This paper addresses Dissolved Air Flotation (DAF) process variables, such as the flocculation parameters and the recycle water addition, as well as the pretreatment chemical variables (coagulation conditions), to determine the optimal values for the flotation of iron ore slimes found in a highly turbid water sample from the Gualaxo do Norte River, a tributary of the Doce River Basin in Minas Gerais, Brazil. This work was conducted using a flotatest batch laboratory-scale device to evaluate the effectiveness of DAF for cleaning the water polluted by the Samarco tailings dam leakage and determine the ability of DAF to reduce the water turbidity from 358 NTU to values below 100 NTU, aiming to comply with current legislation. The results showed that the four types of tested coagulants (PAC, ferric chloride, Tanfloc SG and Tanfloc SL) provided adequate conditions for coagulation, flocculation and flotation (in the range of 90-99.6% turbidity reduction). Although the process variables were optimized and low residual turbidity vales were achieved, results revealed that a portion of the flocs settled at the bottom of the flotatest columns, which indicated that the turbidity results represented removal caused by a combination of flotation and sedimentation processes simultaneously.

No abstract

No abstract

Beneficiation of manganese ores has been conducted around the world by circuits composed basically of crushing, screening and spiral classifier. Therefore, solely the coarse size fractions (and commonly the richest) are actually recovered: "lump" (-75+9mm) and "sinter feed" (-9+0.15mm). In the industrial plant which processes the ore from the Azul Mine-PA, the overflow of the spiral classifier (-0.15mm) is disposed of in a tailings dam because it bears a low Mn content (<10%) and high kaolinite content (70%). This paper reports the findings of a process development conducted on laboratory scale with the aim of concentrating Mn-bearing minerals from the tailings of the Azul Mine-PA (7% of Mn and 70% of kaolinite). The process is composed of desliming (-10μm), followed by reverse cationic flotation of kaolinite (rougher, scavenger) at pH~5. Results indicate that the reverse cationic flotation in the acid medium, when utilizing amide amine as collector and in the presence of a silicate activator and a dispersing agent could be a possible route for the concentration of tailings that had previously been deslimed at 10μm.The results yielded a mass recovery of 18%, a metallurgical recovery of 50% and a 32% Mn concentrate which may be mixed with current high grade products, increasing the overall recovery of the plant.

No abstract

The concern about enhancing the productivity of Salitre phosphate mines has led to an extensive research and development program on new reagents, aimed at the sustainable processing of the ore, with greater production of phosphate concentrate and, consequently, less waste disposal in the tailings dam. Flotation is the most widely applied technique for processing of phosphate rocks, using mainly fatty acids and their salts as collectors. In partnership with BASF Mining Solutions, a new collector system was developed based on soy oil fatty acid (SOFA) and BASF’s co-collector Lupromin® FP A 1210 Base (L.1210, a mix of synthetic surfactants). The laboratory scale flotation results showed that most effective performance was achieved with a ratio SOFA:L.1210 of 70:30, while in the industrial application the ideal proportion was 90:10. This difference in reagent ratio is directly related to the surface tension of the bulk, making the apatite more or less hydrophobic, directly affecting the flotation performance. These proportions were tested by comparing different reagent conditions and pH, while contact angle and surface tension studies provided a theoretical framework to indicate a more effective collector adsorption, allowing for further process optimization. Applying 10% of L.1210 on an industrial scale, a more stable industrial process was observed to the oscillations of the processing plant, corroborating that by using the collector promoter there is a significant performance increase in this circuit of the Serra do Salitre Complex, which increased, on an industrial scale, at least 7.1% in P2O5 recovery and reduced 12.5% of the collector dosage compared to the use of only fatty acid. In this study, it was possible to identify two main process conditions to further optimize the current reagent system applied in industrial scale: pH 9.5 with the mixture SOFA/L.1210 90/10 (θ = 75.8°); and pH 7.5 with SOFA/L.1210 70/30 (θ = 76.8°).

Esta tese busca contribuir para a integração de aspectos macro e micro da cinética de flotação de silicatos, no contexto de dois circuitos industriais brasileiros: Cajati e Brucutu. A cinética do processo de flotação sofre influência de variáveis que atuam em micro escala (tempo de indução, ângulo de contato, tensão superficial, tamanho de bolhas) e em macro escala, como oferta de ar (surface bubble flux -Sb) e qualidade da mistura. Aspectos macro e micro do processo de flotação se integram de modo sinérgico no modelo cinético de primeira ordem, onde a recuperação metalúrgica (R) depende do tempo de residência (t) da polpa no interior da célula de flotação e também da constante cinética (k), que é função da oferta de ar (Sb) e da Eficiência de Coleta (Ek) de partículas minerais por bolhas de ar. No circuito industrial de Brucutu, o mineral que se deseja flotar (quartzo) apresentou elevado ângulo de contato (ϴ=57°), baixo tempo de indução (σi=15ms) e adequada tensão superficial (γ=51mN/m > tensão crítica de molhabilidade do quartzo, γc), indicando alta probabilidade de coleta (Ek). Já no circuito industrial de Cajati, a polpa exibe tensão superficial (γ=28mN/m) inferior à tensão superficial crítica de molhabilidade dos silicatos (γc = 35mN/m), destacando-se o mineral flogopita que exibe baixo ângulo de contato (ϴ=12°) e σi > 6.10 5 ms, o que aponta para uma baixa Ek. O circuito de Brucutu opera com adequada dispersão de gás, exibindo na primeira célula rougher (Wemco#164) Jg = 1,0cm/s, D3,2 = 1,5mm, εg = 14,4% e Sb = 40s -1 . O circuito de Cajati, sob condições de N = 130min -1 , opera com inadequada dispersão de gás, apresentando Jg = 0,7cm/s, D3,2 = 1,6mm, εg = 8,0% e Sb = 26s -1 na primeira célula (Wemco#190) do banco de flotação. A análise de distribuição de tempos de residência (DTR) para 1ª célula de cada banco de flotação resultou em: a) Brucutu: tempo médio (tm) de 4,6min e número de tanques ideais n = 1,2; enquanto que sua polpa passa pelo impelidor 4,6 vezes (nto); b) Cajati (sob uma rotação de N = 130min -1 ): tm = 4,2min, n = 2,4 e nto = 1,7, indicando menor probabilidade de coleta de partículas por bolhas de ar. Os resultados experimentais de recuperação metalúrgica em função do tempo mostraram adequada aderência aos modelos clássicos de cinética de 1ª ordem, tanto em escala industrial quanto em laboratório. A constante cinética (k) do circuito de Brucutu (0,10min -1 ) é 2,5 vezes a do circuito de Cajati (0,04min -1 ). Em relação ao tempo de residência médio total, em Brucutu tem-se 37,6min de flotação, enquanto Cajati observou-se apenas 16,8min. Utilizando-se valores experimentais de k e de Sb, foi possível estimar Ek de Brucutu como sendo 72% maior que Cajati. O sinergismo entre tm, Sb e Ek resulta numa recuperação de SiO2 em Cajati de 35,5% versus 99,2% em Brucutu. O desempenho inferior de Cajati se deve à flotação ocorrer em menor tempo, em um ambiente com menor oferta de ar e cinética mais lenta.