The Effect of Particle Size and Dodecylamine Concentration on the Flotation of Lepidolite in Alkaline Medium

Abstract

Currently, lepidolite is considered an important natural alternative for obtaining lithium, given the difficulty in processing other species containing this metal. However, its mechanical preparation and beneficiation present considerable challenges and play a critical role in its efficient separation by flotation. This study explores the effect of particle size and dodecylamine concentration during flotation in a laboratory Denver cell. The results indicate that particle size significantly affects the finding in which the optimum was −90 + 75 μm, with a separation efficiency of 94%, and with only 2.067 × 10⁻⁵ M of dodecylamine (DDA) (5 g/t) at pH 11.0. The hydrophobicity of lepidolite was generated by the effect of the chemisorption of the cationic collector and the FTIR results indicate detection of the characteristic bands of the adsorption of DDA to the surface of lepidolite.

1. Introduction

Lithium (Li) is an essential metal for humans. It is widely used in all electronic devices that use Li⁺ ion batteries, and its demand is expected to continue increasing [1,2]. Currently, brines are the largest source of lithium extraction. A large percentage is found in countries such as Chile, Argentina, Australia, Canada, China, and Zimbabwe, which lead the extraction of lithium from these non-renewable sources. Therefore, obtaining this element and its compounds, such as LiCO₃, is assured [3]. However, it is necessary to consider that lithium production in 2024 increased by 18% and that this trend will continue in the following years [4,5], so alternative solid natural resources, such as the crystalline minerals described in the literature, must be sought and processed [6].

One of these lithium minerals is lepidolite, with the chemical formula K(Li,Al)₃(Al,Si)₄O₁₀(OH,F)₂ and with a specific gravity of 2.9. This is a mica mineral, where the elemental Li content is variable; it ranges from 1.39% to 3.58% (from 3.0 to 7.7% as Li₂O), has a hardness between 2 and 3 on the Mohs scale, and is much lower than other species such as spodumene, petalite, amblygonite, and eucryptite, which have a hardness of around 6.0. The gangue minerals that accompany lepidolite are mainly calcite, muscovite, feldspars, and quartz, which need to be separated by the most commonly used method for mineral concentration [6,7].

Flotation is the most widely used beneficiation process for concentrating lithium minerals contained in pegmatites; it is especially applicable to minerals of complex chemical nature or low grade. Furthermore, it should be noted that the difference in specific gravity between valuable species and gangue is too limited to require gravimetric separation [6,8]. For this reason, direct or reverse flotation has been used to concentrate lithium species. In the latter, sterile solid phases emerge in the froth generated during the flotation process using cationic collectors, while the minerals of interest remain in solution and represent the tailings [9,10].

It has been reported that the use of mixtures of anionic and cationic collectors, such as sodium oleate (NaOL) and dodecyl ammonium chloride (DTAC), at a respective molar ratio of 9:1, generates high selectivity for lithium ore flotation with respect to feldspars, indicating that NaOL adsorbs on the lithium-containing solid more than on the feldspar, while DTAC enhances this adsorption phenomenon [11].

The importance of lithium as a rare and precious metal has been previously described and is the result of its outstanding physical and chemical properties. It plays an essential role in various industries such as nuclear energy, pharmaceuticals, and lithium batteries [1,11]. This consideration has allowed the synthesis and testing of new collectors, such as hexyloxypropylamine (HPA) and N-dodecyliminodiacetic acid (DIDA). The authors indicate that a respective 1:3 molar mixture of HPA:DIDA produces a concentrate with 85.71% lithium mineral. It is indicated that HPA adsorbs to the surface through hydrogen bonds, while DIDA adsorbs by chemisorption [10].

Typical collectors used in lepidolite flotation described in the literature are, for example, stearyl trimethyl ammonium chloride (STAC), Aeromine 3000C, and Armeen 12D, generally employed at pH 3 and with dosages of 200, 350, and 500 g/t, respectively. The best selectivity of these collectors is achieved at acidic pH and with high collector ratios [12,13,14].

Lepidolite flotation is carried out at pH 2 using a STAC cation collector because the mineral surface exhibits stronger electrostatic interactions. Zeta potential analysis has determined that the point of zero charge, or isoelectric point, for lepidolite minerals is found at pH 2.5; that is, lepidolite particles below pH 2 exhibit a positive zeta potential, while at pH values above the isoelectric point, the lepidolite zeta potential increases to a negative value [12,13].

The highest lepidolite separation efficiency with the STAC collector is reported at 76.3%, while the Armeen 12D has limited potential due to its low water solubility. However, with a collector dosage of 500 g/t, lepidolite recovery is 91.51%; the concentrate has a lithium content of 1.96% [15]. The use of cationic collector mixtures in the flotation of the lepidolite contained in gravity concentration waste allows for a lithium mineral recovery of 70.37% and a concentrate grade of 4.12% lithium when compared with 40% recovery and a grade of 3.5% when using only an amine-type collector [16]. The lepidolite surface is depressed by the dosage of sodium sulfide Na₂S and starch, while sodium silicate and lithium sulfate can be used as surface activators [17].

The lepidolite surface is depressed by the dosage of sodium sulfide Na₂S and starch, while sodium silicate and lithium sulfate can be used as surface activators [17]. It has been previously established that lepidolite can be floated using cationic amine collectors. However, other parameters are required to obtain good flotation properties of lepidolite, such as those investigated in this research, such as particle size, collecting agent dosage, and particle size, among others. As a first step, flotation with single-phase minerals was carried out to determine their behavior in the presence of different collector dosages and particle sizes at alkaline pH 11.

It has previously been reported that efficient lepidolite flotation is also related to pulp density and grinding fineness. Therefore, to improve metallurgical results, lepidolite flotation should be performed with a low pulp density. Furthermore, decreasing particle size improves recovery. Primary amines, such as dodecylamine (DDA or DA), are used for lepidolite flotation. Using amine-based collectors, lepidolite exhibits good flotation properties over a wide pH range, from 2 to 11. Flotation of lepidolite with dodecylamine at pH 3 requires high DDA concentrations, for example, 350 g/t [18].

Once a collector that acts efficiently in flotation is identified, it leads to the optimization of the lithium mineral separation process, as in the work where fatty acid (FA) was used as a collector. Lithium recovery improves by conditioning a high-density pulp (high percentage of solids, 60%), pH, residence time, and collector dosage. In this way, it has been found that the use of 625 g/t of FA at an initial pH of 8.25 helps to obtain a concentrate with 90% recovery and >5.5% Li₂O grade [19].

Notwithstanding the relevant findings reported previously, most of the research on lithium mineral beneficiation has focused on the spodumene species contained in pegmatites and, though scarcely, on the solid crystalline phase of lepidolite [9,19], using expensive collectors not available on a large scale. In addition, factors such as particle size, pH, and collector concentration must be taken into account. Therefore, this research paper addresses the effect of dodecylamine collector dosage at very low concentrations, as well as the role of particle size in the flotation separation of lepidolite mineral at alkaline pH. The solids obtained in the concentrate were characterized by FTIR. These are novel topics that have not been addressed by other researchers.

2. Materials and Methods

2.1. Mineral, Reagents and Equipment

A lepidolite-type lithium mineral composed of a single piece of approximately 100 g with a physical appearance consisting of a majority purple phase was obtained from the state of Sonora, México. The crystalline solid was dry ground in a manufacture Lavallab Laval, Quebec Canada automatic agate mortar, model Pulverisette 2. All of the pulverized mineral was separated by particle size by wet sieving, using water and a series of Tyler brand sieves, using particle sizes of 37, 45, 53, 75 and 90 microns (μm).

The mineral’s elemental chemical composition was analyzed using laser-induced breakdown spectroscopy (LIBS) on a Keyence EA 300 VHX series instrument (KEYENCE CORPORATION OF AMERICA, Itasca, IL 60143, USA).

Table 1. Elemental chemical analysis (%) of lepidolite mineral.

Mineral	Elemental Composition (%)
	O	Si	F	Al	K	Li
Lepidolite	54.5	16.0	9.8	9.8	8.3	1.6

shows the average of 16 LIBS measurements.

For lepidolite mineral flotation tests, pine oil C₁₉H₁₇OH and 98% dodecylamine (DDA) C₁₂H₂₇N with low water solubility (0.018 g/L) were used as frother and cationic collector, respectively. A 2 M sodium hydroxide solution was used to regulate the alkaline pH of the pulp. Deionized water was used in all tests.

Prior to flotation tests, the mineral particles were analyzed using the instrumental technique of X-ray diffraction in INEL brand Equinox 2000 equipment, using an operating current and a voltage of 20 mA and 30 kV, respectively, from 5° to 90° with a step speed of 0.02 and a cobalt radiation source with a wavelength of 1.789010 Å and an analysis time of 15 min. The determination of the phases was carried out with the Match 2.1 software and a PDF2 database.

The morphology, particle size, and semiquantitative chemical composition of the lepidolite mineral solids were identified using a Thermo Scientific Phenom scanning particle XTC scanning electron microscope equipped with a silicon drift detector (SDD) used for the chemical determination of the elements comprising the sample. It should be noted that this technique does not allow the detection of Li and that the equipment is located at Gerdau Corsa S.A. in Sahagún, Hgo., México. The pulverized mineral was analyzed by scanning electron microscopy (SEM) to identify the morphology of the lepidolite particles and their semi-quantitative chemical composition.

The infrared spectra were recorded in a Perkin Elmer Spectrum GX spectrometer (Revvity Inc. Waltham, MA, USA), (400–4000 cm⁻¹; 2 cm⁻¹ resolution). Samples of 0.01 g were mixed with 0.1 g of binder (infrared grade KBr), compressed into pellets (hydraulic press), and immediately analyzed. The spectral band identification was carried out by comparison with a spectral database and with absorption bands reported in the literature.

During the flotation pulp conditioning stage, the pH and oxidation reduction potential (ORP) (mV) are monitored at each chemical modification using a Thermo Scientific Orion 3 Star potentiometer. To the measured value, +242 mV are added to present the data in reference to the standard hydrogen electrode (SHE) [20].

The weighing of all substances and the concentrate-receiving containers were carried out on an Ohaus precision analytical balance; calibrated micropipettes were used to measure volumes. Lepidolite flotation tests were performed on a laboratory flotation machine and a 1 L Denver brand stainless steel sub-aerated cell equipped with a revolutions per minute (RPM) meter. The equipment was operated at 1200 RPM. Air was introduced into the cell through a suction valve, driven by the movement of the impeller, located at the bottom of the cell along with the diffuser.

2.2. Experimental Procedure

Each lepidolite flotation test began with the conditioning stage, in which 1 L of distilled water and 4 g of mineral of −90 + 75 µm, −53 + 45 μm, and −37 μm were added to the cell in individual tests, evaluating C₁₂H₂₇N concentrations of 2.697, 5.395, and 8.092 × 10⁻⁵ M. Once the collector is in the flotation pulp, the pH is adjusted to 11.0 with 2 M sodium hydroxide, and, finally, the frother is added—2.29 × 10⁻⁴ M of pine oil—to give stability to the bubbles formed.

During conditioning and with each change in pulp chemistry, the pH and redox potential (mV) were measured and recorded, allowing a conditioning time of 5 min for each change. Once the conditioning stage was completed, the flotation test began, aerating the cell at the predetermined times of 0.5, 1, 2, 4, 6, 8, and 10 min. At each flotation stage, the foam generated in the cell was poured off using a rubber accessory, and the concentrate was received in a pre-weighed inert plastic tray; this was undertaken for each planned flotation time.

The water was allowed to evaporate from the wet concentrate at room temperature. The plastic container containing the floated mineral was then reweighed, and the exact amount of concentrated mineral in grams under each flotation test condition was obtained by calculating the difference in weight. These were converted to a percentage using Equation (1).

$$\% \mathrm{Flotation} \mathrm{Recovery} ={\displaystyle \frac{C{H}_M-{CH}_v}{C}} \times 100$$

(1)

where $C{H}_M$ is the weight of the tray or container containing the concentrated mineral at each preset flotation time. In total there were 7 time periods used to obtain the concentrate (0.5, 1, 2, 4, 6, 8, and 10 min). ${CH}_v$ is the weight of the empty container without mineral, which was measured in order to obtain the precise weight of the floated mineral, discarding the mineral losses that could occur if it were removed from the tray. C is the mineral head used—in this case 4 g per test—and 100 expresses the percentage of the floated mineral. In this way, the flotation results are graphed for each time period. The reason for using this small sample size is to evaluate the collector’s adsorption on a lepidolite sample with a single, predominantly crystalline phase using FTIR. This would not be possible if a mixture of minerals, as commonly found in nature, were used.

Each test was performed in duplicate, yielding similar results. Therefore, the average of the two tests was used to generate the corresponding graphs and perform the FTIR surface analysis. Error bars are also included for each of the flotation recovery graph points. The redox potential, and pH were graphed to show the behavior of each variable during conditioning and at the end of the flotation test. The potentiometer was calibrated before each test session. The powders obtained in the concentrate were characterized by Fourier transform infrared spectroscopy to determine the surface state of the particles obtained in the concentrate.

3. Results and Discussion

3.1. Mineral Characterization

Lepidolite mineral particles of −37 μm were characterized by X-ray diffraction, with this fraction being representative of the entire sample. Figure 1 shows the obtained spectrum; the characteristic peaks of this lithium phase are observed, identified the species with the International Centre for Diffraction Data (ICDD) database. The solid corresponds to lepidolite with a monoclinic crystal structure. X-ray diffraction (XRD) analysis shows that lepidolite is the only major crystalline phase. The characteristic peaks of the lithium phyllosilicate mineral correspond to the lepidolite components with the following formula: K(Li,Al)₃(Al,Si)₃O₃(OH,F)₂.

It should be noted that SEM does not detect lithium, though the contents of the associated elements can be determined. Figure 2 shows an SEM micrograph, representing a general view of the lepidolite particles, which have a flaky, scaly morphology [6]. This micrograph shows particles of around 45 microns, the largest in size; in addition, they have microparticles of lithium mineral adhered to them due to electrostatic attractions. It is even observed that the micrometric particles tend to separate into thin sheets of lepidolite and tend to orient themselves preferentially.

This predominant arrangement is with the narrowest surface facing upward, with even the smallest particles, less than 45 microns, exhibiting this behavior. This mica-like morphology is indicative and is a principal characteristic of the lithium mineral (lepidolite). The mineral particles shown in the scanning electron microscopy micrograph in Figure 2 have an irregular, smooth-edged morphology. They exhibit perfect cleavage, elastic toughness, and belong to a variety of the mica family [18].

Figure 3A shows the EDS spectrum, Figure (B) shows the SEM micrograph of the analyzed area, and Figure 3C shows the semi-quantitative chemical analysis; the latter shows the presence of the characteristic elements of lepidolite, such as oxygen with 45.2%. It is worth mentioning that silicon (19.8%), aluminum (12.3%), potassium (10.2%), and calcium (0.4%) in the crystalline structure of lepidolite are forming oxides, such as SiO₂, Al₂O₃, K₂O, and CaO, respectively.

The results, with regard to fluorine, correspond to those reported in the literature, indicating a proportion of 6.8% [18,21]. It is expected that the chemical compositions of the LIBS technique do not correspond to those of EDS because calcium and lithium are not considered in both.

Figure 4 shows the infrared (IR) spectrum of lepidolite particles with a particle size of −37 µm. It is worth mentioning that this is a fresh surface without any treatment; the IR absorption bands presented are at 1630 cm⁻¹, which corresponds with the asymmetric vibration of the O–H bonds of the OH⁻ ion that are the result of the absorption of water molecules on the surface of the lepidolite particles.

The signals at 440 cm⁻¹ and 1015 cm⁻¹ correspond to the Si–O bonds in silica. The lithium bonding band forming Li–O is assigned to the peak at 482 cm⁻¹. The values of 534 cm⁻¹ and 752 cm⁻¹ are assigned to the Al–O and Al–O–Si bonds, respectively and the signal at 802 cm⁻¹ is attributed to K–O bonds, which is consistent with findings in previous research [22,23,24,25,26].

3.2. Lepidolite Flotation with DDA: Effect of Particle Size

Figure 5 illustrates the relationship between lepidolite recovery and dodecylamine cation concentration (C₁₂H₂₅NH₃⁺), evaluating molarities of 2.697, 5.395, and 8.092 × 10⁻⁵ M for a particle size of −37 µm, a pH of 11.0, and a pulp potential of around +35 mV, which is a slightly oxidizing value that is close to reducing potentials, i.e., less than zero or negative.

The flotation of lepidolite particles at −37 µm improves with collector concentration and processing time; for example, when using 5.395 × 10⁻⁵ M (10 g/t of C₁₂H₂₅NH₃⁺), the maximum cumulative recovery % in 10 min of experimentation was 70.69%.

This relatively low separation efficiency is attributed to the partial hydrophobicity generated by the electrostatic interaction and hydrogen bonds described above [13,18]. This partial hydrophobicity exists between the mineral surface and the C₁₂H₂₅NH₃⁺ ion and is promoted by the morphology of the lepidolite-flake-like particles, which, despite their small size, maintain their micaceous exfoliation, which makes the interaction between the collector and the mineral surfaces difficult. This dual adsorption mechanism allows for the selective separation of lithium minerals present in alkaline or acid flotation pulp; dodecylamine is preferentially adsorbed on lepidolite. It has been reported in the literature that alkaline conditions favor the recovery of larger particles [1,6,13].

Increasing the dodecylamine concentration to 8.092 × 10⁻⁵ M (15 g/t C₁₂H₂₅NH₃⁺) leads to a decrease in lithium mineral recovery, as shown in Figure 5. Experimentally, it we find that the thickness of the froth bed on the surface of the flotation cell increases, that the bubble–particle aggregate that accumulates in the froth delays its exit, and that the effect of the movement of the pulp around the froth bed causes the release of the collected particles that are returned to the pulp. Additionally, we find the influence of a more hydrated froth, containing a lower proportion of particles. These observations agree with the findings of other investigations [1,18,21].

It is noted that the dissolution reaction of dodecylamine in water increases the pH of the pulp by forming OH⁻ ions, which is a result of the decomposition of an amino group, as indicated by reaction 1. For this reason, the flotation pH selected to carry out the flotation in this experiment was 11.0. However, it has been described in the literature that dodecylamine has the power to collect lepidolite in both alkaline and acidic media [1,6,8]. It is worth mentioning that DDA has limited solubility in water [27].

C₁₂H₂₅NH₂ + H₂O = C₁₂H₂₅NH₃⁺ + OH⁻

(2)

For example, lepidolite flotation at acid pH (3.0) with a combination of the Gemini surfactant hexanediyl-α,ω-bis (Dimethyldodecylammonium bromide) (HBDB) and dodecylamine is reported in the literature, with the authors reporting optimum concentrations of 150 g/ton and 300 g/ton, respectively [28]. In contrast, in the present work, particles of −37 µm, pH 11.0 and 5.395 × 10⁻⁵ M (10 g/t) are used, that is, very dilute concentrations of DDA and a lepidolite flotation of 70% is obtained, as observed in Figure 5.

Due to the physicochemical characteristics of the lepidolite mineral, its surface is hydrophilic in nature; thus, its floatability is limited. Further separation of this phase will occur by inverting its surface conditions to a hydrophobic surface, as described in the literature [21]. Laboratory tests showed that the separation of a single phase of lepidolite mineral depends not only on the pH and concentration of the collector but also on the particle size.

For this reason, the effect of lepidolite particle size was evaluated. Figure 6 shows the curves of % cumulative flotation as a function of processing time for particles −53 + 45 µm For example, using 2.697 and 5.395 × 10⁻⁵ M of DDA, proportions of 59.8 and 54.5%, respectively, are achieved. This slightly larger particle size allows for higher recoveries compared with lepidolite solids of −37 μm, with these efficiencies obtained after 10 min of flotation. This occurs due to the weak adsorption of DDA to lepidolite micas, associated with the scarcity of active sites that facilitate this physicochemical phenomenon, as previously described [21,23].

With the increase of the DDA concentration in the flotation pulp to 8.092 × 10⁻⁵ M (15 g/t), a pH 11, and with particles −53 + 45 µm, the recovery is significant, though it barely exceeds 71%. With regard to this, we have previous findings reported in the literature that the surface of lepidolite in alkaline pulps >10 has a more negative zeta potential, meaning that the increase of electrostatic interactions and hydrogen bonds between the cationic collector and the aluminosilicate structure of the lithium mineral would be expected [13,23]; however, for this particle size, the laminar morphology causes the interactions to remain in low proportion and the hydrophobicity is limited meaning that the recovery, in turn, remains at this average.

During the conditioning stage of the lepidolite flotation test with DDA, the pH and the oxidation-reduction potential (ORP) mV are monitored and referred to as the standard hydrogen electrode (SHE), which is obtained by adding +242 mV to the measured value [29]. Figure 7 shows the obtained behavior. It is observed that the addition of the mineral to the aqueous medium increases the pH. This is due to previous findings related to the oxidation of the potassium (K) contained in the surface of the lepidolite to potassium ion K⁺, forming aqueous potassium hydroxide KOH, as shown in reaction 2. This leaves the mineral surface with a negative character, where O₂²⁻ is the dissolved oxygen in the flotation pulp.

2K + 2H₂O + O₂²⁻ = 2KOH_(a) + 2OH⁻

(3)

Along with this, the active sites on the surface of the lithium mineral, such as silicon (Si) and oxygen (O), interact with hydroxyl groups formed in reaction 2, binding to the surface, and this contributes to the negative character measured by the zeta potential at alkaline pH, as described by previous authors [13,23].

From Figure 7 it can be observed that, after the addition of the mineral to the pulp and after five minutes of conditioning, the pH increases; subsequently, the pulp is brought to the flotation pH (11.0) with 2 M sodium hydroxide solution. The pulp potential (PP) at the beginning of the test is slightly oxidizing in nature, with a value of +35 mV and with a greater tendency to reduce. At the end of the flotation experiment, the PP has a slight increase. Flotation of lepidolite with dodecylamine at an alkaline pH is carried out at potentials with minimal oxidizing conditions.

It is worth mentioning that, in all of the flotation tests carried out, a similar behavior of the pH and the pulp potential is maintained under conditions of particle size −37 µm and −53 + 45 µm, with 8.092 × 10⁻⁵ M of DDA, a pH 11.0, and a slightly oxidizing potential. The adsorption of DDA remains weak due to the effect of morphology and size, where the bubble–particle aggregate formed in the foam bed breaks before leaving the cell and the collected particles return to the pulp. As a result, the recovery does not increase from approximately 70%.

Figure 8 shows the results obtained in the flotation of lepidolite particles −90 + 75 μm at pH 11.0 and testing concentrations of 2.697, 5.395, and 8.092 × 10⁻⁵ M of DDA (5, 10, and 15 g/t, respectively). A clear difference is observed when floating smaller particle sizes, where, for coarser particles, flotation is greatly favored. The morphology of the particles and their type of fracture, in the form of mica, positively influence the success of flotation for this particle size, even when the pulp is only present with 5 g/t of DDA, and, in the first 30 s of start-up, it is concentrated around 75%. The wide difference in reagent consumption of 5 g/t compared with that reported in the literature, where high collector concentrations are required, represents an important advantage at the industrial level of the process due to the high costs of flotation reagents.

The cationic DDA collector adsorbed to the mineral surface reverses the properties from hydrophilic to hydrophobic, and the particles adhere to the bubbles, concentrating them relative to their aqueous medium. These results indicate the positive influence of performing flotation with particles larger than 75 μm, and are consistent with the findings of previous research, which indicates that larger lepidolite grains are easier to concentrate [9,13,27].

These characteristics highlight the importance of establishing specific procedures for each mineral and particle size, ensuring efficient and consistent separation of the desired species. Therefore, the degree of liberation of minerals is a factor. Proper enrichment cannot be achieved for ore that is not crushed/ground to the liberated size. In addition, it must be considered that the grinding must be undertaken dry, using hammer mills. Production of lithium-bearing mineral concentrates such as lepidolite are closely related, firstly, to particle size, followed by collector concentration and flotation time.

Figure 9 shows the IR spectrum of untreated lepidolite particles of −53 + 45 μm as well as of the solids obtained in the presence of 8.092 × 10⁻⁵ M DDA, 2.29 × 10⁻⁴ M pine oil frother agent, and spectrum pure reagent grade 99% Li₂CO₃. This last spectrum serves as a reference for the signals detected in the IR spectra of lepidolite concentrated by flotation.

The IR spectrum of untreated lepidolite shows at 1015 cm⁻¹, 440 cm⁻¹, and at a lower intensity at 696 cm⁻¹, with the peaks corresponding to the asymmetric stretching vibration of the Si–O tetrahedron. Meanwhile, the signal at 752 cm⁻¹ is associated with the vibration of the Al–O bond corresponding to the octahedron of the lepidolite crystal structure. The bonding band at 534 cm⁻¹ corresponds to the asymmetric vibration of Al–O–Si, as reported in the literature [30,31,32]. This spectrum also presents the characteristic peaks of the K–O and Li–O bonds at the vibration positions of 802 cm⁻¹ and 482 cm⁻¹, respectively. It is worth mentioning that only in fresh untreated lepidolite was water adsorption of the O–H bonds of OH⁻ detected with the bending vibration at 1630 cm⁻¹. To analyze the behavior of the Li–O bond of lepidolite in the presence of DDA during flotation tests, the IR spectrum of reagent grade 99% Li₂CO₃ was obtained.

Figure 9 shows the characteristic peaks of lithium carbonate. Here, the band at 863 cm⁻¹ of the C–O bond bending vibrations can be observed, as well as at 1464 cm⁻¹ and with an adsorption shoulder at 1487 cm⁻¹, corresponding to the C–O asymmetric stretching vibrations. The band at 1087 cm⁻¹ indicates the symmetric stretching vibration, while the band at 1434 cm⁻¹ is assigned to the C–O asymmetric stretching, as described in the literature [30,31,32].

The previously described absorption bands in the IR spectrum of Li₂CO₃ correspond to internal vibrations of the isolated anion. The double degeneracy of the normal modes v₃ (asymmetric stretching vibrations) and v₄ (in-plane bending vibrations) is eliminated due to the decrease in the symmetry of lithium carbonate in the crystal lattice relative to the symmetry of the ideal CO₃²⁻ anion, as previously described [30]. The double band corresponding to the asymmetric stretching vibrations ν₃ is observed at 1434 and 1464 cm⁻¹. A band at 743 cm⁻¹ can also be observed from the in-plane bending vibrations. In addition, the stretching vibration ν₁ is activated, giving rise to a small band at 1087 cm⁻¹. The bending vibration ν₂ gives a sharp band at 863 cm⁻¹. The bands observed in the IR spectra at 495 cm⁻¹ and 419 cm⁻¹ can be attributed to vibrations of the Li₂CO₃ quasicrystal reported in the literature [30].

After conditioning and flotation of the −90 + 75 µm lepidolite particles and after drying the sample at room temperature, the powders were analyzed via FTIR, with the aim of detecting the characteristic adsorption bands of the collector; however, none of the floated products with the three different DDA concentrations for this particle size presented the characteristic peaks of dodecylamine-type collector adsorption, instead presenting only the peaks corresponding to lepidolite, indicating that the adsorption of DDA on the mineral surface was minimal. A similar situation has been reported previously [21,23].

Only in the IR spectra of the −53 + 45 µm and −37 µm lepidolite particles, concentrated by flotation, was DDA collector absorption detected. Figure 9 shows the IR of lepidolite particles #325 floated in the presence of 8.092 × 10⁻⁵ M (15 g/t) DDA. The characteristic bonding bands of lepidolite are observed in the IR spectrum at 419 cm⁻¹, 692 cm⁻¹, and 1015 cm⁻¹, corresponding to the Si–O tetrahedral bond vibration bands described in the literature [31,32]. Furthermore, the vibrations of the K–O bonds at 805 cm⁻¹ of the Al–O bond are presented. Previously, it has been mentioned that the collector adsorption occurs primarily on the Al/Si–O sites of the lepidolite surface, making it hydrophobic [31,32,33], a condition favored by the pH modifier (NaOH), which is 11.0. It is worth mentioning that the addition of a frother agent after the addition of DDA was intended to stabilize the bubbles at the water–air interface and facilitate the transport of the particles to the concentrate.

It can be seen from Figure 9 that the band assigned to the Li–O bond vibration in this spectrum was not detected due to collector adsorption. As a first approach, the mechanism by which the Li–O bond signal does not appear in the IR spectrum of floated lepidolite in the presence of 8.092 × 10⁻⁵ M of DDA is the result of the collector preferentially adsorbing to the lithium of the Li–O bond, forming an adsorption layer on the rare metal. This covalent bond is covered and not detected during the FTIR analysis, something which may also explain the reduction and elaboration of the oxide in such a way that the molecular vibrations are not detected in IR. This disappearance of the Li–O bond is attributed to the interaction of DDA with lithium due to electrostatic attractions. Similar behavior has been previously reported in the literature [31,32].

The intensity of the signal associated with the Al–O–Si bond, located at 547 cm⁻¹, diminishes. On the other hand, in the IR spectrum of lepidolite floated in the presence of 8.092 × 10⁻⁵ M DDA, the Al–O–Si bond is detected at 534 cm⁻¹, so the mechanism for the reduced intensity of the Al–O–Si bond is not linked to the competitive effect of DDA adsorption sites (Al/Si–O). The cationic amines, such as the dodecylamine (DDA), are then introduced to selectively adsorb onto the negatively charged lepidolite and form a negative zeta potential, as described in the literature. The mechanism is carried out by electrostatic interactions and hydrogen bonds, improving its hydrophobicity [23,29].

The faint bands at 867 cm⁻¹, 1143 cm⁻¹, and 1412 cm⁻¹ represent the C–O bonds; on the other hand, the signals at 1093 cm⁻¹ and 1458 cm⁻¹ represent C–O–C and C=O, respectively. This indicates the adsorption of DDA to the mineral surface. These adsorption bands have been described previously [30,32]. In summary, the FTIR results show that the surface hydrophobicity generated by DDA adsorption to lepidolite particles alone does not guarantee mineral flotation; furthermore, it requires an optimal particle size to ensure maximum lithium mineral recovery.

4. Conclusions

This study investigated the activation mechanism of DDA in lepidolite flotation, analyzing the impact of cationic collector concentration and particle size at pH 11.0 on separation performance using flotation tests in a laboratory Denver cell. The characterization by EDBS, XRD, SEM, FTIR confirm that the mineral used for the study of the effect of particle size and dodecylamine concentration in the flotation of lepidolite in an alkaline medium corresponds to a lithium species of a single crystalline majority phase of lepidolite determined by XRD, with a lithium percentage of 1.6%. The characteristic elements contained in this species detected by LIBS are O, Si, F, Al and K, with percentages of 54.5, 16.0, 9.8, 9.8 and 8.3 respectively. The morphology of this mineral consists of fine particles, with a morphology of flakes or narrow-edged micas and a large surface area typical of a lepidolite mineral. FTIR analysis of the fresh untreated mineral presents the characteristic positions of bonds present in lepidolite minerals, such as Li–O, Al–O, Al–O–Si, K–O and Si–O, at 482 cm⁻¹, 752 cm⁻¹, 534 cm⁻¹, 802 cm⁻¹ and 440 cm⁻¹, 1015 cm⁻¹ respectively.

Particle size is a key factor for the success of lepidolite flotation. The best efficiencies are obtained when using sections of −90 + 75 µm and, for all used concentrations of DDA—2.697, 5.395 and 8.092 × 10⁻⁵ M of dodecylamine (DDA)—the flotation recovery of lepidolite is greater than 90%, with the separation reaching 75% even after only 30 s of flotation. Particle sizes of −53 + 45 µm and −37 µm negatively affect the flotation process and the recovery decreases to values of around 31%.

FTIR analysis of the mineral floated in the presence of DDA generated superficial changes in the mineral due to the presence of C–O and C=O bonds located at the vibrational positions of 867 cm⁻¹, 1093 cm⁻¹, 1143 cm⁻¹ and 1458 cm⁻¹, respectively, indicating the chemical adsorption and oxidation of the carbon bonds of the DDA, which changes the properties of lepidolite from hydrophilic to hydrophobic and makes its flotation possible. Regarding the Li–O bond, it does not appear in the floated mineral due to the effect of the electrostatic attraction between lithium and the dodecylamine collector, these being the main resulting mechanisms during the flotation of lepidolite with DDA.

The low consumption of collector reagent and the use of coarse grinding necessary to achieve the highest flotation efficiencies found in this research work contribute to the current state of knowledge on the separation of this phase and bring about a reduction in costs in reagent consumption and over-grinding at the industrial scale.