2026 Theses Doctoral

Lewis Acid-Base Interactions of N-Alkoxycarbonyl S-Alkyl Dithiocarbamates and Related S-Based Ligands at the Chalcophile-Aquatic Interface in Complex Ore Systems

Bellusci, Nicholas

In industrial separations of value sulfide minerals and precious metals from ores via froth flotation, S-containing bi-functional ligands (collectors in industry vernacular) are used to selectively impart a hydrophobic character to the value ore particle surfaces. These hydrophobized particles then attach to air bubbles in a stirred aerated suspension and are separated from the preponderance of non-value gangue particles in the ground aqueous ore pulp. In particular, the soft S-donor atoms on these ligands are key in selective and efficient binding to (partially coordinatively saturated) chalcophile metal sites on a given mineral surface.

The main scientific hypotheses substantiated in this thesis work, which have barely received attention in academic flotation literature, are: a) ligand binding to metal surface sites occurs via Lewis acid-base (LAB) interactions and ligand exchange, consistent with Pearson’s hard and soft acid-base (HSAB) principles, and b) ligand adsorption under conditions relevant to flotation plant practice occurs in a patchy, non-uniform manner on heterogeneous mineral surfaces and is influenced by hydration, mineral aquatic chemistry and ligand exchange. These hypotheses serve as a direct challenge to long cherished mechanisms for ligand adsorption at the chalcophile-aquatic interface, which have mostly ignored LAB and HSAB concepts and the inherent complexity of real ore systems.

Amongst the 17 classes of S-based ligands used for flotation processing of sulfide and precious metal ores, the N-alkoxycarbonyl S-alkyl dithiocarbamates are particularly unique and serve as a “poster child” for the hypotheses based on LAB and HSAB concepts stated above. Their coordination chemistry and interfacial properties are quite different compared to the traditional anionic dialkyl dithiocarbamates, which is due to the incorporation of an electron-withdrawing alkoxycarbonyl group and alkylation of the thiolate moiety. These factors subtly influence the electron density and polarizability of the S donors and dramatically alter their metal complexation properties. At the macroscopic level, the n-butyl homologue, viz. N-butoxycarbonyl S-butyl dithiocarbamate (BCBDTC) has empirically demonstrated several benefits in real ore flotation, including versatility across a wide range of ore types and a unique affinity for soft acid chalcophile surface sites (Cu, Ag, Au, and platinum group elements) on mineral surfaces.
Such observations can be rationalized from the subtle molecular level changes in these ligands and their interactions at the mineral-aquatic interface.

However, there is currently no literature describing such interactions of BCBDTC, and there are many gaps and unanswered questions in general concerning how such subtle molecular level changes in the ligand structure result in macro-level outcomes. This is especially because most of the flotation literature on ligand-mineral interactions is woefully inadequate, as they ignore: a) the complex, variable, and heterogeneous mineralogy of real ores, b) the prevailing, non-equilibrium states of the bulk solution phase and the mineral-aquatic interface which the ligands must work under, c) the dynamic process conditions in plant practice, and d) Lewis acid-base and Pearson’s HSAB concepts, which underpin the fundamental science of metal-ligand coordination complexes.

Because of the gaping holes in the knowledgebase, systematic studies – at the molecular, mesoscopic, and macroscopic levels – were conducted to develop a deeper, more meaningful understanding of how S-based ligands interact with chalcophile (mineral and metal) surfaces under conditions relevant to ore flotation plant practice. A top-down approach was adopted, i.e. real ore flotation studies at the macroscopic level were conducted first to set the context, define boundary conditions, and to formulate critical questions and hypotheses. The bulk of the work was focused on BCBDTC, but after conducting a few initial experiments, significant brainstorming, and critical review of existing flotation and non-flotation literature, many of the questions and hypotheses were revised to also include other S-based ligand classes (thiol, thiolate, thioether, thione, thiophosphoryl, etc.) relevant to sulfide and precious metals ore flotation, corrosion inhibition, and self-assembled monolayers. These were then used to design experiments at the mesoscopic and molecular levels for testing hypotheses and answering questions regarding S-based ligand interactions with mineral surfaces. The ore flotation studies also included novel emulsified formulations to facilitate molecular transport of the poorly water-soluble BCBDTC to targeted mineral surfaces, with the specific aim of improving recovery of coarse and fine value particles.

At the macroscopic level, ore flotation experiments demonstrated that BCBDTC is a unique ligand which can selectively interact with Cu, Au, Ag and platinum group mineral (PGM) species which are poorly recovered with the traditional alkyl xanthates and dialkyl dithiophosphates. Emulsification of BCBDTC and other poorly water-soluble ligands in novel direct (oil-in-water) and inverse (water-in-oil) formulations further contributed to improved flotation rates, concentrate metal grades, and recovery of coarse and fine particles in Pb-Zn-Ag, porphyry Cu, and PGM ores. This is attributed to an improved efficiency in molecular diffusion and interfacial transport of oily ligand molecules from micron-sized droplets to the value ore particle surfaces via Ostwald effects (based on Laplace pressure effects similar to Ostwald ripening).

At the mesoscopic level, electroanalytical and contact angle measurements were used to characterize the adsorption of BCBDTC and many S-based ligands for comparison on selected mineral/metal substrates. Via cyclic voltammetry, the primary observation was passivation of the metal or mineral electrode surface using a flotation-relevant potential window and concentration range upon addition of the ligand. No anodic peaks indicating ligand oxidation or any other anodic processes under these conditions were observed. This “shutting down” of the electrode’s redox processes due to ligand binding to metal sites on the surface is attributed to Lewis acid-base (acceptor-donor) interactions. These are described in this work as a “redistribution” of electron density in the surface complex (to distinguish from the ambiguous charge transfer terminology used in other fields) upon overlap of the frontier molecular orbitals of the ligand’s donor atoms and metal (acceptor) sites on the surface. Passivation was the dominant effect across the board for multiple ligands and mineral/metal surfaces, which unequivocally demonstrates that ligand adsorption under flotation-relevant conditions (i.e. low ligand concentrations and a restricted electrochemical potential window of -300 to +100 mV vs Ag/AgCl) is primarily dictated by formation of new chemical bonds (between the ligand and mineral surface) without registering any anodic current that would indicate ligand oxidation.

Contact angle measurements via sessile water droplet and captive bubble methods were conducted to a) test the important hypotheses, b) corroborate ore flotation data, and c) support voltammetric and surface analytical measurements. Specifically, to test the hypothesis that ligand adsorption is non-uniform (patchy) because real mineral (and ore particle) surfaces inherently have significant chemical and physical heterogeneity, contact angle distributions were measured (across the surface) instead of measuring and reporting single values. Advancing and receding sessile water droplet contact angle measurements demonstrated that BCBDTC imparts a more significant degree of contact angle hysteresis compared to its main analogues (N-butoxycarbonyl O-butyl thionocarbamate, BCBTC; N-butoxycarbonyl N’-butyl thiourea, BCBTU) and classical alkanethiols such as n-octyl mercaptan – this implied a less ordered nature of the adlayer structure, but stronger pinning of the three-phase contact line. Data from captive bubble and sessile water droplet contact angle methods tell very different stories about the effects of ligand adsorption. From captive bubble contact angle distributions (each encompassing 100-200 measurements across the mineral or metal surface), it was found that hydrophobicity imparted by adsorbed BCBDTC and other ligands is lessened by hydration and time-dependent evolution at the mineral-aquatic interface. This was not apparent from sessile water droplet measurements. This exemplifies the fact that sessile water droplet and other contact angle measurement methods do not account for the types of pre-wetted surfaces relevant to flotation plant practice. As such, hydration, mineral aquatic chemistry, and ligand exchange processes are critical to account for. In addition, polarization of the mineral surface at various electrochemical potentials demonstrated that BCBDTC and other ligands can impart a hydrophobic character to the surface even under reducing conditions (e.g. -300 mV vs Ag/AgCl), which is contrary to the notion that oxidizing conditions are needed to form a ligand radical or oxidation product to produce sufficient hydrophobicity. The measured contact angle distributions ultimately demonstrate that ideal monolayer coverage, widely believed in the research community, is simply not possible under flotation-relevant conditions and even with high-grade, high purity minerals and metals. Using a single contact angle value to characterize the wetting state of a mineral surface captures neither the significant “patchiness” in ligand-imparted hydrophobicity nor the inherent surface heterogeneity of natural mineral surfaces.

At the molecular level, time-of-flight secondary ion mass spectrometry (ToF-SIMS), provided strong evidence for formation of complexes of BCBDTC with metal surface sites across a variety of substrates ranging from flat metal plates to coarsely ground mineral particles. A wide range of species ranging from the deprotonated parent molecular ion to various metal-BCBDTC complex fragments were detected. Based on this, it is proposed that on a solid mineral/metal surface, BCBDTC is a flexidentate ligand which binds via multi-site attachment (i.e. polynuclear complexation) in a wide range of modes; these range from monodentate to polydentate, primarily involving its C=S donor but also its auxiliary O, N, and S donors. This contradicts the original hypothesis that only the C=S and C=O donors are involved in binding. This is also highly contrary to the conventional thinking that ligands adsorbed on a heterogeneous surface are only of a single coordination mode, restricted to mononuclear complexation, and identical in structure to complexes formed in the bulk solution phase. The secondary ion mass spectra for each substrate exhibited its own unique fragmentation pattern of the original intact metal-BCBDTC complexes, which further implies that its coordination mode is dependent on the softness of its donor atoms and the metal surface sites it binds to. ToF-SIMS imaging analyses also demonstrated that BCBDTC’s adsorption is highly patchy, non-uniform, and correlated with metal-rich areas (or high-energy sites) of mineral surfaces.

Density functional theory (DFT) computations were also used to complement experimental methods. Conformational and frontier molecular orbital (FMO) analysis of BCBDTC demonstrated that the second S donor (i.e. –S–C4H9) in its structure results in enhanced conjugation (p-orbital overlap and delocalization of π electron density), lowered pKa, greater rigidity of the -C(=O)-N-C(=S)- functionality via higher rotational barriers, and a narrowing of the energy gap between its FMOs, all compared to its analogues BCBTC (–O–C4H9) and BCBTU (–NH–C4H9). Partial charge calculations and mapping of localized electronic properties (electrostatic potential and ionization energy) demonstrated that inductive effects from the alkoxycarbonyl group and alkylation of the thiolate moiety result in a lower electron density but highly polarizable nature of the C=S donor on BCBDTC in its deprotonated form, compared to DBDTC. Deprotonation from NH enhances the extent of electron delocalization and is expected to be favorable for metal complexation, though it is not necessarily a requirement. A set of structure-property relationships was also established from DFT results by systematically varying ligand ionic character, the central atom (C or P), and auxiliary S, N, and O donors – the findings from these relationships challenge prior understanding of inductive and resonance effects in these ligands by introducing or excluding certain donor atoms in the structure. Bond dissociation energy calculations and visualization of the frontier molecular orbitals on selected metal-BCBDTC complexes implied that its unique affinity for soft acids is related to its strong conjugation, versatility in adopting various coordination modes, and strong π back-bonding from the filled d orbitals of soft acids to its vacant antibonding p orbitals. were also conducted to study ligand adsorption in real ore pulps and on natural composite mineral surfaces (e.g. co-crystallized chalcocite and pyrite from a supergene Cu ore). The results from these studies clearly demonstrate that BCBDTC’s adsorption processes are influenced by HSAB and ligand exchange, such that its interaction becomes competitive in two fashions: in the presence of two mineral species, BCBDTC will favor interaction with the species with “softer” metal surface sites, and upon adsorption it must displace prevailing ligands at the interface which are derived from the non-equilibrium mineral aquatic chemical processes (i.e. between water, dissolved species, and the preponderance of gangue ore particles) occurring in the ore pulp. Such findings would not have otherwise been obtained using conventional single mineral studies in a clean aqueous phase.

The findings from these studies provide an original and holistic perspective on how BCBDTC and S-ligands in general interact with mineral surfaces under conditions relevant to ore flotation plant practice. Lewis acid-base and HSAB concepts are emphasized as a highly useful and practical scientific foundation for ligand design and in developing a deeper comprehension of the chemical aspects of ore flotation systems. It is also highlighted that ore system complexity can and should be incorporated into fundamental studies whenever possible. Such methodologies are not meant to predict outcomes in real systems, but to address critical questions and hypotheses, make better use of scientific knowledge for developing practical solutions, and advance the practice of sustainability in mineral resource recovery.