Ultrasonic Pulp Conditioning-Induced Nanoparticles: A Critical Driver for Sonication-Assisted Ultrafine Smithsonite Flotation

Abstract

Extensive studies have established that ultrasonic micro-jets and acoustic cavitation selectively intensify interfacial interactions at multiphase boundaries, thereby enhancing the flotation of soluble salt minerals and oxide ores. Although a growing body of evidence shows that pulp-borne nanoparticles (i.e., nanosolids, colloids, and nanoscale gas nuclei) mediate these effects, their role in the flotation of ultrafine smithsonite after collector addition has not yet been systematically examined. To fill this gap, we compared the flotation response of ultrafine smithsonite under conventional stirring (SC) and ultrasonic conditioning (UC), using sodium oleate (NaOL) as the collector, and dissected the governing mechanisms across three pillars, mineral–NaOL interaction, particle aggregation, and frothability, with particular attention paid to how nanoparticles modulate each dimension. The flotation results show that flotation performance under UC is dictated by NaOL concentration. At low NaOL levels (i.e., below 4 × 10⁻⁴ M), UC depresses both recovery and kinetics relative to SC, while at high NaOL levels, the trend reverses and UC outperforms SC. Mechanistic analysis reveals that sonication erodes mineral surfaces and generates cavitation, flooding the pulp with various nanoparticles. When NaOL is scarce, zinc-containing components and zinc-rich nanosolids sequester the collector through non-selective adsorption and precipitation, leaving smithsonite poorly hydrophobized. Consequently, particle aggregation and pulp frothability are markedly inferior to those in the SC system, so the flotation recovery and kinetics remain lower. As the NaOL concentration rises, smithsonite becomes adequately hydrophobized, and the pulp fills with hydrophobic zinc-rich nanosolids, along with cavitation-induced gas nuclei or tiny bubbles. These nanoparticles now act as bridges, accelerating the aggregation of ultrafine smithsonite once sonication stops and agitation begins, while simultaneously improving frothability. Although the strong dispersive action of ultrasound still suppresses initial flotation kinetics, cumulative recovery ultimately surpasses that of SC. The findings delineate a nanoparticle-regulated flotation paradigm and establish a critical NaOL concentration window for effective UC in ultrafine smithsonite flotation. This framework is readily transferable to the beneficiation of other ultrafine, soluble oxidized minerals (rhodochrosite, dolomite, etc.).

1. Introduction

Zinc, the fourth most consumed metal worldwide, is indispensable for galvanizing, brass and zinc alloy production, battery and chemical manufacturing, and emerging high-tech applications such as energy-storage electrodes and pharmaceutical additives [1]. Currently, over 85% of primary zinc is recovered through froth flotation, a process that is highly efficient for sulfide ores [2]. However, as sulfide reserves dwindle, oxidized ores, particularly smithsonite and willemite, have become the primary alternatives [3]. These ores exhibit poor intrinsic hydrophobicity, are finely disseminated, rapidly dissolve, and shed slimes that precipitate heterogeneous coatings. Together, these phenomena foul flotation circuits, yielding low recovery, poor selectivity, and excessive reagent consumption [1,4]. To address these challenges, researchers have explored various physical and chemical treatment methods [5,6,7,8,9,10]. Among these, ultrasonic external-force conditioning has emerged as a promising non-chemical intensification route, due to the unique properties of ultrasonic waves in enhancing interfacial interactions [11].

Recently, ultrasound has been widely applied in the flotation of metal ores [12,13], coal [14,15], and rare-earth minerals [16,17]. This technique leverages acoustic cavitation, which generates micro-jets and shear waves that create transient hot spots reaching temperatures of 5000 K and pressures of 1000 atm [15]. These extreme conditions can strip oxidation films from sulfides and remove clay coatings from coal, thereby restoring hydrophobicity [18,19,20]. For oxide and silicate minerals with specific solubility, such as smithsonite [10], titanaugite [13], dolomite [21], and spodumene [12], ultrasonic pulses selectively dissolve surface-bound metal ions, creating new adsorption sites. Additionally, in the presence of hydrophobic particles, the acoustic field nucleates micro-nanoscale cavitation bubbles (CBs) [22,23]. Collectively, these events generate a dynamic population of nascent nanoparticles, including solid fragments, metal-hydroxy/oleate colloids, and gaseous nuclei. Despite the consensus that these nanoparticles coexist during ultrasonic processes, the complex physicochemical environment of the pulp in the ultrasonic field renders them challenging to differentiate effectively due to the limitations of current research methodologies. However, it is believed that the size, morphology, concentration, and interfacial activity of these nanoparticles play a pivotal role in governing collector equilibria, particle aggregation, bubble–particle encounter rates, and froth stability [24,25,26,27]. However, these parameters remain largely unquantified for ultrafine smithsonite systems. Closing this knowledge gap is essential, as these nanoparticles ultimately dictate pulp chemistry and flotation performance.

Efficient recovery of zinc oxide minerals is essential for securing the global zinc supply and for utilizing low-grade or oxidized resources that would otherwise be discarded. Among emerging intensification strategies, ultrasonic conditioning (UC) has demonstrated broad adaptability and marked effectiveness across sulfide, oxide, halide, and rare-earth systems. Yet its influence within the fatty acid flotation system, especially for the flotation of soluble zinc-oxide minerals, remains largely unexplored. We previously showed that ultrasonic pre-treatment before collector addition alters the interfacial properties and flotation response of ultrafine smithsonite [10]. However, the consequences of applying UC after reagent addition are unknown. To bridge this gap, this study systematically examined the effects of post-dosing UC on ultrafine smithsonite using sodium oleate (NaOL) as the collector, through flotation tests and a series of advanced detection methods. The mechanistic studies primarily focused on three interlocked dimensions: interfacial restructuring and reagent redistribution, nano-mediated aggregation/dispersion, and froth-phase evolution. The insights obtained will not only advance the mechanistic understanding of ultrasound pulp conditioning-enhanced flotation of ultra-fine oxidized zinc ores but also provide a transferable framework for intensifying the beneficiation of other soluble, ultrafine mineral systems.

2. Experimental

2.1. Materials and Reagents

The smithsonite sample was sourced from Yunnan Province, China. Comprehensive characterization via XRD analysis (Figure 1) and XRF (X-ray fluorometer; XRF-1800, Shimadzu Corporation, Kyoto, Japan) analysis (

Table 1. The chemical composition of the smithsonite sample via XRF analysis.

ZnO	Fe	Al2O3	SiO2	CaO	L.O.I
64.1	0.10	0.015	0.75	0.05	34.98

) confirmed the high purity of the ore sample. The ore was first coarsely crushed in a porcelain mortar, then finely ground in an agate mortar, and finally classified in a dry elutriator. The resulting ultrafine fraction (<10 µm), served as the experimental feed. The particle size distribution (PSD) of the experimental smithsonite particles reveals a volume mean diameter (D₄₃) of merely 6.501 μm (Figure 2), attesting to the extreme fineness of the particles. NaOL (chemically pure) was employed as the collector. Hydrochloric acid (HCl, Analytical grade) and sodium hydroxide (NaOH, Analytical grade) were used to adjust the pH. All experimental procedures utilized deionized water with a resistivity of 18.2 MΩ·cm.

2.2. Flotation Procedure

Smithsonite pulp (5 wt%) was first conditioned with NaOL at varying concentrations and pH 8. Conditioning was performed for only 3 min using either conventional stirring (SC) at 1700 rpm or a 28 kHz, 60 W ultrasonic bath (KMD-II, KeMeida) to avoid any significant temperature rise [10]. Immediately after conditioning, the pulp or the corresponding powder after low-temperature vacuum drying was transferred to flotation or further analyses. Single-mineral flotation tests were conducted at room temperature (22 ± 2 °C) in a 50 mL XFG cell at 1700 rpm. For each test, 50 mL of pre-conditioned smithsonite pulp was added into the flotation cell and re-conditioned for 3 min. Flotation then proceeded for 3 min, and froth was collected at 0.5, 1, 2, and 3 min. The floated and unfloated fractions were filtered, dried, and weighed to determine recovery from solid mass distribution. In parallel, water recovery was measured by dividing the 3 min flotation into four intervals (0–0.5, 0.5–1, 1–2 and 2–3 min). Froth solids and entrained water were collected, dried and weighed for each interval. Concentrate yield and water recovery were calculated with Equations (1) and (2):

$$Concentrate yield(\%)={\displaystyle \frac{M_c}{M_f}}\times 100$$

(1)

$$Water recovery(\%)={\displaystyle \frac{W_c}{W_f}}\times 100$$

(2)

where M_c is the concentrate dry mass (g) at each interval, M_f is the initial feed dry mass (g), W_c is the water mass in the concentrate (g), and W_f is the total water mass used (g). The latter includes the water added before conditioning plus any make-up water added during flotation to maintain a constant pulp level.

2.3. Geometrical and Morphological Characterizations of Smithsonite Particles

To assess the impact of sonication on the geometry features of smithsonite particles, the porosity of the samples was determined by an Autosorb-iQ (Quantachrome instruments, Boynton Beach, FL, USA) through N₂ adsorption–desorption isotherms at 77K, yielding specific surface areas (SSAs), pore volumes, and mean pore diameters (MPDs). Moreover, to evaluate the influence of sonication on the morphology of smithsonite particles, scanning electron microscopy (SEM) observations were conducted on smithsonite particles subjected to various conditioning treatments at room temperature using a Nova NanoSEM 230 (FEI Company, Hillsboro, OR, USA). Before testing, the particle surfaces were sputter-coated with a gold layer to enhance conductivity. All geometrical and morphological characterizations were conducted under an environment with a pH of 8 ± 0.1.

2.4. Nanoparticle Tracking Within the Smithsonite Slurry

Nanoparticles in smithsonite slurries were characterized by quantifying their size-concentration profiles. First, a smithsonite pulp was prepared according to the flotation protocols and conditioned for 3 min. After passing the pulp through a 1 µm syringe filter to retain micro-scale particles, the turbidity of the clarified supernatant was measured to obtain a preliminary estimate of nanoparticle content. The filtrate was then analyzed with a ZetaView nanoparticle tracking analysis (NTA) system (Particle Metrix, Malvern Panalytical, Malvern, UK) to determine particle size distributions and number concentrations for different pulps. Each measurement was performed in triplicate, and the mean values are reported.

2.5. Adsorption Experiments

The adsorption of NaOL on smithsonite surfaces was measured using a high-precision Total Organic Carbon (TOC) Analyzer (Multi N/C 3100, Gena, Germany). In each experiment, 2.5 g of smithsonite was mixed with 47.5 mL of NaOL solution of defined concentration and treated by SC or UC for 5 min. The suspension was then filtered through a 0.1 μm or 1 μm syringe-type microporous membrane, and the organic carbon content in the filtrate was measured. Afterward, the adsorption amount of NaOL on solids was calculated using the residual organic carbon method. Specifically, the NaOL concentration was determined via a standard calibration curve, and the adsorbed amount on solids was derived from the residual concentration [28]. Each sample was analyzed in triplicate, and the average value and its standard deviation are reported.

2.6. Zeta Potential Measurements

Zeta potential measurements were conducted at room temperature using a Nano ZS90 analyzer (Malvern Co., Malvern, UK). For each sample, 0.02 g of ultrafine smithsonite particles were dispersed in 40 mL of 0.01 M KNO₃ (background electrolyte solution) solution to create a dilute suspension. The reagent addition sequence used in the flotation tests was followed to prepare the smithsonite pulp. The pulp was then treated for 3 min using either SC or UC, after which the pH was adjusted and measured. The suspension was left to settles for 5 min, and the supernatant was collected for analysis. Each condition was evaluated in at least triplicate, and the mean zeta potential ± one standard deviation is finally reported.

2.7. XPS Analysis

XPS was utilized to examine the surface modifications of smithsonite under various conditions. Samples were prepared in a flotation cell following a procedure similar to the flotation experiments, then dried in a vacuum oven at 40 °C. The resultant powders were analyzed with a K-Alpha⁺ X-ray photoelectron spectrometer (Thermo Fisher, Waltham, MA, USA) using monochromatic Al Kα radiation (200 W, 20 eV pass energy). The analysis chamber was held below 10⁻⁹ mbar, and photoelectrons were collected at a 90° take-off angle. The binding energy scale was calibrated using the C 1s peak from adventitious carbon (approximately 284.8 eV) as an internal reference. Data processing was conducted using Thermo Scientific Advantage 4.52 software.

2.8. Aggregates Size and Mass Fractal Dimension Measurements

The size and structural properties of aggregates were examined using Static Light Scattering (SLS), a method well-suited for characterizing aggregates with low refractive indices and loose structures [29,30]. Measurements were conducted at room temperature using a Malvern Mastersizer 2000 (Malvern Co., Malvern, UK). For each experiment, smithsonite pulp was prepared following the standardized flotation protocol. A precisely metered aliquot of this slurry was then gently introduced through a 0.5 cm-ID pipette into 1 L of ultrapure water, taking care to avoid mechanical disruption of the aggregates. Light obscuration was monitored in real time, and once the signal stabilized at 10% ± 2%, data acquisition began. Each sample was measured five consecutive times, and the arithmetic mean was reported. PSD data were directly obtained using the instrument’s integrated software. To quantify aggregate architecture, the mass-fractal dimension (D_f) was derived from the angular dependence of scattered-light intensity I(q). Specifically, the slope of the log I(q) versus log q plot was analyzed to yield D_f, as detailed in our previous work and illustrated in Figure 3 [29]. Lower D_f values denote open, tenuous aggregates, whereas higher values indicate increasingly compact structures.

2.9. Frothing Ability of Smithsonite Pulp in the Presence of NaOL

The persistent foaming inherent to NaOL renders the conventional dynamic foamability index unsuitable for smithsonite slurries. We therefore adopted an end-point protocol that captures both frothing tendency and stability froth a single aeration cycle [31]. Frothing of smithsonite pulp was continuously monitored under a constant air-flow rate using a custom-designed aeration cell (Figure 4). The experimental setup comprised a 70 cm × 5 cm glass column fitted with an air inlet valve and a mass-flow controller calibrated to 0.2 L·min⁻¹. In each test, 30 mL of smithsonite pulp, prepared under varying pretreatment conditions following standard flotation procedures, was introduced into the column and aerated for 30 s. The peak froth height (H_max) served as the index of frothing. Aeration was then terminated, and the time required for the froth to collapse to 0.5 H_max was taken as the three-phase foam half-life (T_0.5).

3. Results and Discussion

3.1. Flotation of Ultrafine Smithsonite with and Without UC

Figure 5 presents the impact of NaOL concentration on the flotation of ultrafine smithsonite particles under different pretreatments. Irrespective of the pretreatment, recovery rises steeply with collector concentration until a plateau is reached (Figure 5a), confirming NaOL’s strong collecting power [32]. Yet the conditioning method modulates this plateau: below 4 × 10⁻⁴ M NaOL, SC outperforms UC, whereas beyond 5 × 10⁻⁴ M, the situation reverses, with UC achieving a higher ultimate recovery. Thus, sonication is detrimental when NaOL is scarce but beneficial once sufficient collector is present. Figure 5b dissects the kinetics at two representative concentrations. At 2 × 10⁻⁴ M NaOL, SC attains higher stage-wise recoveries while UC exhibited a significantly lower flotation kinetics throughout the test. At 6 × 10⁻⁴ M NaOL, SC enjoys a faster start, especially in the first minute, but its rate decays quickly. UC, though initially slower, overtakes SC and finishes with the higher cumulative recovery. Collectively, these findings demonstrate that UC consistently has a lower initial flotation kinetics than SC but gradually overtakes SC at higher NaOL concentrations. Thus, UC-assisted flotation of ultrafine smithsonite is highly dependent on the initial NaOL concentration.

The divergent responses can be traced to how sonication interacts with both the mineral surface and the collector. Previous studies have shown that ultrasonic comminution-induced size reduction has a negligible effect on the flotation of ultrafine minerals [10]. Instead, surface abrasion and micro-erosion redistribute interfacial species, thereby altering NaOL adsorption and flotation behavior. Specifically, ultrasonic micro-jets can shear physically adsorbed NaOL from the surface, but they cannot detach chemisorbed collector because of its strong mineral–molecule bond [33]. Once desorbed, these molecules are prevented from readsorbing by an electrostatic barrier [34], and surface hydrophobicity therefore drops, and recovery is suppressed when NaOL is limited. At higher NaOL concentrations, additional mechanisms become operative. Sonication promotes gas precipitation and cavitation, and superhydrophobic zinc-oleate nanoparticles or interfaces, formed through Zn²⁺-oleate precipitation, serve as preferential bubble-nucleation sites. The resulting ultrafine CBs intensify particle-bubble encounters, thereby improving collection efficiency [35]. Guided by these hypotheses and analyses, the next phase of our work will systematically quantify how sonication modifies mineral surfaces, NaOL molecules, and their mutual interactions.

3.2. Sonication-Assisted Interaction Between Ultrafine Smithsonite and NaOL

3.2.1. Geometrical and Morphological Characterizations of Ultrafine Smithsonite

Figure 6 compares the porosity of smithsonite after different conditioning routes. The nitrogen adsorption–desorption isotherms of these samples consistently exhibit Type IV behavior with hysteresis loops, indicative of mesoporous materials [36]. The original smithsonite sample exhibits an SSA of 17.272 m²/g and an MPD of 3.835 nm (Figure 6a). After pulp conditioning, the SSA increased to 17.526 m²/g after SC (Figure 6b), attributed to the dissolution and hydrolysis of semi-soluble smithsonite particles during conditioning. In contrast, UC markedly increased the SSA to 18.202 m²/g (Figure 6c). This enhancement is likely due to ultrasonic-assisted surface erosion, including particle crushing, physical surface stripping, and enhanced surface ion dissolution and hydrolysis [37]. The larger SSA generated by UC implies a greater density of exposed surface atoms, which in turn raises total surface energy and strengthens interactions with the aqueous phase and with NaOL [38,39]. In contrast, the average pore diameter decreases marginally after conditioning, and the change is essentially the same for SC and UC. Because smithsonite mesopores are only a few nanometers wide, any slight reduction in MPD likely arises from the partial hydrolysis of zinc oxide ions under the weakly alkaline conditions. Partially resulting nano-colloids can fill existing pores or form surface coatings, producing the observed minor contraction, yet they do not substantially alter the overall pore network [40]. Overall, these minor alterations in porosity are unlikely to significantly impact subsequent flotation.

Figure 7 depicts the surface characteristics of smithsonite particles observed via SEM. In the SC system with 10⁻⁴ M NaOL (Figure 7a), the particles exhibit rough, irregular protrusions formed by adsorbed NaOL and precipitated hydroxyl compounds. This patchy coverage leaves part of the surface hydrophilic and partly hydrophobic [41]. Under the same collector dosage, UC produces smoother, cleaner surfaces with only sparse nanoscale bridges (Figure 7b), evidence that ultrasonic micro-jets effectively strip loosely bound fines and hydrophilic precipitates. However, at low NaOL levels, SC delivers higher recovery than UC, indicating that simple surface cleaning is not the decisive factor when collectors are scarce. At higher NaOL concentrations (e.g., 5 × 10⁻⁴ M), the SC-treated particles display partial surface coverage and some aggregation (Figure 7c), likely due to hydrolyzed zinc hydroxide condensation and hydrophobic aggregation induced by NaOL adsorption [42]. UC-treated particles, however, exhibit a denser and smoother adsorption layer (Figure 7d). Moreover, UC-treated samples display more pronounced surface dissolution and loss of adhering substances, resulting in a smoother surface with smaller particles and larger aggregates. This suggests enhanced mineral surface dissolution and more significant hydrolysis and precipitation reactions under UC conditions. These findings highlight sonication’s ability to alter smithsonite morphology and associated reagent adsorption.

3.2.2. Nanoparticle Detection Results of Different Smithsonite Pulps

Around pH 8, zinc ions precipitate as nanoscale hydroxyl-zinc colloids that either remain suspended or adsorb onto smithsonite surfaces. Ultrasonication accelerates this process by promoting the dissolution–hydrolysis of smithsonite, thereby releasing additional Zn²⁺ and generating extra nanoscale smithsonite fragments. Simultaneously, chemical precipitation with NaOL and oleate micelles further elevates the concentration of nano-colloids. Because these colloids mediate multiphase interactions [43], their size distribution and concentration are critical for interpreting how UC affects ultrafine smithsonite flotation. We therefore used turbidity measurements and dynamic light scattering to quantify nanoparticle abundance and dimensions under varying conditions (Figure 8). Figure 8a presents the macroscopic turbidity of post-treatment supernatants (left panel) and the morphological characteristics of nanoparticles in the bulk solution (right panel). All supernatants are visibly turbid, a result of suspended nanoscale solids (i.e., residual smithsonite, zinc oleate, hydroxyl–zinc species, free oleate micelles, and their complexes) [10]. Generally, turbidity increases with higher NaOL concentrations. However, at equivalent NaOL levels, supernatants from the UC system display consistently greater turbidity, indicating elevated concentrations of slow-settling nanoscale particles. To obtain quantitative size and concentration data, NTA was employed (Figure 8b), and total nanoparticle counts were compiled (Figure 8c). Notably, every suspension contained particles chiefly within 0–300 nm, with negligible larger fractions. In the SC system, nanoparticle concentration increases gradually with rising NaOL concentration, while average particle size and size distribution remain nearly stable, likely due to mild ion dissolution, hydrolysis, smithsonite precipitation, and weak interfacial interactions between NaOL and zinc components during SC [10]. In UC pulps, by contrast, nanoparticle concentration rises sharply, peaks at 5 × 10⁻⁴ M NaOL, and then declines, which is likely attributed to the hydrophobic self-association of nano-ions at elevated sodium oleate levels, which drives the formation of micron-sized aggregates that promptly sediment [13,25]. Compared to the subtle variations in the SC system, the UC system demonstrates a far more pronounced change in nanoparticle concentration. At identical NaOL concentrations, UC-treated pulps consistently show higher nanoparticle counts, smaller particle sizes, and narrower size distributions than SC-treated pulps. This indicates sonication promotes the formation of more numerous, finer nanoscale solid or gas nuclei in the bulk solution.

Solid nuclei primarily arise from ultrasonic fragmentation of smithsonite particles and their dissolution–hydrolysis products, while gas nuclei form via dissolved gas nucleation at NaOL-induced hydrophobic sites/microdomains or the stable adsorption of bulk gas nuclei. It is hypothesized that these nuclei exert a substantial impact on mineral flotation performance, particularly at elevated NaOH concentrations, where the nucleation effects in the pulp are significantly amplified [44]. Based on flotation results, we postulate that the enhanced flotation recovery under high NaOL concentrations with UC treatment is likely linked to the elevated nanoparticle count in the solution, which will be further explored in subsequent studies.

3.2.3. Quantitative Evaluation of NaOL Adsorption on Smithsonite Surfaces Under Variable Conditions

Section 3.2.2 reveals that UC-assisted pulps contain far more nanoparticles than SC-assisted pulps. These nanoparticles, whether solid or gaseous, exhibit a large SSA and strong adsorption capacity, and therefore redistribute NaOL among the bulk solution, the nanoparticle surface, and the smithsonite surface. To quantify this redistribution, we used TOC to measure NaOL in each phase after conditioning at different initial concentrations. Given the broad particle size distribution of nanoparticles in the pulp, we used two microporous membranes with different pore sizes (0.1 μm and 1 μm) for solid–liquid separation. The 0.1-μm membrane effectively removes particles larger than 100 nm, producing a filtrate free of secondary nanoparticles generated during pulp conditioning, while the 1-μm membrane retains nanoparticles in the supernatant. By analyzing the organic content in the filtrates, we elucidated the impact of nanoparticles on NaOL distribution and adsorption behavior during pulp conditioning. These insights shed light on the mechanisms behind the different flotation behaviors caused by varying pulp conditioning methods.

Figure 9 illustrates the residual NaOL concentration in filtrates (Figure 9a), the adsorption quantity of NaOL on filter cake surfaces (Figure 9b), and the adsorption density (Figure 9c). As depicted in Figure 9a, the free NaOL concentration in the filtrate rises with the initial NaOL concentration, irrespective of the conditioning method or filtration technique. However, both the conditioning method and the membrane pore size significantly affect the residual NaOL content. Specifically, with a 0.1-μm membrane, SC-based filtrates have higher residual NaOL concentrations than UC-based filtrates, indicating that UC-based nanoparticles consume more NaOL through adsorption or precipitation. Using a 1-μm membrane results in higher residual NaOL concentrations than with a 0.1-μm membrane, for both SC and UC conditions. This confirms that nanoparticles in the smithsonite pulp consume more NaOL, with UC showing a greater effect, likely due to more nanoparticle generation. Quantitative analysis reveals that nanoparticles significantly consume NaOL during pulp conditioning, especially in UC-treated pulps. This interaction reduces the direct adsorption or chemical coordination of NaOL on large smithsonite particle surfaces, potentially impairing smithsonite hydrophobicity and flotation performance, particularly at low NaOL concentrations. This is likely a key factor contributing to the lower flotation recovery under UC treatment when NaOL is scarce in the pulp. Figure 9b,c elucidate the adsorption characteristics of NaOL on smithsonite. The data reveal an inverse correlation with those in Figure 9a: higher residual NaOL concentrations correspond to lower adsorption on smithsonite surfaces. Higher adsorption quantities and densities are consistently achieved with a 0.1-μm membrane compared to a 1-μm membrane. In the UC system, apparent adsorption is higher when nanoparticle and micrometer-sized smithsonite surface adsorption are not differentiated. However, when these processes are distinguished and the adsorption/precipitation of NaOL on these nanoparticles is accounted for as collector consumption, the UC system demonstrates reduced adsorption on smithsonite surfaces. This reduction is further amplified by the decreased SSA of smithsonite following UC treatment.

These observations highlight the crucial role of nanoparticles in regulating reagent distribution across diverse interfaces within the pulp conditioning system, which is vital for understanding the interfacial behavior of fine-grained minerals, especially soluble salts [45,46]. Additionally, the study highlights that accurately calculating reagent adsorption via the residual concentration method requires considering the materials’ physicochemical properties, such as solubility, hydrolysis, chemical stability, and clay formation tendencies. Proper sample preparation methods are crucial for reliable and accurate results. Importantly, consistent measurement results, derived from extensive replicate testing, validate the experimental patterns and primary hypothesis.

3.2.4. Electrokinetic Potential Results of Smithsonite Particles

Zeta potential measurements can serve as an effective tool for characterizing interfacial phenomena in smithsonite flotation systems. Figure 10 presents the zeta potential-pH profiles of smithsonite particles subjected to SC and UC. SC-treated smithsonite particles exhibit an isoelectric point (IEP) near pH 8, consistent with the findings of Shi et al. [32]. In contrast, UC shifts the curve rightward, raising the IEP to just below pH 9. This shift suggests that sonication promotes the penetration of positively charged ions or polar groups into the Helmholtz layer [8], thereby rendering the surface positively charged at pH 8. This effect is attributed to the combined influence of reverse adsorption of dissolved cations and their hydrolysis products, as well as the alteration of surface charge density by sonication. Specifically, intensified sonication-induced erosion increases the exposure of zinc active sites, thereby exposing more positive charges and elevating the IEP.

The addition of NaOL induces significant negative shifts in the zeta potential curve, primarily due to the specific chemisorption of oleate anions onto zinc active sites within the Helmholtz layer [32]. Comparative analysis of SC- and UC-treated samples reveals that NaOL concentration critically modulates this surface potential response. At low NaOL concentrations (e.g., 10⁻⁴ M), both SC and UC exhibit negative shifts, but SC-treated particles are more electronegative. Although UC enhances NaOL-smithsonite interactions, low NaOL levels result in a higher degree of unsaturation of the adsorption sites. Consequently, the increased exposure of zinc cationic active sites under UC conditions leads to a lower negative charge on the particles [38]. In contrast, at high NaOL concentrations (e.g., 5 × 10⁻⁴ M), the zeta potentials converge between SC and UC systems, suggesting that the smithsonite surfaces are approaching adsorption saturation with NaOL.

The zeta potential data further confirm that UC promotes more extensive surface dissolution and erosion than SC. The elevated concentration of dissolved zinc species and their hydrolysis products alters the surface potential of smithsonite, thereby increasing the NaOL concentration required to reach adsorption saturation. This observation aligns with the findings presented in Section 3.2.2 and Section 3.2.3. Although unsaturated NaOL adsorption impacts surface electrostatics, its direct effect on flotation remains uncertain. Some studies have shown that even with optimal NaOL dosages that significantly enhance smithsonite floatability, many zinc sites on the surface remain unoccupied [47]. However, a prevailing view among researchers is that inefficient collector precipitation and consumption are key factors contributing to the deteriorated flotation performance of various oxide minerals [48,49], particularly at low NaOL concentrations.

3.2.5. XPS Analysis Results

XPS analysis was conducted to investigate the elemental composition and surface properties of smithsonite under various conditions (Figure 11). Figure 11a illustrates distinct trends in the relative atomic concentrations of C1s, O1s, and Zn2p, influenced by NaOL concentration and pulp conditioning methods. At low NaOL concentrations, UC exhibited a slightly higher Zn 2p abundance than SC, whereas this relationship reversed at high NaOL concentrations. These variations reflect a competition between zinc exposure and NaOL coverage. Under UC, enhanced dissolution and hydration liberate additional Zn²⁺ sites, so zinc dominates the outermost layer when NaOL is scarce. At high NaOL concentrations, the surplus of oleate anions drives extensive chemisorption, and the more vigorous UC further promotes this process, resulting in a thicker oleate layer that attenuates the Zn 2p signal. The same trend is mirrored in the C 1s content and corroborated by the electrophoretic data.

High-resolution C1s spectra (Figure 11b) were deconvoluted into four carbon environments: CO₃²⁻ (289.4 eV), C=O (288.4 eV), C–OH (286.5 eV), and C–C/C–H (284.8 eV). At 2 × 10⁻⁴ M NaOL, UC shifts the CO₃²⁻, C=O, and C–OH peaks by −0.28 eV, −0.86 eV, and +0.03 eV, respectively, relative to SC. The proportion of C–OH decreases, while CO₃²⁻, C=O, and C–C/C–H increase. The reduction in the C–OH proportion implies weaker hydrogen bonding between RCOO⁻ and the smithsonite surface, whereas the rise in C=O points to stronger bidentate coordination with Zn sites [50]. Moreover, the elevated C–C/C–H proportion suggests that UC enhances the physical deposition of NaOL on the smithsonite surface compared to SC, partially due to stronger electrostatic interactions [6]. This is consistent with the UC process, which exposes more zinc active sites, as shown by the increased CO₃²⁻ content. At 6 × 10⁻⁴ M NaOL, SC–treated samples show a further drop in C–OH and gains in CO₃²⁻, C=O, and C–C/C–H, consistent with denser chemisorption and co-adsorption layers reported for oxide minerals [51]. The higher CO₃²⁻ proportions are likely due to NaOL-assisted surface dissolution of smithsonite and patchy NaOL adsorption, as observed in the research by Zheng et al. [38,39]. UC–treated samples, however, display the opposite trend: C=O decreases and C–OH increases, implying that coordination weakens while hydrogen bonding strengthens. Under these circumstances, NaOL molecules and zinc oleate are believed to facilitate the formation of nanoscale nuclei during pulp conditioning [23]. Consequently, some NaOL molecules may physically adsorb onto the mineral surface through gas–liquid mass transfer, enabled by hydrophobic micro-regions formed from gas nuclei anchoring [52,53]. Furthermore, a high proportion of exposed CO₃²⁻ is achieved, attributed to several factors. First, NaOL-assisted migration of surface components and ultrasonic erosion lead to the removal of hydroxy zinc compounds from the smithsonite surface and the exposure of more zinc active sites [10]. Second, the enhanced NaOL orientation at the solid–liquid interface promotes the coordination role [41]. Additionally, zinc-oleate precipitation and NaOL orientation can enrich local metal ions and boost complexation reactivity [48]. Collectively, these factors increase the adsorption heterogeneity of NaOL on smithsonite, thereby elevating the proportion of CO₃²⁻ on the surface.

Narrow-scan Zn 2p spectra (Figure 11c) reveal that peak positions remain fixed regardless of conditioning method or NaOL dosage, indicating that the intrinsic Zn coordination environment is preserved. Peak-area ratios, however, vary systematically. At 2 × 10⁻⁴ M NaOL, the UC sample registers 78.16% Zn 2p₃/₂ intensity versus 73.95% for SC, evidencing more Zn sites available for NaOL binding. Moreover, under SC conditions, the Zn coordination ratio increased to 77.93% at a high NaOL concentration level (6 × 10⁻⁴ M). This increase highlights the stronger coordination interaction between NaOL and the smithsonite surface at higher concentrations, aligning with previous research showing that NaOL promotes chemical adsorption on oxidized mineral surfaces at medium-high concentrations [41,54]. Conversely, the UC system at 6 × 10⁻⁴ M NaOL shows the lowest Zn coordination ratio among the tested samples at 73.81%, indicating weaker chemical interaction between NaOL and the zinc active sites. As previously discussed, the enhanced physical adsorption or hydrogen bonding of NaOL, largely induced by solid and gaseous nanoparticles in the solution, is probably a key factor contributing to this outcome.

Overall, the interaction between NaOL and smithsonite increases the relative proportion of the Zn 2p₃/₂ peak, driven by changes in the coordination environment of zinc atoms and electron transfer. XPS analysis reconfirms that both reagent concentration and pulp conditioning method affect NaOL adsorption on the smithsonite surface. Nanoparticles in the pulp system are a key factor causing differences in adsorption characteristics. At low concentrations, NaOL adsorption is relatively low. While UC promotes more coordination bonding and heterogeneous adsorption than SC, ultrasonic-induced dissolution of interfacial ions and surface erosion expose more zinc active sites. At high concentrations, the SC system experiences enhanced physical and chemical adsorption of NaOL, with more zinc active sites exposed than at low concentrations. In contrast, the difference in reagent adsorption between UC and SC surfaces is significantly reduced at high concentrations compared to low ones. This is evident from the varying proportions of CO₃²⁻ and supported by zeta potential measurements. Significantly, in the UC system, the proportion of NaOL’s coordination action on the mineral surface decreases, while hydrogen bonding increases. This shift may be related to UC promoting the generation of more hydrophobic nanoparticles and their interactions among smithsonite particles and excess NaOL molecules in slurries [48,55].

3.3. Aggregation/Dispersion of Smithsonite Particles Under Different Conditions

The effective flotation of ultrafine smithsonite particles hinges on hydrophobic aggregation induced by pre-adsorbed NaOL. Because conditioning history governs both NaOL uptake and subsequent interparticle interactions, we monitored aggregation with SLS technique, extracting PSD and D_f to quantify the process. Figure 12 summarizes the PSD evolution evolves with NaOL dosage and conditioning style. Two variables dominate the apparent particle size: the initial NaOL dosage and the conditioning method. Within the tested NaOL range, the median smithsonite size grows monotonically with collector dose: PSD curves shift right and the D₄₃ expands. The mechanism is straightforward: higher NaOL coverage converts the inherently hydrophilic surface into a hydrophobic one, so hydrophobic association and interlocked hydrocarbon chains drive increasingly strong aggregation [56]. However, UC breaks this trend. Regardless of NaOL concentration, UC locks the apparent size near its original value, demonstrating that micro-jet turbulence neutralizes NaOL adsorption-induced attraction. SC, conversely, amplifies aggregation, and extending SC from 3 min to 6 min further enlarges the aggregates. Quantitatively, at 2 × 10⁻⁴ M NaOL, an extra 3 min of SC lifts the D₄₃ of aggregate from 11.99 μm (SC only) to 15.39 μm, whereas “UC + SC” inches from 6.56 μm to only 7.38 μm. At 6 × 10⁻⁴ M NaOL, the gap narrows: the extra SC shifts the size from 31.61 μm to 35.66 μm, while “UC + SC” surges from 7.56 μm to 27.05 μm. Thus, even under high NaOL coverage, a brief “UC + SC” sequence secures substantial hydrophobic aggregation. Classical aggregation kinetics regard this process as a cascade from primary particles through clusters to final aggregates [57]. The retarded size growth observed at a higher NaOL concentration accords with the late-stage “cluster-to-aggregate” transition. Although “UC + SC” still yields slightly smaller aggregates than prolonged SC alone, the residual difference is small enough that its adverse impact on flotation is largely mitigated. Overall, UC possesses a significantly greater dispersion capability than SC and therefore suppresses smithsonite hydrophobic aggregation. This strong dispersion effect aligns well with the subsequent flotation performance. At low NaOL concentrations, UC’s intense dispersion leads to inefficient collector use and irreversible inhibition of particle aggregation, producing far lower recovery and initial rate constants than the mild aggregation-flotation sequence obtained with SC alone. At high NaOL concentrations, however, ultrasonication ceases before flotation, and the subsequent SC stage allows rapid re-aggregation of ultrafine solids [29,58]. Consequently, the detrimental influence of UC on flotation is largely offset. Nevertheless, because the initial dispersion is so strong, the UC system’s kinetic constant remains slightly lower even after SC.

In addition to the PSD, the D_f of smithsonite flocs is a key descriptor of aggregation architecture that controls both sedimentation and bubble–aggregate interactions during flotation. Figure 13 tracks how the D_f of smithsonite flocs evolves with NaOL concentration and conditioning style. Within the tested range, D_f declines generally as NaOL increases, mirroring the inverse trend in floc size, a trend previously reported for similar systems [29]. We next contrast the structural signatures imposed by UC and SC. At low NaOL, SC yields D_f > 2, signaling the reaction-limited-cluster-aggregation (RLCA) driven by patchy, chemisorbed NaOL. UC gives a slightly lower but still RLCA-level D_f, suggesting either near-complete dispersion or only minimal hydrophobic clustering. As NaOL rises, both SC and UC depress D_f, yet to different extents. After 6 min of SC, aggregates reach the lowest D_f, whereas 3 min of UC yields the highest, reflecting the ultrasound field’s suppression of interparticle aggregations. At 4–6 × 10⁻⁴ M NaOL, SC flocs collapse to D_f around 1.95, a value approaching the diffusion-limited-cluster-aggregation (DLCA) and indicative of physically adsorbed NaOL layers. Beyond this concentration, SC flocs cease to densify and even rebound slightly, plausibly because shear-induced rupture is followed by re-aggregation into denser networks [59,60]. Meanwhile, under the “UC + SC” sequence, D_f plunges to the overall minimum. This sharp drop confirms that UC pre-treatment primes the pulp for pronounced DLCA-type aggregation once SC resumes. This is most likely due to physically driven nanoparticle bridging that involves both solid and gaseous phases [29,61]. For instance, in a pulp containing a high concentration of NaOL, sonication continuously nucleates a large population of tiny CBs stabilized at solid–liquid and liquid–liquid interfaces. These CBs form complexes with particles, acting as low-density linkages. The resulting particle–CBs complexes yield fragile, open flocs characterized by low D_f values [62]. A lower D_f lengthens pulp residence time and facilitates bubble–particle attachment, while it also increases the risk of non-selective gangue entrainment [63].

3.4. Frothing of Smithsonite Pulp Under Different NaOL Concentrations

Besides hydrophobic aggregation, the frothability of the pulp is decisive for ultrafine smithsonite flotation. Figure 14a illustrates how the frothing index (H_max) and froth half-life (T_0.5) of smithsonite pulp evolve with NaOL concentration after SC and UC. H_max rises steeply at low NaOL concentrations, then plateaus for SC or levels off gradually for UC. Below 4 × 10⁻⁴ M NaOL, UC yields the lower H_max, and beyond this threshold it surpasses SC. T_0.5 follows a similar crossover: UC produces less-stable froth than SC at low NaOL, yet generates more persistent froth beyond 3 × 10⁻⁴ M. This reversal likely stems from the ultrasound-driven surface erosion and cavitation, which accelerate dissolution–hydrolysis–precipitation and flood the pulp with Zn-containing nanosolids together with CBs. These extra surfaces scavenge NaOL, so free-collector concentration and effective particle hydrophobicity remain depressed in the UC system until NaOL is in surplus. Consequently, at low NaOL, the UC pulp produces smaller H_max and shorter T_0.5, mirroring its poorer flotation recovery. As NaOL concentration rises, these constraints are progressively lifted. A surplus of free NaOL not only boosts frothing capacity but also, at elevated levels, triggers the ultrasonic process to generate abundant stable CBs. They fulfill two critical roles: enlarging the froth volume and thickening the froth layer; suppressing coalescence and collapse, and markedly enhancing froth stability, by lodging between conventional bubbles [64]. Concurrently, the nanosolids generated during UC become increasingly hydrophobic as NaOL concentration increases. Once their surface hydrophobicity exceeds a critical threshold, they further stabilize the froth layer [65]. This synergy explains why the UC system attains greater froth stability at a lower critical NaOL concentration: at 4 × 10⁻⁴ M, UC records a slightly lower maximum froth height than SC but a significantly longer froth half-life. In flotation practice, the elevated frothing capacity and extended froth residence time translate into substantially higher recoveries of fine smithsonite particles. Although UC induces weaker hydrophobic aggregation of smithsonite than SC (Section 3.3), its superior froth performance outweighs this drawback at elevated NaOL concentrations. At this stage, nanoparticles generated by UC act as both froth thickeners and stabilizers, underpinning the enhanced recovery of ultrafine smithsonite.

Figure 14b reveals that smithsonite recovery scales linearly with water recovery, a signature of ultrafine particle systems [63]. This linear relationship underscores that the efficiency of smithsonite flotation is intrinsically linked to the water recovery dynamics within the flotation process. Notably, for any given smithsonite recovery, UC yields markedly higher water recovery than SC. This disparity can be attributed to the unique effects of UC under high NaOL concentrations. UC not only generates smaller, more porous smithsonite flocs but also simultaneously produces a denser and more stable froth layer. The smaller and more porous flocs created by UC increase the surface area available for water entrapment, thereby enhancing the amount of water recovered alongside the smithsonite particles. The denser and more stable froth layer ensures that fine particles remaining suspended can be effectively transported to the froth phase. These two phenomena—smaller, more porous flocs and a denser, more stable froth layer—collectively amplify water recovery. The increased water content in the froth phase can lead to higher entrainment of fine particles, which may impact the final concentrate grade. As a result, UC-assisted flotation produces wetter froths with a higher potential for particle entrainment. When scaling UC to industrial applications in fine-grained mineral flotation, operators must carefully account for this additional water loading to ensure the final concentrate grade remains within desired specifications.

The data and analyses presented herein demonstrate that relative to SC, UC exerts more pronounced changes on the physicochemical signature of smithsonite. These changes manifest as alterations in particle geometry and morphology, electrokinetic behavior, and the dissolution-hydrolysis-precipitation of surface ions. Consequently, UC enriches the pulp with zinc ions and zinc-bearing nanosolids, thereby enhancing NaOL adsorption. Flotation outcome, however, hinges on NaOL dosage. At low NaOL concentrations, zinc ions and nanosolids compete for the collector, leaving an insufficient amount to hydrophobize the smithsonite surface. Concurrently, the vigorous ultrasonic field disperses ultrafine smithsonite particles and suppresses their aggregation. The triple handicap of depleted free NaOL, poorly hydrophobized surfaces, and a deflocculated pulp precludes the formation of a stable froth layer, driving both recovery and kinetics below SC values. Raising the NaOL concentration removes the collector-limitation bottleneck. Zinc-bearing interfaces become adequately hydrophobized. Some hydrophobic zinc nanoparticles function as precipitated or bridging collectors, while others reinforce the froth. Simultaneously, high NaOL levels promote acoustic cavitation, generating a profusion of CBs that act as secondary collectors or bridging agents. As a result, the dispersed ultrafine smithsonite particles can quickly aggregate in a relatively mild stirring flow environment when sonication ceases. Like their solid counterparts, these gaseous nanoparticles not only stabilize the froth but also thicken it. Hence, elevated NaOL concentrations mitigate hydrophobic-aggregation inhibition, yielding a deeper, more persistent froth and higher recovery than SC, though with greater water recovery. Nevertheless, the strong dispersive effect of ultrasonics still suppresses the initial flotation rate in UC, even at high NaOL levels.

4. Conclusions

This study explores the influence of UC on the flotation behavior of ultrafine smithsonite using NaOL as the collector. The interactions between smithsonite particles and NaOL, the aggregation of mineral solids, and the froth characteristics were systematically examined to elucidate the underlying mechanisms, with particular emphasis on the role of various nanoparticles generated during UC in mediating these phenomena. The key findings are summarized as follows:

(1)

UC significantly alters ultrafine smithsonite flotation, but its influence is governed by the NaOL concentration. Below 4 × 10⁻⁴ M NaOL, UC lowers both recovery and kinetics relative to SC. Once the concentration exceeds this threshold, UC delivers higher recovery than SC, although the initial flotation kinetics remain slower.

(2)

UC reshapes the smithsonite surface far more drastically than SC. The intense acoustic field dissolves lattice ions and nucleates zinc-rich nanoparticles, thereby creating extra adsorption sites for NaOL. At low NaOL concentrations, sonication marginally enhances NaOL chemisorption, yet the simultaneous exposure of fresh Zn-rich sites aggravates collector undersaturation. When NaOL is ample, both the mineral surface and dissolved zinc species are fully hydrophobized by oleate, forming a hydrophobic shell that likely promotes the deposition of nanosolids and the interfacial anchoring of gaseous nanoparticles.

(3)

During SC, gentle shear steadily enlarges smithsonite flocs as the NaOL concentration rises, whereas UC violently disperses all particles regardless of the NaOL level. Although sonication destroys flocs, re-starting SC allows the dispersed particles to re-flocculate, and the extent of re-aggregation is set by the prevailing NaOL concentration. At low NaOL levels, only partial surface hydrophobization occurs, yielding weak and slow re-aggregation that accounts for the poor recovery and sluggish kinetics observed after UC. When NaOL concentration is high, two parallel changes occur: complete hydrophobization of both solid particles and dissolved zinc species, and a sharp rise in CBs. These synergistic effects accelerate hydrophobic aggregation. Once NaOL reaches 6 × 10⁻⁴ M, nanoparticles bridging triggers the rapid formation of large, loose aggregates and flotation performance improves dramatically.

(4)

The frothability of smithsonite pulp is controlled by two inter-related variables: the NaOL concentration and the conditioning route. At low NaOL concentrations, UC produces froths that are markedly weaker and less stable than those generated by SC. As NaOL rises, smithsonite surfaces become more hydrophobized and the pulp accumulates both solid and gaseous nanoparticles. Together these changes enlarge and stabilize the froth layer. Once NaOL exceeds 4 × 10⁻⁴ M, UC surpasses SC in overall smithsonite recovery, albeit with greater water recovery. Nevertheless, the vigorous dispersion imparted by ultrasound still suppresses UC’s initial flotation rate below that of SC, even at elevated NaOL levels.