Photoelectric pre-separation and flotation combined high-calcium fluorite ore separation process

Abstract

The invention discloses a photoelectric pre-separation and flotation combined high-calcium fluorite ore separation process which comprises the following steps: firstly, crushing and grading raw ores, pre-enriching coarse and medium-sized ores based on multispectral photoelectric separation, AI identification and near infrared spectrum identification, and removing more than 40% of low-calcium waste rocks in advance, so that the subsequent treatment load is obviously reduced. And the photoelectric concentrate and the fine ore are combined and then enter an optimized flotation process, and the final high-quality fluorite concentrate is obtained through an accurate reagent system. Particularly, a synergistic agent system composed of a compound inhibitor, a compound synergist and a collecting agent is adopted in the flotation stage, and the floatability difference of fluorite and calcite can be remarkably enlarged through the system. According to the method, the problem of efficient separation of fluorite and calcium carbonate is successfully solved, the CaF2 grade of the final concentrate is larger than or equal to 98%, the recovery rate is larger than or equal to 85%, energy consumption and agent cost are greatly reduced, and efficient and green development of resources is achieved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of mineral separation, and particularly relates to a high-calcium fluorite ore separation process combining photoelectric pre-separation with flotation. BACKGROUND

[0002] High-calcium fluorite ore is mainly composed of calcium fluoride, which is an important raw material in the fields of metallurgy, aluminum smelting, glass, ceramics, cement, and chemical industry. Its products are widely used in aerospace, medicine and pesticide, machinery and electronics, air conditioning and refrigeration, and other fields, and have become an important strategic material. High-calcium fluorite ore has complex composition, and fluorite is finely disseminated (0.005-0.030 mm) and closely associated with calcite, spessartine, quartz and other high-calcium gangue minerals. There are as many as 143 kinds of minerals. Fluorite recovery needs to be carried out at the end of the multi-metal flotation process, but residual reagents (such as tungsten flotation depressants) interfere with the floatability of fluorite, and conventional fatty acid collectors have strong collecting ability for fluorite and calcite, resulting in low separation efficiency. In addition, the current separation method of high-calcium fluorite ore generally uses single flotation method, which cannot realize the separation of full-size high-calcium fluorite ore, resulting in waste.

[0003] Photoelectric separation technology has become one of the commonly used pretreatment methods in ore separation due to its core advantages of precision, efficiency, environmental protection, and low cost. However, the traditional photoelectric separation equipment has obvious limitations: on the one hand, the intelligent level is insufficient, and it is difficult to identify the subtle spectral differences between fluorite (CaF2) and calcareous gangue minerals such as calcite (CaCO3) with highly similar apparent properties; on the other hand, the effective separation particle size range is narrow, and the middling treatment efficiency is low; in addition, the fixed threshold method used has poor adaptability, resulting in low overall separation accuracy. In recent years, patents such as near-infrared multi-spectral separation equipment (CN201710890985.3), material separator containing intelligent scanning and identification structure (CN202421143089.2), optical separator (CN202180018602.8), and photoelectric analysis and detection device of separator (CN202421507794.6) have improved the separation efficiency from the equipment itself, but they usually only handle coarse particles (such as >10 mm), and the treatment of medium and fine particles is limited, and the middling treatment method is relatively simple and rough.

[0004] In the aspect of reagents in the flotation stage, the existing technology is mostly focused on the improvement or conventional combination of single reagent, such as fatty acid collectors such as oleic acid, and conventional depressants represented by water glass and modified water glass. It is difficult to achieve effective separation of fluorite and calcium-containing gangue minerals, resulting in the problem of unqualified fluorite concentrate product or low recovery rate, and there is a lack of high-selectivity composite reagent system capable of deeply separating fluorite and calcite. In addition, the selectivity of conventional single depressant is poor, which makes it difficult to balance the concentrate grade and recovery rate, and restricts the production of high-quality fluorite concentrate.

[0005] Therefore, it is necessary to develop a sorting process with high sorting accuracy, high separation efficiency, and high recovery rate. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a high-calcium fluorite ore separation process that combines photoelectric pre-selection and flotation. This process effectively solves the problems of low separation accuracy and low separation efficiency in high-calcium fluorite ore. While expanding the effective particle size range of photoelectric separation, it maximizes the overall recovery rate by relying on middlings recycling, thus making up for the shortcomings of a single process. It also has the significant advantages of a simple process flow and excellent separation indicators.

[0007] To achieve the above objectives, the present invention provides a high-calcium fluorite ore separation process combining photoelectric pre-selection and flotation, characterized by comprising the following steps: (1) Crushing and grading of raw ore: The high-calcium fluorite raw ore is crushed to a particle size ≤60mm, and then screened to obtain coarse-grained ore with a particle size of 8-50mm, medium-grained ore with a particle size of 1-8mm and fine-grained ore with a particle size ≤1mm. (2) Photoelectric separation: Multispectral photoelectric separation is performed on coarse-grained ore, and AI recognition is used to dynamically distinguish high-calcium fluorite particles from low-calcium fluorite particles to obtain photoelectric concentrate I, photoelectric tailings I and photoelectric middlings I; among them, photoelectric middlings I is returned to step (1) for recycling. (3) Gradient separation: The medium-sized ore is subjected to multispectral photoelectric separation and near-infrared spectroscopy to identify surface texture features, resulting in photoelectric concentrate II and photoelectric tailings II; (4) Grinding: Take photoelectric concentrate I and photoelectric concentrate II and fine-grained ore and grind them together to obtain flotation feed; (5) Roughing: After the flotation feed is prepared into a slurry, the pH is adjusted and then compound inhibitors, compound synergists and collectors are added for roughing to obtain roughing concentrate and roughing tailings; (6) Fine treatment: Add composite inhibitors to the roughing concentrate and perform multiple fine treatments to obtain fluorite concentrate and fine tailings; (7) After the roughing tailings are scavenged once, scavenged tailings I and scavenged tailings I are obtained. Scavenged tailings I and cleaned tailings are combined for a second scavenging to obtain scavenged tailings II and scavenged tailings II. Scavenged tailings II are returned to step (5) for repeated roughing and cleaning. Scavenged tailings I, scavenged tailings II, photoelectric tailings I and photoelectric tailings II are discarded.

[0008] Preferably, the wavelength range of the multispectral photoelectric sorting in step (2) is 300-2700nm; the AI ​​recognition adopts a deep learning algorithm, which dynamically optimizes the sorting threshold through the training set, with a recognition accuracy of ≥95% and a processing speed of ≥0.5 seconds / particle.

[0009] Preferably, the band of the multi-spectrum photoelectric sorting in step (3) is 200-2800 nm; and the band of the near-infrared spectrum recognition is 1300-2500 nm.

[0010] Preferably, the particle size of the floatation feed ore in step (4) is 80%-85% of ≤0.074 mm.

[0011] Preferably, the pH value of the ore slurry after pH adjustment in step (5) is 8-10; the dosage of the compounded inhibitor is 200-800 g / t, the dosage of the compounded synergist is 50-200 g / t, and the dosage of the collector is 200-600 g / t.

[0012] Further preferably, the pH adjustment agent used in the pH adjustment is sodium carbonate; and the collector is one of sodium oleate, linoleic acid and oleic acid.

[0013] Further preferably, the compounded inhibitor is compounded by the following components in weight parts: 40-60 parts of sodium lignosulfonate, 5-15 parts of fluosilicic acid, 20-30 parts of carboxymethyl cellulose, 15-25 parts of sodium silicate and 5-10 parts of sodium tripolyphosphate; and the compounded synergist is compounded by the following components in weight percentage: 30%-50% of trisodium citrate, 20%-40% of polyacrylamide, 10%-20% of sodium dodecylsulfate and 10%-20% of sodium silicate.

[0014] More preferably, the molecular weight of the polyacrylamide is 5-15 million.

[0015] Preferably, the dosage of the composite inhibitor in step (6) is 100-200 g / t; the number of cleaning is 4-5 times, and the pH value of the ore slurry during cleaning is 8-10.

[0016] Preferably, the composite inhibitor is compounded by the following components in weight parts: 40-60 parts of sodium lignosulfonate, 5-15 parts of fluosilicic acid, 20-30 parts of carboxymethyl cellulose, 15-25 parts of sodium silicate and 5-10 parts of sodium tripolyphosphate.

[0017] The present application has the following beneficial effects: 1. The combination of multi-spectrum photoelectric sorting and AI artificial intelligence recognition can effectively capture the subtle spectral differences of fluorite and calcite and other gangue minerals, because calcite (CaCO3) has a CO3 2-The characteristic absorption peak caused by group vibration, while fluorite (CaF2) has no such characteristic absorption in this waveband. The multi-spectral sorting uses a wide waveband of 300-2700 nm scanning, and based on the characteristic absorption peak of calcium carbonate near 2300 nm, through the particle size adaptive sorting strategy, the spectral analysis is focused on the coarse particle size of 8-50 mm, the texture recognition is focused on the medium particle size of 1-8 mm, the full particle size coverage is realized, and the high calcium fluorite particles (CaF2 grade high) and low calcium waste rock (mainly calcite) are accurately sorted out, realizing pre-concentration. The tailings control rate in multi-spectral photoelectric sorting is ≥40%, which greatly reduces the flotation throughput, and the photoelectric middlings are returned for reselection, realizing resource cascade recovery, and the total recovery rate is ≥85%.

[0018] 2. The complex inhibitor is obtained by compounding sodium lignosulfonate, fluorosilicic acid, carboxymethyl cellulose, sodium silicate and sodium tripolyphosphate. The F - and SiF6 2- ions dissociated from the fluorosilicic acid specifically react with the surface of calcite (CaCO3) to generate CaF2 or fluorosilicate-containing precipitates, which firmly cover the surface of calcite, making it hydrophilic and strongly inhibited, thereby significantly reducing the floatability of calcite, facilitating separation from fluorite and realizing high selectivity inhibition. The fluorosilicic acid and sodium silicate can react to generate fluorosilicate, further inhibiting the gangue mineral calcite. In addition, sodium lignosulfonate acts as a dispersant to prevent fine mud from covering the surface of useful minerals and has a weak inhibitory effect on calcite; carboxymethyl cellulose can enhance the adsorption strength and coverage of the inhibitor on the surface of the gangue mineral; sodium silicate is not only a dispersant but also can generate fluorosilicate with fluorosilicic acid to inhibit calcite; sodium tripolyphosphate can complex with the inevitable Ca 2+ and other ions in the ore pulp, reducing their activation effect on the gangue mineral.

[0019] 3. The complex synergist is obtained by compounding trisodium citrate, polyacrylamide, sodium dodecyl sulfonate and sodium silicate, which can selectively activate the surface of fluorite minerals and complex with calcium ions, thereby enhancing the adsorption selectivity and efficiency of the collector on the surface of fluorite, and optimizing the stability and fluidity of the foam. In addition, the complex inhibitor and the complex synergist synergize in the alkaline flotation environment; in addition, the fluorosilicic acid in the complex inhibitor and the sodium silicate generate fluorosilicate to selectively cover calcite; the trisodium citrate in the complex synergist complexes with the inevitable ions (Ca 2+ , Fe 3+ , Al 3+ , Mg 2+ ), polyacrylamide stabilizes the foam, and sodium dodecyl sulfonate enhances the diffusion of the collector, together expanding the floatability difference between fluorite and calcite, realizing deep separation, ensuring high recovery rate, and obtaining fluorite concentrate with extremely high grade, effectively solving the separation problem of high calcium fluorite ore due to low grade and similar gangue. BRIEF DESCRIPTION OF DRAWINGS Figure 1 Flowchart of the sorting process of the present application. DETAILED DESCRIPTION

[0020] The technical solutions of the present application will be further explained in combination with the drawings and specific embodiments. It should be noted that the following embodiments are only preferred embodiments of the present application and should not be construed as limiting the present application. The protection scope of the present application should be based on the content recorded in the claims. The modifications and replacements of the technical solutions of the present application made by those skilled in the art without creative labor fall within the protection scope of the present application.

[0021] The high-calcium fluorite ore in the following examples comes from a certain place in Hunan Province, with fluorite grade of 33.39% (expressed in terms of CaF2 content) and calcite grade of 50.13% (expressed in terms of CaCO3 content).

[0022] Example 1 (1) Crushing and screening: The high-calcium fluorite ore is crushed to a particle size of ≤60 mm, and then screened to obtain coarse-grained ore (55% of the total ore) with a particle size of 8-50 mm, medium-grained ore (30% of the total ore) with a particle size of 1-8 mm, and fine-grained ore (15% of the total ore) with a particle size of ≤1 mm; (2) Photoelectric sorting: The coarse-grained ore is fed into a multi-spectral photoelectric sorting machine, which uses multi-spectral scanning technology in the 300-2700 nm waveband and an AI recognition model based on deep learning to dynamically distinguish high-calcium fluorite particles from low-calcium fluorite particles, obtaining photoelectric concentrate I (CaF2 grade 45%, calcium carbonate 12%), photoelectric tailings I (CaF2 grade 7.12%, calcium carbonate 76.81%), and photoelectric middlings I (CaF2 grade 25.37%, calcium carbonate 42.35%). The photoelectric middlings I is returned to step (1) for recycling crushing and screening. The multi-spectral photoelectric sorting machine has a tailings rejection rate of 40%. The AI recognition model dynamically optimizes the sorting threshold through the training set, with an accuracy of 95% and a processing speed of 0.5 seconds per particle; (3) Gradient sorting: The medium-grained ore in step (1) is subjected to multi-spectral photoelectric sorting and near-infrared spectral recognition of surface texture features, obtaining photoelectric concentrate II and photoelectric tailings II. The multi-spectral photoelectric sorting has a waveband of 300-1700 nm and a tailings rejection rate of 35%. The near-infrared spectral recognition has a resolution of ≤10 nm; (4) Grinding: Collecting photoelectric concentrate I, photoelectric concentrate II, and fine-grained ore, using a high-pressure roller mill (working pressure: 5.0 N / mm 2 ) and a ball mill (grinding concentration: 75%) in combination to grind the ore to a particle size of ≤0.074 mm accounting for 80%, forming a flotation feed; (5) Roughing: the flotation feed ore is configured with water to obtain an ore slurry with a concentration of 30%, then sodium carbonate (concentration 10%) is added to adjust the pH to 8, then 200 g / t of a compounded inhibitor, 50 g / t of a compounded synergist and 200 g / t of sodium oleate are added as a collector to perform roughing to obtain a roughing concentrate; (6) Cleaning: the roughing concentrate is sequentially subjected to four cleaning, wherein 100 g / t of a compounded inhibitor is added each time, and the pH of the ore slurry is maintained at 8, to finally obtain a fluorite concentrate and a cleaning tailings; wherein the CaF2 grade of the fluorite concentrate is 98.28%, and the calcium carbonate content is 0.91%; (7) The roughing tailings are subjected to one-time scavenging to obtain a scavenging concentrate I and a scavenging tailings I, the scavenging concentrate I is combined with the cleaning tailings to perform two-time scavenging to obtain a scavenging concentrate II and a scavenging tailings II, the scavenging concentrate II is returned to step (5) for repeated roughing and cleaning, and the scavenging tailings I and II and the photoelectric tailings I and II are discarded; The compounded inhibitor is compounded by the following components in parts by weight: sodium lignosulfonate 40 parts, fluorosilicic acid 5 parts, carboxymethyl cellulose 20 parts, sodium silicate 15 parts, and sodium tripolyphosphate 5 parts. The compounded synergist is compounded by the following components in percentage by weight: trisodium citrate 30%, polyacrylamide (PAM, molecular weight 8 million) 20 parts, sodium dodecyl sulfonate (SDS) 10 parts, and sodium silicate 10 parts.

[0023] Example 2 (1) Crushing and screening: the high-calcium fluorite ore is crushed to a particle size of ≤60 mm, and then screened to obtain a coarse-grained ore with a particle size of 8-50 mm (accounting for 55% of the total ore), a medium-grained ore with a particle size of 1-8 mm (accounting for 30% of the total ore), and a fine-grained ore with a particle size of ≤1 mm (accounting for 15% of the total ore); (2) Photoelectric separation: the coarse-grained ore is fed into a multi-spectrum photoelectric separator, a multi-spectrum scanning technology with a wave band of 300-2700 nm is used, and an AI recognition model based on deep learning is combined to dynamically distinguish high-calcium fluorite particles and low-calcium fluorite particles, to obtain photoelectric concentrate I (CaF2 grade 45%, calcium carbonate 12%), photoelectric tailings I (CaF2 grade 7.28%, calcium carbonate 77.98%) and photoelectric middlings I (CaF2 grade 28.46%, calcium carbonate 38.57%), and the photoelectric middlings I is returned to step (1) for cyclic crushing and screening treatment; wherein the tailings discarding rate of the multi-spectrum photoelectric separator is 45%; wherein the AI recognition model dynamically optimizes the separation threshold through the training set, and the recognition accuracy is 97% and the processing speed is 1.0 second per particle; (3) Gradient separation: the medium-grained ore in step (1) is subjected to multi-spectral photoelectric separation and near-infrared spectrum identification of surface texture features to obtain photoelectric concentrate II and photoelectric tailings II; the wave band of multi-spectral photoelectric separation is 300-1700 nm, and the tailing throwing rate is 38%; the resolution of near-infrared spectrum identification is ≤10 nm; (4) Grinding: collect photoelectric concentrate I, photoelectric concentrate II and fine-grained ore, and use high-pressure roller mill (working pressure: 4.0 N / mm 2 ) and ball mill (grinding concentration: 65%) to jointly operate to grind the ore to a particle size of ≤0.074 mm accounting for 82%, forming a flotation feed; (5) Roughing: the flotation feed is added with water to obtain an ore slurry with a concentration of 32%, then sodium carbonate (concentration 10%) is added to adjust the pH to 9, then 400 g / t of a complex inhibitor, 100 g / t of a complex synergist and 400 g / t of sodium oleate are added as a collector to perform roughing to obtain a roughing concentrate; (6) Cleaning: the roughing concentrate is sequentially subjected to four times of cleaning, wherein 150 g / t of a complex inhibitor is added each time, and the pH of the ore slurry is maintained at 9, and finally a fluorite concentrate and a cleaning tailings are obtained, wherein the CaF2 grade of the fluorite concentrate is 98.52%, and the calcium carbonate content is 0.73%; (7) The roughing tailings are subjected to one-time scavenging to obtain a scavenging concentrate I and a scavenging tailings I, the scavenging concentrate I and the cleaning tailings are combined to perform two-time scavenging to obtain a scavenging concentrate II and a scavenging tailings II, the scavenging concentrate II is returned to step (5) to perform repeated roughing and cleaning, and the scavenging tailings I, the scavenging tailings II, the photoelectric tailings I and the photoelectric tailings II are discarded; The complex inhibitor is compounded by the following components in parts by weight: sodium lignosulfonate 50 parts, fluorosilicic acid 10 parts, carboxymethyl cellulose 25 parts, sodium silicate 20 parts, and sodium tripolyphosphate 8 parts; The complex synergist is compounded by the following components in percentage by weight: trisodium citrate 40%, polyacrylamide (PAM, molecular weight 100 million) 30%, sodium dodecyl sulfonate (SDS) 15%, and sodium silicate 15%.

[0024] Example 3 (1) Breaking and screening: take high-calcium fluorite ore and break it to a particle size of ≤60 mm, and then screen it to obtain coarse-grained ore (55% of the total amount of ore) with a particle size of 8-50 mm, medium-grained ore (30% of the total amount of ore) with a particle size of 1-8 mm, and fine-grained ore (15% of the total amount of ore) with a particle size of ≤1 mm; (2) Photoelectric sorting: take the coarse-grained ore into a multi-spectral photoelectric sorting machine, adopt multi-spectral scanning technology of 300-2700 nm wave band, combine with AI recognition model based on deep learning, dynamically distinguish high-calcium fluorite particles and low-calcium fluorite particles, and sort to obtain photoelectric concentrate I (CaF2 grade 45%, calcium carbonate 12%), photoelectric tailings I (CaF2 grade 6.18%, calcium carbonate 79.34%) and photoelectric middlings I (CaF2 grade 30.98%, calcium carbonate 35.86%), and the photoelectric middlings I returns to step (1) for cyclic crushing and screening treatment; the multi-spectral photoelectric sorting machine has a tailing throwing rate of 50%; the AI recognition model dynamically optimizes the sorting threshold through the training set, the recognition accuracy is 99%, and the processing speed is 2.0 seconds / particle; (3) Gradient sorting: multi-spectral photoelectric sorting and near-infrared spectral recognition of surface texture features are performed on the medium-grained ore in step (1) to obtain photoelectric concentrate II and photoelectric tailings II; the multi-spectral photoelectric sorting has a wave band of 300-1700 nm and a tailing throwing rate of 40%; the near-infrared spectral recognition has a resolution of ≤10 nm; (4) Grinding: collect the photoelectric concentrate I, photoelectric concentrate II and fine-grained ore, and adopt high-pressure roller mill (working pressure: 4.5 N / mm 2 ) and ball mill (grinding concentration: 70%) to jointly operate to grind to a particle size of ≤0.074 mm accounting for 85%, forming a flotation feed; (5) Roughing: the flotation feed is added with water to obtain an ore slurry with a concentration of 35%, then sodium carbonate (concentration 10%) is added to adjust the pH to 10, then 800 g / t of a complex inhibitor, 200 g / t of a complex synergist and 600 g / t of sodium oleate are added as a collector to perform roughing to obtain a roughing concentrate; (6) Cleaning: the roughing concentrate is sequentially cleaned four times, each time adding 200 g / t of a complex inhibitor and maintaining the pH of the ore slurry at 10, and finally obtaining a fluorite concentrate and a cleaning tailings, wherein the CaF2 grade of the fluorite concentrate is 98.81%, and the calcium carbonate content is 0.52%; (7) The roughing tailings are subjected to one-time scavenging to obtain scavenging middlings I and scavenging tailings I, the scavenging middlings I and the cleaning tailings are combined to perform two-time scavenging to obtain scavenging middlings II and scavenging tailings II, the scavenging middlings II returns to step (5) for repeated roughing and cleaning, and the scavenging tailings I, scavenging tailings II, photoelectric tailings I and photoelectric tailings II are discarded; The complex inhibitor is compounded by the following components in parts by weight: sodium lignosulfonate 60 parts, fluorosilicic acid 15 parts, carboxymethyl cellulose 30 parts, sodium silicate 25 parts, and sodium tripolyphosphate 10 parts; The complex synergist is compounded by the following weight percentage: 50% of trisodium citrate, 40% of polyacrylamide (PAM, molecular weight of 1200), 20% of sodium dodecyl sulfonate (SDS), and 20% of sodium silicate.

[0025] Comparative Example 1 The method and steps are the same as those in Example 1, except that step (2) is omitted, and the coarse-grained ore, fine-grained ore, and photoelectric concentrate II are mixed and separated to obtain a fluorite concentrate with a CaF2 grade of 92.09% and a calcium carbonate content of 3.51%.

[0026] Comparative Example 2 The method and steps are the same as those in Example 1, except that step (3) is omitted, and the photoelectric concentrate I, medium-grained ore, and fine-grained ore are mixed and separated to obtain a fluorite concentrate with a CaF2 grade of 90.18% and a calcium carbonate content of 4.03%.

[0027] Comparative Example 3 The method and steps are the same as those in Example 1, except that the complex synergist is not used in the roughing, and a fluorite concentrate with a CaF2 grade of 94.24% and a calcium carbonate content of 2.56% is obtained.

[0028] Comparative Example 4 The method and steps are the same as those in Example 1, except that the formula of the complex inhibitor in steps (5) and (6) is changed to not contain fluorosilicic acid, and a fluorite concentrate with a CaF2 grade of 88.11% and a calcium carbonate content of 5.02% is obtained.

[0029] Comparative Example 5 The method and steps are the same as those in Example 1, except that the amount of fluorosilicic acid in the complex inhibitor in steps (5) and (6) is changed to 20 parts, and a fluorite concentrate with a CaF2 grade of 89.09% and a calcium carbonate content of 4.82% is obtained.

[0030] Comparative Example 6 The method and steps are the same as those in Example 1, except that the 5 parts of fluorosilicic acid and 15 parts of sodium silicate in the complex inhibitor in steps (5) and (6) are replaced by 20 parts of sodium fluorosilicate, and a fluorite concentrate with a CaF2 grade of 87.24% and a calcium carbonate content of 5.27% is obtained.

[0031] Comparative Example 7 The method and steps are the same as those in Example 1, except that the formula of the complex inhibitor in steps (5) and (6) is changed to not contain sodium silicate, and a fluorite concentrate with a CaF2 grade of 86.35% and a calcium carbonate content of 5.52% is obtained.

[0032] Comparative Example 8 The method and steps are the same as those of Example 1, only the amount of sodium silicate in the compounded inhibitor of steps (5) and (6) is changed to 30 parts, and the fluorite concentrate is obtained by sorting, with a CaF2 grade of 88.13% and a calcium carbonate content of 4.95%.

[0033] Comparative Example 9 The method and steps are the same as those of Example 1, only the formula of the compounded synergist of step (5) is changed, with sodium silicate replacing an equal amount of trisodium citrate, and the fluorite concentrate is obtained by sorting, with a CaF2 grade of 91.34% and a calcium carbonate content of 3.02%.

[0034] Comparative Example 10 The method and steps are the same as those of Example 1, only the amount of trisodium citrate in the compounded synergist of step (5) is changed to 55%, and the amount of polyacrylamide is correspondingly reduced by 5% to 15%, and the fluorite concentrate is obtained by sorting, with a CaF2 grade of 90.02% and a calcium carbonate content of 3.22%.

[0035] Comparative Example 11 The method and steps are the same as those of Example 1, only the formula of the compounded synergist of step (5) is changed, with sodium silicate replacing sodium dodecyl sulfonate, and the fluorite concentrate is obtained by sorting, with a CaF2 grade of 92.47% and a calcium carbonate content of 2.84%.

[0036] Comparative Example 12 The method and steps are the same as those of Example 1, only the amount of sodium dodecyl sulfonate in the compounded synergist of step (5) is changed to 25%, and the amount of polyacrylamide is correspondingly reduced by 5% to 15%, and the fluorite concentrate is obtained by sorting, with a CaF2 grade of 91.56% and a calcium carbonate content of 3.15%.

[0037] Comparative Example 13 The method and steps are the same as those of Example 1, only the compounded inhibitor of step (5) is changed to water glass, and the fluorite concentrate is obtained by sorting, with a CaF2 grade of 85.27% and a calcium carbonate content of 6.03%.

[0038] Comparative Example 14 (1) Crushing and screening: the high-calcium fluorite ore is crushed to a particle size of ≤60 mm; (2) Grinding: the crushed ore is placed in a high-pressure roller mill (working pressure: 4.0-5.0 N / mm 2 ) and a ball mill (grinding concentration: 65-75%) for joint operation, and the ore is ground to a particle size of ≤0.074 mm accounting for 80%, forming a flotation feed; (3) Roughing: the flotation feed is configured with water to obtain a slurry with a concentration of 32%, then sodium carbonate (concentration 10%) is added to adjust the pH to 8, then 200 g / t of the inhibitor water glass and 200 g / t of sodium oleate as the collector are added, and roughing is carried out to obtain a roughing concentrate; (4) The rough concentrate is sequentially subjected to four times of cleaning, wherein 100 g / t of inhibitor water glass is added in each cleaning, and the pulp pH is maintained at 8, and finally the fluorite concentrate with CaF2 grade of 80.23% and calcium carbonate content of 8.50% is obtained.

[0039] Example 4 The fluorite concentrates prepared in the above examples and comparative examples are collected, and the yield (yield = mass of fluorite concentrate / mass of raw ore), CaF2 recovery rate (ε = (β × γ) / α × 100%, wherein α = CaF2 grade of raw ore (33.39%), β = CaF2 grade of concentrate (%), and γ = yield of concentrate (%)) are calculated. The results are shown in Table 1: Table 1: Data of separation effect

[0040] The results show that: by the combined process of photoelectric preselection and optimized flotation reagents, the high-quality fluorite concentrate with CaF2 grade higher than 98.00%, CaCO3 content lower than 1.00%, and recovery rate higher than 84.00% is successfully obtained in the examples 1-3 of the present application. The indicators of all comparative examples are worse than those of the examples of the present application to different degrees, which fully proves the key role and synergistic value of photoelectric preselection, complex inhibitor and complex synergist in the present process.

Claims

1. A high-calcium fluorite ore separation process combining photoelectric pre-selection and flotation, characterized in that: Includes the following steps: (1) Crushing and grading of raw ore: The high-calcium fluorite raw ore is crushed and screened to obtain coarse-grained ore with a particle size of 8-50 mm, medium-grained ore with a particle size of 1-8 mm and fine-grained ore with a particle size of ≤1 mm. (2) Photoelectric separation: Multispectral photoelectric separation and AI identification are performed on coarse-grained ore to obtain photoelectric concentrate I, photoelectric tailings I and photoelectric middlings I; (3) Gradient separation: The medium-sized ore is subjected to multispectral photoelectric separation and near-infrared spectral identification to obtain photoelectric concentrate II and photoelectric tailings II; (4) Grinding: Take photoelectric concentrate I and photoelectric concentrate II and fine-grained ore and grind them together to obtain flotation feed; (5) Roughing: After the flotation feed is prepared into a slurry, the pH is adjusted and then compound inhibitors, compound synergists and collectors are added for roughing to obtain roughing concentrate and roughing tailings; (6) Fine treatment: Add composite inhibitors to the roughing concentrate and perform multiple fine treatments to obtain fluorite concentrate and fine tailings; (7) After the roughing tailings are scavenged once, scavenged tailings I and scavenged tailings I are obtained. Scavenged tailings I and cleaned tailings are combined for a second scavenging to obtain scavenged tailings II and scavenged tailings II. Scavenged tailings II are returned to step (5) for repeated roughing and cleaning. Scavenged tailings I, scavenged tailings II, photoelectric tailings I and photoelectric tailings II are discarded.

2. The high-calcium fluorite ore separation process combining photoelectric pre-selection and flotation according to claim 1, characterized in that: The multispectral photoelectric sorting in step (2) has a wavelength range of 300-2700nm; the AI ​​recognition uses a deep learning algorithm, dynamically optimizes the sorting threshold through the training set, and has a recognition accuracy of ≥95% and a processing speed of ≥0.5 seconds / particle.

3. The high-calcium fluorite ore separation process combining photoelectric pre-selection and flotation according to claim 1, characterized in that: The multispectral photoelectric sorting band in step (3) is 300-2800nm; the near-infrared spectral identification band is 1300-2500nm.

4. The high-calcium fluorite ore separation process combining photoelectric pre-selection and flotation according to claim 1, characterized in that: The particle size of the flotation feed in step (4) is ≤0.074mm, accounting for 80%-85%.

5. The high-calcium fluorite ore separation process combining photoelectric pre-selection and flotation according to claim 1, characterized in that: The pH value of the slurry after pH adjustment in step (5) is 8-10; the dosage of the compound inhibitor is 200-800 g / t, the dosage of the compound synergist is 50-200 g / t, and the dosage of the collector is 200-600 g / t.

6. The high-calcium fluorite ore separation process combining photoelectric pre-selection and flotation according to claim 5, characterized in that: The pH adjustment is achieved using a pH adjuster, which is sodium carbonate; the collector is sodium oleate.

7. The high-calcium fluorite ore separation process combining photoelectric pre-selection and flotation according to claim 5, characterized in that: The compound inhibitor is composed of the following components in parts by weight: sodium lignosulfonate 40-60 parts, fluorosilicic acid 5-15 parts, carboxymethyl cellulose 20-30 parts, sodium silicate 15-25 parts, and sodium tripolyphosphate 5-10 parts; the compound synergist is composed of the following components in weight percentage: trisodium citrate 30%-50%, polyacrylamide 20%-40%, sodium dodecyl sulfonate 10%-20%, and sodium silicate 10%-20%.

8. The high-calcium fluorite ore separation process combining photoelectric pre-selection and flotation according to claim 7, characterized in that: The molecular weight of the polyacrylamide is 5 million to 15 million.

9. The high-calcium fluorite ore separation process combining photoelectric pre-selection and flotation according to claim 1, characterized in that: The dosage of the compound inhibitor in step (6) is 100-200 g / t; the number of refining operations is 4-5 times, and the pH value of the slurry during refining is 8-10.

10. The high-calcium fluorite ore separation process combining photoelectric pre-selection and flotation according to claim 9, characterized in that: The compound inhibitor is composed of the following components in parts by weight: 40-60 parts sodium lignosulfonate, 5-15 parts fluorosilicic acid, 20-30 parts carboxymethyl cellulose, 15-25 parts sodium silicate, and 5-10 parts sodium tripolyphosphate.

Images