XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 2407
on flat, polished surfaces do not corroborate with the inher-
ent dynamics of particle-bubble interactions and are gener-
ally inadequate for predicting flotation outcomes in real ore
systems (Nagaraj &Ravishankar, 2007). The main realiza-
tion from these studies, however, is that even well-polished
and relatively “pure” minerals can display significant sur-
face heterogeneity, which demonstrates that the practice of
reporting single contact angles is highly unsatisfactory.
Cyclic Voltammetry
For all voltammetry experiments, absolute limits for the
potential scan range were –300 to 150 mV (with reference
to Ag/AgCl) these were chosen with respect to empirically
observed potential ranges in sulfide ore pulps. In addition
to those generated using BCBDTC, voltammograms were
also recorded with DIBDTPI. In contrast to BCBDTC,
DIBDTPI is an anionic, hydrolytically stable ligand whose
two active S donors and alkyl substituents are directly
attached to a central P atom. Thus, its phosphorus-based
structure and lack of electron-withdrawing O donors make
its coordination chemistry radically different. These fea-
tures, in addition to the fact that it does not oxidize (to a
dimer) under flotation-relevant conditions, make it a good
candidate for comparison to BCBDTC.
The voltammograms for Cu and chalcocite are pre-
sented in Figure 6. For Cu, in the presence of BCBDTC,
there is evident passivation of the electrode as indicated
by the sharp decrease in current the passivation is slightly
stronger for the same concentration of DIBDTPI.
In the case of chalcocite (narrower potential window
of –80 to 80 mV), the passivation with either ligand is
extremely strong, indicating that redox processes at the
interface are significantly reduced or shut down by ligand
binding to metal sites (akin to the behavior of corrosion
inhibitors). This also clearly shows that both ligands show
strong affinity for soft acid Cu sites.
The notable feature in the above voltammograms is the
absence of a current peak indicative of electron transfer out
of the complex or oxidation of the ligand this contrasts
with the case of xanthate, which oxidizes to dixanthogen.
This demonstrates that prescribing the adsorption mecha-
nism of a ligand like xanthate to all others is inaccurate.
Most importantly, this ignores the importance of LAB con-
cepts, in which electron redistribution upon formation of a
metal-ligand complex occurs.
The voltammograms for pyrite are shown in Figure 7.
In a range of –100 to 100 mV, the interaction of either
ligand with pyrite, as judged by the extent of passivation,
is much weaker very little change is observed on the oxi-
dizing side for both ligands, although there is slight pas-
sivation on the reducing side. The voltammogram of pyrite
with DIBDTPI is consistent with its empirically observed
selectivity against it however, in the case of BCBDTC, the
lack of passivation is not consistent with the contact angle
studies. This is perhaps related to mineralogical differences
between the two pyrite samples used in these studies.
For Ag, in the range of –300 to 150 mV the extent
of passivation with BCBDTC is greater than that of
DIBDTPI, especially on the reducing side (Figure 8). The
voltammetric behavior for Ag in Figure 8 was near-identical
to Au. The sharp cathodic drop on the reducing side may be
attributed to dissolved oxygen that was present. Regardless,
it is apparent that passivation of Au and Ag with BCBDTC
is greater than that for DIBDTPI.
Figure 6. Voltammograms of Cu and chalcocite with and without 5 × 10 –5 M BCBDTC and DIBDTPI, borate-buffered
on flat, polished surfaces do not corroborate with the inher-
ent dynamics of particle-bubble interactions and are gener-
ally inadequate for predicting flotation outcomes in real ore
systems (Nagaraj &Ravishankar, 2007). The main realiza-
tion from these studies, however, is that even well-polished
and relatively “pure” minerals can display significant sur-
face heterogeneity, which demonstrates that the practice of
reporting single contact angles is highly unsatisfactory.
Cyclic Voltammetry
For all voltammetry experiments, absolute limits for the
potential scan range were –300 to 150 mV (with reference
to Ag/AgCl) these were chosen with respect to empirically
observed potential ranges in sulfide ore pulps. In addition
to those generated using BCBDTC, voltammograms were
also recorded with DIBDTPI. In contrast to BCBDTC,
DIBDTPI is an anionic, hydrolytically stable ligand whose
two active S donors and alkyl substituents are directly
attached to a central P atom. Thus, its phosphorus-based
structure and lack of electron-withdrawing O donors make
its coordination chemistry radically different. These fea-
tures, in addition to the fact that it does not oxidize (to a
dimer) under flotation-relevant conditions, make it a good
candidate for comparison to BCBDTC.
The voltammograms for Cu and chalcocite are pre-
sented in Figure 6. For Cu, in the presence of BCBDTC,
there is evident passivation of the electrode as indicated
by the sharp decrease in current the passivation is slightly
stronger for the same concentration of DIBDTPI.
In the case of chalcocite (narrower potential window
of –80 to 80 mV), the passivation with either ligand is
extremely strong, indicating that redox processes at the
interface are significantly reduced or shut down by ligand
binding to metal sites (akin to the behavior of corrosion
inhibitors). This also clearly shows that both ligands show
strong affinity for soft acid Cu sites.
The notable feature in the above voltammograms is the
absence of a current peak indicative of electron transfer out
of the complex or oxidation of the ligand this contrasts
with the case of xanthate, which oxidizes to dixanthogen.
This demonstrates that prescribing the adsorption mecha-
nism of a ligand like xanthate to all others is inaccurate.
Most importantly, this ignores the importance of LAB con-
cepts, in which electron redistribution upon formation of a
metal-ligand complex occurs.
The voltammograms for pyrite are shown in Figure 7.
In a range of –100 to 100 mV, the interaction of either
ligand with pyrite, as judged by the extent of passivation,
is much weaker very little change is observed on the oxi-
dizing side for both ligands, although there is slight pas-
sivation on the reducing side. The voltammogram of pyrite
with DIBDTPI is consistent with its empirically observed
selectivity against it however, in the case of BCBDTC, the
lack of passivation is not consistent with the contact angle
studies. This is perhaps related to mineralogical differences
between the two pyrite samples used in these studies.
For Ag, in the range of –300 to 150 mV the extent
of passivation with BCBDTC is greater than that of
DIBDTPI, especially on the reducing side (Figure 8). The
voltammetric behavior for Ag in Figure 8 was near-identical
to Au. The sharp cathodic drop on the reducing side may be
attributed to dissolved oxygen that was present. Regardless,
it is apparent that passivation of Au and Ag with BCBDTC
is greater than that for DIBDTPI.
Figure 6. Voltammograms of Cu and chalcocite with and without 5 × 10 –5 M BCBDTC and DIBDTPI, borate-buffered