High-hardness nano-aluminum nitride ceramic, and preparation method and application thereof

Abstract

The application discloses high-hardness nano-aluminum nitride ceramic and a preparation method and application thereof, and relates to the field of advanced ceramic material preparation technology.The method comprises the following steps: pre-dispersing and grading nano-aluminum nitride powder, obtaining primary dispersed phase and hard agglomerate phase; independently wet-deagglomerating the hard agglomerate phase, obtaining recycled dispersed phase; respectively detecting the total oxygen content of the two types of powder and preparing yttrium oxide; respectively granulating and adjusting the granulation state through a linear shrinkage curve, so that the shrinkage starting temperature and cumulative linear shrinkage rate of the two types of test bodies are matched; low-shear dry-flowing the aluminum nitride according to the dry mass proportion, performing forming, degreasing, auxiliary agent reaction homogenizing sintering and densification sintering, and obtaining aluminum nitride ceramic.The application can reduce local yttrium distribution and shrinkage difference caused by hard agglomerates, improve the relative density, average hardness and local hardness uniformity of the ceramic, reduce abrasion, collapse and island-shaped peeling, and can be used for wear-resistant lining, wafer carrier plates and insulating heat-dissipating substrates.

Description

Technical Field

[0001] This invention relates to the field of advanced ceramic material preparation technology, and in particular to high-hardness nano-aluminum nitride ceramics, their preparation methods, and applications. Background Technology

[0002] Aluminum nitride ceramics possess high thermal conductivity, low coefficient of thermal expansion, good electrical insulation properties, and good high-temperature and corrosion resistance. They can be used to manufacture products such as wear-resistant liners, guide plates, sealing rings, semiconductor wafer carriers, insulating components for semiconductor equipment, and insulating heat dissipation substrates for power electronic devices. Using nano-aluminum nitride powder as a sintering raw material is beneficial for improving powder sintering activity, reducing the diffusion distance required for densification, and improving the overall density and mechanical properties of the sintered ceramic. Because the surface of aluminum nitride powder easily adsorbs oxygen and moisture, sintering aids such as yttrium oxide are usually added during actual preparation. These aids react with the oxygen-containing substances carried by the powder, promoting sintering mass transfer between aluminum nitride particles and pore discharge.

[0003] However, nano-aluminum nitride powder has a high specific surface area and surface energy, making it prone to forming agglomerates with significant size variations during preparation, drying, storage, and transfer. Some soft agglomerates can be disintegrated through conventional stirring or low-intensity dispersion, while some hard agglomerates persist during subsequent mixing, granulation, and molding. The total oxygen content, actual yttrium oxide distribution, granulation particle state, and early sintering shrinkage behavior of the powder inside and around the hard agglomerates may differ from those of the bulk dispersed powder. When yttrium oxide is added uniformly to the entire batch of powder based solely on average oxygen content, and a uniform granulation and sintering regime is adopted, the areas corresponding to the hard agglomerates are prone to localized concentration of yttrium-containing components, different shrinkage initiation temperatures, and inconsistent cumulative shrinkage, thus forming localized low-hardness regions within the sintered ceramic. These localized low-hardness regions may be continuously distributed along the periphery of the normal hardness region, while the overall relative density and average hardness of the ceramic remain at a high level, making them difficult to detect promptly through conventional single-point hardness testing or average performance testing. Under wear, cutting, grinding, or localized contact loads, the aforementioned localized low-hardness areas may further cause accelerated wear, edge chipping, or island-like spalling, affecting the service life and processing reliability of aluminum nitride ceramic products.

[0004] Therefore, it is necessary to provide an aluminum nitride ceramic and its preparation method that can separately treat the oxygen content and shrinkage differences of powders from hard agglomerates and improve the local hardness uniformity at different locations after sintering. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies, such as the difficulty in fully depolymerizing hard agglomerates in nano-aluminum nitride powder, the differences in oxygen content, sintering aid distribution, and early sintering shrinkage behavior among powders from different sources, which lead to localized low-hardness regions and uneven hardness spatial distribution in sintered aluminum nitride ceramics. Furthermore, these ceramics are prone to accelerated wear, edge chipping, and island-like spalling during wear resistance, cutting, or localized stress. The invention proposes high-hardness nano-aluminum nitride ceramics, their preparation method, and applications.

[0006] To address the problems existing in the prior art, the present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides a high-hardness nano-aluminum nitride ceramic prepared by using yttrium oxide as a sintering aid, wherein the relative density of the high-hardness nano-aluminum nitride ceramic is not less than 98%, and the relative density is the ratio of the volume density measured by Archimedes displacement method to the true density measured by helium displacement method.

[0008] A rectangular array of 21×21 detection points is set in the center, middle and edge regions of the polished cross-section of the high-hardness nano-aluminum nitride ceramic. The spacing between the detection points in each rectangular array is 30 micrometers. Vickers microhardness testing is performed on each detection point using a detection load of 0.05 kgf and a holding time of 10 seconds. The arithmetic mean of the Vickers microhardness of all effective detection points in the three rectangular arrays is taken as the cross-sectional average hardness. The Vickers microhardness of each effective detection point is not less than 90% of the cross-sectional average hardness. Detection points with Vickers microhardness less than 95% of the cross-sectional average hardness do not form a closed or semi-closed chain of detection points surrounding at least one detection point with Vickers microhardness not less than 95% of the cross-sectional average hardness.

[0009] Preferably, a radial reference line is set from the geometric center of the polished section to the nearest edge, the position where the geometric center is located is defined as the central region, the position where the distance from the geometric center to the nearest edge is half the position is defined as the intermediate region, and the position 1 to 2 mm away from the nearest edge is defined as the edge region;

[0010] The closed detection point chain is formed by sequentially connecting adjacent detection points with a Vickers microhardness lower than 95% of the average cross-sectional hardness along the transverse, longitudinal, or diagonal direction of the rectangular array, with the ends connected together. The semi-closed detection point chain is formed by sequentially connecting adjacent detection points with a Vickers microhardness lower than 95% of the average cross-sectional hardness, with the straight-line distance between the first and last detection points not exceeding 60 micrometers. Furthermore, the semi-closed detection point chain, together with the lines connecting the first and last detection points, encloses at least one detection point with a Vickers microhardness not lower than 95% of the average cross-sectional hardness.

[0011] Preferably, the average hardness of the cross section is not less than 1500HV0.05.

[0012] Secondly, the present invention also provides a method for preparing high-hardness nano-aluminum nitride ceramics, comprising the following steps:

[0013] S1. Add nano-aluminum nitride raw material powder to an anhydrous organic dispersion medium and pre-dispersettle under inert gas protection. Detect and classify the particle size of the pre-dispersion suspension to obtain the primary dispersed phase and the hard agglomerate phase, and record the original dry basis mass ratio of aluminum nitride contained in the primary dispersed phase and the hard agglomerate phase.

[0014] S2. Perform independent wet depolymerization on the hard agglomerate phase to obtain a recovered dispersed phase until the volume cumulative distribution of the recovered dispersed phase reaches 90%, and the corresponding equivalent particle size is not greater than 1.20 times the corresponding equivalent particle size of the original dispersed phase.

[0015] S3. Dry the primary dispersed phase and the recovered dispersed phase respectively and test the total oxygen content. Determine the oxygen mass of the primary dispersed phase based on the dry basis of aluminum nitride and the total oxygen content. Determine the oxygen mass of the recovered dispersed phase based on the dry basis of aluminum nitride and the total oxygen content. Determine the amount of first yttrium oxide to be added based on the oxygen mass of the primary dispersed phase. Determine the amount of second yttrium oxide to be added based on the oxygen mass of the recovered dispersed phase. Add yttrium oxide and mix them to obtain the first oxygen correction powder and the second oxygen correction powder.

[0016] S4. The first oxygen-correcting powder and the second oxygen-correcting powder are respectively pulped, granulated and pressed to obtain a first test blank and a second test blank. The linear shrinkage curves of the first test blank and the second test blank are measured under the same atmosphere and the same temperature rise conditions. The granulation state of the second oxygen-correcting powder is adjusted with the linear shrinkage curve of the first test blank as the matching benchmark so that the difference between the shrinkage initiation temperature of the two test blanks is not greater than 30 degrees Celsius and the difference in the cumulative linear shrinkage rate when the temperature is raised to 1700 degrees Celsius is not greater than 0.5 percentage points. The first oxygen-correcting powder and the second oxygen-correcting powder are granulated in batches according to the granulation parameters corresponding to the above conditions to obtain the first shrinkage matching granulated powder and the second shrinkage matching granulated powder.

[0017] S5. Weigh two types of granulated powders in such a way that the dry basis mass ratio of aluminum nitride contained in the first shrinkage matching granulated powder and the second shrinkage matching granulated powder is the same as the original dry basis mass ratio. Mix the two types of granulated powders by low shear dry mixing to obtain composite granulated powder. Then, perform molding preforming, cold isostatic pressing and debinding on the composite granulated powder in sequence to obtain aluminum nitride ceramic green body.

[0018] S6. The aluminum nitride ceramic green body is placed in a nitrogen protective atmosphere and subjected to additive reaction homogenization sintering and densification sintering in sequence. After cooling, high-hardness nano-aluminum nitride ceramic is obtained.

[0019] Preferably, in step S1, the pre-dispersed suspension is pre-graded to obtain a primary dispersed phase prediction sample. The equivalent particle size corresponding to the cumulative volume distribution of the primary dispersed phase prediction sample reaching 90% is used as the basic particle size, and 1.3 to 2.0 times the basic particle size is used as the grading cut-off point. The grading adopts at least one of centrifugal grading and sedimentation grading, or firstly, wet sieving is used to collect the coarse agglomerates on the sieve, and then the suspension under the sieve is centrifuged and graded, and the coarse agglomerates on the sieve and the sedimented particles obtained by centrifugal grading are combined into a hard agglomerate phase.

[0020] Preferably, in step S2, the independent wet depolymerization employs at least one of narrow-gap shearing, wet jet depolymerization, bead milling depolymerization, and high-pressure homogenization; when the equivalent particle size corresponding to the cumulative volume distribution of the recovered dispersed phase reaches 90% is not greater than 1.10 times the corresponding equivalent particle size of the original dispersed phase, the independent wet depolymerization is stopped.

[0021] Preferably, in step S3, the mass of oxygen in the primary dispersed phase and the mass of oxygen in the recovered dispersed phase are converted into alumina equivalents, respectively. The theoretical amount of yttrium oxide to be added is determined according to the ratio of 5 moles of alumina equivalent to 3 moles of yttrium oxide. The theoretical amount of yttrium oxide to be added is then multiplied by a correction factor of 0.85 to 1.15 to obtain the first amount of yttrium oxide to be added and the second amount of yttrium oxide to be added. The alumina equivalent is a stoichiometric standard for determining the amount of yttrium oxide to be added in the process, and does not mean that all oxygen in the nano-aluminum nitride powder exists in the alumina phase.

[0022] Preferably, in step S4, by adjusting at least one of the slurry solid content, binder content, granulation particle size distribution, and loose packing density of the second oxygen-correcting powder, the difference in shrinkage initiation temperature between the first test blank and the second test blank is not greater than 20 degrees Celsius, and the difference in cumulative linear shrinkage rate when heated to 1700 degrees Celsius is not greater than 0.3 percentage points.

[0023] Preferably, in step S5, the low-shear dry mixing is performed using a three-dimensional mixing device, with a mixing speed of 10-20 rpm and a mixing time of 20-40 minutes. Samples are taken from different locations of the mixed composite granulated powder, and the mass fraction of yttrium and the median equivalent particle size of each sample are measured. The low-shear dry mixing is terminated when the relative standard deviation of the mass fraction of yttrium is not greater than 3% and the relative standard deviation of the median equivalent particle size is not greater than 3%. In step S6, the temperature for homogenization and sintering of the additive reaction is 1450-1700 degrees Celsius, and the holding time is 0.5-3 hours. The temperature for densification and sintering is 1700-1900 degrees Celsius, and the holding time is 1-6 hours.

[0024] Thirdly, the present invention also provides the application of the above-mentioned high-hardness nano-aluminum nitride ceramic, wherein the high-hardness nano-aluminum nitride ceramic is the aforementioned product or the high-hardness nano-aluminum nitride ceramic prepared by the aforementioned preparation method, which can be used to prepare wear-resistant linings, guide sheets, sealing rings, precision ceramic bearing components, semiconductor wafer bearing disks, semiconductor equipment insulating components, power electronic device insulating heat dissipation substrates or high-power electronic device ceramic substrates.

[0025] Compared with the prior art, the beneficial effects of the present invention are:

[0026] 1. This invention classifies the pre-dispersed nano-aluminum nitride powder, separating the primary dispersed phase from the hard agglomerates that are difficult to dissolve using conventional dispersion methods. The hard agglomerates are then independently depolymerized using a wet process, allowing the nano-aluminum nitride particles within the hard agglomerates to redisperse, reducing the likelihood of large hard agglomerates continuing into the granulation, molding, and sintering processes. Furthermore, the total oxygen content of both the primary and recovered dispersed phases is measured, and the yttrium oxide addition amount is determined based on the dry basis weight of the aluminum nitride and the total oxygen content of each phase. This reduces the possibility of localized insufficient or excessive yttrium oxide caused by uniformly dispensing yttrium based on the average oxygen content of the entire batch of powder. It ensures that the yttrium oxide addition state in powders from different sources is adapted to the oxygen-containing substances they carry, thereby reducing the possibility of localized concentration of yttrium-containing components in the corresponding regions of the original hard agglomerates and uneven sintering reaction.

[0027] 2. This invention prepares test green bodies and measures linear shrinkage curves before re-merging the two types of powders. By adjusting the granulation state of the corresponding powders in the recovery dispersion, the shrinkage initiation temperature and cumulative linear shrinkage rate of the two types of powders are matched. Then, low-shear dry merging is carried out according to the original dry mass ratio of aluminum nitride, which can preserve the particle size distribution, loose density, and binder distribution of the two types of shrinkage-matched granulated powders. Combined with additive reaction homogenization sintering and densification sintering, the local shrinkage difference between powders from different sources can be reduced. The resulting aluminum nitride ceramics have high relative density and average hardness, while improving the local hardness uniformity of the central, intermediate, and edge regions. This reduces the formation of closed or semi-closed weakened detection point chains, thereby reducing the possibility of accelerated wear, edge chipping, and island spalling during wear resistance, cutting, and local stress processes, and improving the processing reliability and service life of ceramic products. Attached Figure Description

[0028] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0029] Figure 1 The flowchart of the method for preparing high-hardness nano-aluminum nitride ceramics provided by the present invention;

[0030] Figure 2 A schematic diagram illustrating the principles of hard agglomerate phase separation, independent depolymerization, and separate yttrium preparation;

[0031] Figure 3 This diagram illustrates the shrinkage matching, dry confluence, and hardness array detection of two types of powders. Detailed Implementation

[0032] The present invention will be further described below with reference to specific embodiments. It should be noted that the following embodiments are used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention.

[0033] like Figure 1 As shown, the method for preparing high-hardness nano-aluminum nitride ceramics of the present invention firstly pre-disperses and classifies nano-aluminum nitride powder to obtain a primary dispersed phase and a hard agglomerated phase, and then independently wet-depolymerizes the hard agglomerated phase to obtain a recovered dispersed phase. The total oxygen content of the primary dispersed phase and the recovered dispersed phase are detected respectively, and yttrium oxide is added to prepare a first oxygen-correcting powder and a second oxygen-correcting powder. Through slurry preparation, granulation, testing of green body preparation and linear shrinkage curve detection, the granulation state of the two types of powders is adjusted to achieve shrinkage matching, resulting in a first shrinkage-matched granulated powder and a second shrinkage-matched granulated powder. Then, according to the dry basis mass ratio of aluminum nitride in the two types of granulated powders, low-shear dry confluence is carried out, followed by molding preforming, cold isostatic pressing, degreasing, additive reaction homogenization sintering, and densification sintering to finally obtain high-hardness nano-aluminum nitride ceramics.

[0034] Example 1

[0035] This embodiment prepares a high-hardness nano-aluminum nitride ceramic for wear-resistant linings, specifically including the following steps.

[0036] S1, Pre-dispersion and splitting of nano-aluminum nitride powder;

[0037] Nano-aluminum nitride raw material powder with an average particle size of 80 nanometers and a purity of not less than 99.5% was selected, and 100 kg of aluminum nitride dry basis mass was used as the process measurement standard in this embodiment.

[0038] The primary particles refer to the basic particles that constitute the secondary agglomerates of nano-aluminum nitride powder. During the preparation, drying, storage, and transfer of nano-aluminum nitride powder, multiple primary particles can form secondary agglomerates with an equivalent particle size larger than that of the primary particles through electrostatic adsorption, van der Waals forces, surface adhesion, or local chemical bonding.

[0039] Isopropanol was dehydrated using 3A molecular sieves to obtain anhydrous isopropanol with a moisture content not exceeding 0.05%. Nano-aluminum nitride powder was then added to the anhydrous isopropanol to prepare an aluminum nitride suspension with a solid content of 30%.

[0040] The aluminum nitride suspension was added to a closed dispersion tank with an effective volume of 500 liters. The dispersion tank was equipped with a four-bladed impeller with a diameter of 240 mm, and the distance between the impeller and the bottom of the tank was 120 mm. Nitrogen gas with a purity of not less than 99.99% was introduced into the dispersion tank to maintain a positive pressure of 0.01–0.02 MPa.

[0041] Without loading abrasive media, the aluminum nitride suspension was stirred by a four-bladed oblique propeller at a speed of 320 rpm, with a propeller tip linear velocity of 4.02 m / s and a stirring time of 20 minutes.

[0042] The aforementioned pre-dispersion is used to dissolve soft agglomerates mainly formed by electrostatic adsorption, van der Waals forces, and drying adhesion, while retaining hard agglomerates that cannot be dissolved even under the aforementioned pre-dispersion conditions. The hard agglomerates refer to aggregates of particles that retain a relatively large equivalent particle size after the aforementioned pre-dispersion and can be separated from the bulk dispersion powder through subsequent classification.

[0043] During pre-dispersion, samples were taken from the center of the dispersion tank every 5 minutes, and the particle size distribution was measured using a laser particle size analyzer. Anhydrous isopropanol was used as the detection medium for each measurement, and the sample concentration, circulation flow rate, particle refractive index, and medium refractive index parameters were kept constant. High-power ultrasonic treatment sufficient to further depolymerize hard agglomerates was not performed before the measurement.

[0044] In this embodiment, the equivalent particle size corresponding to a cumulative powder volume distribution of 10% is denoted as D. 10 The median equivalent particle size corresponding to when the cumulative volume distribution of powder reaches 50% is denoted as D. 50 The equivalent particle size corresponding to a cumulative volume distribution of 90% is denoted as D. 90 .

[0045] When the results of two consecutive tests are D 90 When the rate of change is no greater than 3%, it is determined that the soft aggregates that can be disintegrated under the current pre-dispersion intensity have been basically disintegrated, and the pre-dispersion ends.

[0046] A representative sample of no less than 0.5% of the total mass of the pre-dispersed aluminum nitride suspension was obtained and pre-separated by centrifugation using the same effective separation radius, relative centrifugal force, and average residence time as the subsequent formal centrifugation fractionation.

[0047] Detection of D in particles in the pre-fractionated dispersion 90 The particle size is 0.42 micrometers, which is used as the base particle size for determining the grading cut point. The grading cut point is determined to be 1.5 times the base particle size, i.e., 0.63 micrometers in this embodiment.

[0048] All pre-dispersed suspensions were formally classified using a closed continuous centrifugal classifier. The effective separation radius of the centrifugal classifier was 150 mm, the rotation speed was 1080 rpm, and the corresponding relative centrifugal force was 196 times the acceleration due to gravity. The average residence time of the suspension in the centrifugal classifier was 4 minutes, and the feed flow rate was 75 liters / hour.

[0049] After the formal classification is completed, the D of the particles in the dispersion is measured. 90 The formal classification yielded D. 90 D obtained from pre-grading 90 If the difference is no more than 5%, it indicates that the basic particle size determined by the pre-grading and the formal grading conditions are effective.

[0050] The particle aggregates in the dispersion obtained by centrifugation were defined as the primary dispersed phase, and the precipitated particle aggregates obtained by centrifugation were defined as the hard aggregated phase. The primary dispersed phase refers to the particle aggregates that can maintain stable dispersion after pre-dispersion and remain in the dispersion under the above-mentioned centrifugation conditions.

[0051] Three representative samples were obtained from the primary dispersed phase and the hard aggregated phase, respectively. They were vacuum dried for 8 hours at 80 degrees Celsius and an absolute pressure not exceeding 10 kPa. After cooling under nitrogen protection, the dry basis mass was measured.

[0052] Based on the material mass balance before and after the split, the dry basis mass of aluminum nitride in the primary dispersed phase and the hard agglomerate phase is converted to the 100 kg aluminum nitride raw material measurement standard. The converted dry basis mass of aluminum nitride in the primary dispersed phase is 92 kg, and the converted dry basis mass of aluminum nitride in the hard agglomerate phase is 8 kg. The original dry basis mass ratio of aluminum nitride contained in the two is 92:8.

[0053] The small amount of powder consumed during sampling and testing is not included in the subsequent actual feed amount. During subsequent merging, the aluminum nitride dry basis mass in the actual remaining powder is measured at 92:8.

[0054] The particle size of the hard agglomerate phase was measured using the same detection conditions as described above, and the D of the hard agglomerate phase was obtained. 50 It is 62 micrometers, D 90 It is 118 micrometers.

[0055] S2, Independent wet depolymerization of hard aggregated phases;

[0056] The hard agglomerate obtained in step S1 is added to anhydrous isopropanol to prepare a hard agglomerate suspension with a solid content of 15%.

[0057] The hard agglomerated phase suspension is fed into a closed wet jet depolymerization device. The flow channels inside the wet jet depolymerization device that come into contact with the suspension are lined with ceramic to reduce the introduction of metallic impurities into the nano-aluminum nitride powder.

[0058] The wet jet depolymerization operates at a pressure of 80 MPa, with a nozzle orifice diameter of 0.20 mm and a single-pass flow rate of 18 L / h. The temperature of the hard agglomerate suspension before entering the depolymerization equipment is controlled at 20 ± 2 degrees Celsius, and during the depolymerization process, circulating cooling ensures that the outlet temperature does not exceed 35 degrees Celsius.

[0059] The hard agglomerate suspension was subjected to four consecutive cycles of wet jet depolymerization. After each depolymerization cycle, three representative samples were obtained from the discharge end of the equipment, and the particle size distribution (D) was measured using the same particle size analysis conditions as in step S1. 90 .

[0060] D after the first cycle 90 The value is 6.8 micrometers, and the D value after the second cycle is... 90 The value is 1.9 micrometers, and the D value after the third cycle is... 90 The value is 0.73 micrometers, and the D value after the fourth cycle is... 90 It is 0.45 micrometers.

[0061] D after the fourth cycle 90 Primary dispersed phase D 90 It is 1.071 times that of the primary dispersed phase D, which is smaller than that of the primary dispersed phase D. 90 The concentration is 1.10 times that of the target concentration, therefore the wet jet depolymerization is stopped, and the recovered dispersed phase is obtained.

[0062] The recovered dispersed phase refers to the phase formed by independent wet depolymerization of the hard agglomerated phase, and D 90 Approximately native dispersed phase D 90 A collection of particles.

[0063] Through the aforementioned independent wet depolymerization process, the original closed outer contour of the hard agglomerates is disrupted, causing the primary particles within the hard agglomerates to redisperse. The depolymerization endpoint is achieved by recovering the dispersed phase D. 90 With the primary dispersed phase D 90 The ratio is limited to avoid insufficient depolymerization, while also preventing unnecessary excessive breakage of the primary nano-aluminum nitride particles due to further increases in depolymerization energy.

[0064] The recovered dispersed phase and the primary dispersed phase will undergo total oxygen content detection and yttrium oxide preparation in subsequent steps, respectively, and will not be combined in this step.

[0065] like Figure 2As shown, the pre-dispersed suspension contains both dispersed nano-aluminum nitride particles and hard agglomerates that have not been pre-dispersed. Particle size classification separates these two phases into a primary dispersed phase and a hard agglomerate phase, respectively. The hard agglomerate phase is then subjected to independent wet deagglomeration, allowing the particles within the hard agglomerates to redisperse and form a recovered dispersed phase. The total oxygen content of the primary and recovered dispersed phases is measured separately. Based on the dry basis mass of aluminum nitride and the total oxygen content of each type of powder, the first and second yttrium oxides are determined and added, respectively, to obtain the first and second oxygen-correcting powders. This reduces the localized under- or over-addition of yttrium oxide caused by uniformly adding yttrium oxide according to the average oxygen content of the entire batch of powder.

[0066] S3. Detection of total oxygen content in the two types of powders and preparation of yttrium powders separately;

[0067] The primary dispersed phase obtained in step S1 and the recovered dispersed phase obtained in step S2 were placed in a vacuum drying device of the same specifications and dried for 8 hours at 80 degrees Celsius and an absolute pressure not exceeding 10 kPa.

[0068] After drying, nitrogen gas is introduced into the vacuum drying equipment to restore the pressure to normal, and the original dispersed phase and the recovered dispersed phase are cooled to room temperature under nitrogen protection.

[0069] Three parallel samples were randomly obtained from both the primary and recovered dispersed phases. The samples were weighed according to the specified loading mass of the calibrated oxygen and nitrogen analyzer, and the total oxygen content was determined by inert gas melting infrared absorption method. Each sample was tested for oxygen content only once.

[0070] The total oxygen content of the primary dispersed phase was measured at 0.91%, 0.92%, and 0.93%, respectively. The range of the three results was 0.02 percentage points. The arithmetic mean of 0.92% was taken as the total oxygen content of the primary dispersed phase.

[0071] The three total oxygen content test results for the recovered dispersed phase were 1.32%, 1.34%, and 1.35%, respectively. The range of the three test results was 0.03 percentage points. The arithmetic mean of 1.34% was taken as the total oxygen content of the recovered dispersed phase.

[0072] Based on a dry basis of 100 kg of aluminum nitride, the equivalent dry basis mass of aluminum nitride in the primary dispersed phase is 92 kg, corresponding to the oxygen mass... Calculate according to the following formula:

[0073]

[0074] The dry basis mass of recovered aluminum nitride from the dispersed phase is 8 kg, corresponding to the mass of oxygen. Calculate according to the following formula:

[0075]

[0076] The total oxygen content of the primary dispersed phase and the total oxygen content of the recovered dispersed phase were converted into alumina equivalents, and the theoretical amount of yttrium oxide to be added was determined according to the ratio of 5 moles of alumina equivalent to 3 moles of yttrium oxide.

[0077] The alumina equivalent is only used as a stoichiometric standard for determining the amount of yttrium oxide added in the process, and does not indicate that all oxygen in the nano-aluminum nitride powder exists in the alumina phase. Surface oxygen, lattice oxygen, and other oxygen-containing substances in the nano-aluminum nitride powder that can affect the reaction of sintering aids are uniformly measured by the total oxygen mass, and process corrections are performed to account for differences in reaction states caused by different oxygen states.

[0078] Since 1 mole of alumina contains 3 moles of oxygen atoms, corresponding to an oxygen mass of 48 grams, and the molar mass of yttrium oxide is 225.81 g / mol, the theoretical mass of added yttrium oxide is... Calculate according to the following formula:

[0079]

[0080] After simplification, it becomes:

[0081]

[0082] Actual mass of yttrium oxide added Calculate according to the following formula:

[0083]

[0084] In the formula, This represents a correction factor for the amount of yttrium oxide added, in this embodiment. Take 0.95.

[0085] The amount of yttrium oxide added is calculated based on the oxygen mass of the primary dispersed phase:

[0086]

[0087] The amount of yttrium oxide added is calculated based on the oxygen mass of the recovered dispersed phase:

[0088]

[0089] The amounts of yttrium oxide added, both primary and secondary, are calculated based on a dry weight of 100 kg of aluminum nitride. In actual production, the actual amounts added are proportionally calculated based on the actual dry weights of the primary and recovered dispersed phases of aluminum nitride.

[0090] Yttrium oxide (calculated based on the actual dry weight of aluminum nitride) was added to the primary dispersed phase, and anhydrous isopropanol was added to prepare a first mixed slurry with a solid content of 30%. Under nitrogen protection, the mixture was stirred at 300 rpm for 2 hours to obtain the first oxygen-correcting slurry.

[0091] Yttrium oxide, calculated based on its actual dry aluminum nitride content, was added to the recovered dispersed phase, and anhydrous isopropanol was added to prepare a second mixed slurry with a solid content of 30%. Under nitrogen protection, the mixture was stirred at 300 rpm for 2 hours to obtain the second oxygen-correcting slurry.

[0092] The first and second oxygen slurries were placed in a vacuum drying equipment and dried for 8 hours at 80 degrees Celsius and an absolute pressure not exceeding 10 kPa to obtain the first and second oxygen slurries.

[0093] The primary dispersed phase and the recovered dispersed phase are each formulated with yttrium oxide according to their respective total oxygen content. The two are not pre-combined and yttrium oxide is added uniformly according to the average total oxygen content of the mixed powder.

[0094] S4. Matching of early sintering shrinkage between the two types of oxygen-correcting powders;

[0095] The first and second oxygen-correcting powders were added to anhydrous isopropanol to prepare a spray granulation slurry.

[0096] The solid content of the first oxygen-correcting powder spray granulation slurry is 48%, and the amount of polyvinyl butyral added is 1.50% of the dry basis mass of the first oxygen-correcting powder.

[0097] The solid content of the initial spray granulation slurry of the second oxygen-correcting powder is 44%, and the amount of polyvinyl butyral added is 1.50% of the dry basis mass of the second oxygen-correcting powder.

[0098] Two types of spray-granulated slurries were granulated using a nitrogen-based closed-loop spray dryer. The inlet air temperature of the spray dryer was 90 degrees Celsius, and the outlet air temperature was 55 degrees Celsius.

[0099] The D of the initial granulated particles obtained from the first oxygen-correcting powder 10 D 50 and D 90 The micrometers are 27 micrometers, 63 micrometers and 106 micrometers respectively, and the loose packing density is 0.83 g / cm³.

[0100] The D of the particles obtained from the initial granulation of the second oxygen-purifying powder 10 D 50 and D 90 The micrometers are 18 micrometers, 46 micrometers and 84 micrometers respectively, and the loose packing density is 0.74 g / cm³.

[0101] Granulated powder of the first oxygen-correcting powder and initial granulated powder of the second oxygen-correcting powder were respectively molded with the same mold at 60 MPa and then cold isostatically pressed at 200 MPa to obtain a first test blank with a diameter of 10 mm and a length of 20 mm and an initial second test blank.

[0102] The green density of the first test blank was 1.87 g / cm³, and the green density of the initial second test blank was 1.75 g / cm³.

[0103] The first test blank and the initial second test blank were placed in a thermal expansion apparatus and linear shrinkage tests were performed under a nitrogen atmosphere with a purity of not less than 99.999%. The temperature was increased from room temperature to 1000 degrees Celsius at a rate of 5 degrees Celsius / minute, and then increased to 1700 degrees Celsius at a rate of 3 degrees Celsius / minute.

[0104] Cumulative linear shrinkage at test temperature T Calculate according to the following formula:

[0105]

[0106] In the formula, This indicates the initial length of the test blank at the start of the test; This indicates the length of the test blank when it is heated to the test temperature T.

[0107] The temperature at which the cumulative linear shrinkage rate first reaches 0.10% and maintains a continuous shrinkage trend within a subsequent temperature increase range of 20 degrees Celsius is determined as the shrinkage initiation temperature.

[0108] The shrinkage initiation temperature of the first test blank was 1420 degrees Celsius, and the cumulative linear shrinkage rate when the temperature was raised to 1700 degrees Celsius was 11.6%.

[0109] The initial shrinkage temperature of the second test blank was 1376 degrees Celsius, and the cumulative linear shrinkage rate when the temperature was raised to 1700 degrees Celsius was 12.5%.

[0110] In the initial state, the shrinkage initiation temperature of the first test blank and the initial second test blank differed by 44 degrees Celsius, and the cumulative linear shrinkage rate at 1700 degrees Celsius differed by 0.9 percentage points, which did not meet the shrinkage matching condition.

[0111] The spray granulation slurry of the second oxygen-correcting powder was re-formulated, with the solid content of the slurry adjusted from 44% to 48%, the amount of polyvinyl butyral added adjusted from 1.50% to 1.35% of the dry basis mass of the second oxygen-correcting powder, and the atomization pressure of the spray drying equipment was adjusted.

[0112] The adjusted D of the second oxygen-correcting powder granules 10 D 50 and D90 The micrometers are 26 micrometers, 61 micrometers and 103 micrometers respectively, and the loose packing density is 0.82 g / cm³.

[0113] A second test blank, after adjustment, was prepared using the same molding conditions as the first test blank. The green density of the adjusted second test blank was 1.85 g / cm³.

[0114] The adjusted second test billet was subjected to linear shrinkage testing using the same atmosphere and heating procedure as the first test billet. The shrinkage initiation temperature of the adjusted second test billet was 1408 degrees Celsius, and the cumulative linear shrinkage rate when heated to 1700 degrees Celsius was 11.8%.

[0115] After adjustment, the shrinkage initiation temperature of the first test blank and the second test blank differ by 12 degrees Celsius, and the cumulative linear shrinkage rate at 1700 degrees Celsius differs by 0.2 percentage points, satisfying the matching condition that the difference in shrinkage initiation temperature is no more than 20 degrees Celsius and the difference in cumulative linear shrinkage rate is no more than 0.3 percentage points.

[0116] Based on the slurry solid content, polyvinyl butyral addition amount, and spray drying parameters corresponding to the preparation of the first test preform, the remaining first oxygen-correcting powder was batch-processed into slurry and spray-granulated.

[0117] Based on the slurry solid content, polyvinyl butyral addition amount, and spray drying parameters corresponding to the preparation of the adjusted second test preform, the remaining second oxygen-correcting powder was batch-processed into slurry and spray-granulated.

[0118] Samples were taken at the beginning, middle and end stages of the two types of granulated powder production to test the particle size and bulk density of the granulated particles.

[0119] The D of the particles obtained from the first batch granulation of oxygen-correcting powder 10 It is 27 micrometers, D 50 It is 63 micrometers, D 90 It has a diameter of 106 micrometers and a loose packing density of 0.83 g / cm³.

[0120] The D of the particles obtained from the second batch granulation of oxygen-correcting powder 10 It is 26 micrometers, D 50 It is 61 micrometers, D 90 It has a diameter of 103 micrometers and a loose packing density of 0.82 g / cm³.

[0121] The relative standard deviation described in this invention refers to the ratio of the standard deviation of each parallel measurement to the corresponding arithmetic mean, and is expressed as a percentage.

[0122] D of samples taken from the initial, intermediate, and final stages of the two types of granulated powder 50The relative standard deviations are no greater than 3%, and the relative standard deviations of loose packing density are no greater than 2%.

[0123] The batch granulated powder obtained from the primary dispersed phase is designated as the first shrinkage-matched granulated powder, and the batch granulated powder obtained from the recycled dispersed phase is designated as the second shrinkage-matched granulated powder.

[0124] S5. Low-shear dry confluence, molding and degreasing of two types of shrinkage-matched granulated powders;

[0125] In step S1, the original dry basis mass ratio of aluminum nitride contained in the primary dispersed phase and the hard agglomerated phase is 92:8.

[0126] Based on the actual dry weight of aluminum nitride contained in the first shrinkage-matched granulation powder and the second shrinkage-matched granulation powder, the two types of granulation powder were weighed in a ratio of 92:8.

[0127] The metering ratio of the two types of granulated powder is based on the dry basis mass of the aluminum nitride it contains, and the 92:8 mixing ratio is not determined according to the total mass of the granulated powder after adding yttrium oxide and polyvinyl butyral.

[0128] The weighed first and second shrinkage-matched granulated powders were added to a three-dimensional mixer with an effective volume of 300 liters. The loading volume was calculated based on the natural packing state of the granulated powders and was 50% of the effective volume of the three-dimensional mixer.

[0129] Nitrogen gas is introduced into the 3D mixer to replace the air, thus placing the equipment under nitrogen protection. The 3D mixer rotates at 15 revolutions per minute, and the mixing time is 30 minutes.

[0130] The mixing process does not involve the addition of anhydrous isopropanol or other liquid media, does not involve wet pulping, does not involve secondary spray granulation, and does not employ high-shear mixing methods that could damage the structure of the two types of granulated particles.

[0131] By using low-shear dry confluence, the first shrinkage-matched granulated powder and the second shrinkage-matched granulated powder are evenly distributed, while retaining the granulation particle size distribution, loose density and binder content states formed in step S4.

[0132] After mixing, a sample with a mass of no less than 50 grams was obtained from the top, middle, and bottom of the three-dimensional mixer, as well as from a position near two different side walls, for a total of five samples.

[0133] The mass fraction of yttrium in five samples was determined by X-ray fluorescence spectrometry. Each sample was pulverized at low energy to a particle size suitable for preparing detection pellets, and then pressed into X-ray fluorescence detection pellets without altering the sample composition.

[0134] The component mixing homogeneity of the two types of granulated powders was evaluated by the relative standard deviation of the yttrium mass fraction in five samples. The Dt of the five samples was also measured. 50 To evaluate whether significant breakage of granulated particles occurs during low-shear dry mixing.

[0135] When the relative standard deviation of the mass fraction of yttrium in the five samples is no greater than 3%, and D 50 When the relative standard deviation is no greater than 3%, it is determined that the two types of shrinkage-matched granulated powders have completed uniform merging.

[0136] In this embodiment, after mixing for 30 minutes, the mass fraction of yttrium and D in the five samples were determined. 50 All of the above-mentioned criteria are met, and composite granulated powder is obtained.

[0137] D of compound granulated powder 10 It is 26 micrometers, D 50 It is 62 micrometers, D 90 It has a density of 105 micrometers and a loose packing density of 0.83 g / cm³.

[0138] The composite granulated powder was directly loaded into a wear-resistant liner mold, and pre-formed by pressing at 60 MPa for 30 seconds, followed by cold isostatic pressing at 200 MPa for 3 minutes to obtain an aluminum nitride ceramic green body. The bulk density of the aluminum nitride ceramic green body was 1.86 g / cm³.

[0139] The aluminum nitride ceramic green body is placed in a covered boron nitride crucible, with the green body loading volume in the crucible not exceeding 70% of the crucible's effective volume. The boron nitride crucible is then placed in an atmosphere sintering furnace.

[0140] Nitrogen gas with a purity of not less than 99.999% is introduced into the atmosphere sintering furnace at a flow rate of 15 liters / minute, and the furnace is maintained at a positive pressure of 0.02 MPa.

[0141] The temperature is increased from room temperature to 300 degrees Celsius at a rate of 0.5 degrees Celsius per minute and held for 1 hour; then the temperature is increased to 600 degrees Celsius at a rate of 1 degree Celsius per minute and held for 2 hours, so that polyvinyl butyral and other organic components are gradually discharged, thus completing the degreasing process.

[0142] like Figure 3As shown, the linear shrinkage curves of the first and second test green bodies were measured respectively. By adjusting the slurry solid content, binder content, granulation particle size distribution, and loose packing density of the second oxygen-correcting powder, the shrinkage initiation temperature and cumulative linear shrinkage rate of the two test green bodies were matched to obtain the first shrinkage-matched granulated powder and the second shrinkage-matched granulated powder. Subsequently, according to the original dry basis mass ratio of aluminum nitride contained in the two types of granulated powder, low-shear dry confluence was carried out to form composite granulated powder without secondary slurry preparation and secondary spray granulation, so as to retain the shrinkage matching state of the two types of granulated powder. After sintering, Vickers microhardness array tests were performed in the central, middle, and edge regions of the ceramic polished section. The weakened test points were dispersed and did not form closed or semi-closed test point chains, indicating that the obtained aluminum nitride ceramic has good local hardness uniformity.

[0143] S6, additive reaction homogenization sintering and densification sintering;

[0144] After degreasing, while continuously introducing nitrogen and maintaining positive pressure inside the furnace, the temperature is increased to 1200 degrees Celsius at a rate of 5 degrees Celsius per minute, and then increased to 1550 degrees Celsius at a rate of 3 degrees Celsius per minute.

[0145] The mixture was kept at 1550 degrees Celsius for 1.5 hours to homogenize and sinter the additive reaction, so that yttrium oxide reacted with the oxygen-containing substances carried by the primary dispersed phase and the recovered dispersed phase, and promoted the diffusion and redistribution of yttrium oxide and the primary yttrium-containing reaction phase in the contact area of ​​aluminum nitride particles.

[0146] The additive reaction homogenization sintering described in this embodiment does not require the formation of a continuous flowable liquid phase at 1550 degrees Celsius as a necessary condition. Instead, it achieves a more uniform interfacial reaction state between the two types of powders through additive reaction and diffusion before densification.

[0147] After the additive reaction homogenization sintering is completed, the temperature is increased to 1820 degrees Celsius at 2 degrees Celsius / minute and held at 1820 degrees Celsius for 3 hours to carry out densification sintering, so that the residual pores in the aluminum nitride ceramic green body shrink and the sintering bonding between aluminum nitride particles is completed.

[0148] At the densification sintering temperature, yttrium oxide and oxygen-containing substances in the powder form yttrium-containing components that participate in the sintering mass transfer process, and after sintering, a yttrium-containing phase is formed, including yttrium-containing reaction products and residual yttrium-containing additive components.

[0149] After densification sintering is completed, the furnace is cooled to 800 degrees Celsius at a cooling rate of no more than 5 degrees Celsius per minute, and then cooled to room temperature with the furnace.

[0150] Nitrogen gas was continuously introduced during the heating, holding and cooling processes, and positive pressure was maintained inside the furnace to obtain the aluminum nitride ceramic wear-resistant liner of Example 1.

[0151] Example 2

[0152] In another embodiment, the aluminum nitride ceramic of the present invention can be made into a semiconductor wafer carrier disk.

[0153] Nano-aluminum nitride powder with an average primary particle size of 100 nanometers was selected, and anhydrous ethanol was used as the dispersion medium to prepare a suspension with a solid content of 20% to 30%. The suspension was pre-dispersed under argon protection using a combination of low-power ultrasound and mechanical stirring.

[0154] After pre-dispersion, the suspension is allowed to stand, and the rapidly settling coarse agglomerates are collected. Then, the upper suspension is centrifuged and classified. The rapidly settling coarse agglomerates are combined with the settled particles obtained from centrifugation and classified as the hard agglomerate phase; the particle aggregates in the dispersion obtained from centrifugation and classified are identified as the primary dispersed phase.

[0155] Therefore, the rapidly settling coarse agglomerates are not discarded, and the sum of the dry basis mass of aluminum nitride contained in the primary dispersed phase and the hard agglomerate phase corresponds to the dry basis mass of aluminum nitride entering the fractionation step.

[0156] The hard agglomerated phase is added to anhydrous ethanol and depolymerized independently using a bead milling method. The bead milling media can be zirconia ceramic beads with a diameter of 0.08–0.15 mm, with a grinding media filling rate of 55%–70% and a rotor linear speed of 6–10 m / s.

[0157] When the dispersed phase D is recovered 90 Not greater than the primary dispersed phase D 90 When the concentration reaches 1.20 times the standard value, independent wet depolymerization is stopped.

[0158] The primary dispersed phase and the recovered dispersed phase were subjected to vacuum drying and total oxygen content testing under the same conditions. The amount of yttrium oxide added was determined according to the alumina equivalent conversion method described in Example 1, and the first oxygen-correcting powder and the second oxygen-correcting powder were obtained respectively.

[0159] First and second test blanks were prepared separately, and the linear shrinkage curves of the two test blanks were measured. By adjusting at least one of the following factors of the slurry solid content, binder content, granulation particle size distribution, and loose packing density of the second oxygen-correcting powder, the difference in the shrinkage initiation temperature of the two test blanks was made not greater than 30 degrees Celsius, and the difference in the cumulative linear shrinkage rate when heated to 1700 degrees Celsius was not greater than 0.5 percentage points.

[0160] The remaining two types of oxygen-correcting powders were granulated in batches according to the granulation parameters corresponding to the shrinkage matching conditions, to obtain the first shrinkage-matched granulated powder and the second shrinkage-matched granulated powder.

[0161] Weigh the aluminum nitride in the two types of granulated powder according to the original dry basis mass ratio, and mix them using a three-dimensional mixer at a speed of 10-20 rpm using a low-shear dry method for 20-40 minutes. The mass fraction of yttrium and the D content of the granulated powder are then used as the basis for the final determination. 50 The relative standard deviations are used to evaluate the uniformity of merging.

[0162] The mixed composite granulated powder is directly subjected to molding preforming and cold isostatic pressing to produce a bearing plate green body.

[0163] After degreasing, the additive reaction homogenization sintering is carried out at 1500-1650 degrees Celsius, followed by densification sintering at 1800-1880 degrees Celsius. After cooling, the surface is ground and polished to obtain the semiconductor wafer carrier disk.

[0164] This embodiment illustrates that the present invention can be applied to the fabrication of semiconductor wafer carrier disks.

[0165] Example 3

[0166] In another embodiment, the aluminum nitride ceramic of the present invention can be used to make an insulating heat dissipation substrate for power electronic devices.

[0167] Nano-aluminum nitride powder with an average primary particle size of 50–100 nm was selected, and a suspension with a solid content of 30%–40% was prepared using anhydrous isopropanol as the dispersion medium. The suspension was then pre-dispersed under nitrogen protection.

[0168] After pre-dispersion, a wet sieve with a pore size of 20 micrometers is used for sieving to collect the coarse aggregates on the sieve and the suspension below the sieve.

[0169] The suspension under the sieve was centrifuged to obtain a centrifuged dispersion and centrifuged sedimented particles. The coarse agglomerates on the sieve were combined with the centrifuged sedimented particles to form the hard agglomerate phase; the particle aggregates in the centrifuged dispersion were identified as the primary dispersed phase.

[0170] Therefore, the large agglomerates on the wet screening screen are not discarded, but can enter the hard agglomerate phase treatment process together with the subsequent centrifugal sedimentation particles, thus maintaining the quality closure of the aluminum nitride material.

[0171] The hard agglomerated phase is added to anhydrous isopropanol, and independent wet depolymerization is carried out using a narrow-gap shearing device. The shear gap can be 0.15–0.30 mm, and the rotor linear velocity can be 8–15 m / s.

[0172] When the dispersed phase D is recovered 90 Not greater than the primary dispersed phase D 90 When the concentration reaches 1.20 times the standard value, independent wet depolymerization is stopped.

[0173] The total oxygen content of the primary dispersed phase and the recovered dispersed phase were measured separately, and the amount of first yttrium oxide and second yttrium oxide added was determined based on the dry basis mass of aluminum nitride and the total oxygen content of the two types of powders.

[0174] Oxygen mixing, spray granulation, test green body preparation, and linear shrinkage testing were performed separately to ensure that the difference in shrinkage initiation temperature and the difference in cumulative linear shrinkage rate at 1700 degrees Celsius between the two test green bodies met the preset shrinkage matching conditions.

[0175] The two types of shrinkage-matched granulated powders were combined using a low-shear dry method according to the original dry basis mass ratio of the aluminum nitride they contained, without secondary pulping or secondary spray granulation during the mixing process.

[0176] The obtained composite granulated powder was pre-formed by plate molding and cold isostatic pressing. After degreasing, it was homogenized and sintered at 1450-1600 degrees Celsius with additives, and then densified and sintered at 1750-1850 degrees Celsius to obtain plate-shaped aluminum nitride ceramic.

[0177] Double-sided grinding and polishing of plate-shaped aluminum nitride ceramics can produce insulating and heat-dissipating substrates for power electronic devices.

[0178] This embodiment is used to illustrate the material flow direction of the combination of wet screening and centrifugal grading, and the application of the present invention in plate-shaped aluminum nitride ceramic products.

[0179] Comparative Example 1

[0180] This comparative example uses the same batch of nano-aluminum nitride raw material powder, anhydrous isopropanol, yttrium oxide, polyvinyl butyral, molding equipment, and sintering equipment as Example 1.

[0181] All nano-aluminum nitride raw material powder was added to anhydrous isopropanol to prepare a suspension with a solid content of 30%. Pre-dispersion was carried out using the same dispersion tank, four-bladed impeller, stirring speed of 320 rpm, and processing time as in Example 1, but the primary dispersed phase and hard agglomerate phase were not separated in the pre-dispersion suspension, nor were the hard agglomerates subjected to independent wet depolymerization.

[0182] The total oxygen content of the entire batch of powder was determined using the same drying and total oxygen content detection conditions as in Example 1.

[0183] To ensure that the total amount of yttrium oxide added is the same between Comparative Example 1 and Example 1, Comparative Example 1 does not calculate the amount of yttrium oxide added separately based on the total oxygen content of the whole batch of powder, but directly adds 2.557 kg of yttrium oxide based on 100 kg of aluminum nitride dry basis.

[0184] 2.557 kg of yttrium oxide was added to all the nano-aluminum nitride powder at once, and anhydrous isopropanol was added for wet mixing.

[0185] The resulting mixed powder was formulated into a spray granulation slurry with a solid content of 48%, and the amount of polyvinyl butyral added was 1.50% of the dry basis mass of the powder.

[0186] Granulated powder was prepared using the same spray granulation conditions as in Example 1. The granulated powder was pre-molded at 60 MPa and then cold isostatically pressed at 200 MPa.

[0187] Using the same degreasing, additive reaction homogenization sintering, and densification sintering procedures as in Example 1, aluminum nitride ceramic samples of Comparative Example 1 were prepared.

[0188] The total amount of yttrium oxide added, molding conditions, and sintering conditions were the same for Comparative Example 1 and Example 1. The difference between the two is that Comparative Example 1 did not perform hard agglomerate phase separation and independent deagglomeration, did not separately prepare yttrium oxide according to the total oxygen content of the primary dispersed phase and the powder from which the hard agglomerates originated, and did not perform early sintering shrinkage matching between the two types of powders.

[0189] Comparative Example 2

[0190] The difference between this comparative example and Example 1 is that the amount of yttrium oxide added is not determined based on the total oxygen content of the primary dispersed phase and the recovered dispersed phase.

[0191] Using the same pre-dispersion, centrifugal fractionation, and wet jet depolymerization conditions as in Example 1, the primary dispersed phase and the recovered dispersed phase were obtained.

[0192] The same total amount of yttrium oxide as in Example 1 was distributed according to the dry mass ratio of aluminum nitride in the two types of powders of 92:8, so that the two types of powders used the same proportion of yttrium oxide per unit mass of aluminum nitride.

[0193] The two types of powders were mixed, dried, slurried, and granulated respectively. By adjusting the granulation state of the second type of powder, the two test preforms were made to meet the shrinkage matching conditions described in Example 1.

[0194] The two types of granulated powders were combined using a low-shear dry method with a dry mass ratio of aluminum nitride of 92:8. The subsequent molding, degreasing, additive reaction homogenization sintering, and densification sintering conditions were the same as in Example 1.

[0195] This comparative example is used to illustrate the difference between it and Example 1 in the distribution of yttrium oxide.

[0196] Comparative Example 3

[0197] The difference between this comparative example and Example 1 is that the early sintering shrinkage behavior of the first and second oxygen-correcting powders is not matched.

[0198] Using the same pre-dispersion, centrifugal classification, independent wet depolymerization of hard agglomerates, total oxygen content detection, and separate yttrium preparation steps as in Example 1, the first oxygen-correcting powder and the second oxygen-correcting powder were obtained.

[0199] The first and second test blanks are not prepared, the linear shrinkage curves of the two types of oxygen-correcting powders are not measured, and the slurry solid content, granulation particle size distribution, granulation particle loose density and polyvinyl butyral addition amount of the second oxygen-correcting powder are not adjusted according to the linear shrinkage curves.

[0200] The first oxygen-correcting powder was spray-granulated with a solid content of 48% and a polyvinyl butyral content of 1.50%; the second oxygen-correcting powder was spray-granulated with an initial solid content of 44% and a polyvinyl butyral content of 1.50%.

[0201] The two types of granulated powders were combined using a low-shear dry method with a dry mass ratio of aluminum nitride of 92:8. The subsequent molding, degreasing, additive reaction homogenization sintering, and densification sintering conditions were the same as in Example 1.

[0202] This comparative example is used to illustrate the difference between it and Example 1 in the early sintering shrinkage matching step.

[0203] Detection methods

[0204] 1. Relative density detection

[0205] Three parallel samples were obtained from the aluminum nitride ceramics obtained in the examples and comparative examples, respectively.

[0206] Before testing, the sample was vacuum dried at 150 degrees Celsius to constant weight. Constant weight was determined when the mass difference between two consecutive weighings was no greater than 0.01% of the mass of the previous weighing.

[0207] Anhydrous ethanol was used as the displacer medium in the Archimedes displacement method, and the measured density of anhydrous ethanol was determined at 23±2 degrees Celsius. .

[0208] Determine the dry mass of the sample in air. Immerse the sample completely in anhydrous ethanol and perform low-intensity vacuum degassing until no more continuous bubbles appear on the sample surface. The vacuum degassing time should not exceed 5 minutes.

[0209] Determine the saturated mass of the sample after impregnation with anhydrous ethanol And the suspended mass of the sample completely immersed in anhydrous ethanol .

[0210] Sample bulk density Calculate according to the following formula:

[0211]

[0212] In the formula, This indicates the mass of the sample after it has dried in air. This indicates the saturated mass of the sample after it has been immersed in anhydrous ethanol. This indicates the suspended mass of the sample when it is completely submerged in anhydrous ethanol. This indicates the measured density of anhydrous ethanol at the detection temperature.

[0213] At least 5 grams of material were cut from the same batch of ceramic samples, and the material was pulverized until it passed through a 75-micron sieve. After drying under vacuum at 150 degrees Celsius for 4 hours, the true density of the same batch of ceramics was determined by the helium displacement method. .

[0214] relative density Calculate according to the following formula:

[0215]

[0216] In the formula, This indicates the bulk density measured using the Archimedes displacement method; This indicates the true density measured using the helium displacement method; This indicates relative density.

[0217] The relative densities of the three parallel samples were calculated separately, and their arithmetic mean was used as the relative density of the corresponding embodiment or comparative example.

[0218] 2. Array-based Vickers microhardness testing

[0219] Polished inspection surfaces were prepared from representative cross sections of aluminum nitride ceramic products.

[0220] For circular, rectangular, or other products with an identifiable geometric center, draw a radial reference line from the geometric center of the inspection surface to the nearest edge.

[0221] The location of the geometric center of the detection surface is defined as the central region; the location at half the distance from the geometric center to the nearest edge is defined as the intermediate region; and the location 1 to 2 millimeters from the nearest edge is defined as the edge region.

[0222] The array centers of the central, intermediate, and edge regions lie on the same radial reference line. For irregular articles, the three detection regions are determined based on the geometric center of their largest inscribed region and the nearest edge.

[0223] The test section is successively ground and polished using a diamond grinding wheel and diamond polishing fluid to reduce its surface roughness. No larger than 0.05 micrometers.

[0224] A rectangular array of 21×21 detection points is set in the central area, the middle area and the edge area respectively. Each rectangular array includes 441 detection points, and a total of 1323 detection points are set in the three detection areas.

[0225] The spacing between detection points in each rectangular array is fixed at 30 micrometers. A detection load of 0.05 kgf is used, with a holding time of 10 seconds. Under this detection load, the diagonal length of the Vickers indentation is 7.4–8.2 micrometers, and adjacent indentations do not overlap.

[0226] Invalid test points are allowed only when the indentation overlaps with an existing indentation, the indentation crosses the edge of the sample, the indentation outline cannot be identified due to polishing scratches, or the testing instrument fails to form a complete indentation.

[0227] A test point should not be excluded solely because of its low hardness. The number of valid test points in each 21×21 rectangular array should not be less than 420; if there are fewer than 420, the test surface should be re-prepared and the area should be re-tested.

[0228] The arithmetic mean of the Vickers microhardness of all valid test points in the three test areas is taken as the cross-sectional average hardness. Calculate according to the following formula:

[0229]

[0230] In the formula, This represents the Vickers microhardness of the i-th valid test point; This represents the total number of valid detection points within the three detection areas.

[0231] Record the average hardness of the central area, the middle area, and the edge area, the minimum hardness of the area, the average cross-sectional hardness of the entire test area, the minimum hardness of all valid test points, the ratio of the minimum hardness to the average cross-sectional hardness, and the standard deviation of the hardness values ​​of all valid test points.

[0232] Test points with Vickers microhardness lower than 95% of the average cross-sectional hardness are defined as weakening test points, while test points with Vickers microhardness not lower than 95% of the average cross-sectional hardness are defined as non-weakening test points.

[0233] Closed detection point chains and semi-closed detection point chains are judged separately within each 21×21 rectangular array and do not connect across different detection areas.

[0234] Each detection point is treated as a grid node. Two adjacent weakened detection points, whether horizontally, vertically, or diagonally adjacent, are connected according to an eight-neighbor relationship.

[0235] When weakened detection points form a continuous path that connects end to end, and at least one non-weakened detection point located inside the path cannot be connected to the boundary of the rectangular array along an eight-neighborhood path consisting only of non-weakened detection points, a closed detection point chain is determined to be formed.

[0236] When a weakened detection point forms a continuous path that is not connected end to end, and the straight-line distance between the first and last detection points of the continuous path is no more than 60 micrometers, and after being closed by the straight-line segment between the first and last detection points, it surrounds at least one non-weakened detection point, it is determined that a semi-closed detection point chain has been formed.

[0237] The straight line segments used to supplement the closed and semi-closed detection point chains should be located inside the corresponding rectangular array and should not cross another independent weakened detection point chain.

[0238] The aluminum nitride ceramic of the present invention satisfies the following conditions: the Vickers microhardness of any effective detection point is not less than 90% of the average cross-sectional hardness, and the weakened detection points in each detection area do not form a closed detection point chain or a semi-closed detection point chain.

[0239] 3. Abrasion resistance and island-like spalling test

[0240] The aluminum nitride ceramics obtained in Example 1 and Comparative Example 1 were processed into wear-resistant specimens with dimensions of 30 mm × 20 mm × 5 mm, and three parallel specimens were set up in each group.

[0241] A silicon nitride ball with a diameter of 6 mm is used as the friction pair, with a normal load of 10 Newtons, a reciprocating stroke of 5 mm, a reciprocating frequency of 5 Hz, and a cycle count of 20,000.

[0242] The reciprocating stroke of 5 mm refers to the one-way distance the friction pair moves from one end of the motion path to the other. One cycle includes the forward and return strokes, therefore the total sliding distance is... Calculate according to the following formula:

[0243]

[0244] The test environment temperature was 23±2 degrees Celsius and the relative humidity was 50%±5%.

[0245] After the test, the wear-resistant samples were cleaned with anhydrous ethanol and dried. The cross-section of the wear track was measured using a three-dimensional optical profilometer, and the wear volume was obtained by integrating along the length of the wear track. .

[0246] specific wear rate Calculate according to the following formula:

[0247]

[0248] In the formula, This indicates the wear volume, expressed in cubic millimeters. Indicates the normal load, measured in Newtons; This represents the total sliding distance, in meters.

[0249] The entire wear area was observed using an optical microscope. Block-shaped spalling pits with closed or nearly closed contours and an equivalent diameter of not less than 50 micrometers were identified as island-shaped spalling pits, and the number and maximum equivalent diameter of the island-shaped spalling pits were recorded.

[0250] Test results

[0251] The aluminum nitride ceramics obtained in Example 1 and Comparative Example 1 were tested according to the above testing methods, and the test results are shown in Table 1.

[0252] Table 1. Detection results of Example 1 and Comparative Example 1

[0253] Helium displacement true density, g / cm³ 3.255 3.254 Archimedes' volume density, grams per cubic centimeter 3.229 3.211 Relative density, % 99.20 98.68 Total number of array detection points 1323 1323 Average Vickers microhardness, HV 0.05 1548 1512 Minimum Vickers microhardness, HV 0.05 1452 1247 The ratio of minimum hardness to average hardness, % 93.80 82.47 Hardness standard deviation, HV 0.05 37 86 Number of closed detection point chains 0 1 Number of semi-closed detection point chains 0 2 Specific wear rate, cubic millimeters / Newton-meter <![CDATA[2.5×10⁻ 6 ]]> <![CDATA[7.1×10⁻ 6 ]]> Number of island-shaped spalling pits 0 6 The equivalent diameter of the largest island-shaped spalling pit is in micrometers. Not detected 115

[0254] The hardness test results of the three test areas of Example 1 and Comparative Example 1 are shown in Table 2.

[0255] Table 2. Hardness test results of different regions in Example 1 and Comparative Example 1

[0256] Example 1 Central area 1552 1468 0 0 Example 1 Middle area 1547 1459 0 0 Example 1 Edge area 1545 1452 0 0 Comparative Example 1 Central area 1520 1328 0 1 Comparative Example 1 Middle area 1514 1286 1 0 Comparative Example 1 Edge area 1502 1247 0 1

[0257] As shown in Tables 1 and 2, the relative densities of both Example 1 and Comparative Example 1 are higher than 98%, indicating that both preparation processes can achieve a high overall density of aluminum nitride ceramics.

[0258] In Example 1, the average Vickers microhardness was 1548 HV0.05, the lowest Vickers microhardness was 1452 HV0.05, and the ratio of the lowest hardness to the average hardness was 93.80%. All effective test points met the condition of not less than 90% of the average hardness of the cross section.

[0259] In Example 1, no closed or semi-closed detection point chains were formed in the central, middle, and edge regions.

[0260] The average Vickers microhardness of Comparative Example 1 was 1512 HV0.05, which was in a similar range to that of Example 1, but the lowest Vickers microhardness was 1247 HV0.05, and the ratio of the lowest hardness to the average hardness was 82.47%.

[0261] Comparative Example 1 forms a semi-closed chain of detection points in the central region, a closed chain of detection points in the middle region, and a semi-closed chain of detection points in the edge region.

[0262] The above results indicate that the existence of localized hardness valleys in aluminum nitride ceramics cannot be ruled out based solely on average hardness and overall relative density.

[0263] Example 1 achieves relatively uniform local hardness in the central, middle, and edge regions through hard agglomerate phase splitting and independent deagglomeration, yttrium formulation of primary and recycled dispersed phases, early sintering shrinkage matching of the two types of powders, low-shear dry merging of the two types of shrinkage-matched granulated powders, and homogenization sintering by additive reaction.

[0264] Under the same frictional wear conditions, the specific wear rate of Example 1 was 2.5 × 10⁻ 6 cubic millimeters / Newton-meter, no island-shaped spalling pits detected; the specific wear rate of Comparative Example 1 was 7.1 × 10⁻⁻⁻⁶. 6 The volume was measured in cubic millimeters per Newton-meter. Six island-shaped spalling pits were detected, with the largest island-shaped spalling pit having an equivalent diameter of 115 micrometers.

[0265] Test results show that the present invention can not only enable aluminum nitride ceramics to achieve a high overall density and average hardness, but also reduce the possibility of local low hardness areas being continuously distributed along the periphery of normal hardness areas, thereby reducing island-like overall peeling caused by local weakening boundaries.

[0266] The aluminum nitride ceramic prepared in Example 1 can be processed into wear-resistant linings, guide plates, or sealing rings, and can be used in mechanical components subject to sliding friction, particle erosion, or localized contact loads.

[0267] Using the process described in Example 2, aluminum nitride ceramics can be made into semiconductor wafer carriers, semiconductor device insulating components, or precision ceramic carrier components.

[0268] Using the process described in Example 3, aluminum nitride ceramics can be made into insulating and heat-dissipating substrates for power electronic devices or ceramic substrates for high-power electronic devices.

[0269] During the processing of the above-mentioned products, diamond grinding, precision grinding, and polishing are employed according to the product's shape, size, and precision requirements. The processed products can then be tested for dimensional accuracy, flatness, edge chipping, wear resistance, insulation performance, or thermal conductivity, depending on actual usage requirements.

[0270] This invention separates the hard agglomerated phase from the primary dispersed phase and deagglomerates it independently. Yttrium oxide is then added based on the total oxygen content of both the primary and recovered dispersed phases. Before the two types of powders merge, their early sintering shrinkage behavior is matched. A low-shear dry merging method is used to preserve the granulation state of the two types of shrinkage-matched granulated powders. This reduces the possibility that the original hard agglomerated boundary simultaneously becomes a localized concentration area of ​​yttrium-containing components and a differential shrinkage area. The resulting aluminum nitride ceramic achieves high relative density and average hardness while improving the uniformity of local hardness at different locations, reducing the formation of closed or semi-closed weakened detection point chains, and lowering the possibility of island-like spalling during wear, cutting, or impact.

[0271] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A high-hardness nano-aluminum nitride ceramic prepared using yttrium oxide as a sintering aid, characterized in that, The high-hardness nano-aluminum nitride ceramic has a relative density of not less than 98%, and the relative density is the ratio of the volume density measured by Archimedes' displacement method to the true density measured by helium displacement method. A rectangular array of 21×21 detection points is set in the center, middle and edge regions of the polished cross-section of the high-hardness nano-aluminum nitride ceramic. The spacing between the detection points in each rectangular array is 30 micrometers. Vickers microhardness testing is performed on each detection point using a detection load of 0.05 kgf and a holding time of 10 seconds. The arithmetic mean of the Vickers microhardness of all effective detection points in the three rectangular arrays is taken as the cross-sectional average hardness. The Vickers microhardness of each effective detection point is not less than 90% of the cross-sectional average hardness. Detection points with Vickers microhardness less than 95% of the cross-sectional average hardness do not form a closed or semi-closed chain of detection points surrounding at least one detection point with Vickers microhardness not less than 95% of the cross-sectional average hardness.

2. The high-hardness nano-aluminum nitride ceramic according to claim 1, characterized in that, A radial reference line is set from the geometric center of the polished section to the nearest edge. The position where the geometric center is located is defined as the central region, the position where the distance from the geometric center to the nearest edge is half the position is defined as the middle region, and the position 1 to 2 mm away from the nearest edge is defined as the edge region. The closed detection point chain is formed by sequentially connecting adjacent detection points with a Vickers microhardness lower than 95% of the average cross-sectional hardness along the transverse, longitudinal, or diagonal direction of the rectangular array, with the ends connected together. The semi-closed detection point chain is formed by sequentially connecting adjacent detection points with a Vickers microhardness lower than 95% of the average cross-sectional hardness, with the straight-line distance between the first and last detection points not exceeding 60 micrometers. Furthermore, the semi-closed detection point chain, together with the lines connecting the first and last detection points, encloses at least one detection point with a Vickers microhardness not lower than 95% of the average cross-sectional hardness.

3. The high-hardness nano-aluminum nitride ceramic according to claim 1 or 2, characterized in that, The average hardness of the cross section is not less than 1500HV0.

05.

4. A method for preparing high-hardness nano-aluminum nitride ceramics, characterized in that, Includes the following steps: S1. Add nano-aluminum nitride raw material powder to an anhydrous organic dispersion medium and pre-dispersettle under inert gas protection. Detect and classify the particle size of the pre-dispersion suspension to obtain the primary dispersed phase and the hard agglomerate phase, and record the original dry basis mass ratio of aluminum nitride contained in the primary dispersed phase and the hard agglomerate phase. S2. Perform independent wet depolymerization on the hard agglomerate phase to obtain a recovered dispersed phase until the volume cumulative distribution of the recovered dispersed phase reaches 90%, and the corresponding equivalent particle size is not greater than 1.20 times the corresponding equivalent particle size of the original dispersed phase. S3. Dry the primary dispersed phase and the recovered dispersed phase respectively and test the total oxygen content. Determine the oxygen mass of the primary dispersed phase based on the dry basis of aluminum nitride and the total oxygen content. Determine the oxygen mass of the recovered dispersed phase based on the dry basis of aluminum nitride and the total oxygen content. Determine the amount of first yttrium oxide to be added based on the oxygen mass of the primary dispersed phase. Determine the amount of second yttrium oxide to be added based on the oxygen mass of the recovered dispersed phase. Add yttrium oxide and mix them to obtain the first oxygen correction powder and the second oxygen correction powder. S4. The first oxygen-correcting powder and the second oxygen-correcting powder are respectively pulped, granulated and pressed to obtain a first test blank and a second test blank. The linear shrinkage curves of the first test blank and the second test blank are measured under the same atmosphere and the same temperature rise conditions. The granulation state of the second oxygen-correcting powder is adjusted with the linear shrinkage curve of the first test blank as the matching benchmark so that the difference between the shrinkage initiation temperature of the two test blanks is not greater than 30 degrees Celsius and the difference in the cumulative linear shrinkage rate when the temperature is raised to 1700 degrees Celsius is not greater than 0.5 percentage points. The first oxygen-correcting powder and the second oxygen-correcting powder are granulated in batches according to the granulation parameters corresponding to the above conditions to obtain the first shrinkage matching granulated powder and the second shrinkage matching granulated powder. S5. Weigh two types of granulated powders in such a way that the dry basis mass ratio of aluminum nitride contained in the first shrinkage matching granulated powder and the second shrinkage matching granulated powder is the same as the original dry basis mass ratio. Mix the two types of granulated powders by low shear dry mixing to obtain composite granulated powder. Then, perform molding preforming, cold isostatic pressing and debinding on the composite granulated powder in sequence to obtain aluminum nitride ceramic green body. S6. The aluminum nitride ceramic green body is placed in a nitrogen protective atmosphere and subjected to additive reaction homogenization sintering and densification sintering in sequence. After cooling, high-hardness nano-aluminum nitride ceramic is obtained.

5. The method for preparing high-hardness nano-aluminum nitride ceramics according to claim 4, characterized in that, In step S1, the pre-dispersed suspension is pre-graded to obtain a primary dispersed phase prediction sample. The equivalent particle size corresponding to the cumulative volume distribution of the primary dispersed phase prediction sample reaching 90% is used as the basic particle size, and 1.3 to 2.0 times the basic particle size is used as the grading cut-off point. The grading adopts at least one of centrifugal grading and sedimentation grading, or firstly, wet sieving is used to collect the coarse agglomerates on the sieve, and then the suspension under the sieve is centrifuged and graded, and the coarse agglomerates on the sieve and the sedimented particles obtained by centrifugal grading are combined into a hard agglomerate phase.

6. The method for preparing high-hardness nano-aluminum nitride ceramics according to claim 4 or 5, characterized in that, In step S2, the independent wet depolymerization employs at least one of narrow-gap shearing, wet jet depolymerization, bead milling depolymerization, and high-pressure homogenization; when the equivalent particle size corresponding to the cumulative volume distribution of the recovered dispersed phase reaches 90% is not greater than 1.10 times the corresponding equivalent particle size of the original dispersed phase, the independent wet depolymerization is stopped.

7. The method for preparing high-hardness nano-aluminum nitride ceramics according to any one of claims 4 to 6, characterized in that, In step S3, the mass of oxygen in the primary dispersed phase and the mass of oxygen in the recovered dispersed phase are converted into alumina equivalents, respectively. The theoretical amount of yttrium oxide to be added is determined according to the ratio of 5 moles of alumina equivalent to 3 moles of yttrium oxide. The theoretical amount of yttrium oxide to be added is then multiplied by a correction factor of 0.85 to 1.15 to obtain the first amount of yttrium oxide to be added and the second amount of yttrium oxide to be added. The alumina equivalent is a stoichiometric standard for determining the amount of yttrium oxide to be added in the process, and does not mean that all oxygen in the nano-aluminum nitride powder exists in the alumina phase.

8. The method for preparing high-hardness nano-aluminum nitride ceramics according to any one of claims 4 to 7, characterized in that, In step S4, by adjusting at least one of the following: slurry solid content, binder content, granulation particle size distribution, and loose packing density of the second oxygen-correcting powder, the difference in shrinkage initiation temperature between the first test blank and the second test blank is no greater than 20 degrees Celsius, and the difference in cumulative linear shrinkage rate when heated to 1700 degrees Celsius is no greater than 0.3 percentage points.

9. The method for preparing high-hardness nano-aluminum nitride ceramics according to any one of claims 4 to 8, characterized in that, In step S5, the low-shear dry mixing is carried out using a three-dimensional mixing device at a mixing speed of 10-20 rpm and a mixing time of 20-40 minutes. Samples are taken from different locations of the mixed composite granulated powder, and the mass fraction of yttrium and the median equivalent particle size of each sample are measured. The low-shear dry mixing ends when the relative standard deviation of the mass fraction of yttrium is not greater than 3% and the relative standard deviation of the median equivalent particle size is not greater than 3%. In step S6, the temperature for homogenization and sintering of the additive reaction is 1450-1700 degrees Celsius, and the holding time is 0.5-3 hours. The temperature for densification and sintering is 1700-1900 degrees Celsius, and the holding time is 1-6 hours.

10. The application of the high-hardness nano-aluminum nitride ceramic according to any one of claims 1 to 3 or the high-hardness nano-aluminum nitride ceramic prepared by the preparation method according to any one of claims 4 to 9 in the preparation of wear-resistant linings, guide plates, sealing rings, precision ceramic bearing components, semiconductor wafer bearing disks, semiconductor equipment insulating components, power electronic device insulating heat dissipation substrates or high-power electronic device ceramic substrates.

Images