Lithium extraction from α-spodumene by low-temperature fluorination with NH4HF2

A.C.J. RESENTERA; G.D.ROSALES; M.R. ESQUIVEL; M.H.RODRIGUEZ

Atamacama Chile

Congreso; 7 th IWLiMe 2020 International Workshop on Lithium Industrial Minerals and Energy; 2020

Universidad de Atacama

Currently, lithium isconsidered a "critical energy element" worldwide, mainly due to itsapplications in energy storage technologies. [1] The "security of lithiumsupply" is a top priority for technology companies due to the continuedincrease in the consumption of mobile electronic devices, the demand for hybridand electric vehicles, and the policies of adoption of energy storage systemsin different countries of the world. [2] Furthermore, according to the growingworld demand for lithium, world production is expected to be insufficient inthe next years. This situation has competitively positioned lithium-bearingmineral sources in the market. [2]α-spodumene (4-8.03%Li2O) is the main mineral considered for the extraction of Li. Theonly industrial scale? Li extraction process is digestion with concentratedsulfuric acid. [2] The mineral is calcined at 1100°C to generate a phase change(β-spodumene) and digested with concentrated H2SO4 at250°C. The main disadvantages of the process include the high consumption ofenergy required and that only 5% of the mineral is used, generatingenvironmental liabilities. [2]This work describes anovel method for Li extraction from α-spodumene by thermal treatment with NH4HF2at low-temperature. The thermal analysis of the reactive mixture was done analyzedby thermogravimetry and differential thermal analysis (TG-DTA). The operationalparameters of the thermal treatment (temperature, α-spodumene/NH4HF2molar ratio, and reaction time) are analyzed and modeled using artificialneural networks (ANN) to maximize Li extraction.The α-spodumene (7.54%Li2O), from the province of Catamarca Argentine, was mixed with NH4HF2(commercial grade, ≥98%) in different molar ratios (1:10.5 to 1:21,respectively). The samples were thermal-treated in an oven equipped with a gasrecovery system at 2°C/min between 100 and 170°C for reaction times of 60 to120 min. The solids obtained were leached first with water, to eliminate theunreacted NH4HF2 and the ammonium fluorosilicatesgenerated, and secondly with 100 mL of H2SO4 at 10% v/v(Alkemit) to dissolve the extracted Li (Rosales et al., 2019). The solids weredried and characterized by XRD, and Li was determined in the liquors by flamephotometry.Figure 1 shows theresults of the TG-DTA analysis. The curves indicate six apparent endothermicstages with associated mass losses during the process. The first event (124°C)agrees with the NH4HF2 fusion. The second peak appears at145°C as a shoulder and it is associated with the reaction between α-spodumeneand NH4HF2 (reaction (1)). [2] The third peak (152°C) ismainly due to the maximum decomposition of NH4HF2. Theevents observed at 158, 212, and 227°C correspond to the decomposition of theproducts of Si and Al, obtained in reaction (1). The mineral fluorination andthe NH4HF2 decomposition occur simultaneously in alimited range of temperatures, between 100 and 170 ° C, approximately. [2]Figure 2 presents theresults of the Li extraction analyzed ANN. The data training results showed agood fit, with values of R2 = 0.99801, SD = 1.4701 and CV% = 1.9747.According to themodel, the temperature has the greatest influence on Li extraction because itcontrols the rate of fluorination reaction and NH4HF2 decomposition.The amount of NH4HF2 and time also have a positiveinfluence on the process. This is since they improve the contact surface,greater diffusion of NH4HF2 in a liquid-state through themineral structure, and an increase in the interaction time.   Li extraction valuesof 99% are obtained with a molar ratio of α-spodumene:NH4HF2of 1:17.5 at 155°C for 120 min. Also, the water leach liquor can be evaporatedto obtain (NH4)3SiF6·F as by-products. Thesecan be used directly in the synthesis of mesoporous zeolites or amorphoussilica. Keywords:Lithium,spodumene, ammonium bifluoride, Artificial neural networks References: [1] B. Jaskula, 2016Minerals yearbook: Lithium (advance release). USGS, (2018), URL https://minerals.usgs.gov/minerals/pubs/commodity/lithium/myb1-2016-lithi.pdf[2] A. Resentera, G.Rosales, M. Esquivel, M. Rodriguez, Thermal and structural analysis of thereaction pathways of α-spodumene with NH4HF2. Thermochim. Acta 689 (2020) https://doi.org/10.1016/j.tca.2020.178609