Preparation method of metal powder injection molding brass alloy

Abstract

The invention discloses a metal powder injection molding brass alloy preparation method, and relates to the field of metal powder injection molding, feedstock preparation: spherical brass alloy powder and a binder are mixed and granulated to prepare feedstock; the binder comprises polyformaldehyde, high density polyethylene and an ethylene-vinyl acetate copolymer; injection molding: carrying out injection molding on the feed to obtain a green body; catalytic degreasing: the green body is placed in a nitric acid steam atmosphere for catalytic degreasing, and a brown blank is obtained; performing one-step sintering: performing further thermal degreasing and sintering on the brown blank in a sintering furnace in a mixed atmosphere consisting of nitrogen and hydrogen with the volume fraction of 1-5%; wherein in the process of raising the temperature from the room temperature to the final sintering temperature of 880-920 DEG C, a medium-temperature heat preservation platform specially used for reducing the metal oxide is not arranged; compared with a traditional process, the method has the advantages that the production period is shortened, the energy consumption is reduced, and the problem of excessive volatilization of zinc in a medium-temperature stage is avoided.

Description

Technical Field

[0001] This invention relates to the field of metal powder injection molding technology, and more specifically, to a method for preparing brass alloys by metal powder injection molding. Background Technology

[0002] Metal powder injection molding is a near-net-shape forming technology that combines plastic injection molding with powder metallurgy. It is particularly suitable for producing small, complex, and precision metal parts. Brass alloys, due to their excellent mechanical properties, corrosion resistance, electrical conductivity, and attractive color, have broad application prospects in fields such as locks, valves, hardware, decorative items, and microelectronic components.

[0003] Traditional brass powder injection molding typically involves four basic steps: feedstock preparation, injection molding, debinding, and sintering. However, this technology faces two key technical bottlenecks in industrial applications: First, zinc in brass has a high saturated vapor pressure and is chemically reactive, making it highly volatile and oxidizable during high-temperature sintering. This leads to deviations in alloy composition from design values, decreased mechanical properties of the product, surface defects, and contamination of the sintering equipment. Second, existing debinding and sintering processes are time-consuming, energy-intensive, and heavily reliant on operator experience, making it difficult to guarantee consistent quality across different batches.

[0004] To suppress the volatilization and oxidation of zinc, existing technologies often employ the pre-addition of excess zinc to the powder or the use of a high-concentration hydrogen atmosphere (usually >5% by volume) during sintering, and the setting of a long-term heat preservation platform in the range of 500°C to 600°C to fully reduce the oxides.

[0005] However, the former leads to inaccurate component control and increased costs; the latter brings safety risks and cost pressures, and the prolonged medium-temperature holding actually widens the volatilization window of zinc, failing to fundamentally solve the problem. Furthermore, traditional hot degreasing processes take tens of hours, and the decomposition of the binder during the process may generate a localized atmosphere unfavorable to zinc protection. Although catalytic degreasing technology can significantly shorten degreasing time, its process parameters (such as time) are currently mostly set based on experience and fail to correlate with the geometric characteristics of the green blank. For complex structural parts, uneven or excessive degreasing is prone to occur. In the injection molding stage, process parameters are also mostly fixed values, making it difficult to adapt to the slight fluctuations in feed batches, resulting in uneven green blank density, increased defects, and affecting the dimensional accuracy and performance consistency of the final product.

[0006] Therefore, there is an urgent need in this field to develop a novel method for preparing brass alloy powder by injection molding to solve the above problems. Summary of the Invention

[0007] To address the above problems, this invention provides a method for preparing brass alloys by metal powder injection molding.

[0008] This invention provides a method for preparing brass alloys by metal powder injection molding, comprising the following steps:

[0009] S1. Feed preparation: Spherical brass alloy powder is mixed with a binder and granulated to prepare a feed; the binder contains polyoxymethylene, high-density polyethylene and ethylene-vinyl acetate copolymer.

[0010] S2, Injection Molding: The feed material is injection molded to obtain a green body;

[0011] S3. Catalytic degreasing: The green body is placed in a nitric acid vapor atmosphere for catalytic degreasing to obtain a brown body;

[0012] S4. One-step sintering: The brown blank is further degreased and sintered in a sintering furnace under a mixed atmosphere of nitrogen and hydrogen with a volume fraction of 1% to 5%; wherein, during the process of heating from room temperature to the final sintering temperature of 880-920℃, no medium-temperature heat preservation platform specifically for reducing metal oxides is set up.

[0013] Preferably, in step S1:

[0014] The spherical brass alloy powder is prepared by gas atomization, with an oxygen content of no more than 300 ppm and a particle size distribution satisfying: D50 = 8-15 μm;

[0015] The adhesive, by weight percentage, consists of the following components: 70%–85% polyoxymethylene, 10%–20% high-density polyethylene, 3%–8% ethylene-vinyl acetate copolymer, and 1%–3% stearic acid.

[0016] Preferably, in step S4, the temperature program for the one-step sintering includes: first heating to 450-500℃ at a rate of 0.5-2℃ / min and holding at that temperature for 60-90 minutes, then heating to 895-905℃ at a rate of 3-8℃ / min and holding at that temperature for 60-120 minutes.

[0017] Preferably, before performing step S4, a step S0 is included to determine the sintering parameters:

[0018] S01. Obtain the specific surface area (SSA) and actual powder loading (φ) of the current batch of brass feed, according to the formula: Calculate the dynamic sintering factor (DSF), where Where K is the theoretical density of brass, and K is a process equipment constant.

[0019] S02, based on the calculation The optimal heating rate (R) and the optimal holding time (t) at the final sintering temperature are determined in step S4 by using a pre-established mapping relationship.

[0020] Furthermore, in step S4, the sintering process of the final sintering stage is performed using the optimal heating rate (R) and optimal holding time (t).

[0021] Preferably, in step S02, the optimal heating rate (R) is determined by the following formula: The optimal heat preservation time (t) is determined by the following formula: , where a, b, c, d are constants fitted through a historical process database.

[0022] Preferably, in step S02, the pre-established mapping relationship is implemented by a machine learning model.

[0023] Preferably, in step S01, the specific surface area (SSA) is measured by the BET method, and the actual powder loading (φ) is measured by thermogravimetric analysis.

[0024] Preferably, during the injection molding process in step S2, the holding pressure ( ) and holding time ( The following methods were used to optimize and determine the following:

[0025] S21. Obtaining State Parameters: Obtain the melt pressure (Pcavity) in the mold cavity at the end of the injection stage and the maximum flow length to thickness ratio (L / T) of the product.

[0026] S22. Calculation of pressure holding parameters: Pressure holding pressure ( The formula for calculating ) is: Holding time ( The formula for calculating ) is: ,in , This is a compensation coefficient related to the material's shrinkage characteristics;

[0027] Furthermore, the pressure holding stage in step S2 uses the calculated... and implement.

[0028] Preferably, in step S3, the catalytic degreasing treatment time ( The determination is made dynamically in the following ways:

[0029] S31. Geometric parameter acquisition: Obtain the critical wall thickness (D) of the green body, which is the maximum thickness of the green body in the direction of degreasing atmosphere diffusion.

[0030] S32. Calculation of defatting time: Calculation of defatting time: Based on the formula... Calculate the required catalytic defatting time (Tc), where These are process constants related to nitric acid vapor concentration and temperature;

[0031] Furthermore, the catalytic degreasing treatment time in step S3 is based on the calculated... implement.

[0032] Preferably, the method is executed by a smart manufacturing system, which includes:

[0033] The parameter measurement module is used to measure the specific surface area of ​​powder, the actual feed loading, the melt pressure, and the geometric parameters of green body.

[0034] The data processing and calculation module includes a dynamic sintering factor calculation model, a pressure holding parameter optimization model, a catalytic debinding time calculation model, and a sintering parameter mapping relationship.

[0035] The injection molding machine and sintering furnace control module is communicatively connected to the data processing module and is used to receive instructions and control the corresponding equipment to execute the process.

[0036] Beneficial effects:

[0037] The core of this method's innovation lies in: First, the use of customized low-oxygen spherical powder and composite binder in S1 not only improves feed flowability but, more importantly, reduces the need for oxide reduction in S4, providing a prerequisite for eliminating the need for medium-temperature insulation. Combined with the 80%–85% degreasing rate controlled by dynamic degreasing in S3, it reduces zinc volatilization from the source, ensuring accurate brass alloy composition and significantly improving mechanical and corrosion resistance compared to traditional processes.

[0038] Secondly, S2 is designed based on dynamic pressure holding parameters of melt pressure and flow length-to-thickness ratio, which precisely controls the green compact density of 65% to 70%. This density is not only suitable for the binder content of S1 feed, but also creates the optimal channel for the diffusion of nitric acid vapor in S3, forming a closed-loop control of "feeding-forming", which makes the green compact size accuracy and consistency significantly better than traditional fixed parameter forming.

[0039] Third, S3 uses dynamic degreasing time calculation based on key wall thickness and process constant k to adapt to the structure of S2 molded products and the degradation characteristics of S1 binder, ensuring that the strength of the brown blank is sufficient to withstand the temperature rise stress of S4 and avoid cracking defects.

[0040] Fourth, the S4 one-step sintering process eliminates the need for a medium-temperature holding platform. This is not simply a matter of omitting steps, but rather relies on the synergistic effect of the preceding low-oxygen powder and thorough degreasing to simultaneously complete the remaining degreasing and slight reduction during the 450-500℃ holding stage. This achieves the integration of "degreasing-reduction-sintering," which shortens the production cycle compared to traditional processes, reduces energy consumption, and avoids the problem of excessive zinc volatilization during the medium-temperature stage. Attached Figure Description

[0041] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0042] Application scenario: In the preparation of brass alloys by metal powder injection molding, the "catalytic degreasing + segmented sintering" process is often used. That is, after catalytic degreasing, a heat preservation platform needs to be set up in the medium temperature stage (600-700℃) to reduce the metal oxide.

[0043] This method not only extends the production cycle, but also easily leads to excessive volatilization of zinc in brass due to medium-temperature heat preservation, causing deviations in alloy composition. At the same time, segmented heating increases energy consumption and equipment control complexity. To solve the above problems...

[0044] like Figure 1 As shown: This invention provides a method for preparing brass alloy by metal powder injection molding, comprising the following steps:

[0045] The first step is to prepare the S1 feed: spherical brass alloy powder is mixed with a binder and granulated to form the feed; the binder contains polyoxymethylene, high-density polyethylene and ethylene-vinyl acetate copolymer.

[0046] In this embodiment, the spherical brass alloy powder is prepared by gas atomization. Its oxygen content is no more than 300ppm, the particle size distribution meets the requirements of D50=8-15μm, sphericity ≥0.9, and loose density ≥4.2g / cm³. These parameters can significantly improve the powder flowability and forming density. T2 brass atomized powder can meet the requirements. Compared with irregular powder, spherical powder makes the feeding filling of the mold cavity more full, effectively improving the dimensional accuracy of the subsequent green blank.

[0047] The binder, by weight percentage, consists of the following components: 70%–85% polyoxymethylene (POM), 10%–20% high-density polyethylene (HDPE), 3%–8% ethylene-vinyl acetate copolymer (EVA), and 1%–3% stearic acid. POM acts as the main binder, providing molding strength; HDPE enhances feed toughness; EVA strengthens the compatibility between the powder and the binder; and stearic acid acts as a lubricant, reducing mixing resistance and improving feed injection flowability. This composite binder system prevents feed from separating during storage, significantly extending shelf life and outperforming single POM binders.

[0048] The mixing process uses a twin-screw mixer with a mixing temperature of 160-180℃, a speed of 80-120 r / min, and a mixing time of 30-40 min to ensure that the binder uniformly coats the powder. Granulation is carried out by underwater pelletizing with a particle size of 2-5 mm. After granulation, the pellets are dried at 40-50℃ for 2-3 hours to remove surface moisture and avoid air bubble defects during subsequent injection. The uniformly coated powder and drying treatment can reduce the fluctuation of injection flowability of the feed and further improve the stability of green body quality.

[0049] The core value of this step lies in laying a foundation for the adaptability of subsequent whole-chain processes: the low-oxygen (≤300ppm) spherical powder prepared by gas atomization not only improves fluidity, but more importantly, reduces the oxide reduction requirement during S4 sintering, providing a prerequisite for eliminating the need for a medium-temperature heat preservation platform.

[0050] The lubricating effect of stearic acid in the composite binder and the toughness regulation of polyethylene not only achieve dense molding by coordinating with the dynamic holding pressure parameters during S2 injection, but also adapt to the rate requirements of S3 nitric acid vapor catalytic degreasing through the controllable degradation characteristics of polyoxymethylene. This pre-adaptation to subsequent design breaks through the limitation of traditional feed only focusing on its own performance, highlighting its non-obviousness.

[0051] The proportions are all based on process parameters explicitly defined in the patent, combined with material properties and industry-standard selections, as detailed below:

[0052] 1. Metal powder ratio

[0053] Spherical brass alloy powder (prepared by gas atomization method), oxygen content ≤300ppm, D50=8-15μm;

[0054] Low-oxygen spherical powder is a common choice for improving the flowability of MIM processes;

[0055] Polyoxymethylene 70%-85%, high-density polyethylene 10%-20%, ethylene-vinyl acetate copolymer 3%-8%, stearic acid 1%-3%;

[0056] Polyoxymethylene (POM) is the main binder, polyethylene enhances toughness, and stearic acid optimizes flowability. It is a commonly used composite binder system in MIM process, and the formulation is designed based on the compatibility of feeding, mixing, and degreasing.

[0057] Nitrogen gas + 1%-5% hydrogen gas;

[0058] A low-hydrogen atmosphere is a key design feature to suppress zinc volatilization, and it represents the safe atmosphere blending range for brass MIMs within the industry.

[0059] The second step is to perform S2 injection molding: the feed material is injected into the mold to obtain a green body.

[0060] In this embodiment, a horizontal injection molding machine is used for injection molding, and the injection temperature is set to 170-190℃. The temperature is controlled in three sections: 170-175℃ at the front of the barrel, 175-185℃ at the middle section, and 185-190℃ at the nozzle. The segmented temperature control can avoid local overheating and degradation of the feed material in the barrel, and greatly reduce the scorch defects on the surface of the green body.

[0061] Injection pressure 80-120MPa, holding pressure during the holding phase ( ) and holding time ( After optimization and confirmation, the process is as follows:

[0062] S21. Obtaining State Parameters: Obtain the melt pressure (Pcavity) in the mold cavity at the end of the injection stage and the maximum flow length to thickness ratio (L / T) of the product.

[0063] The melt pressure is collected in real time by a pressure sensor built into the mold. The collection time is 1 second before the pressure holding switching signal is triggered during the injection stage, so as to ensure that steady-state pressure data is obtained at the end of the injection.

[0064] The maximum flow length to thickness ratio (L / T) is the ratio of the maximum flow length (L) in the forming direction of the product to the minimum thickness (T) at the corresponding position. It is obtained by measuring the three-dimensional model of the product using CAD software, with a measurement accuracy of 0.1 mm.

[0065] S22. Pressure holding parameter calculation: Calculate the optimal pressure holding pressure based on an empirical model. ) and holding time ( ),in , This is a compensation coefficient related to the material's shrinkage characteristics;

[0066] Based on an empirical model of material shrinkage characteristics, combined with real-time process parameter acquisition.

[0067] In the empirical model, the holding pressure ( The calculation formula is as follows: Holding time ( The calculation formula is as follows: ;

[0068] Among them, the compensation coefficient , Calibration is required based on the shrinkage characteristics of the brass feedstock. The calibration method is as follows:

[0069] Using target feed, a standard stretching spline mold is injected at a fixed injection temperature, pressure, and mold temperature.

[0070] First, set a set of initial pressure holding parameters, such as , .

[0071] A large L / T ratio means a longer manufacturing process, thinner wall thickness, more difficult melt filling and holding pressure, and faster cooling, thus requiring a longer holding pressure time.

[0072] A smaller L / T ratio means a shorter manufacturing process, thicker walls, easier pressure holding, and a shorter time required.

[0073] 1.0: This is an empirical coefficient. In practical applications, this coefficient may need to be adjusted based on material properties (such as specific heat capacity, crystallinity) and mold temperature (e.g., it may be between 0.8 and 1.5).

[0074] After injection, measure the critical dimensions of the green sample (such as length L0) and compare them with the corresponding dimensions (Ldie) of the mold cavity.

[0075] Then, using the formula: Calculate the volume shrinkage rate.

[0076] Through iterative adjustments and The value (for example, if the green body shrinks insufficiently, increase it appropriately) If dents or shrinkage cavities appear, increase the size appropriately. This process continues until the dimensions of the green body are stable and the volume shrinkage rate is controlled within the ideal range of 1.5% to 2.5%. At this point, the method used... and This value is the calibrated value for that batch of feed. Typically, The value range is 0.8-1.2. The value range is 0.9-1.3.

[0077] During injection, the mold temperature is 40-60℃ and the cooling time is 10-20s. The specific cooling time can be adjusted according to the size of the part. For example, the cooling time for micro gears (outer diameter ≤ 5mm) is set to 10s, and for larger parts (outer diameter ≥ 20mm), it is extended to 20s.

[0078] The appropriate parameter settings can improve the consistency of the density of various parts blanks, enhance the stability of mass production, avoid stress concentration inside the blanks caused by uneven cooling, and reduce the risk of subsequent degreasing and cracking.

[0079] After injection, the green body is ejected smoothly using an ejection mechanism at a speed of 5-10 mm / s to prevent cracking due to uneven stress. The green body density is strictly controlled at 65%-70%. This density range is the key link between S1 and S3: it is necessary to match the binder content of the S1 feed to ensure molding, and to provide sufficient diffusion channels for the nitric acid vapor of S3.

[0080] Density is checked in real time using the drainage method. If it deviates, the injection pressure of S2 or the mixing parameters of S1 are adjusted in reverse to form a closed-loop control of feeding and molding.

[0081] The third step is to perform S3 catalytic degreasing: the green body is placed in a nitric acid vapor atmosphere for catalytic degreasing to obtain a brown body.

[0082] In this embodiment, the catalytic degreasing equipment is a sealed degreasing furnace, the nitric acid vapor concentration is controlled at 5% to 10%, the degreasing temperature is 80-100℃, and the catalytic degreasing treatment time is ( The process is as follows: The execution is dynamically determined and executed in the following way.

[0083] S31. Geometric parameter acquisition: Obtain the critical wall thickness (D) of the green body, which is the maximum thickness of the green body in the direction of degreasing atmosphere diffusion.

[0084] It should be noted that the 3D model of the green billet is measured using CAD software. During the measurement, the main diffusion path of nitric acid vapor inside the green billet is used as the reference, and the farthest distance in the diffusion direction is taken as the critical wall thickness (D). The measurement accuracy is accurate to 0.1 mm. For example, the diffusion direction of a plate-shaped green billet is the thickness direction, and its thickness is the critical wall thickness. For irregularly shaped green billets, the maximum diffusion thickness needs to be determined through multi-directional diffusion path analysis.

[0085] S32. Degreasing Time Calculation: Calculate the required catalytic degreasing time according to the formula. ), where α, β, and γ are process constants related to nitric acid vapor concentration and temperature;

[0086] The formula is process constants , , Orthogonal experiments were conducted beforehand to calibrate the process. The experimental factors were nitric acid vapor concentration (5%, 7.5%, 10%) and degreasing temperature (80℃, 90℃, 100℃). The complete degreasing time of green bodies with different critical wall thicknesses was tested under each factor combination. The results were obtained under each working condition through data fitting. Values ​​and establish a database;

[0087] To facilitate reproduction by those skilled in the art, the following table provides two sets of typical constants obtained through experimental fitting under the adhesive system described in this invention:

[0088] Nitric acid vapor concentration Degreasing temperature α(h / mm²) β(h / mm) γ(h) 6% 85℃ 1.15 0.65 1.5 9% 95℃ 0.82 0.45 1.0

[0089] Those skilled in the art can implement the present invention based on the above example constants, and can fine-tune the constants through a small number of conventional experiments according to their specific equipment and raw material conditions.

[0090] Furthermore, the defatting time is calculated dynamically ( Catalytic degreasing avoids the problems of excessive degreasing and cracking of thin parts and insufficient degreasing of thick parts, thus improving the yield of palm blanks. During the degreasing process, the furnace pressure is maintained at 0.1-0.15MPa, and nitrogen purging is used to carry out the decomposition products at a flow rate of 5-10L / min. This avoids residual decomposition products from contaminating the palm blanks. Nitrogen purging also improves the surface cleanliness of the palm blanks, resulting in smoother surfaces of parts after sintering, reducing the workload of subsequent polishing processes, and lowering the overall labor and time costs of production.

[0091] After degreasing, the degreasing rate of the palm blank needs to reach 80% to 85%. The degreasing rate is calculated by the difference in quality before and after degreasing. If the degreasing rate is less than 80%, the concentration of nitric acid vapor can be appropriately increased (not exceeding 10%) or the degreasing time can be extended to prevent excessive degreasing from causing a decrease in the strength of the palm blank. This degreasing rate range can ensure that the amount of binder residue is small during subsequent sintering, and can maintain sufficient strength of the palm blank to resist the temperature rise stress during sintering.

[0092] It should be noted that the critical wall thickness of the green body obtained in S31 is directly related to the structural design of the product formed in S2, ensuring that the degreasing time calculation is adapted to the actual forming size.

[0093] S32 passes The degreasing time is dynamically calculated to match the degradation characteristics of polyoxymethylene in S1 binder and to keep the degreasing rate of the palm blank stable at 80% to 85%. This degreasing rate can ensure that there is no excessive decomposition of binder during the heating of S4, which will lead to cracking, and can also maintain the strength of the palm blank to resist the thermal stress of the rapid heating of S4.

[0094] If the degreasing rate deviates, the S2 holding pressure parameter (adjusting the density) or the S1 binder ratio is optimized in reverse to form a molding-degreasing linkage control.

[0095] The fourth step is to perform S4 one-step sintering: the brown blank is further degreased and sintered in a sintering furnace under a mixed atmosphere of nitrogen and hydrogen with a volume fraction of 1% to 5%.

[0096] During the process of heating from room temperature to the final sintering temperature of 880-920℃, no medium-temperature heat preservation platform specifically for reducing metal oxides is set up.

[0097] In this embodiment, an atmosphere-protected sintering furnace is selected, and the mixed atmosphere flow rate is controlled at 15-25 L / min. The hydrogen gas fraction is preferably 3%, which can provide a reducing atmosphere while avoiding the safety risks caused by excessive hydrogen content. The 3% hydrogen gas fraction can effectively reduce the explosion-proof cost of the sintering process while ensuring the reduction effect.

[0098] The specific heating process is as follows: first, heat the temperature to 450-500℃ at a rate of 0.5-2℃ / min and hold it for 60-90 minutes. This stage can fully complete the thermal degreasing of the remaining adhesive and avoid the rapid decomposition of the residual adhesive during the subsequent heating process, which may cause the parts to crack.

[0099] Then, raise the temperature to 895-905℃ at a rate of 3-8℃ / min and hold for 60-120 minutes. When the zinc content of the brass alloy is high, the holding temperature can be set to the lower limit and the holding time can be extended appropriately. When the zinc content is low, the holding temperature can be set to the upper limit. Setting the sintering parameters according to the zinc content can be adapted to brass parts with different compositions, thus expanding the applicability of the method.

[0100] During the heat preservation period, maintain the furnace pressure at 0.05-0.1MPa to ensure sufficient sintering and densification. After the heat preservation period, cool down to 500℃ at a rate of 3-5℃ / min, and then let it cool naturally to room temperature. This avoids rapid cooling that could cause deformation of the parts. Gradual cooling can effectively control the warping and deformation of the parts, meet the dimensional requirements of precision parts, reduce the scrap rate caused by deformation, and improve production efficiency.

[0101] The innovative design that omits the medium-temperature insulation platform is not simply an omission of steps, but a systematic optimization based on the synergistic effect of the preceding steps:

[0102] S1 powder has an oxygen content of ≤300ppm, which greatly reduces the amount of oxides generated. S3 fully degreases and reduces impurities during sintering. Together, they provide a prerequisite for eliminating the need for medium-temperature reduction.

[0103] Meanwhile, during the 450-500℃ holding stage in the S4 heating process, the remaining degreasing is completed and a slight reduction is achieved simultaneously using a 3% hydrogen atmosphere, realizing the integration of degreasing-reduction-sintering. This design, in turn, requires precise matching of parameters from S1 to S3 to form a closed-loop synergy across the entire process, solving the problems of zinc volatilization and cycle time in traditional processes.

[0104] This method improves the forming quality by optimizing the feeding ratio and ensures the strength of the brown blank by catalytic degreasing. The key is that one-step sintering eliminates the need for a medium-temperature heat preservation platform.

[0105] The core of this method's innovation lies in: First, the use of customized low-oxygen spherical powder and composite binder in S1 not only improves feed flowability but, more importantly, reduces the need for oxide reduction in S4, providing a prerequisite for eliminating the need for medium-temperature insulation. Combined with the 80%–85% degreasing rate controlled by dynamic degreasing in S3, it reduces zinc volatilization from the source, ensuring accurate brass alloy composition and improving mechanical and corrosion resistance compared to traditional processes.

[0106] Secondly, S2 is designed based on dynamic pressure holding parameters of melt pressure and flow length-to-thickness ratio, which precisely controls the green compact density of 65% to 70%. This density is not only suitable for the binder content of S1 feed, but also creates the optimal channel for the diffusion of nitric acid vapor in S3, forming a closed-loop control of "feeding-forming", which makes the green compact size accuracy and consistency significantly better than traditional fixed parameter forming.

[0107] Third, S3 uses dynamic degreasing time calculation based on key wall thickness and process constant k to adapt to the structure of S2 molded products and the degradation characteristics of S1 binder, ensuring that the strength of the brown blank is sufficient to withstand the temperature rise stress of S4 and avoid cracking defects.

[0108] Fourth, the S4 one-step sintering process eliminates the need for a medium-temperature holding platform. This is not simply a matter of omitting steps, but rather relies on the synergistic effect of the preceding low-oxygen powder and thorough degreasing to simultaneously complete the remaining degreasing and slight reduction during the 450-500℃ holding stage. This achieves the integration of "degreasing-reduction-sintering," which shortens the production cycle compared to traditional processes, reduces energy consumption, and avoids the problem of excessive zinc volatilization during the medium-temperature stage.

[0109] It should be understood that traditional processes do not take into account the subtle differences in specific surface area (reflecting the surface activity of powder particles) and actual powder loading (reflecting the proportion of powder in the feed) between different batches of feed. These differences will directly affect the atomic diffusion rate and densification efficiency during the sintering process, ultimately resulting in large fluctuations in key properties such as density and tensile strength of parts from different batches, making it difficult to meet the consistency requirements of mass production of precision parts.

[0110] As a further embodiment, before performing step S4, a step S0 is included to determine the sintering parameters:

[0111] S01. Parameter Acquisition and Calculation: Obtain the specific surface area (SSA) and actual powder loading (φ) of the current batch of brass feed, according to the formula... Calculate the dynamic sintering factor (DSF), where The theoretical density of brass, For process equipment constants;

[0112] Specific surface area (SSA) was determined using the BET method (liquid nitrogen adsorption-specific surface area determination method, industry standard measurement method), and actual powder loading (φ) was determined using thermogravimetric analysis (simultaneous thermal analyzer test, to accurately monitor the mass loss due to binder decomposition).

[0113] In this embodiment, the two core parameters are obtained through precise measurement to ensure the accuracy of subsequent calculations: Specific surface area (SSA) is measured by the BET method (liquid nitrogen adsorption-specific surface area determination method) using a fully automated BET specific surface area analyzer. Before testing, the feed sample needs to be vacuum dried at 100°C for 2 hours to remove adsorbed moisture and avoid moisture affecting the adsorption data. During testing, the multi-point BET method is used for calculation, and the liquid nitrogen adsorption time is set to 30 minutes to ensure sufficient adsorption-desorption balance. The test accuracy reaches 0.01 m² / g. This parameter directly reflects the surface activity of the powder particles. The higher the surface activity, the faster the atomic diffusion rate during sintering, and the more significant the densification effect.

[0114] Furthermore, the actual powder loading (φ) was measured by thermogravimetric analysis. Specifically, 10 mg ± 0.001 mg of feed sample was weighed and placed in an alumina crucible, and tested using a synchronous thermal analyzer. The heating rate was set to 10 °C / min, and the heating range was from room temperature to 600 °C. Nitrogen atmosphere protection was used (flow rate 50 mL / min). The mass loss due to complete decomposition of the binder was analyzed by thermogravimetric curve analysis, and the ratio of the remaining powder mass to the original feed mass was calculated. This method can monitor mass changes in real time, ensuring complete decomposition of the binder and avoiding errors caused by incomplete local combustion in the traditional ignition method. This parameter reflects the actual proportion of powder in the feed and directly affects the green compact density of S2 injection molding and the shrinkage rate of subsequent sintering, thus improving the accuracy of parameter measurement.

[0115] In this embodiment, the formula is used for calculation. A fixed value of 8.5 g / cm³ is used to ensure that the density benchmark is consistent when calculating different batches;

[0116] k is a process equipment constant used to calibrate the effect of the heat transfer efficiency of a specific sintering furnace on the diffusion rate of powder surfaces. It is calibrated through the following steps:

[0117] A characteristic feed with known specific surface area (SSA) and actual powder loading (φ) is selected, and its optimal sintering parameters (Ractual, tactual) are determined experimentally in the target sintering furnace, and the corresponding DSFeffective value is derived.

[0118] Using formula Calculate the k-value. For the same sintering furnace, this calibration process only needs to be performed once after the equipment is started up or overhauled, and the obtained k-value is usually between 0.85 and 0.95.

[0119] It should be noted that the core innovation of this formula lies in organically integrating the microscopic characteristics (specific surface area) of the feed with the macroscopic proportion (powder loading) into a quantifiable sintering factor, which solves the problem that microscopic characteristics cannot directly guide the setting of macroscopic sintering parameters in traditional processes, and realizes the scientific correlation between feed characteristics and sintering parameters.

[0120] S02. Parameter determination: Based on the calculated DSF value, the optimal heating rate (R) from 550℃ to the final sintering temperature and the optimal holding time (t) at the final sintering temperature are determined through the pre-established mapping relationship in step S4.

[0121] In this embodiment, the pre-established mapping relationship is implemented through a machine learning model, and the model building process combines data adaptability and prediction accuracy:

[0122] First, more than 100 batches of production process data were collected as a training set. The data should cover the feeding characteristics range of specific surface area 0.5 to 1.5 m² / g and actual powder loading 60% to 70%. Each data sample includes input features and output labels. The input feature is the dynamic sintering factor (DSF), and the output labels are the production-verified optimal heating rate (R) and optimal holding time (t). The criteria for determining the optimal parameters are that the density of the sintered product is ≥95% and the tensile strength is ≥300 MPa.

[0123] In the data preprocessing stage, outliers (such as substandard data due to equipment failure) are removed, and the DSF values ​​are scaled to the 0-1 range using a normalization method to improve the model training convergence speed.

[0124] In this embodiment, Gradient Boosting Regression Tree (GBRT) is selected as the machine learning model. This model has a strong ability to fit nonlinear relationships and can adapt to the complex correlation between DSF values ​​and sintering parameters. The model structure is set as follows:

[0125] The number of decision trees was set to 100, the maximum depth of each tree was set to 5, and the learning rate was set to 0.1 to avoid overfitting or underfitting. During training, 5-fold cross-validation was used, and the dataset was divided into training and test sets in an 8:2 ratio. The mean absolute error (MAE) was used as the model evaluation metric to ensure that the predicted heating rate MAE was ≤0.3℃ / min and the predicted holding time MAE was ≤5 minutes.

[0126] To illustrate the data composition and ensure model reproducibility, the following table provides five sets of exemplary training data extracted from historical databases:

[0127] DSF Optimal heating rate R (°C / min) Optimal heat preservation time t (min) 0.35 7.5 115 0.48 6.0 100 0.65 5.2 85 0.79 4.5 75 0.92 3.8 65

[0128] During model training, the goal is to minimize the mean absolute error (MAE) between the predicted values ​​and the actual optimal values.

[0129] After training, the model is deployed to the production control system. During production, the calculated DSF value is input, and the model can output the optimal heating rate (R) and optimal holding time (t) in real time.

[0130] It should be noted that, compared with traditional data fitting methods, machine learning models can autonomously learn the implicit correlation between DSF values ​​and sintering parameters, and can incorporate new batch production data through incremental learning to continuously optimize prediction accuracy and reduce the cost of manual intervention. This intelligent mapping relationship establishment method is non-obvious.

[0131] In this embodiment, when the input DSF value is in the low range of 0.3-0.5, the model outputs a heating rate of 6-8℃ / min and a holding time of 100-120 minutes, which is suitable for the characteristics of low surface activity and small loading of the feed powder in this range. The higher heating rate and longer holding time promote densification.

[0132] When the DSF value is in the high range of 0.8-1.0, the model outputs a heating rate of 3-5℃ / min and a holding time of 60-80 minutes, which matches the atomic diffusion characteristics of powders with high surface activity and high loading, and avoids deformation caused by excessive sintering.

[0133] When the DSF value is in the middle range of 0.5-0.8, the output heating rate is 5-6℃ / min and the holding time is 80-100 minutes, which is in perfect agreement with the optimal parameters for actual production.

[0134] Specifically, considering the high surface activity of low-oxygen powder prepared by S1 gas atomization in the high DSF value range, the model learns the sintering rules of this type of powder from historical data and autonomously outputs low heating rate parameters to accurately solve the problem of coarse grains caused by traditional fixed parameters; and the model supports incremental training every month with new production data to dynamically adapt to subtle changes in feeding characteristics.

[0135] In this embodiment, in the heating process of step S4, the temperature is first raised to 450-500℃ at a rate of 0.5-2℃ / min and held for 60-90 minutes to complete the remaining degreasing. Then, the temperature is raised to 550℃, and the system automatically calls the optimal heating rate (R) calculated by S0 to raise the temperature to the final sintering temperature (895-905℃) and holds for the optimal holding time (t).

[0136] It should be noted that this connection method takes into account the state of the palm blank after S3 catalytic degreasing. The 80% to 85% degreasing rate of S3 dynamic degreasing, combined with the customized sintering parameters of S0, can not only avoid the palm blank cracking caused by excessive heating, but also ensure sufficient densification. This achieves closed-loop control of "feed batch characteristics - sintering parameters", solving the problem of large performance fluctuations of parts from different batches in traditional processes from the source.

[0137] As a further embodiment, the method is executed by a smart manufacturing system, which includes:

[0138] The parameter measurement module is used to measure powder specific surface area, actual feed loading, melt pressure and green body geometric parameters; it integrates a BET specific surface area analyzer, thermogravimetric analyzer, mold pressure sensor and three-dimensional optical measuring instrument to collect core parameters across steps in real time.

[0139] In this embodiment, the BET method is used to measure the powder specific surface area (SSA) and the thermogravimetric method is used to measure the feed loading (φ) to provide data for SO1;

[0140] The mold sensor collects the melt pressure (Pcavity) at the end of injection, adapting to the S21 requirements; the 3D optical instrument scans the green blank to obtain the critical wall thickness (D), supporting S31 calculations. The module's innovation lies in the integrated synchronous acquisition of multiple devices, avoiding the parameter time difference problem of traditional independent measurements.

[0141] The data processing and calculation module includes a dynamic sintering factor calculation model, a pressure holding parameter optimization model, a catalytic degreasing time calculation model, and a sintering parameter mapping relationship. Specifically, the industrial-grade processor has four embedded collaborative models to realize the conversion of raw data into process instructions.

[0142] The dynamic sintering factor model calculates the DSF value according to the S01 formula;

[0143] The pressure holding parameter model is output according to formula S22. , Automatic matching , coefficient;

[0144] The defatting time model is calculated according to formula S32. retrieve the corresponding constant;

[0145] The sintering parameters were determined using the GBRT machine learning model, which outputs R and t based on the DSF values.

[0146] The injection molding machine and sintering furnace control module is communicatively connected to the data processing module, and is used to receive instructions and control the corresponding equipment to execute the process. In this embodiment, communication with the data module is achieved through industrial Ethernet to realize automated instruction execution and closed-loop feedback.

[0147] Injection molding machine , The command adjusts the pressure valve and timer, with a control accuracy of 0.1MPa and 1s. The sintering furnace adjusts the heating power according to the R and t commands, accurately raising and holding the temperature after reaching 550℃, with a temperature rise accuracy of ±1℃. When the actual parameters deviate from the command by more than ±5%, a recalculation is triggered. For example, if the S4 temperature rise deviates, the power is automatically adjusted and the holding time is finely adjusted to avoid defects.

[0148] It should be noted that the measurement module collects parameters and uploads them to the processing module. The processing module then outputs instructions to the control module, which executes the instructions and provides feedback on the deviation, triggering a re-measurement and calculation. This system enables innovative implementation of the process, significantly improving production efficiency compared to traditional manual methods. Batch density fluctuations are reduced to ±1%, ensuring the quality of mass-produced precision brass parts.

[0149] To enable those skilled in the art to more clearly understand the technical solution of the present invention and its beneficial effects, the following comparative description is provided through specific embodiments and comparative examples.

[0150] Example 1 represents the optimal implementation of the present invention, while Comparative Examples 1 to 4 are used to demonstrate the technical problems caused by not adopting all or part of the key technical features of the present invention.

[0151] Example 1

[0152] S1. Feed preparation: Spherical brass alloy powder (oxygen content 250ppm, D50=12μm) prepared by gas atomization is mixed with binder (78% polyoxymethylene, 15% high-density polyethylene, 5% ethylene-vinyl acetate copolymer, 2% stearic acid) in a twin-screw mixer at 175°C for 35 minutes, and then granulated to obtain the feed.

[0153] S2. Injection Molding: Injection temperature 180℃. Obtain the melt pressure (Pcavity) at the end of injection and the ratio of the maximum flow length to thickness of the product (L / T). Calculate dynamically based on the model and use the optimized holding pressure parameters. =100MPa, Injection was performed at 12s to obtain a green body.

[0154] S3. Catalytic Degreasing: Under conditions of 8% nitric acid vapor concentration and 90℃, the degreasing time is dynamically calculated based on the critical wall thickness of the green body. Catalytic degreasing is performed over a period of 3.5 hours.

[0155] S4, One-step sintering: Before executing S4, execute step S0 first. Measure the specific surface area (SSA) and actual powder loading (φ) of the current batch feed, calculate the dynamic sintering factor (DSF), and determine the optimal heating rate (R=5℃ / min) and optimal holding time (t=90 minutes) for the final sintering stage through a pre-established machine learning model.

[0156] Subsequently, under a mixed atmosphere of nitrogen and 3% hydrogen, the temperature was first increased to 480℃ at 1℃ / min and held for 75 minutes, then increased to 900℃ at 5℃ / min (i.e., the optimal heating rate R) and held for 90 minutes (i.e., the optimal holding time t) to complete the sintering. No intermediate temperature holding platform was set up throughout the process.

[0157] Comparative Example 1

[0158] This comparative example uses the traditional brass MIM process and does not contain the core features of this invention.

[0159] Feed preparation: Conventional water-atomized irregular brass powder (oxygen content 850ppm) and paraffin-polyethylene binder were used.

[0160] Degreasing: Solvent degreasing is used, which takes about 8 hours.

[0161] Sintering: Segmented sintering is carried out in a mixed atmosphere of nitrogen and 8% hydrogen. A heat preservation platform is set at 600℃ for 120 minutes to reduce oxides. The total sintering cycle is about 18 hours.

[0162] Comparative Example 2

[0163] The only difference between this comparative example and Example 1 is that the step S0 of determining the sintering parameters is omitted.

[0164] Process: In the S4 sintering stage, the dynamic sintering factor (DSF) is not calculated. Instead, a heating rate of 5℃ / min and a holding time of 90 minutes are used for sintering.

[0165] Comparative Example 3

[0166] The only difference between this comparative example and Example 1 is that the S4 one-step sintering process has been modified.

[0167] Process: Use 8% hydrogen by volume and add a 120-minute holding platform at 600°C during the heating process.

[0168] Comparative Example 4

[0169] The only difference between this comparative example and Example 1 is that the S3 catalytic degreasing process has been modified.

[0170] Process: Instead of using nitric acid vapor for catalytic degreasing, a traditional thermal degreasing process is adopted, in which the temperature is slowly raised to above 500°C under a protective atmosphere to decompose the binder. The degreasing cycle takes about 24 hours.

[0171] The brass alloy parts obtained in the above embodiments and comparative examples were subjected to performance tests, and the results are recorded in the table below.

[0172] Table 1: Performance Comparison of Examples and Comparative Examples

[0173] Test Project Example 1 Comparative Example 1 Comparative Example 2 Comparative Example 3 Comparative Example 4 Zinc volatilization loss (wt%) 0.4% 2.3% 0.9% 1.7% 0.7% Product relative density 96.5% 91.0% 93.8% 94.5% 92.8% Tensile strength (MPa) 320 240 285 295 265 Dimensional accuracy (dimensional deviation %) ±0.25% ±0.85% ±0.55% ±0.45% ±0.65% Total production cycle (hours) 10 >30 10 14 >30

[0174] The following conclusions can be clearly drawn from the comparative data in Table 1:

[0175] Example 1 of this invention is significantly superior to all comparative examples in controlling zinc volatilization, demonstrating the effectiveness of the synergistic solution of "low-oxygen powder + low-hydrogen atmosphere + no intermediate temperature platform".

[0176] Comparative Example 2, due to the omission of dynamic sintering parameter adjustment, resulted in lower product density and strength than Example 1 with greater fluctuations, demonstrating the crucial role of parameter optimization based on the dynamic sintering factor (DSF) in ensuring batch consistency.

[0177] The comparison between Example 1 and Comparative Example 4 shows that using nitric acid vapor catalytic degreasing can greatly shorten the production cycle compared to traditional thermal degreasing.

[0178] Example 1 of this invention uses a low concentration of 3% hydrogen gas, which, while ensuring performance, is safer and more economical than the 8% high concentration of hydrogen gas used in Comparative Examples 1 and 3.

[0179] In summary, this invention, through the organic combination of technical features, synergistically solves the long-standing technical problems in the brass MIM process, such as zinc volatilization, long cycle time, and poor consistency, and has achieved significant technical progress.

[0180] Produced using the above formula and proportions, the zinc volatilization loss is only 0.4%, the product relative density is 96.5%, the tensile strength is 320MPa, and the production cycle is only 10 hours, which is significantly better than the traditional process (Comparative Example 1: zinc volatilization 2.3%, strength 240MPa, cycle >30 hours).

[0181] Constants in the formula (such as) All have been calibrated through experiments. The machine learning model can be optimized through incremental learning to adapt to the slight fluctuations in different batches of feed, meeting the needs of industrial mass production.

[0182] Industry technology adaptation: BET method, thermogravimetric analysis, atmosphere-protected mixing / sintering, etc. are all mature technologies in the MIM industry. The patents only optimize the process based on them and do not exceed the scope of industry technology.

[0183] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for producing a metal powder injection molded brass alloy, characterized by, The method comprises the following steps: S1, feed preparation: mixing and granulating spherical brass alloy powder with a binder to prepare a feed; the binder comprises polyformaldehyde, high-density polyethylene and ethylene-vinyl acetate copolymer; S2, injection molding: injection molding the feed to obtain a green body; S3, catalytic debinding: placing the green body in a nitric acid vapor atmosphere for catalytic debinding to obtain a brown body; S4, one-step sintering: further heat debinding and sintering the brown body in a sintering furnace under a mixed atmosphere composed of nitrogen and hydrogen with a volume fraction of 1-5%; no intermediate temperature holding platform for reducing metal oxides is provided during the process of heating from room temperature to the final sintering temperature of 880-920℃.

2. A method of producing a metal powder injection molded brass alloy according to claim 1, characterized in that, In step S1: The spherical brass alloy powder is prepared by gas atomization method, and the oxygen content is not more than 300ppm, and the particle size distribution satisfies D50=8-15μm; The binder consists of the following components in mass percentage: polyformaldehyde 70%-85%, high-density polyethylene 10%-20%, ethylene-vinyl acetate copolymer 3%-8%, and stearic acid 1%-3%.

3. The method of claim 1, wherein the metal powder injection molding brass alloy is prepared by the steps of: preparing a brass alloy powder; and mixing the brass alloy powder with a binder to form a feedstock. In step S4, the temperature program of the one-step sintering includes: first heating at a rate of 0.5-2℃ / min to 450-500℃ and holding for 60-90 minutes, and then heating at a rate of 3-8℃ / min to 895-905℃ and holding for 60-120 minutes.

4. The method of claim 1, wherein the metal powder injection molding brass alloy is prepared by the steps of: preparing a brass alloy powder; and mixing the brass alloy powder with a binder to form a feedstock. Before step S4 is performed, step S0 of determining sintering parameters is further included: S01, obtain the specific surface area (SSA) and the actual powder loading (φ) of the current batch of brass feedstock, according to the formula: calculate the dynamic sintering factor (DSF), wherein is the theoretical density of brass, is a process equipment constant; S02、according to the calculated values, through a pre-established mapping relationship, determine the optimal heating rate (R) from 550°C to the final sintering temperature and the optimal holding time (t) at the final sintering temperature in the final sintering stage in step S4. In step S4, the optimal heating rate (R) and the optimal holding time (t) are used to perform the sintering process of the final sintering stage.

5. A method of producing a metal powder injection molded brass alloy according to claim 4, wherein In step S02, the optimal heating rate (R) is determined by the relation where a, b, c, d are constants fitted by the historical process database. where a, b, c, d are constants fitted by the historical process database.

6. A method of producing a metal powder injection molded brass alloy according to claim 4, wherein In step S02, the pre-established mapping relationship is realized by a machine learning model.

7. A method of producing a metal powder injection molded brass alloy according to claim 4, characterized by: In step S01, the specific surface area (SSA) is measured by the BET method, and the actual powder loading (φ) is measured by the thermogravimetric analysis method.

8. The method for preparing a brass alloy by metal powder injection molding according to claim 1, characterized in that, The holding pressure (P ) and the holding time (t ) in the injection molding process of step S2 are determined optimally by means of: S21, state parameter acquisition: acquiring the melt pressure (Pcavity) in the mold cavity at the end of the injection stage and the maximum flow length thickness ratio (L / T) of the product; S22, pressure maintaining parameter calculation: the calculation formula of the pressure maintaining pressure (Pp) ; the calculation formula of the pressure maintaining time (Tp) , wherein , is a compensation coefficient related to the material shrinkage characteristic;​​ Also, the pressure holding phase in step S2 employs the calculated and is performed.

9. The method for preparing a brass alloy by metal powder injection molding according to claim 1, characterized in that, In step S3, the treatment time of the catalytic defatting ) is dynamically determined by: S31, geometric parameter acquisition: acquiring the key wall thickness (D) of the green body, which is the maximum thickness of the green body in the diffusion direction of the debinding atmosphere; S32, defatting time calculation: according to the formula the required catalytic defatting time (tcat) is calculated wherein K is a process constant related to the concentration and temperature of the nitric acid vapor; Also, the catalytic debinding treatment time in step S3 is determined based on the calculated is executed.

10. The method for preparing a brass alloy by metal powder injection molding according to claim 1, characterized in that, The method is performed by an intelligent manufacturing system, which comprises: A parameter measurement module for measuring the specific surface area of the powder, the actual loading of the feed, the melt pressure and the geometric parameters of the green body; A data processing and calculation module embedded with a dynamic sintering factor calculation model, a holding pressure parameter optimization model, a catalytic debinding time calculation model and a sintering parameter mapping relationship; An injection molding machine and sintering furnace control module in communication connection with the data processing module for receiving instructions and controlling the corresponding equipment to perform the process.

Images