graph for detecting protein oxidation products; B is a graph for detecting NFK content; C is a graph for detecting dityrosine content. Figure 11 Curve fitting results for homogenization cycles and homogenization pressures related to different detection indicators; AH represents average particle size (D), zeta potential (Zeta), emulsifying activity (EAI), emulsifying stability (ESI), adsorbed protein (AP), malondialdehyde (MDA), protein oxidation product N'-formylkynurenine (NFK), and protein oxidation product (DT), respectively. Figure 12 The relationship between the emulsifying ability and physicochemical properties of sterilization samples under different homogenization parameters; Figure 13 Vertical clustering for homogenization parameters and horizontal clustering for physicochemical properties. Detailed Implementation

[0024] Those skilled in the art can refer to the content of this document and appropriately improve the process parameters to achieve the desired results. It should be particularly noted that all similar substitutions and modifications are obvious to those skilled in the art and are considered to be included in this invention. The methods and applications of this invention have been described through preferred embodiments, and those skilled in the art can obviously make modifications or appropriate alterations and combinations to the methods and applications described herein without departing from the content and scope of this invention to implement and apply the technology of this invention.

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. Unless otherwise specified, the experimental methods used in the following embodiments are conventional methods, and the materials, reagents, methods, and instruments used are all conventional materials, reagents, methods, and instruments in the art, and can be obtained commercially by those skilled in the art.

[0026] The sources of materials involved in the following embodiments: The whey protein powder was purchased from Fonterra Australia Pty Ltd (Richmond, Victoria, Australia); the lactose was purchased from Murray Goulburn Ingredients (MG Co-operative Co. Limited, Melbourne, Australia); the maltodextrin was purchased from Xiwang Group (Shandong, China); and the edible vegetable oil was purchased from Bungelord (Xiamen, China).

[0027] Example 1: S1: Dissolve whey protein powder (protein content of 80%) and carbohydrates (the carbohydrates are obtained by mixing lactose and maltodextrin in a mass ratio of 7:3) in deionized water at 60°C, stir at 600 rpm for 2 hours, and store at 4°C overnight to obtain hydrated whey protein powder and carbohydrates respectively. S2: The hydrated whey protein powder, carbohydrates and edible vegetable oil in S1 are mixed evenly at a mass-volume ratio of 7:21:12, and processed using an IKA high-shear homogenizer at 10000 rpm for 1 min to obtain an emulsified mixture. S3: The emulsified mixture in S2 is placed in a high-pressure homogenizer and homogenized 1, 2 and 3 times under a pressure of 20 MPa to obtain homogenized mixtures. The homogenized liquids are then sterilized in a 95℃ oil bath for 15 s and then dried into powder using a vacuum spray drying tower (vacuum pressure -20 kPa, inlet air temperature 115℃, outlet air temperature 60℃, gas flow rate 30 L / min, peristaltic pump speed 30 r / min) to obtain premature infant / low birth weight infant formula milk powder, abbreviated as: S-1-20, S-2-20 and S-3-20 respectively.

[0028] The naming principle for the premature / low birth weight infant formula obtained above is: Snp, where n and p represent the number of homogenization processes and the homogenization pressure, respectively.

[0029] Example 2: S1: Dissolve whey protein powder (protein content of 80%) and carbohydrates (the carbohydrates are obtained by mixing lactose and maltodextrin in a mass ratio of 7:3) in deionized water at 60°C, stir at 600 rpm for 2 hours, and store at 4°C overnight to obtain hydrated whey protein powder and carbohydrates respectively. S2: The hydrated whey protein powder, carbohydrates and edible vegetable oil in S1 are mixed evenly at a mass-volume ratio of 7:21:12, and processed using an IKA high-shear homogenizer at 10000 rpm for 1 min to obtain an emulsified mixture. S3: The emulsified mixture in S2 is placed in a high-pressure homogenizer and homogenized 1, 2 and 3 times under a pressure of 40 MPa to obtain homogenized mixtures. The homogenized liquids are sterilized in a 95℃ oil bath for 15 s and then dried into dry powder using a vacuum spray drying tower (vacuum pressure -20 kPa, inlet air temperature 115℃, outlet air temperature 60℃, gas flow rate 30 L / min, peristaltic pump speed 30 r / min) to obtain premature infant / low birth weight infant formula milk powder, abbreviated as: S-1-40, S-2-40 and S-3-40 respectively.

[0030] Example 3: S1: Dissolve whey protein powder (protein content of 80%) and carbohydrates (the carbohydrates are obtained by mixing lactose and maltodextrin in a mass ratio of 7:3) in deionized water at 60°C, stir at 600 rpm for 2 hours, and store at 4°C overnight to obtain hydrated whey protein powder and carbohydrates respectively. S2: The hydrated whey protein powder, carbohydrates and edible vegetable oil in S1 are mixed evenly at a mass-volume ratio of 7:21:12, and processed using an IKA high-shear homogenizer at 10000 rpm for 1 min to obtain an emulsified mixture. S3: The emulsified mixture in S2 is placed in a high-pressure homogenizer and homogenized 1, 2 and 3 times under a pressure of 60 MPa to obtain homogenized mixtures. The homogenized liquids are sterilized in a 95℃ oil bath for 15 s and then dried into dry powder using a vacuum spray drying tower (vacuum pressure -20 kPa, inlet air temperature 115℃, outlet air temperature 60℃, gas flow rate 30 L / min, peristaltic pump speed 30 r / min) to obtain premature infant / low birth weight infant formula milk powder, abbreviated as: S-1-60, S-2-60 and S-3-60 respectively.

[0031] Comparative Example 1: S1: Dissolve whey protein powder (protein content of 80%) and carbohydrates (the carbohydrates are obtained by mixing lactose and maltodextrin in a mass ratio of 7:3) in deionized water at 60°C, stir at 600 rpm for 2 hours, and store at 4°C overnight to obtain hydrated whey protein powder and carbohydrates respectively. S2: The hydrated whey protein powder, carbohydrates and edible vegetable oil in S1 are mixed evenly at a mass-volume ratio of 7:21:12, and processed using an IKA high-shear homogenizer at 10000 rpm for 1 min to obtain an emulsified mixture. S3: The emulsified mixture in S2 is placed in a high-pressure homogenizer and homogenized 1, 2 and 3 times under a pressure of 80 MPa to obtain homogenized mixtures. The homogenized liquids are sterilized in a 95℃ oil bath for 15 s and then dried into dry powder using a vacuum spray drying tower (vacuum pressure -20 kPa, inlet air temperature 115℃, outlet air temperature 60℃, gas flow rate 30 L / min, peristaltic pump speed 30 r / min) to obtain premature infant / low birth weight infant formula milk powder, abbreviated as: S-1-80, S-2-80 and S-3-80 respectively.

[0032] Comparative Example 2: S1: Dissolve whey protein powder (protein content of 80%) and carbohydrates (the carbohydrates are obtained by mixing lactose and maltodextrin in a mass ratio of 7:3) in deionized water at 60°C, stir at 600 rpm for 2 hours, and store at 4°C overnight to obtain hydrated whey protein powder and carbohydrates respectively. S2: The hydrated whey protein powder, carbohydrates and edible vegetable oil in S1 are mixed evenly at a mass-volume ratio of 7:21:12, and processed using an IKA high-shear homogenizer at 10000 rpm for 1 min to obtain an emulsified mixture. S3: The emulsified mixture in S2 is placed in a high-pressure homogenizer and homogenized 1, 2 and 3 times under a pressure of 100 MPa to obtain homogenized mixtures. The homogenized liquids are sterilized in a 95℃ oil bath for 15 s and then dried into dry powder using a vacuum spray drying tower (vacuum pressure -20 kPa, inlet air temperature 115℃, outlet air temperature 60℃, gas flow rate 30 L / min, peristaltic pump speed 30 r / min) to obtain premature infant / low birth weight infant formula milk powder, abbreviated as: S-1-100, S-2-100 and S-3-100 respectively.

[0033] Effect Experiment: To investigate the oxidative stability characteristics of the premature / low birth weight infant formula provided by the present invention, the formulas prepared in Examples 1-3 and Comparative Examples 1-2 were subjected to the following effect experiments; wherein, S-1 is the single homogenization treatment group, S-2 is the double homogenization treatment group, and S-3 is the triple homogenization treatment group.

[0034] 1. Detection of particle size and zeta potential At 25°C, the particle size and zeta potential of the formula milk powder under test were measured using a Nano ZS 90 (purchased from Malvern Instruments Ltd., Worcestershire, UK).

[0035]

[0036] In the formula: DI is the diameter corresponding to different particle size ranges; NI is the number of particles in that particle size range; ∑ is the summation.

[0037] like Figure 1As shown in section AC, the homogenized formula milk powder samples under low pressure conditions of 20 and 40 MPa exhibited a bimodal size distribution, indicating that the droplet distribution of the formula milk powder was not uniform. When the homogenization pressure exceeded 60 MPa, all the formula milk powder samples showed a unimodal distribution, although the widths varied. The bimodal distribution of sample S-2-80 may be due to over-processing under high pressure, inducing droplet aggregation. The droplet size ranges of samples S-1-100, S-2-100, and S-3-100 were 59.07-1209.00 nm, 92.89-1209.00 nm, and 92.89-660.90 nm, respectively. It can be seen that with the increase of homogenization cycles, the droplet size distribution gradually narrowed, and the uniformity increased. This indicates that homogenization at 100 MPa effectively emulsified the formula milk powder, forming a uniform and stable droplet size distribution.

[0038] like Figure 1 As shown in section D, the formula milk powder samples homogenized three times under low pressure conditions of 20 and 40 MPa exhibited significantly larger volume-average droplet sizes, indicating the occurrence of a large number of droplet aggregations. This is because the increased shear force and cavitation during high-pressure homogenization lead to increased protein kinetic energy and Brownian motion, promoting collisions and aggregation; shear force and cavitation also disrupt the secondary structure of proteins, exposing hydrophobic groups and enhancing interfacial adsorption. Conversely, the formula milk powder homogenized at 100 MPa exhibited significantly smaller volume-average droplet sizes. This is because the reduced shear force and cavitation during high-pressure homogenization allowed the formula milk powder proteins to rapidly adsorb onto new interfaces, forming stable interfacial films, preventing droplet re-aggregation, and achieving smaller droplet sizes.

[0039] like Figure 2 As shown, all tested formula milk powder samples exhibited a negative surface charge, indicating that the molecules adsorbed on the surface were negatively charged, and their pH value was higher than the isoelectric point of whey protein. The absolute values ​​of the zeta potential of samples S-1, S-2, and S-3 showed a trend of first increasing and then decreasing, with the lowest values ​​corresponding to different homogenization pressures of 40, 80, and 60 MPa, respectively; indicating that the number of homogenization cycles and homogenization pressure jointly affect the surface charge density of proteins in formula milk powder. The absolute zeta potential of sample S-1-40 reached the lowest at 15.57 mV, while that of sample S-3-100 reached the highest at 23.70 mV; indicating that homogenization under these conditions established a larger energy barrier between formula milk powder droplets, thereby promoting effective electrostatic repulsion. It is evident that high-intensity homogenization induces partial unfolding of the protein structure in formula milk powder, and synergistic sterilization and vacuum spray drying further promote this unfolding, exposing a large number of hydrophobic side chains, thus leading to an increase in absolute zeta potential; resulting in strong anti-aggregation properties of formula milk powder droplets and producing smaller particle sizes.

[0040] 2. Measurement of Emulsification Index Add 50 μL of the formula milk powder to be tested to a reaction tube containing 5 mL of SDS solution (0.10 g / L), measure the absorbance at 500 nm, let stand for 10 min, and repeat the above treatment 3 times; calculate the emulsification activity index (EAI) and emulsification stability index (ESI) by equation.

[0041]

[0042] In the formula: A 500 φ is the absorbance at 500 nm; φ is the oil phase volume fraction (v / v) (φ=0.2); C is the protein concentration (g / mL); A0, A 10 The absorbance values ​​of the emulsion at 0 min and 10 min are respectively.

[0043] During emulsification, proteins can be rapidly adsorbed at the oil-water interface, forming a dense interfacial film, thereby improving the emulsion stability index. EAI and ESI are key parameters for evaluating the emulsifying properties of proteins; EAI measures the adsorption capacity of proteins at the interface between the aqueous phase and oil droplets, while ESI assesses the ability of proteins to maintain their position at the oil-water interface during storage.

[0044] like Figure 3 As shown, sample S-1 exhibited increased EAI and ESI, indicating that increased homogenization pressure enhanced the emulsifying properties of the stability index. Sample S-2 showed a trend of first increasing and then decreasing EAI and ESI, reaching a peak at a homogenization pressure of 60 MPa; this indicates that appropriate homogenization pressure treatment can significantly reduce the interfacial tension between formula milk powder droplets and enhance electrostatic repulsion, thereby preventing droplet aggregation and improving the emulsification stability of formula milk powder. For sample S-3, within the homogenization pressure range of 20-80 MPa, the EAI value decreased significantly with increasing homogenization pressure; this is because excessive homogenization induces protein aggregation, hindering the exposure of hydrophobic regions of proteins, thus reducing emulsifying activity. The EAI value of sample S-3-100 was 14.85. This invention, through the synergistic effect of homogenization treatment with sterilization and vacuum spray drying technology, strengthens the interaction between various components in formula milk powder, resulting in the formation of carbohydrate-protein complexes within the formula milk powder system. Polysaccharides provide spatial stability, while proteins facilitate interfacial adsorption. Through their synergistic effect, droplet aggregation is effectively prevented and the emulsifying activity of the formula milk powder is enhanced.

[0045] 3. Detection of Adsorbed Protein (AP) Content Mix 1 mL of the formula milk powder to be tested with 1 mL of isopropanol into a centrifuge tube, centrifuge at 4℃ and 12000×g for 15 min, and repeat the above treatment 3 times; mix the cream layer (the opaque layer after centrifugation) with PBS solution to restore the total volume to 1 mL, and shake continuously for 24 hours. The cream layer (opaque layer) is the adsorbed protein, and the protein concentration is determined by the Coomassie brilliant blue method.

[0046] Proteins adsorbed at the oil-water interface (adsorbed proteins, APs) form a protective film that stabilizes formula milk powder by preventing droplet aggregation and flocculation through steric hindrance and electrostatic repulsion. Therefore, higher AP content is usually associated with better emulsification stability of formula milk powder.

[0047] like Figure 4 As shown, the interfacial protein content of samples S-1 and S-2 increased with increasing homogenization pressure. This indicates that the increased homogenization pressure leads to a decrease in the size of the emulsion droplets, which in turn leads to a decrease in the size of the protein particles, exposing more hydrophobic groups at the interface and enhancing protein adsorption at the oil-water interface. In contrast, the interfacial protein content of sample S-3 decreased, with S-3-100 containing 2.14, a 17.68% decrease compared to S-3-20 (2.60). This is because the high-intensity multiple homogenization parameters caused droplets to continuously merge, resulting in a decrease in the adsorption surface area and protein separation from the oil-water interface, thus reducing the number of proteins adsorbed at the interface.

[0048] 4. Detection of Peroxide Value (POV) 2 mL of the formula milk powder to be tested was mixed with 10 mL of isooctane / 2-propanol (3:1, v / v), vortexed, and centrifuged at 5000 rpm for 5 min. 1 mL of the supernatant was mixed with 2.8 mL of methanol / n-butanol solution (2:1, v / v), 15 μL of ammonium thiocyanate, and 15 μL of ferrous solution, vortexed, and incubated at room temperature for 20 min. After incubation, the absorbance was measured at 510 nm using a BioTek Synergy Neo2 multimode microplate reader (purchased from Agilent Technologies, USA), and the peroxide value was obtained from a standard curve constructed using cumene hydroperoxide.

[0049] Peroxides are byproducts of lipid oxidation, and their content can be used as an indicator of the degree of lipid oxidation in formula milk powder.

[0050] like Figure 5As shown, in samples S-1 and S-2, the POV content initially increased and then decreased with increasing homogenization pressure. This is because the increased mechanical stress and lipid droplet disruption led to greater exposure of lipid substrates to pro-oxidants present in the system, such as dissolved oxygen and metal ions. This accelerated the lipid oxidation rate, resulting in increased peroxide concentration and ultimately an increase in POV. The subsequent decrease in POV content is a more complex phenomenon, caused by the decomposition or consumption of peroxides through secondary reactions. For example, peroxides may react with other components in the formula milk powder (such as antioxidants or proteins), leading to their depletion and a decrease in POV content. In sample S-3, the POV content decreased with increasing homogenization pressure. As oxidation reactions occurred, peroxides accumulated, promoting their reaction with lipids. Simultaneously, hydrogen peroxide degraded to form alkoxy radicals, which further initiated chain reactions, promoting the formation of various oxidation products (such as aldehydes) and correspondingly increasing MDA content.

[0051] 5. Detection of malondialdehyde (MDA) content 100 mg of the formula milk powder to be tested was heated to 100℃ for 15 min, cooled to room temperature, and centrifuged at 10000×g for 10 min to obtain the supernatant. The absorbance of the supernatant was measured at 532 nm and 600 nm using an ELISA reader. The MDA content of the supernatant was detected using a detection kit (purchased from Nanjing Jiancheng Co., Ltd.).

[0052]

[0053] In the formula: V1 is the amount of sample added; V2 is the total volume of the sample added and the working solution; ε is the molar extinction coefficient of MDA; d is the optical path length.

[0054] like Figure 6As shown, MDA is one of the most common secondary oxidation products during lipid peroxidation. The MDA content in samples S-1 and S-2 showed an increasing trend, rising from 0.92 to 1.03 and from 1.06 to 1.31, respectively, indicating that increased homogenization pressure reduced droplet size and increased the contact area between lipids and oxidants. In sample S-3, the MDA content first increased and then decreased with increasing homogenization pressure. Sample S-3-40 had the highest MDA content (1.88), which eventually decreased to 1.38 with further increases in homogenization pressure, but was still significantly higher than samples S-1 and S-2. This is because the reaction between protein carbonyl groups and aldehydes (such as 4-hydroxy-2-enyl and acrolein) generated during lipid peroxidation leads to the formation of reactive carbonyl derivatives (such as ketamine, ketones, and deoxyglucose) within the protein. Therefore, it is speculated that during protein-lipid peroxidation, carbonyl groups and acrolein further react to form reactive carbonyl derivatives.

[0055] 6. Detection of carbonyl content The extracted protein, i.e., adsorbed protein AP, was dissolved in 10 mM phosphate buffer (pH=7.0) and stirred at 25℃ for 2 h. The supernatant was centrifuged at 10000 rpm for 10 min to obtain a protein suspension. 700 μL of the protein suspension was mixed with 2 mL of 2,4-dinitrophenylhydrazine (10 mM) and reacted at room temperature in the dark for 2 h (a protein sample containing 2 M HCl was used as a blank group). 900 μL of 40% TCA was added to precipitate the protein, and the mixture was centrifuged at 4℃ and 10000 rpm for 10 min. The supernatant was discarded. The precipitate was washed with 3.5 mL of TCA, and the supernatant was discarded again. The precipitate was then washed three times with 3 mL of ethanol / ethyl acetate (1:1, v / v) solution. The precipitate was then dissolved in 2 mL of 6M guanidine hydrochloride solution, and the absorbance was measured at 370 nm using a microplate reader to obtain the carbonyl content.

[0056]

[0057] In the formula: V_T is the total reaction volume (mL); ε is the molar extinction coefficient (22000 M⁻¹·cm⁻¹); d is the optical path length of the cuvette (cm); C_protein is the protein concentration (mg·mL⁻¹).

[0058] Protein carbonyl compounds are early biomarkers of oxidative modification. The content of protein carbonyl compounds indicates the degree of oxidative damage to proteins and is positively correlated with the degree of protein oxidation.

[0059] like Figure 7As shown, the carbonyl content in sample S-1 increased significantly with increasing homogenization pressure, indicating that protein oxidation became more severe with increasing homogenization pressure. The carbonyl content in sample S-2 decreased slightly with increasing homogenization pressure, while the carbonyl content in sample S-3 showed a significant trend of first increasing and then stabilizing, reaching a peak at a homogenization pressure of 40 MPa. This is because, with increasing homogenization cycles and homogenization pressure, protein carbonyl compounds further reacted with amino groups to form Schiff base adducts, leading to protein-protein crosslinking and consequently a decrease in carbonyl content.

[0060] 7. Detection of free thiol content The test formula milk powder was diluted to a concentration of 0.1 mg / mL using Tris-Gly buffer with a pH of 7.0. 0.5 mL of the diluted test formula milk powder sample was added to 20 μL of Ellman reagent and placed in a water bath at 25±1℃ for 30 min. The absorbance at 412 nm was then measured.

[0061]

[0062] Where: V_T: total reaction volume (mL); ε: molar extinction coefficient (13600 M⁻¹·cm⁻¹); d: optical path length of the cuvette (cm); C_protein: protein concentration (mg·mL⁻¹).

[0063] Free thiol groups (-SH groups and disulfide bonds -SS-) are key functional components of proteins, and their interconversion can regulate the spatial conformation and functional properties of proteins.

[0064] like Figure 8 As shown, in samples S-1 and S-2, the content of -SH groups increased with increasing homogenization pressure, indicating that increased homogenization pressure improved the oxidation level of whey protein, gradually increasing the content of -SH groups. In sample S-3, the content of -SH groups showed a trend of first increasing and then decreasing. This is because the electrophilic properties of MDA generated from lipid oxidation preferentially extract hydrogen atoms from the -SH groups of cysteine, thereby forming sulfur radicals and disulfide bonds.

[0065] 8. Detection of endogenous fluorescence spectroscopy The formula milk powder to be tested was diluted 10-fold using phosphate buffer (10 mmol / L, pH 7.0). The fluorescence spectrophotometer was calibrated using the following parameters: excitation wavelength of 280 nm, wavelength scan range of 310 to 400 nm, and slit width of 5 min. The intrinsic fluorescence spectrum of the diluted formula milk powder sample was detected using the fluorescence spectrophotometer.

[0066] In whey proteins, most of the fluorescence observed is primarily due to tryptophan, as it is an abundant essential amino acid in whey proteins and has the highest quantum yield among aromatic amino acids.

[0067] like Figure 9 As shown, the fluorescence intensity of sample S-1 increased with increasing homogenization pressure, indicating that high homogenization pressure promoted the complete unfolding of interfacial protein molecules, exposing hydrophobic groups. This invention further exposes protein fluorophores by combining homogenization treatment with sterilization and vacuum spray drying, enhancing the intrinsic fluorescence intensity. The fluorescence intensity of samples S-2 and S-3 did not show a significant increasing trend with increasing homogenization pressure. Sample S-2-60 exhibited the highest fluorescence intensity, while the fluorescence intensity of samples S-2-80 and S-2-100 decreased. This is because excessive homogenization induces protein-protein interactions, causing tryptophan to be mainly located on the protein surface, in close contact with oxidized lipids. Therefore, it is easily attacked during oxidation reactions, inducing conformational changes in protein molecules and leading to a decrease in the intrinsic fluorescence of tryptophan.

[0068] 9. Detection of protein oxidation products The content of N′-formyl-l-kynurenine (N′-formylkynurenine) in 0.1 mg / mL of the tested formula milk powder was determined using a fluorescence spectrophotometer. The fluorescence spectrophotometer was calibrated using the following parameters: excitation wavelength of 330 nm, emission wavelength of 440 nm, and slit width of 10 nm. The content of dityrosine in the 0.1 mg / mL formula milk powder was determined using a fluorescence spectrophotometer. The fluorescence spectrophotometer was calibrated using the following parameters: excitation wavelength of 325 nm, emission wavelength of 420 nm, and slit width of 10 nm.

[0069] N'-Formylkynurenine is a protein oxidation product and an oxidative degradation product of tryptophan. It is quantified using a fluorescence-based method to assess the degree of oxidation in infant formula. The formation of dityrosine is due to the dimerization of two tyrosine residues induced by hydrogen peroxide-derived lipid free radicals. An increase in dityrosine levels indicates that lipid peroxidation has triggered protein oxidation.

[0070] like Figure 10 As shown in Part A, within the homogenization pressure range of 20-40 MPa, the number of homogenization cycles is positively correlated with the content of N'-formylkynurenine. Under a homogenization pressure of 100 MPa, the content of N'-formylkynurenine decreases with the increase of the number of homogenization cycles. This is because the aggregation or separation of interfacial proteins leads to a decrease in the level of N'-formylkynurenine within the adsorbed proteins.

[0071] like Figure 10 As shown in Part B, within the homogenization pressure range of 20-40 MPa, the number of homogenization cycles was positively correlated with the fluorescence intensity of dityrosine. At a homogenization pressure of 100 MPa, the fluorescence intensity of dityrosine decreased with increasing homogenization cycles. The fluorescence intensity of dityrosine exhibited the same trend as that of N'-formylkynurenine, indicating a positive quantitative correlation between the degree of oxidation and the levels of both dityrosine and N'-formylkynurenine.

[0072] Example 4: This embodiment provides an evaluation and analysis method. The evaluation and analysis method is based on the QSAR model and uses homogenization parameters and interface properties as engineering descriptors to predict oxidative stability. It designs the first ternary descriptor-response framework for formula milk powder system, which integrates homogenization parameters (processing descriptor), interface properties (structural descriptor), and oxidative stability (response activity). This allows for the evaluation and analysis of the oxidative stability of a premature / low birth weight infant formula milk powder prepared in the above embodiment.

[0073] The evaluation and analysis method is as follows: Step 1: Obtain the key parameters of the formula milk powder; Step 2: Construct a quartic polynomial function and fit the key parameters into the quartic polynomial function. At the same time, use the coefficient of determination and root mean square error to evaluate the goodness of fit. Step 3: Construct a QSAR-inspired multivariate model and analyze the fourth-order polynomial function that fits the key parameters to regress the optimal scaling.

[0074] In practical application, the key parameters of formula milk powder were first compiled, including average particle size (D), zeta potential (Zeta), emulsifying activity (EAI), emulsifying stability (ESI), adsorbed protein (AP), malondialdehyde (MDA), protein oxidation product N'-formylkynurenine (NFK), and protein oxidation product (DT). Data were expressed as mean ± standard deviation (SD) for each technical replication. One-way ANOVA and Duncan analysis were performed using SPSS 26.0 software to further determine differences between groups. GraphPad Prism 9 and Origin 2024 were used to generate graphs. Differences were considered significant when p-values ​​< 0.05. Next, the data were imported into MATLAB 2024 software, and the software's built-in curve fitter tool was used to adjust the x and y degrees using polynomials to achieve R0. 2 The results were relatively high, and a quadratic polynomial was finally determined. A quartic polynomial function was constructed, and the coefficients and goodness of fit were sorted out.

[0075] The quartic polynomial function is specifically:

[0076] Where x is the number of homogenization cycles, y is the homogenization pressure, and the coefficients are determined by an algorithm using the coefficient of determination (R²). 2 The goodness of fit was evaluated using the root mean square error (RMSE), and the results were presented in a visual manner and statistically evaluated, thus realizing the correlation between homogenization parameters, interface characteristics and oxidation stability.

[0077] A QSAR-inspired multivariate model was then constructed using the CATREG program with optimal scaling in SPSS 26.0 (IBM, Armonk, NY). The optimal scaling was analyzed by regressing a fourth-order polynomial function that fitted the key parameters. Prior to the analysis, time, pressure, AP, and EAI were monotonically transformed using ordinal splines (degree = 2, section = 3) to generate a fully monotonic nonlinear CATREG regression. Finally, correlation analysis and hierarchical cluster analysis were performed. Correlation analysis established a quantitative relationship between homogenized parameters and oxidation characteristics, while hierarchical cluster analysis highlighted the vertical clustering of homogenized parameters and the horizontal clustering of physicochemical properties.

[0078] Example 5: To accurately describe trends and relationships in the data and reveal potential patterns, this embodiment focuses on the relationship between homogenization pressure, homogenization cycle, and the emulsification and physicochemical properties of the formula milk powder samples. Among various curve fitting methods, the least squares method is chosen to determine the coefficients of the quadratic polynomial fitting function for the data curve. This method is based on minimizing the sum of squares of the vertical differences between actual data points and their corresponding points on the fitted curve, thereby effectively identifying the fitting function that best approximates the data. After adjusting multiple parameters, Figure 11 Curve fitting results for homogenization cycling and homogenization pressure related to different detection indices are described. Both homogenization cycling and homogenization pressure can be modeled as fourth-order polynomial functions of mean particle size (D), zeta potential (Zeta), emulsifying activity (EAI), emulsifying stability (ESI), adsorbed protein (AP), malondialdehyde (MDA), protein oxidation product N'-formylkynurenine (NFK), and protein oxidation product (DT). Specific fitting formulas are shown in Table 1. The coefficient of determination (R²) is also presented. 2The coefficient of determination (COD) and root mean square error (RMSE) are key indicators for evaluating the performance of the fit. The COD measures the contribution of the fitted curve to the data; a value close to 1 indicates excellent fit performance. The RMSE reflects the average error between the fitted curve and the actual data points; a smaller value indicates higher accuracy. In our in-depth study of the homogenization process of infant formula, detailed analysis and processing of existing datasets revealed that the quadratic polynomial functions fitted by EAI and ESI regarding homogenization cycles and homogenization pressure exhibited outstanding characteristics, with CODs of 0.9944 and 0.9941, respectively. These values ​​are very close to 1, clearly indicating a high degree of consistency between the constructed fitted model and the actual data. Simultaneously, the RMSE values ​​were only 0.5572 and 0.1969, meaning that the relatively low error further confirms the accuracy of the fitted curves in describing data changes. Among the many potential forms of quadratic polynomial functions, these two fitted curves are particularly adept at capturing complex trends embedded in the data, thus demonstrating the best fit performance and the highest level of accuracy within the current research scope. The curve fitting results can successfully predict the processing parameters and formula milk powder characteristics, but no quantitative relationship is formed with oxidation characteristics. Therefore, the following correlation analysis was also performed in this embodiment.

[0079] Figure 12 The heatmaps provided illustrate the relationship between emulsifying ability and physicochemical properties of sterilized samples under different homogenization parameters. Blue tones indicate a negative correlation, with darker shades representing stronger negative correlations. Conversely, red tones indicate a positive correlation, with brighter colors indicating more pronounced positive correlations. Notably, the particle size of the formula milk powder showed a significant positive correlation with zeta potential (ζ-potential), with a correlation coefficient of 0.75. This implies a significant negative correlation between the absolute value of the formula milk powder particle size and the ζ-potential. This phenomenon may be attributed to the fact that smaller particle sizes typically have larger specific surface areas. Under the same total charge, this leads to an increased surface charge density, resulting in either a higher absolute value or a lower ζ-potential. The finding that formula milk powder exhibiting small particle size and high electrostatic repulsion improves emulsification stability index has been confirmed by numerous studies. Emulsifying activity showed a significant positive correlation with interfacial protein content (correlation coefficient of 0.53) and a significant negative correlation with MDA content (correlation coefficient of -0.56). This indicates that lower levels of lipid oxidation within the droplets, as shown by lower MDA levels, mitigate the oxidative aggregation of interfacial proteins, thereby preventing their separation and ensuring greater adsorption of these proteins. Given the hydrophilicity of proteins, this ultimately enhances emulsifying activity. Dityrosine and N'-formylkynurenine, both products of protein and lipid oxidation, particularly MDA, all showed significant positive correlations, with correlation coefficients of 0.76, 0.54, and 0.66, respectively. This suggests a mutually reinforcing effect between lipid and protein oxidation in infant formula, a phenomenon also highlighted in several other studies.

[0080] Based on the above results, it can be seen that by clarifying the correlation between the characteristics and oxidative properties of formula milk powder, the interfacial protein content and emulsifying activity can be increased by modifying the homogenization process parameters, thereby minimizing protein-lipid co-oxidation and enhancing the emulsification stability of formula milk powder. Based on this finding, the homogenization process parameters (number of homogenization cycles and homogenization pressure) were set as X1 and X2, the interfacial characteristics (interfacial protein content and emulsifying activity) as Y1 and Y2, and the key oxidative property indicator MDA as Z. A monotonic nonlinear regression equation was constructed to obtain:

[0081] It should be noted that the unit of the homogeneous pressure parameter in this regression equation has been changed from MPa to GPa. A quantitative structure-activity relationship model of "processing parameters-interfacial properties-oxidative stability" has been established.

[0082] However, the current ternary regression model is built on a finite experimental matrix (n = 15). Although the optimal scaling regression provides satisfactory in-sample predictions (R² = 0.967), the model remains a proof of concept. Future work should include: 1. Expand the dataset to cover a wider range of processing and compositional variables; 2. Explore machine learning or ensemble methods to handle potential non-cumulative effects; 3. External validation was conducted using independent batches of preterm formula milk powder to refine or reparameterize the prediction equation.

[0083] Furthermore, this embodiment also performs hierarchical cluster analysis. Figure 13The results were presented as heatmaps, highlighting the more prominent vertical clustering of homogenization parameters and the horizontal clustering of physicochemical properties. The horizontal clustering of emulsifying properties and physicochemical properties revealed a high degree of similarity between particle size and zeta potential, as well as between dityrosine and N'-formylkynurenine content. Furthermore, some similarity was observed between emulsifying activity and interfacial protein content. The clustering results of homogenization parameters divided the samples into two groups. The first group comprised nine samples—S-1-20, S-1-80, S-2-60, S-2-40, S-1-100, S-1-40, S-1-60, S-2-80, and S-2-20—which collectively exhibited relatively low levels of lipid oxidation products, particularly MDA. The second group comprises six samples, notably S-2-100, S-3-20, S-3-40, S-3-60, S-3-80, and S-3-100, which exhibited low emulsifying activity accompanied by high levels of protein oxidation products, such as dityrosine and N'-formylkynurenine, and high levels of lipid oxidation product MDA. This suggests that high homogenization intensity (e.g., three homogenization cycles and two homogenization cycles under high pressure) may adversely affect formula milk quality. These results provide valuable insights for the reference and optimization of homogenization parameters. Generally, our primary objective is to achieve superior emulsifying properties and reduce the levels of protein-lipid co-oxidation products. Therefore, the nine samples in the first group are more conducive to meeting product quality requirements, with sample S-2-40 standing out particularly for its high emulsifying activity and low content of protein-lipid co-oxidation products, making it a prime candidate for further research.

[0084] In summary, this invention plays a crucial role in improving interfacial protein adsorption and emulsification activity by optimizing homogenization process parameters, effectively inhibiting protein-lipid co-oxidation and enhancing the emulsification stability of formula milk powder. Appropriate homogenization pressure helps form suitable droplet size distribution and zeta potential, promoting the construction of a stable interfacial protein layer. However, excessive homogenization (e.g., 2-3 cycles under high pressure) has multiple negative effects: it induces droplet aggregation and increased particle size, reduces the absolute value of the zeta potential, leading to desorption and reduced content of interfacial proteins; simultaneously, it induces protein oxidative damage, manifested as conformational changes (decreased intrinsic fluorescence intensity), accumulation of oxidation products (N'-formylkynurenine and dityrosine), and a significant increase in lipid oxidation indicators (peroxide value and malondialdehyde). Notably, under these conditions, the degree of protein-lipid co-oxidation is significantly higher than in other homogenization parameter groups. These findings highlight the importance of precisely controlling homogenization intensity in balancing the emulsification stability and oxidative protection of formula milk powder in preterm / low birth weight infant formula systems, providing an important theoretical basis for the optimization of related product processes. Through systematic data modeling and analysis, the quantitative relationship between homogenization process parameters and the oxidative stability of formula milk powder was revealed. Curve fitting and correlation analysis results showed that there was a very high correlation (R²>0.99) between homogenization cycle number, homogenization pressure, and emulsifying activity index (EAI) and emulsifying stability index (ESI), demonstrating excellent fitting accuracy. Further research found that the degree of protein-lipid co-oxidation is closely related to the characteristics of formula milk powder. Based on this, this invention innovatively applied QSAR to the formula milk powder system for premature / low birth weight infants, constructing a monotonic nonlinear regression model that includes homogenization process parameters (homogenization cycle number X1 and homogenization pressure X2), interfacial characteristics (interfacial protein content Y1 and emulsifying activity Y2), and oxidation index (malondialdehyde Z), achieving for the first time a quantitative prediction of the complete action chain of "processing parameters-interfacial characteristics-oxidative stability". Hierarchical cluster analysis results further showed that the S-2-100 and S-3 sample groups, due to their poor emulsification characteristics and high oxidation degree, are not suitable for the preparation of formula milk powder for premature / low birth weight infants. This invention provides a new method for the precise manufacturing of formula for premature / low birth weight infants. This discovery not only identifies the optimal homogenization process window for subsequent research but also provides crucial data for optimizing process parameters in the industrial production of high-quality premature / low birth weight infant formula, demonstrating significant practical application value.

[0085] In this embodiment, Table 1 shows the fourth-degree polynomial functions with different coefficients.

[0086] Table 1

[0087] Where D(4,3) represents the average particle size, ζP represents the ζ potential, EAI represents the emulsifying activity, ESI represents the emulsifying stability, AP represents the adsorbed protein, MDA represents malondialdehyde (lipid oxidation product), NFK represents N′-formyl-l-kynurenine (protein oxidation product), and DT represents dityrosine (protein oxidation product).

[0088] The specific embodiments of the present invention disclosed above are merely illustrative of the invention. These embodiments do not exhaustively describe all details, nor do they limit the invention to the specific embodiments described. Many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention.

Claims

1. A low side effect production method of a preterm infant / low birth weight infant formula, characterized by, The preparation method comprises the following steps: S1: Dissolve the whey protein powder and the carbohydrate in 60°C deionized water respectively, stir for 2h under the condition of 600 rpm, and store overnight under the condition of 4°C to obtain the hydrated whey protein powder and the carbohydrate respectively; S2: Mix the hydrated whey protein powder and the carbohydrate in S1 with edible vegetable oil uniformly, treat the mixture in an IKA high-shear homogenizer under the condition of 10000 rpm for 1 min to obtain an emulsified mixture; S3: Homogenize the emulsified mixture in S2 in a high-pressure homogenizer to obtain a homogenized mixture, sterilize the homogenized mixture in a 95°C oil bath for 15 s, and then spray dry the homogenized mixture in a vacuum spray drying tower to obtain a dry powder, thereby obtaining a premature infant / low birth weight infant formula milk powder.

2. The low side reaction production method according to claim 1, characterized by, The whey protein powder in S1 has a protein content of 80%; and the carbohydrate is a mixture of lactose and maltodextrin in a mass ratio of 7:

3.

3. The low side reaction production method according to claim 1, characterized by, The mixture of the whey protein powder, the carbohydrate and the edible vegetable oil in S2 has a mass-volume ratio of 7:21:

12.

4. The low side reaction production method according to claim 1, characterized by, The homogenization in S3 is performed under the following conditions: the pressure is 20-60 MPa, and the number of homogenization is 1-2.

5. The low side reaction production method according to claim 1, characterized by, The process parameters of the vacuum spray drying tower in S3 are as follows: the vacuum pressure is -20 kPa, the inlet air temperature is 115°C, the outlet air temperature is 60°C, the gas flow rate is 30 L / min, and the peristaltic pump speed is 30 r / min.

6. A formula for preterm / low birth weight infants, characterized by comprising, The premature infant / low birth weight infant formula milk powder is obtained by the low side reaction preparation method according to any one of claims 1-5.

7. An evaluation analysis method characterized by, The evaluation and analysis method is based on a QSAR model, which takes the homogenization parameters and the interfacial properties as engineering descriptors for predicting the oxidative stability, thereby evaluating and analyzing the oxidative stability of the premature infant / low birth weight infant formula milk powder according to claim 6.

8. The evaluation analysis method according to claim 7, characterized by, The method specifically comprises the following steps: Step 1: Obtain the key parameters of the formula milk powder; Step 2: Construct a quartic polynomial function, fit the key parameters into the quartic polynomial function, and evaluate the fitting degree by using the determination coefficient and the root mean square error; The quartic polynomial function is as follows: wherein x is the number of homogenization cycles, and y is the homogenization pressure; Step 3: Construct a QSAR-based multivariate model to establish a multiple regression analysis of the number of homogenization, the homogenization pressure, the interfacial protein and the emulsification activity.

9. The evaluation analysis method according to claim 8, characterized by, The key parameters include the average particle size, the zeta potential, the emulsification activity EAI, the emulsification stability ESI, the adsorbed protein AP, the malondialdehyde MDA, the protein oxidation product N'-formylkynurenine and the protein oxidation product DT.

10. The evaluation analysis method according to claim 9, characterized by, The multiple regression analysis specifically comprises the following steps: Firstly, convert all the prediction variables and the response variables into monotonic ordered number spline curves before analysis to generate completely monotonic nonlinear CATREG regression; Then, perform correlation analysis to form a quantitative relationship between the homogenization parameters and the oxidative properties; Finally, perform hierarchical cluster analysis to highlight the vertical clustering of the homogenization parameters and the horizontal clustering of the physicochemical properties.

Images