Advertisement

Mining, Metallurgy & Exploration

, Volume 36, Issue 1, pp 99–110 | Cite as

A Critical Overview of Dithiophosphinate and Dithiophosphate Interactions with Base Metal Sulfides and Precious Metals

  • Napoleon Tercero
  • D. R. NagarajEmail author
  • Raymond Farinato
Review Article
  • 26 Downloads

Abstract

Historically, as for most reagents used in flotation, the wide-spread use of dialkyl dithiophosphinates as collectors in practical applications has preceded a detailed scientific understanding of how they work. Dialkyl dithiophosphinates are a technologically and commercially important class of collectors used in the flotation processing of Cu-Au, complex polymetallic, and precious metal ores. Plant usage has demonstrated that dithiophosphinates exhibit higher selectivity and flotation rates for Cu sulfides, galena, and precious metal values in the presence of other gangue sulfides, than the traditionally used collectors such as dialkyl dithiophosphates and xanthates. They also have an attractive health, safety, and environmental (HSE) profile. Although dithiophosphinates received little attention in the past in the flotation research community, recently, there has been significant interest in this chemistry, and numerous studies have been conducted to understand the fundamental interactions of dithiophosphinates with several base metal sulfides and precious metals. A critical overview of fundamental studies, laboratory ore flotation, and plant practice of dithiophosphinates is given in this paper. The findings from these studies are rationalized and explained by drawing on concepts and insights from coordination chemistry. A comparison between the interactions with various minerals and metals of dithiophosphinate and dithiophosphate analogues brings out some of the subtleties that seemingly minor changes in collector chemistry can make. The critical overview and analysis given here is useful in identifying gaps in knowledge regarding this important class of compounds. This in turn can provide a basis for the design of studies to advance our understanding of dithiophosphinate flotation performance.

Keywords

Dithiophosphinates AEROPHINE 3418A Dithiophosphates Coordination chemistry Electrochemistry Adsorption Flotation Sulfide minerals Sulfide ores Cu ores Cu-Au ores Precious metals ores Polymetallic ores 

1 Introduction

Dialkyl dithiophosphinates are an important member of the class of thiophosphorus compounds that are widely used in flotation beneficiation of base metal sulfide and precious metals ores. Ore flotation studies and plant practice experience have demonstrated their unique properties compared to the traditionally used sulfide collectors such as dialkyl dithiophosphates (their closest analogues) and xanthates. The predominant dialkyl dithiophosphinate used in the industry today is sodium diisobutyl dithiophosphinate (a.k.a. DIBDTPi; AEROPHINE® 3418A1 introduced by American Cyanamid in the late 1970s; Fig. 1). The widest application of AEROPHINE® 3418A is found in complex polymetallic ores of Cu-Pb-Zn-Ag, Pb-Zn-Ag, and Cu-Zn where it has demonstrated improved recoveries (and flotation rates) of sulfide minerals of Cu, Pb, and Ag and selectivity against sphalerite, iron sulfides, and penalty elements [1]. More recently, it has also been used as a collector in the processing of Cu, Cu-Ni, and Ni ores [2] and Cu-Au ores [3]. In general, it is considered to possess the collecting power of xanthates together with the selectivity (against pyrite) of dithiophosphates [4, 5], while requiring lower dosages than those reagents [2]. AEROPHINE® 3418A has also found an important place in several targeted formulations to improve flotation performance. Other benefits of this reagent include higher flotation rates, having a better HSE (health, safety, and environmental) profile2 compared to xanthate, and exhibiting resistance to hydrolysis and oxidation [2]. Compared to diisobutyl dithiophosphate (DIBDTP), which is known to contribute significantly to frothing (often leading to over-frothing), AEROPHINE® 3418A is observed to be froth neutral (i.e., it does not alter the existing froth characteristics). As is the case for other flotation reagents, the design and application of AEROPHINE® 3418A has preceded understanding of its behavior at a fundamental level. Recently, there have been numerous fundamental and application studies designed to develop a better understanding of its interaction with sulfide minerals and precious metals. This should help in optimizing its application in flotation and provide a basis to better rationalize its unique selectivity and activity characteristics. Past studies on the interaction of DIBDTPi and DTPs with sulfide minerals and precious metals do not offer any explanation for the observed differences between these two ligands, and a discussion as to the origin of these differences is still lacking. The goal of this paper is to provide a critical overview of the chemistry of dithiophosphinates and dithiophosphates, their interactions with metals and sulfide minerals, and to discuss the origin of the unique properties of dithiophosphinate by drawing from information available on coordination chemistry.
Fig. 1

Structures of sodium diisobutyldithiophosphate (NaDIBDTP), sodium diisobutyldithiophosphinate (NaDIBDTPi), and the structure of the disulfide (aka, dimer)

2 Discussion

2.1 Overview of Dithiophosphinate and Dithiophosphate Chemistry

Dialkyl (or diaryl) dithiophosphinates are structurally similar to dithiophosphates. However, in dialkyl dithiophosphinates, the substituents are attached directly to the central phosphorus atom, which has a profound effect on its chemical properties and interactions with metals and minerals relative to dithiophosphate for which substituents are attached through oxygen atoms (see Fig. 1). Interactions of thiophosphorus compounds with metal and metal sulfides can be partially understood in terms of hard-soft acid-base concepts, where the soft sulfur donors will tend to bind to soft acceptor metal ions or metal sites on minerals. In a dithiophosphorus anion, the softness of the donating sulfur atoms is related to its effective polarizability which, in turn, determines its selectivity towards a given metal ion. The polarizability of the sulfur atoms of the functional group is affected by the nature of the substituents attached to the central P atom through inductive effects as well as resonance or mesomeric effects (delocalization of electron density) [6, 7, 8]; there is more discussion in a later section. Although reasonably stable especially under conditions relevant to flotation, in general, thiophosphorus compounds are known to decompose under strong oxidizing and hydrolytic conditions, leading to the conversion of both P=S and SH groups to their O-analogues and generation of toxic H2S gas [9]. However, dithiophosphinic acids exhibit far greater hydrolytic stability than the dithiophosphoric acids under a wide range of conditions, both mild and harsh; diethyl dithiophosphoric acid decomposes completely after 20 h in 15 N H2SO4 [10], whereas diethyl dithiophosphinic acid stays practically unchanged for weeks in 15 N H2SO4 [11]. The greater hydrolytic stability of dithiophosphinic acids allows them to be used commercially in solvent extraction of metal species in acidic solution [12, 13]. Both dithiophosphinic acids and dithiophosphoric acids oxidize readily to disulfides in the presence of nitric acid, peroxide, bromine, and iron (III) [14]. Reliable data on rest potentials for oxidation in alkaline solutions, especially under conditions relevant to flotation, are not available. Similarly, reliable data for hydrolytic stability under conditions relevant to plant practice, namely in the alkaline pH range 8–12, are not available.

2.2 Fundamental Studies on Dithiophosphinates and Related Chemicals

2.2.1 Adsorption, Surface Chemical Speciation, and Flotation Studies

Recently, several studies on the fundamental aspects of interaction of dialkyl dithiophosphinates with metals, metal ions, and minerals, as well as the nature of the species formed on the metal/mineral surfaces have emerged, with the goal of gaining insight as to reasons for their unique performance in ore flotation. The methods used in these studies include (i) spectroscopic techniques: Raman, surface-enhanced Raman, infrared spectroscopy, time-of-flight secondary ion mass spectrometry (ToF-SIMS); nuclear magnetic resonance (NMR) spectroscopy, (ii) electrochemical techniques: voltammetry and open circuit potential measurements, (iii) microflotation, and differential adsorption onto mineral particles, and (iv) contact angle measurements. The following is a discussion providing a summary of the results of these investigations.

2.2.2 Galena, Chalcopyrite, Pyrite, and Sphalerite Systems

The selectivity of AEROPHINE® 3418A for galena and pyrite particles, extent of adsorption, and effects of surface contamination by metal ions were investigated by Piantadosi and Smart [15] using ToF-SIMS on both concentrate and tails particles. While most fundamental studies focus on single minerals in isolation, this study is unique in that it used a 30:70 mixture of galena and pyrite, ground and floated together, thus allowing for competitive adsorption that exists in ore flotation systems. It was observed that the amount of collector adsorbed on galena was three times that adsorbed on pyrite in the concentrate. The major contribution to efficient galena flotation was from hydrophobic DIBDTPi and Pb-DIBDTPi complexes. The observed high flotation recovery of pyrite was attributed to inadvertent activation of pyrite by Pb ions (as much as 50% of the lead released from galena transferred to pyrite); this observation is in agreement with the results obtained by Pecina-Treviño et al. [16, 17, 18]. In ore flotation, it is a common practice to use modifiers to enhance separation efficiency; one of the functions of the modifier may be to mitigate inadvertent activation. The selectivity of DIBDTPi (AEROPHINE® 3418A) vs. sodium diethyl dithiocarbamate and ammonium dibutyl dithiophosphate was demonstrated for galena against sphalerite and chalcopyrite against pyrite through single mineral flotation and adsorption capacity experiments as a function of pH [19]. This particular study is noteworthy in that the single-mineral studies were validated by ore flotation.

The role of pulp potential on the flotation of complex polymetallic and precious metal ores has also been assessed for DIBDTPi [20, 21, 22]. Gorken et al. [20] considered the flotation performance of a Pb-Zn ore at varying pulp potentials, namely, 119 mV, 129 mV, 197 mV, and 206 mV (vs. SHE).3 Galena flotation from the lead ore was best at potentials more positive than 140 mV; however, the onset of galena flotation occurred at much lower potentials. The flotation performance of a Cu-Zn-Ag-Fe massive sulfide ore at varying potentials was also probed, where it was observed that copper recovery and concentrate grade were the highest at 60 mV indicating higher selectivity for copper against iron. Uribe-Salas et al. [22] studied the role of pulp potential on the recoveries of sulfides from a fine-grained massive, Zn-Pb-Cu-Ag-Au sulfide ore. For all metals, in both continuous and batch flotation, the recoveries clearly peaked at potentials in the range of 310–360 mV. In view of the fact that multiple collectors were used, it is difficult to learn the particular contribution from DIBDTPi. Another complication is that in ore flotation studies, the pulp potential can only be controlled by the addition of redox chemicals, in contrast to model single-mineral-electrode systems for which the desired surface potential can be imposed directly. Use of chemicals to control pulp potential (which may not be the same as the mineral surface potential) will undoubtedly create a complex solution and mineral-solution interfacial chemistry which confounds electrochemical and chemical factors. Consequently, often results from single-mineral studies do not translate to ore flotation [23, 24, 25, 26]. Therefore caution must be used in attempting to apply results from mineral electrode experiments to predict behavior in ore flotation.

Pecina-Treviño et al. [27] used open circuit potential (OCP) measurements and cyclic voltammetry together with wettability measurements to study the interaction of DIBDTPi with galena and pyrite. Their results showed that DIBDTPi passivated the galena surface independent of pH and dissolved oxygen concentration, indicating a chemical mechanism for adsorption of DIBDTPi with a strong affinity for galena. It was suggested that the species formed at the surface was a 1:2 lead-dithiophosphinate surface complex (Pb (DIBDTPi)2); no mention was made of a 1:1 Pb-DIBDTPi complex as part of a mechanism involving an electrochemical step (oxidation of the galena metal site) and a chemical step (exchange between hydroxyl ions and collector ions at the surface). Though not mentioned in this specific work, the formation of a 1:1 Pb-DIBDTPi complex on galena has been demonstrated by ToF-SIMS [15, 28]. The same investigators [16] demonstrated, by microflotation and cyclic voltammetry, that DIBDTPi was more selective against copper- and lead-activated pyrite when compared to isopropyl xanthate. According to the authors, lead activation favored the oxidation of xanthate to dixanthogen on pyrite whereas such oxidation was deemed to be less likely for DIBDTPi as supported by the higher oxidation potential of DIBDTPi observed through cyclic voltammetry. The oxidation of DIBDTPi at potentials of 500–600 mV (vs. SHE) in turn lead to hydrophobization of the pyrite surface as evidenced from contact angle measurements at said potential. The species formed, however, could not be identified but speculated to be a dimer in analogy with isopropyl xanthate. The authors therefore reasoned that the relatively higher propensity of isopropyl xanthate to dimerize as compared to DIBDTPi may account for observed selectivity against pyrite. On the other hand, a slight increase in lead-activated pyrite recovery was observed with DIBDTPi and ascribed to the formation of some sort of metal-collector compound, presumed to be a Pb (DIBDTPi)2 complex, based on zeta potential and cyclic voltammetric measurements on galena. This complex did not seem to be present at the surface to an extent that would result in an appreciable passivation of current as evidenced by the indistinguishability of voltammograms generated with unactivated or activated pyrite.

For pyrite, electrochemical and wettability experiments suggested a weak interaction of DIBDTPi with the mineral [17], leading to the conclusion that the collector did not interact chemically with the surface and that pyrite only served as an electron sink for the oxidation of the collector to its dimer. These findings are consistent with those from ToF-SIMS analysis, which showed that the iron-collector complex was virtually absent from the surface, but dithiophosphinate (presumably as dimer) was present [29]. The presence of adsorbed dithiophosphinate and its dimer on pyrite was identified using diffuse reflectance infrared spectroscopy (DRIFT) [30]; adsorbed collector amount decreased significantly in neutral to alkaline conditions as was independently observed through adsorption measurements [19]. In contrast, Hicyilmaz et al. [31, 32] using cyclic voltammetry, contact angle and DRIFT measurements, concluded that dithiophosphinates interacted chemically with pyrite, as opposed to electrochemically, and that the species formed on the surface consisted of dithiophosphinate dimers and the ferric dithiophosphinate complex. These conclusions are difficult to rationalize together since the formation of the dimer (and possibly Fe complex) involves electron transfer. Caution must be exercised when drawing conclusions from DRIFT results in view of the fact that this technique is not sufficiently sensitive, often requiring the use of very high concentrations of collector to observe signals. Also, spectra of multiple species may be difficult to deconvolute. Nevertheless, there appears to be a consensus for the presence of a dimer on pyrite, but not for Fe complex, though such Fe complexes have been suggested in other collector systems. SIMS studies for dithiophosphate adsorption on pyrite showed evidence for an iron-hydroxy complex, Fe (OH) DTP on pyrite [33]. Iron-hydroxy xanthate complexes have been suggested in the case of xanthate adsorption on pyrite [34, 35]. Thus the presence of iron complexes cannot be entirely ruled out for DIBDTPi on pyrite; perhaps it forms under certain conditions yet unidentified.

The adsorption mechanism of DIBDTPi and potassium diethyl dithiophosphate (DEDTP) on chalcopyrite was compared by conducting wettability measurements under potential control [36] and showed that the hydrophobicity of chalcopyrite with DIBDTPi was greater than that with DEDTP in a wide range of potentials for all pH values tested. By analogy with interactions between DEDTP and chalcopyrite [37], it was speculated that formation of DIBDTPi dimer occurred and that it was responsible for the enhancement of hydrophobicity. Microflotation results indicated that the DIBDTPi was a much stronger collector than DEDTP (though part of this may be due to the difference in chain length). DRIFT spectroscopy at varying potentials and solution pH was used in an attempt to provide evidence for the presence of DIBDTPi dimer [38]. From the spectra, it was inferred that DIBDTPi chemisorbed at the surface as a dimer species together with cupric DIBDTPi. Formation of cupric DIBDTPi rather than the expected cuprous complex is surprising since with thiol and other S ligands, the cuprous complex is the most stable and favored, and the cupric complex formed initially will change over to the cuprous complex via a redox process with the concurrent formation of ligand dimer. The cautionary statement about DRIFT mentioned above applies here also.

Table 1 provides a summary of the key findings from the above-described studies.
Table 1

Summary of findings from fundamental studies on DTPi interaction with mineral surfaces by adsorption, wettability, voltammetry, infrared spectroscopy, and flotation [17, 18, 19, 31, 32, 36, 38]

Technique

Galena

Chalcopyrite

Pyrite

Voltammetry/OCP*

OCP lower and independent of pH

Significant passivation

Peak for disulfide; no change in OCP

Adsorption

Strong; independent of pH

Strong; independent of pH

Weak, especially at pH > 7

Contact angles

85° in the range of − 465 mV to 645 mV

Greater than that of DTP at − 100 mV to 400 mV

Lower than that of DTP at -100 mV to 400 mV

Surface species

Chemisorption and Pb complex

Cu complex and disulfide dimer

Disulfide (and Fe complex)

Single mineral Flotation

Strong; independent of pH

Strong; independent of pH

Poor

Ore flotation

High recovery and grade

High recovery and grade

Poor

*OCP is open circuit potential

Bagci et al. [39] studied the adsorption behavior of mixtures of DIBDTPi and sodium isopropyl xanthate (SIPX) by differential adsorption measurements and voltammetry. It was shown that for a given ratio of DIBDITPi to SIPX, the collector that was added first would dominate the adsorption and would always adsorb to a larger extent compared to the collector added second. This was interpreted to reflect the preferential adsorption of the first-added collector at high activity sites, thus only leaving the less active sites available for the collector that was added subsequently. It should be pointed out that the concentration used in electrochemical experiments was three orders of magnitude larger than that used in differential adsorption, perhaps obscuring effects of collector activity on adsorption extent under conditions relevant to flotation. Although DIBDTPi is a more selective reagent than SIPX, its extent of adsorption on chalcopyrite was higher when added as the sole collector. Interestingly, it was observed that synergistic adsorption took place at the given ratios of both collectors whereby the total amount of collector adsorbed was higher than that of either collector added on its own. As a plausible scenario, the authors proposed the adsorption of dixanthogen onto a previously formed DIBDTPi layer through van der Waals interactions. Unfortunately, the kinetics of the competitive adsorption between collectors was not explored.

The only study that does not agree with any of the other studies or with plant experience was by Ignatkina et al. [40]. They compared rate and extent of adsorption of AEROPHINE® 3418A onto pyrite, chalcopyrite, pyrrhotite, and sphalerite with that of butyl- and isobutyl- xanthate, and dithiophosphate at pH 9 and concluded that AEROPHINE® 3418A adsorbed slower onto all minerals and also to a lesser extent on all minerals except pyrrhotite, and that adsorption rate constant for pyrite was greater than that for other sulfides. In single-mineral flotation tests, AEROPHINE® 3418A was only slightly more selective than butyl xanthate and less selective than the other collectors. Based on these results, it was concluded that AEROPHINE® 3418A was not a selective collector in copper-zinc ores. It is not possible to rationalize these peculiar and contradictory findings and conclusions. There are many inconsistencies and deficiencies in this study with respect to the experimental procedure, results, and interpretations, all of which make this study and the conclusions therefrom highly unreliable.

2.2.3 Silver, Gold, Platinum, and Copper

The interaction of AEROPHINE® 3418A with gold, silver, and gold-silver alloys has been investigated using X-ray photoelectron spectroscopy, infrared spectroscopy, and voltammetry [4, 5, 41, 42]. Hope et al. [5], using surface-enhanced Raman spectroscopy (SERS) and voltammetry, concluded that on gold, the dithiophosphinate dimer formed above 540 mV; Au complex formed at much higher potentials. On silver, the bulk Ag-dithiophosphinate complex was present, characterized by an anodic current above 240 mV, which was supported by SERS (after ex situ polarization of Ag at potentials > +410 mV in the presence of DIBDTPi). This is in agreement with Farinato and Nagaraj [43] who suggested the formation of a Ag-DIBDTPi layer on silver surfaces at open circuit potential (≈ 340 mV). No current due to chemisorption on Ag was observed below 40 mV, in contrast to xanthate on silver which shows a pre-wave (underpotential adsorption) before the onset of the signal due to the metal-thiolate. Such pre-waves have been observed for xanthate on argentite, lead, galena, copper, and chalcocite [44] and for dithiophosphate on chalcocite [45]. SERS spectra, however, showed evidence for chemisorbed DIBDTPi at potentials down to −1 V (vs. SHE).

Hope et al. [5] offered possible reasons for the absence of a pre-wave for DIBDTPi on silver: (a) low rate of chemisorption, (b) coverage of the chemisorbed species was perhaps too low for detection, and (c) chemisorption occurred at highly negative potentials, below the negative limit of the scan. This last reason is supported by the observation that hydrogen evolution on silver was strongly inhibited in the presence of DIBDTPi even before the initial positive-going scan was initiated, which indicates that a DIBDTPi species was present at the interface. These findings are again consistent with the fact that DIBDTPi adsorption (and the species formed) on silver vary within a wide range of potentials.

On Cu metal, SERS spectra show bulk Cu-DIBDTPi formed after ex situ polarization at potentials > − 190 mV. No electrochemical or SERS evidence of chemisorbed DIBDTPi was observed, as would be indicated from the appearance of a voltammetric pre-wave or a set of discernible SERS bands, respectively. It was therefore concluded that perhaps significant chemisorption of DIBDTPi on Cu does not occur at underpotentials and that formation of bulk Cu-DIBDTPi is more thermodynamically favorable in turn. The adsorption of dithiophosphate on Cu was shown to be similar. It was also emphasized that the adsorption of DTP is in general not the same on Cu metal as it is on copper sulfides, in that an electrochemical chemisorption pre-wave appears for the sulfide but is not present for Cu metal, as reported previously by Chander and Fuerstenau [37]. However, underpotential chemisorption of DIBDTPi or DIBDTP on Cu cannot be ruled out. Absence of a SERS signal for the chemisorbed DIBDTPi does not preclude its presence since Cu exhibits only weak surface Raman enhancement. Also, reduction of cathodic currents due to passivation at − 800 mV, as was observed for Ag, cannot be ignored.

In the case of Au, even after extended ex situ polarization at 650 mV, the major species identified was (DIBDITPi)2 with a small amount of AuDIBDTPi also forming. In this case, it was rationalized that by analogy with the dimerization behavior of xanthate, extended polarization is required due to the requirement for an adsorbed intermediate in dimer formation.

For Au-Ag alloys, FTIR indicated the presence of Ag-DIBDTPi complex; the DIBDTPi dimer is formed along with the Ag-DIBDTPi complex at high potentials. For Au, the dimer was present only at high potentials [42]. FTIR and contact angle results were in excellent agreement with the voltammetry results.

There is limited literature on the interactions of dithiophosphorus and dithiophosphinic acids with platinum group metals. Chanturiya et al. investigated the interaction of DIBDTPi with Pt ions in solution and Pt metal coated on a galena mineral powder surface [46]. The metal complex formed when DIBDTPi solution was contacted with hexachloroplatinic acid or sodium hexachloroplatinate at 1:1 and 1:2 ratios of Pt to ligand was characterized to be a Pt (DIBDTPi)2 complex (note: structure shown in the paper is incorrect) wherein Pt was in +2 oxidation state. This complex formation proceeded via reduction of Pt (IV) to Pt (II); a reducing agent was used to accelerate formation of the complex. In order to study interaction with a Pt surface, the authors conditioned galena powder (− 150 + 75 μm) in hexachloroplatinic acid solution for 52 h and the galena surface was characterized by SEM-EDX, which showed Pt patches on galena grains. Two types of Pt species were proposed from SEM data—Pt metal and Pt sulfide. These coated particles were then contacted with 0.1% DIBDTPi solution at pH 7.5 for 3.5 min. Again, based on SEM-EDX analysis of the mineral particles and TLC analysis of the carbon tetrachloride extract of the surface species, the authors concluded that the surface species was the Pt (DIBDTPi)2. There are several problems with this study. SEM-EDX is not strictly a surface-analysis technique. It samples about a micron-thick layer of mineral. Pt coating on galena particles occurred in patches thereby leaving uncoated galena surfaces exposed which are expected to interact with DIBDTPi. The authors mentioned, however, that there was no interaction with galena, which is at odds with other fundamental studies and plant observations. It is quite possible that the Pt (DIBDTPi)2 complex observed in SEM/EDX was in fact the complex formed as a result of the interaction of DIBDTPi and residual hexachloroplatinic acid that remained with the galena particles. There are other issues as well. In view of these, the results of the study are questionable.

Waterson et al. [47] conducted voltammetry experiments as part of a larger computational chemistry study to investigate interactions of ethyl xanthate, diethyl dithiocarbamate, and DIBDTPi with Pt, pentlandite, and sperrylite. Voltammograms were characterized by a prominent anodic peak attributed to the oxidation of the ligand molecule at the surface. While the peaks for xanthate and dithiocarbamate on all surfaces were distinct, those for DIBDTPi were “less clear-cut” (not distinct) for pentlandite and sperrylite. It is possible that (underpotential) chemisorption of DIBDTPi on the substrates is responsible for this possible obscuration.

Table 2 summarizes the findings of all the studies discussed above.
Table 2

Summary of identified DTPi surface species on metals

Interaction

Cu

Ag

Au-Ag

Au

Pt

Chemisorption

Yes (?)

Yes

Yes

No

?

Chemisorption (underpotential)

No

No

No

No

No

Surface complex

Cu-DTPi

Ag-DTPi

Ag-DTPi

Au-DTPi at high potentials

?

Dimer (disulfide)

?

?

Yes

Yes at high potentials

Yes at high potentials

2.2.4 Surface Coordination Modes

One relevant, but often overlooked, aspect of dithiophosphorus anion adsorption on minerals is determination of coordination modes with the surface lattice. For example, chemisorption can result in at least three modes of coordination with the surface: monodentate, bidentate (or terminal chelating), and bridging (see Fig. 2). Note that a surface complex such as 1:1 metal-ligand, observed for Cu and Ag both in solution and on mineral surfaces, and which could be oligomeric species, is not represented in Fig. 2. Larsson et al. [48] determined, by comparison with reference 31P CP/MAS (Cross Polarization—Magic Angle Spinning) NMR spectra of solid state metal complexes, that dithiophosphate (DTP) ions of varying alkyl chains chemisorbed onto synthetic galena by terminal chelation with single lead metal sites. No DTP dimer was detected at the surface, consistent with other reports on DTP species on galena. A similar analysis using 31P CP/MAS NMR has not been carried out for dithiophosphinates, and it would be valuable to characterize using this technique the surface species formed with DIBDTPi, and compare these with findings from other techniques (Raman, SERS, and FTIR).
Fig. 2

Adsorption modes for DIBDTP on PbS (adapted from [48])

The mechanisms of adsorption of dithiophosphinates and dithiophosphates on mineral or metal surfaces, the electrochemical and chemical processes involved, as well as the nature of the species formed, are very much dependent on a variety of chemical factors, conditions used, the chemical state of the surface, and the stability of the metal-collector complexes formed. Knowledge of adsorption per se does not carry any information on the hydrophobicity of the surface. Although DIBDTPi and DIBDTP are structurally similar and may share general features of adsorption, there are considerable differences in the selectivity, properties of the complexes and surface species, and certainly flotation outcome. While the studies discussed above have highlighted such differences, they do not shed enough light as to their origin. The following section is an attempt to discern the origin of these differences by drawing on existing knowledge of physico-chemical properties and coordination chemistry.

2.3 Coordination Chemistry of Dithiophosphorus Compounds

2.3.1 Acid Dissociation Constant (pKa)

In dithiophosphates, the electron-withdrawing effect of the electronegative oxygen atoms makes the sulfur atoms less polarizable, thereby decreasing their softness character, bond strength with soft cations, and the pKa. A review of the literature on organophosphorus compounds reveals that there is much scatter and disagreement among reported pKa values, though trends in pKa resulting from changes in substituents are similar. Kabachnik et al. [49] used Hammett-type free energy relationships to correlate effects of substituents at the phosphorus atom (as characterized by their corresponding electronic substituent constants, e.g. σ-parameter) with the ionization constants of both monothio- and dithiophosphorus acids. They showed that monothio acids were very sensitive to the nature of the substituents while dithio acids were not. SciFinder® offers a prediction tool powered by Advanced Chemistry Development software (v. 11.02; ACD/Labs) that allows one to estimate pKa of organic compounds based on Hammett-type analyses. This software has been tested against other available predictors and was deemed to be acceptable in estimating pKa values [50]; however, discrepancies are observed for thiophosphorus compounds. For instance, for the commercial dithiophosphinic acids: CYANEX® 301 (bis(2,4,4-trimethylpentyl) dithiophosphinic acid), CYANEX® 302 (bis(2,4,4-trimethylpentyl) monothiophosphinic acid), and CYANEX® 272 (bis(2,4,4-trimethylpentyl) phosphinic acid), the reported experimental pKa values are as follows: 2.84 ± 0.04, 3.12 ± 0.20, and 3.73 ± 0.1 in one study [51]; and 2.61, 5.63, and 6.37 in another study [52]. For comparison, the values estimated through the SciFinder® predictor tool are 4.45 ± 0.10, 4.45 ± 0.10, and 3.26 ± 0.50, which do not follow the expected order. The discrepancies between predicted and experimental values may be attributed to the difficulty of experimentally measuring pKa values due to their limited aqueous solubility [53] as well as their hydrolytic and oxidative stability and the inadequacy of models for this class of compounds. In fact, significant efforts are still being made to develop predictive models from first principles [54, 55]. Inconsistencies in computed pKa values of CYANEX® 301, CYANEX® 302, and CYANEX® 272 with what is observed experimentally show the limitations of the predictive tool for this set of compounds. By contrast, the calculated pKa of diisobutyl- dithiophosphate and dithiophosphinate are − 0.07 ± 0.34 and 4.45 ± 0.10, respectively (as per SciFinder®’s predictor tool) which are consistent with the relative acidity observed experimentally for these compounds; however, the validity of these numbers is not known.

In summary, there are no reliable data for pKa of dithiophosphoric or dithiophosphinic acids; nevertheless, altogether, as shown in Table 3, available experimental data seem to support the higher acidity of dithiophosphinic acid versus phosphinic acids and their weaker acidity compared to the corresponding dithiophosphoric acids [56]. Given the connection between pKa, metal-complex stability in solution [54] and surface complexation/flotation [7], it is no surprise that dithiophosphinates are observed to be more selective than dithiophosphates in flotation, which normally occurs under alkaline conditions.
Table 3

Experimental pKa values of thiophosphorus acids from [56]

Compound

pKa at 25 °C

Diethylphosphoric acid

1.37

Diethylphosphinic acid

3.29

Diethyldithiophosphoric acid

1.62

Diethyldithiophosphinic acid

1.71

2.3.2 Stability Constants of Metal-Ligand Complexes

The degree of selectivity of a complexing (or chelating) agent for a metal atom is closely connected with the stability of metal complexes [7], which is affected by the redistribution of charge on the functional group upon complex formation. DIBDTPi possesses chemical affinity for certain metal atoms from which its selectivity is derived. This concept was demonstrated by Pecina-Treviño et al. [18] through a study of complexation of DIBDTPi with lead and iron ions in solution as a means of gaining insight regarding its adsorption behavior on the corresponding mineral surfaces and macroscopic behavior observed in industrial flotation operations. Their results revealed that DIBDTPi had a more pronounced affinity for lead species than for iron; lead ions were completely consumed through complex formation whereas iron ions remained in solution. This translated into a pronounced affinity of DIBDTPi for galena and little affinity for pyrite; the molar concentration of DIBDTPi required to achieve a floatability of close to 100% with galena was two orders of magnitude lower than that for pyrite.

Although a reasonable correspondence may often exist in complexation with metal atoms in solution (characterized by the stability constants β1, β2, etc. depending upon the number of ligands attached to metal), complexation on mineral surfaces (for which stability constants are difficult to determine) and flotation outcome, it must be recognized that the process and extent of adsorption are far more complicated and may involve several different modes and species, and different parking areas. Also, thermodynamic stability constants do not provide any information on rate of complexation or adsorption; such kinetics information, which is important for flotation applications, is not readily available. In some cases, the extent of adsorption may not show the expected correspondence with flotation outcome, which relies on differential wettability of surface films and subsequent fruitful and stable attachment to, and transport by, bubbles. Surface complexation is believed to be dependent on the correct orientation of mineral lattice ion orbitals in such a way that steric hindrance between ligands is minimized during bond formation with the surface and coordination is, therefore, highly dependent on surface topography and composition [7, 29]. The softness of the PSS functional group in dithiophosphorus compounds depends on the effective charge on the sulfur atoms. Depending on the nature of the substituents bound to phosphorus, this effective charge and “softness” can be tuned [7, 8, 57]. For instance, the oxygen atoms in the RO substituents of dithiophosphates attract electron density from the S atoms. This results in a decrease in softness of the S atoms, sigma bond strength, and pKa, and an increase in π-bonding in the formed complex. Compared to dialkyl dithiophosphates, therefore, the S donors in dialkyl dithiophosphinates are softer (resulting in increased sigma bonding, greater basicity, and lesser π-bonding). As a result, dithiophosphinates would have a greater tendency to form stable complexes with softer acceptors (e.g., Cu+ instead of Fe3+). The effect is more pronounced for the later transition metals. In fact, the softness character of organophosphorus compounds increases as (OR)2PSS < R (OR)PSS < R2PSS [8, 58]. These observations for complex formation with metal ions can be extended to ligand binding to sulfide mineral surfaces. The greater softness of sulfur atoms may also reduce DIBDTPi’s affinity for water. Such a reduced water affinity would promote DIBDTPi (relative to DIBDTP) adsorption onto sulfide mineral surfaces by virtue of lessening the enthalpic penalty for dehydrating the ligand portion of the collector, which is a necessary step in surface binding and complex formation in this instance. The relative magnitude of such an effect is not known, but seems worthy of investigation.

Attempts at correlating structure of organophosphorus compound and stability of the metal-complex have been made by the use of Hammett-Taft free energy relationships [49, 59]. Ovchinnikov et al. [59] determined the stability of dithiophosphinate, dithiophosphonate, and dithiophosphate complexes of Ni (II), Co (II), and Ag(I) of varying alkyl substituent structure. The relative order of stability was clearly Ag(I) ≫ Ni (II) > Co (II); also evident was the order R2PSS > R (OR)PSS > (OR)2PSS which is consistent with the findings of Toropova et al. [60] and the relative softness character of the sulfur donors as mentioned previously. For a given metal, stability constant increased with alkyl chain length, and for a given chain length, higher stability correlated with higher branching. Determination of Cu (II) complex properties was not possible given their tendency to undergo redox reactions to form Cu(I) compounds (as expected from the greater softness of S donors in R2PSS).

Stability constants in aqueous solution for complexes between dithiophosphorus compounds and metal ions relevant to flotation are mostly unavailable due in part to difficulties in their measurement as a result of either the low solubility of the complexes or their instability in media of high dielectric constant [60, 61]. Thus, the crucial information on reliable metal-complex stability constants for dithiophosphinates in aqueous media is not readily available for many metals of interest to flotation such as Cu, Pb, Zn, or Fe. A thorough review of the literature revealed that all such information for organophosphorus-metal complexes is scant, scattered, and oftentimes difficult to acquire. The paucity of specific stability constant data was made obvious by an exhaustive review of authoritative databases such as the “Critical Stability Constants” database of NIST and the Mini-SC-Database by IUPAC, where only a limited number of relevant compounds were found and an even smaller number of specific metal-complex data were present. Nevertheless, from the limited data available, it is apparent that for a given dithiophosphorus ligand, the order of stability is Ag(I) > Pb (II) > Cd (II) > Zn (II) ~ Ni (II) > Co (II) and that, for a given ion, dithiophosphinate ligands are in general more stable compared to their dithiophosphate analogues. These trends are supported by observation in flotation practice as well as in fundamental studies as discussed in the previous sections. Table 4 summarizes the key chemical differences between DIBDTPi and DIBDTP.
Table 4

Key differences between DIBDTPi and DIBDTP

Property

DIBDTPi

DIBDTP

pKa

3 to 4 (weaker acid)

1 to 2 (stronger acid)

Hydrolytic stability

Higher

Lower

Oxidative stability*

Higher

Lower

Desolvation of S

Higher

Lower

Metal Complexation

Soft acids but more selective

Soft acids but less selective

pK and Ksp

Higher stability constants, more hydrophobic, lower solubility

Lower stability constants, less hydrophobic, higher solubility

Organic group density

Smaller

Larger

Parking area

Smaller

Larger

*Propensity for oxidation on the mineral/metal surface will be quite different from that in solutions. Oxidation of collector is less likely and may occur at very high potentials when there is underpotential chemisorption as is noted for DIBDTPi

†Metal-complex stability constant

‡Solubility product

2.3.3 Computational Chemistry

Insight regarding the relative affinity of organophosphorus acids for given metals and minerals can be obtained with the aid of computational chemical tools. For example, in more recent studies, quantum mechanical calculations which take into account the presence of an aqueous medium have been used to calculate properties of organophosphorus acids such as their pKa and structure-dependent electron distributions as well as to rationalize their interaction with mineral and metal surfaces [53, 62].

Liu et al. [62] evaluated diisobutyldithiophosphinate (DIBDTPi), monothiophosphinate (DIBMTPi), dithiophosphate (DIBDTP), and monothiophosphate (DIBMTP) in both acid and ionic form on the basis of quantum-chemical parameters such as frontier molecular orbital energies, bond lengths, and dipole moments. Based on atomic charges of the donating group X–(P=S), where X=S or O, and the composition and energies of occupied and unoccupied molecular orbitals, they reached the conclusion that the reactive power of these compounds for mineral surfaces of Cu, Au, Ag, and Pb can be rationalized on the basis of the electron back-donating ability of these latter ions through their d-orbitals as well as the differential energetics between collector frontier orbitals and mineral bands. For example, frontier orbitals of the collectors donate their electrons to the lowest unoccupied molecular orbital (LUMO) of the mineral or metal (i.e., its conduction band) to form a sigma bond and the metal ions in the mineral can donate their d-electrons to the collector’s LUMO to form a dative pi-bond. The higher electron-donating and electron-accepting ability of DIBDTPi ranked it as the strongest collector of all four compound types where collecting strength followed the order DIBDTPi ≫ DIBMTPi ≥ DIBDTP > DIBMTP. In addition, selectivity against iron- and zinc-minerals (sphalerite in the case of zinc) compared to metal ions such as Cu, Au, Ag, and Pb was explained through the above-described concept. For instance, due to their electron configuration, Fe (III) ions can only accept frontier orbital electrons from the collector to form a sigma bond, with no back-bonding ability thus forming weaker bonds compared to say, Pb ions; on the other hand, sphalerite has one of the highest band gaps among the typical sulfide minerals and thus electrons donated from the collector to the mineral LUMO have to overcome a higher energy barrier presumably leading to a weaker mineral-collector interaction. On this basis, it was deduced that the order of selectivity would be DIBDTP < DIBDTPi < DIBMTPi < DIBMTP which is in fair agreement with what is known of the compounds in general.

Benson et al. [53], using DFT to predict structure and pKa values for a series of substituted diphenyldithiophosphinic acids, concluded that the trends in pKa values were reasonable although the absolute values could not be confirmed without experimental validation.

Finally, Waterson et al. [47] conducted computational and experimental studies of the binding of ethyl xanthate, diethyldithiocarbamate, and DIBDTPi ligands to sperrylite (PtAs2), pentlandite, and platinum metal. Binding energies calculated from first principles showed that binding of all ligands studied was least favorable for sperrylite, followed by pentlandite. All ligands showed the strongest binding affinity for platinum.

3 Conclusions

Much progress has been made in fundamental studies to understand the mechanisms of interaction of dithiophosphinates and dithiophosphates with many mineral/metal surfaces and the correlation with flotation results. The donor-acceptor paradigm goes a long way to rationalize the chemical behavior of dithiophosphates, dithiophosphinates, and other collectors in sulfide mineral systems. Fundamental and computation studies support this assertion. However, though powerful and insightful, computational tools should serve only as guides for molecular design and not as reliable predictors of flotation.

There are still gaps in our understanding and a few questions remain concerning the identity of dithiophosphinate species present at surfaces, quantifying the relevant physico-chemical parameters, as well as determining which species are responsible for conferring the hydrophobic properties needed for flotation. A few studies have even reached opposing conclusions; therefore, there is much ground to cover in this respect.

Empirically, diisobutyl dithiophosphinate (AEROPHINE® 3418A) has been demonstrated to be a strong collector for sulfides of Cu, Pb, and Ag, with selectivity against gangue sulfides, which has also been seen in fundamental studies of their interaction with minerals. In comparison to xanthates and dithiophosphate (its closest analogue), dithiophosphinate possesses a more attractive safety, health and environmental profile while, in many respects, demonstrating better performance in flotation with respect to selectivity against gangue sulfides and collecting strength.

Except for a few examples, outstanding omissions in fundamental studies are (a) kinetics of adsorption and electrochemical processes for dithiophosphinates interacting with mineral surfaces and how these compare to other collectors, (b) competitive adsorption and electrochemical processes in a multi-mineral system (as found in ore systems) and how they compare with other collectors, (c) studies on DIBDTPi adsorption and electrochemistry on commercially important minerals (chalcocite, covellite, bornite, cubanite, pentlandite, penalty element sulfides, activated, and unactivated sphalerite); (d) more systematic ore flotation studies; (e) a better understanding of the connection between mineral wettability changes and collector adsorption, surface speciation and interface coverage. In addition, there is a lack of information on key properties for organophosphorus compounds, such as reliable stability constants of their complexes with various metals and reliable pKa values, as well as methods to readily measure or estimate them, particularly with respect to their application in flotation. This would perhaps lead to further understanding of the greater selectivity exhibited by dithiophosphinates as well as the differences in selectivity between monothio- and dithio-phosphates and the corresponding phosphinates for given metals and mineral surfaces, thus providing a basis for rationalizing flotation performance. Unfortunately, as it stands, we have insufficient understanding to be able to rationalize the flotation properties of dithiophosphinates observed in plant practice.

Footnotes

  1. 1.

    DIBDTPi and AEROPHINE® 3418A are often used interchangeably in this paper; we have attempted to use the appropriate term wherever possible.

  2. 2.

    Based on Safety Data Sheet: For AEROPHINE® 3418A, the oral and dermal exposure limits are 3.35 and 5 g/kg respectively; those for dithiophosphates are 4.5 and 2 g/kg; for potassium amyl xanthate, the oral exposure limit is 1.2 g/kg. AEROPHINE® 3418A is not a water pollutant. It does not release any toxic gases in flotation use, unlike xanthate which can generate CS2 and COS due to hydrolysis. AEROPHINE® 3418A has an NFPA rating of 2 (health), 1 (fire), and 0 (reactivity)—all of which are low. Potassium Amyl xanthate has a rating of 3, 4, and 0, and diisobutyl dithiophosphate 3, 2, and 1.

  3. 3.

    All potentials reported hereafter are in reference to the standard hydrogen electrode. Therefore, for any given work described, values cited in terms of other references were converted accordingly.

Notes

Acknowledgements

We thank the management of Solvay for their support and permission to publish this work.

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest.

References

  1. 1.
    Mingione PA (1991) Performance characteristics and application of Aerophine 3418A promoter for base and precious metals flotation. In: SME Annual Meeting, Denver Colorado, February 25–28. Society for Mining, Metallurgy, and Exploration Inc,Google Scholar
  2. 2.
    Guitard E, Bruce T, Bruey F, Nagaraj DR, Riccio P, Thomas W (2015) AEROPHINE® 3418A Promoter – the Canadian collector – 50 years of improved metallurgy in various applications. In: Blatter P, Zinck J (eds) 47th Annual Meeting of the Canadian Mineral Processors, Ottawa, Canada, 20–22 January 2015. Canadian Institute of Mining Metallurgy & Petroleum, pp 255–270Google Scholar
  3. 3.
    Dunne R (2005) Flotation of gold and gold-bearing ores. In: Adams MD, Wills BA (eds) Developments in mineral processing, vol 15. Elsevier, pp 309–344. doi: https://doi.org/10.1016/S0167-4528(05)15014-5
  4. 4.
    Hope GA, Woods R, Boyd S, Watling K (2003) A spectroelectrochemical investigation of the interaction of diisobutyldithiophosphinate with copper, silver and gold surfaces: I. Raman and NMR spectra of diisobutyldithiophosphinate compounds. Colloids Surf A Physicochem Eng Asp 214:77–85.  https://doi.org/10.1016/S0927-7757(02)00383-7 CrossRefGoogle Scholar
  5. 5.
    Hope GA, Woods R, Watling K (2003) A spectroelectrochemical investigation of the interaction of diisobutyldithiophosphinate with copper, silver and gold surfaces: II. Electrochemistry and Raman spectroscopy. Colloids Surf A Physicochem Eng Asp 214:87–97.  https://doi.org/10.1016/S0927-7757(02)00384-9 CrossRefGoogle Scholar
  6. 6.
    Rudzinski WE (1977) Stability of transition metal complexes of O,O′-dialkyldithiophosphates and their adducts. PhD Dissertation, University of ArizonaGoogle Scholar
  7. 7.
    Nagaraj DR (1987) The chemistry and applications of chelating or complexing agents in mineral separations. In: Somasundaran P, Moudgil BM (eds) Reagents in mineral technology, Ch. 9. Marcel Dekker, New York, pp 257–334Google Scholar
  8. 8.
    Ionova G, Ionov S, Rabbe C, Hill C, Madic C, Guillaumont R, Krupa JC (2001) Mechanism of trivalent actinide/lanthanide separation using synergistic mixtures of di (chlorophenyl) dithiophosphinic acid and neutral O-bearing co-extractants. New J Chem 25:491–501.  https://doi.org/10.1039/B006745H CrossRefGoogle Scholar
  9. 9.
    Quin LD (2000) A guide to organophosphorus chemistry. John Wiley & Sons, CanadaGoogle Scholar
  10. 10.
    Bode H, Arnswald W (1962) Untersuchungen über substituierte dithiophosphate II. Bildung der metall-diäthyldithiophosphate und ihre extrahierbarkeit aus mineralsäuren lösungen. Z Analyst Chem 185:179–201CrossRefGoogle Scholar
  11. 11.
    Handley TH, Dean JA (1962) O,O′-dialkyl phosphorodithioic acids as extractants for metals. Anal Chem 34:1312–1315CrossRefGoogle Scholar
  12. 12.
    Rickelton WA, Boyle RJ (1988) Solvent extraction with organophosphines—commercial & potential applications—. Sep Sci Technol 23:1227–1250.  https://doi.org/10.1080/01496398808075627 CrossRefGoogle Scholar
  13. 13.
    Sole KC, Hiskey JB, Ferguson TL (1993) An assessment of the long-term stabilities of Cyanex 302 and Cyanex 301 in sulfuric and nitric acids. Solvent Extr Ion Exch 11:783–796.  https://doi.org/10.1080/07366299308918186 CrossRefGoogle Scholar
  14. 14.
    Schlesinger ME, King MJ, Sole KC, Davenport WG (2011) Production of Cu concentrate from finely ground Cu ore. In: Extractive metallurgy of copper (Fifth Edition). Editor(s): Mark E. Schlesinger, Matthew J. King, Kathryn C. Sole, William G. Davenport, pp 51–71Google Scholar
  15. 15.
    Piantadosi C, Smart RSC (2002) Statistical comparison of hydrophobic and hydrophilic species on galena and pyrite particles in flotation concentrates and tails from TOF-SIMS evidence. Int J Miner Process 64:43–54.  https://doi.org/10.1016/S0301-7516(01)00075-8 CrossRefGoogle Scholar
  16. 16.
    Pecina ET, Uribe A, Nava F, Finch JA (2006) The role of copper and lead in the activation of pyrite in xanthate and non-xanthate systems. Miner Eng 19:172–179.  https://doi.org/10.1016/j.mineng.2005.09.024 CrossRefGoogle Scholar
  17. 17.
    Pecina-Treviño ET, Uribe-Salas A, Nava-Alonso F (2003) Effect of dissolved oxygen and galvanic contact on the floatability of galena and pyrite with Aerophine 3418A. Miner Eng 16:359–367.  https://doi.org/10.1016/S0892-6875(03)00022-0 CrossRefGoogle Scholar
  18. 18.
    Pecina-Treviño ET, Uribe-Salas A, Nava-Alonso F, Pérez-Garibay R (2003) On the sodium-diisobutyl dithiophosphinate (Aerophine 3418A) interaction with activated and unactivated galena and pyrite. Int J Miner Process 71:201–217.  https://doi.org/10.1016/S0301-7516(03)00059-0 CrossRefGoogle Scholar
  19. 19.
    Zhong H, Huang Z, Zhao G, Wang S, Liu G, Cao Z (2015) The collecting performance and interaction mechanism of sodium diisobutyl dithiophosphinate in sulfide minerals flotation. J Mater Res Technol 4:151–161.  https://doi.org/10.1016/j.jmrt.2014.12.003 CrossRefGoogle Scholar
  20. 20.
    Gorken A, Nagaraj DR, Riccio PJ (1992) The influence of pulp redox potentials and modifiers in the flotation separation of complex sulfide ores with dithiophosphinate. In: Woods R, Richardson PE (eds) Electrochemistry in mineral and metal processing III. The Electrochemical Society, Princeton, Proceedings Vol. 92–17, pp 95–107Google Scholar
  21. 21.
    Hintikka VV, Leppinen JO (1995) Potential control in the flotation of sulphide minerals and precious metals. Miner Eng 8:1151–1158.  https://doi.org/10.1016/0892-6875(95)00080-A CrossRefGoogle Scholar
  22. 22.
    Uribe-Salas A, Martínez-Cavazos TE, Nava-Alonso FC, Méndez-Nonell J, Lara-Valenzuela C (2000) Metallurgical improvement of a lead/copper flotation stage by pulp potential control. Int J Miner Process 59:69–83.  https://doi.org/10.1016/S0301-7516(99)00059-9 CrossRefGoogle Scholar
  23. 23.
    Sheridan MS, Nagaraj DR, Fornasiero D, Ralston J (2002) The use of a factorial experimental design to study collector properties of N-allyl-O-alkyl thionocarbamate collector in the flotation of a copper ore. Miner Eng 15:333–340.  https://doi.org/10.1016/S0892-6875(02)00037-7 CrossRefGoogle Scholar
  24. 24.
    Nagaraj DR, Farinato RS (2016) Evolution of flotation chemistry and chemicals: a century of innovations and the lingering challenges. Miner Eng 96-97:2–14.  https://doi.org/10.1016/j.mineng.2016.06.019 CrossRefGoogle Scholar
  25. 25.
    Nagaraj DR (2005) Reagent selection and optimization - the case for a holistic approach. Miner Eng 18:151–158.  https://doi.org/10.1016/j.mineng.2004.10.017 CrossRefGoogle Scholar
  26. 26.
    Woods R (2003) Electrochemical potential controlling flotation. Int J Miner Process 72:151–162.  https://doi.org/10.1016/S0301-7516(03)00095-4 CrossRefGoogle Scholar
  27. 27.
    Pecina ET, Uribe A, Finch JA, Nava F (2006) Mechanism of di-isobutyl dithiophosphinate adsorption onto galena and pyrite. Miner Eng 19:904–911.  https://doi.org/10.1016/j.mineng.2005.10.004 CrossRefGoogle Scholar
  28. 28.
    Brinen JS, Nagaraj DR (1994) Direct observation of a Pb–dithiophosphinate complex on galena mineral surfaces using SIMS. Surf Interface Anal 21:874–876.  https://doi.org/10.1002/sia.740211210 CrossRefGoogle Scholar
  29. 29.
    Piantadosi C, Jasieniak M, Skinner WM, Smart RSC (2000) Statistical comparison of surface species in flotation concentrates and tails from TOF-SIMS evidence. Miner Eng 13:1377–1394.  https://doi.org/10.1016/S0892-6875(00)00120-5 CrossRefGoogle Scholar
  30. 30.
    Güler T (2005) Dithiophosphinate–pyrite interaction: voltammetry and DRIFT spectroscopy investigations at oxidizing potentials. J Colloid Interface Sci 288:319–324.  https://doi.org/10.1016/j.jcis.2005.03.022 CrossRefGoogle Scholar
  31. 31.
    Hicyilmaz C, Altun NE, Ekmekci, Gökağaç G (2004) Pyrite–DTPI interaction as a function of pulp potential and pH. Colloids Surf A Physicochem Eng Asp 233:11–24.  https://doi.org/10.1016/j.colsurfa.2003.11.004 CrossRefGoogle Scholar
  32. 32.
    Hicyilmaz C, Altun NE, Ekmekci Z, Gökağaç G (2004) Quantifying hydrophobicity of pyrite after copper activation and DTPI addition under electrochemically controlled conditions. Miner Eng 17:879–890.  https://doi.org/10.1016/j.mineng.2004.02.007 CrossRefGoogle Scholar
  33. 33.
    Nagaraj DR, Brinen JS (2001) SIMS study of adsorption of collectors on pyrite. Int J Miner Process 63:45–57CrossRefGoogle Scholar
  34. 34.
    Wang X, Forssberg E, Bolin KS, Johan N (1989) Thermodynamic calculations on iron-containing sulphide mineral flotation systems, I. The stability of iron-xanthates. Int J Miner Process 27:1–19CrossRefGoogle Scholar
  35. 35.
    Leppinen JO (1990) FTIR and flotation investigation of the adsorption of ethyl xanthate on activated and non-activated sulfide minerals. Int J Miner Process 30:245–263CrossRefGoogle Scholar
  36. 36.
    Güler T, Hiçyilmaz C (2004) Hydrophobicity of chalcopyrite with dithiophosphate and dithiophosphinate in electrochemically controlled condition. Colloids Surf A Physicochem Eng Asp 235:11–15.  https://doi.org/10.1016/j.colsurfa.2004.01.009 CrossRefGoogle Scholar
  37. 37.
    Chander S, Fuerstenau DW (1975) Electrochemical reaction control of contact angles on copper and synthetic chalcocite in aqueous potassium diethyldithiophosphate solutions. Int J Miner Process 2:333–352CrossRefGoogle Scholar
  38. 38.
    Güler T, Hiçyilmaz C, Gökağaç G, Ekmeçi Z (2006) Adsorption of dithiophosphate and dithiophosphinate on chalcopyrite. Miner Eng 19:62–71.  https://doi.org/10.1016/j.mineng.2005.06.007 CrossRefGoogle Scholar
  39. 39.
    Bagci E, Ekmekci Z, Bradshaw D (2007) Adsorption behaviour of xanthate and dithiophosphinate from their mixtures on chalcopyrite. Miner Eng 20:1047–1053.  https://doi.org/10.1016/j.mineng.2007.04.011 CrossRefGoogle Scholar
  40. 40.
    Ignatkina VA, Bocharov VA, D’yachkov FG (2013) Collecting properties of diisobutyl dithiophosphinate in sulfide minerals flotation from sulfide ore. J Min Sci 49:795–802.  https://doi.org/10.1134/S1062739149050146 CrossRefGoogle Scholar
  41. 41.
    Hope GA, Woods R, Parker GK, Watling KM, Buckley AN (2006) Spectroelectrochemical investigations of flotation reagent-surface interaction. Miner Eng 19:561–570CrossRefGoogle Scholar
  42. 42.
    Basilio CI, Kim DS, Yoon RH, Leppinen JO, Nagaraj DR (1992) Interaction of thiophosphinate collectors with precious metals. In: SME Annual Meeting, Phoenix, Arizona, February 24–27, 1992. vol Preprint Number 92–174. Society for Mining, Metallurgy, and Exploration, Inc.,Google Scholar
  43. 43.
    Farinato R, Nagaraj DR (1992) Time dependent wettability of mineral and metal surfaces in the presence of thiol surfactants. J Adhes Sci Technol 6:1331–1346CrossRefGoogle Scholar
  44. 44.
    Buckley AN, Woods R (1997) Chemisorption - the thermodynamically favoured process in the interaction of thiol collectors with sulphide minerals. Int J Miner Process 51:15–26CrossRefGoogle Scholar
  45. 45.
    Buckley AN, Woods R (1992) Underpotential deposition of dithiophosphate on chalcocite. J Electroanal Chem 357:387–405CrossRefGoogle Scholar
  46. 46.
    Chanturia VA, Ivanova TA, Koporulina EV (2009) Interaction of sodium diisobutyl dithiophosphinate and platinum in aqueous solutions and on sulphide surface. J Min Sci 45:164–172CrossRefGoogle Scholar
  47. 47.
    Waterson CN, Tasker PA, Farinato R, Nagaraj DR, Shackelton N, Morrison CA (2016) A computational and experimental study on the binding of dithio ligands to sperrylite, pentlandite, and platinum. J Phys Chem C 120:22476–22488CrossRefGoogle Scholar
  48. 48.
    Larsson AC, Ivanov AV, Antzutkin ON, Forsling W (2008) A 31P CP/MAS NMR study of PbS surface-dialkyldithiophosphate lead (II) complexes. J Colloid Interface Sci 327:379–376CrossRefGoogle Scholar
  49. 49.
    Kabachnik MI, Mastrukova TA, Shipov AE, Melentyeva TA (1960) The application of the hammett equation to the theory of tautomeric equilibrium: thione-thiol equilibrium, acidity, and structure of phosphorus thio-acids. Tetrahedron 9:10–28.  https://doi.org/10.1016/0040-4020(60)80048-8 CrossRefGoogle Scholar
  50. 50.
    Meloun M, Bordovská S (2007) Benchmarking and validating algorithms that estimate pKa values of drugs based on their molecular structures. Anal Bioanal Chem 389:1267–1281CrossRefGoogle Scholar
  51. 51.
    Xun F, Yahong X, Shuyun X, Shaona Z, Zhengshui H (2002) Study on the thiophosphinic extractants. I. The basic properties of the extractants and the phase behavior in their saponified systems. Solvent Extr Ion Exch 20:331–344CrossRefGoogle Scholar
  52. 52.
    Jia Q, Zhan C, Li D, Niu C (2005) Extraction of zinc (II) and cadmium (II) by using mixtures of primary amine N1923 an organophosphorus acids. Sep Sci Technol 39:1111–1123CrossRefGoogle Scholar
  53. 53.
    Benson MT, Moser ML, Peterman DR, Dinescu A (2008) Determination of pKa for dithiophosphinic acids using density functional theory. J Mol Struct THEOCHEM 867:71–77CrossRefGoogle Scholar
  54. 54.
    Casasnovas R, Ortega-Castro J, Donoso J, Frau J, Muñoz F (2013) Theoretical calculations of stability constants and pKa values of metal complexes in solution: application to pyridoxamine-copper (II) complexes and their biological implications in AGE inhibition. Phys Chem Chem Phys 15:16303–16313CrossRefGoogle Scholar
  55. 55.
    Carbonaro RF, Atalay YB, Di Toro DM (2011) Linear free energy relationships for metal–ligand complexation: bidentate binding to negatively-charged oxygen donor atoms. Geochim Cosmochim Acta 75:2499–2511CrossRefGoogle Scholar
  56. 56.
    Wang SS, Avotins PV (1982) The use of dialkyldithiophosphinates in sulfide flotation. Paper presented at the SME-AIME Annual Meeting, Dallas, Texas, 14–18 February 1982. Preprint number is 82-155Google Scholar
  57. 57.
    Larin GM, Solozhenkin PM, Dyatkina ME, Kopitsya NI (1971) Hyperfine structure of ligands studied in EPR spectra of complexes. V. Copper (II) dithiophosphinate and dithiophosphates. Zh Strukt Khim 12:26–33Google Scholar
  58. 58.
    Yordanov ND (1997) A spectroscopic study of the self-redox reaction of sulfur-containing copper (II) complexes. Transition Met Chem 22(2):200–207Google Scholar
  59. 59.
    Ovchinnikov VV, Toropova VF, Garifzyanov AR, Cherkasov RA, Pudovik AN (1985) Complexing, redox and extractive properties of cyclic and acyclic derivatives of phosphorus dithioacids. Phosphorus Sulfur 22:199–210CrossRefGoogle Scholar
  60. 60.
    Toropova VF, Garifzyanov AR, Panfilova IE (1987) Steric and hydrophobic effects of substituents in extraction of metal complexes with O,O-dialkyldithiophosphoric acids. Talanta 34:211–214CrossRefGoogle Scholar
  61. 61.
    Shetty PS, Fernando Q (1967) A polarographic study of certain metal chelates of diethylphosphorodithioic acid in ethanol solutions. J Inorg Nucl Chem 29:1921–1930CrossRefGoogle Scholar
  62. 62.
    Liu G, Xiao J, Zhou D, Zhong H, Choi P, Xu Z (2013) A DFT study on the structure-reactivity relationship of thiophosphorus acids as flotation collectors with sulfide minerals: implication of surface adsorption. Colloids Surf A Physicochem Eng Asp 434:243–252CrossRefGoogle Scholar

Copyright information

© Society for Mining, Metallurgy & Exploration Inc. 2019

Authors and Affiliations

  1. 1.Solvay Technology SolutionsStamfordUSA

Personalised recommendations