XXXI International Mineral Processing Congress 2024 Proceedings/Washington, DC/Sep 29–Oct 3 1647
Notably, silica gel formation has little to do with the
strength of the individual bonds and much more to do with
the high degree of polymerization (75%) in the tetrahe-
dral sheet. This is obvious from comparisons between the
phyllosilicates and less polymerized silicate minerals such as
willemite and hemimorphite. Both of these dissolve com-
pletely in dilute sulfuric acid without forming silica gels
(Matthew and Elsner, 1976). The Si-O bond strength is
the same what differs is that the soluble silicates have a
lower degree of polymerization, with at most 25% of the
silica tetrahedra bonded to other silica tetrahedra through
bridging oxide ions. This lack of polymerization effectively
puts the silica, though itself strongly bonded, into digest-
ible units, which exist in the solution as silicic acid species
with unbroken Si-O bonds (Terry, 1983b). Simply put, tet-
rahedrally coordinated silica in silicate minerals only enters
solution if everything else dissolves from around the tetra-
hedron. This happens in low-polymerized silicates, where
the Si-O units are attached to weaker-bonded species, but
not in phyllosilicates, where tetrahedra are mostly attached
to each other and the high average bond strength protects
the entire layer.
Metallurgical Implications
Applying these two crystal-chemical parameters – bond
strength and silica polymerization – to the different types
of phyllosilicates explains their variability in acid consump-
tion and silica gel formation.
The main control on acid consumption is the extent
of dissolution in the octahedral layer. Under leaching
conditions, interlayers are exchanged and alkali layers
dissolve more or less equally from all phyllosilicates that
possess them, but phyllosilicates whose octahedral layers
are Al-dominated tend to have far lower acid consump-
tion than those with more Mg and Fe. Even though the
total amount of charge in the layer is equal (6+ per for-
mula unit), the higher average bond strength and conse-
quent lower dissolution of Al means that less H+ has to be
exchanged. This also explains the higher acid consumption
of chlorite compared to a biotite with equivalent composi-
tion: where biotite is a T-O-T phyllosilicate, chlorite has a
T-O structure with twice as many octahedral layers per unit
(Baum, 1999). If these are dominated by Mg, Fe, and other
divalent cations, their dissolution leads to extreme molar
acid consumption over the long term.
The tendency to form a silica gel is a function of phyl-
losilicate polymerization, which is the same 75% no matter
the type of phyllosilicate. If that were the only major influ-
ence, all phyllosilicates would form gels during leaching.
Here again, the observed variable behavior can be traced
back to the octahedral layer. If it stays more or less intact
during leaching, the phyllosilicate’s crystal structure is pre-
served and the tetrahedral layer does not rearrange into
a gel (Figure 3). If the octahedral layer dissolves to any
significant extent, the resulting underbonding makes it
impossible for the tetrahedral layer to maintain both charge
balance and its layered structure. It therefore crumples and
forms a hydrous gel with Si-O bonds still intact, but rear-
ranged into a noncrystalline framework.
PHYLLOSILICATE DISSOLUTION AND
METAL EXTRACTION
For zinc, nickel, vanadium, and some other base metals,
phyllosilicates are an important ore type. Because of their
refractory nature, it is common to treat them via roasting,
vat leaching, or other process options more powerful than
standard room-temperature, relatively dilute leaching. Even
with these options, though, low recoveries from phyllosili-
cate ores are common. This is partly due to unrecognized
mineralogical complexity: not all metals in phyllosilicates
actually occur on the phyllosilicate lattice proper (Barton
et al., 2023 Drexler et al., 2023). For those that do, it is
worth exploring the implications of phyllosilicate dissolu-
tion as discussed above.
The same crystal-chemical factors that determine
leaching behavior control metal recovery from phyllosili-
cates. The strength of the individual metal bond determines
whether or not it will dissolve during leaching, and the
average bond strength of the layer that hosts it determines
whether or not a strongly bonded species will be forced into
solution by the dissolution of its surroundings. Thus recov-
ery of a divalent metal in octahedral coordination (e.g.,
zinc) is typically high, while higher-valent (and thus more
strongly bonded) metals such as vanadium pose more of a
problem. If they are trivalent and hosted in an otherwise
divalent octahedral layer, they will probably leach as every-
thing around them dissolves. If they are trivalent and so is
the rest of the layer, leaching is unlikely to recover much
of them. At least in the case of vanadium, the target metal
can exist in a mix of valences and coordination states in
phyllosilicates, leading to highly variable recovery (Drexler
et al., 2023).
UNANSWERED QUESTIONS AND
FUTURE RESEARCH
The dissolution of phyllosilicates is seldom studied in detail
in practical leaching contexts. For that reason, much of the
above is conjectural: based on the known crystal chemis-
try and observed variations in e.g., acid consumption, but
not actually demonstrated at a mechanistic level. There
Notably, silica gel formation has little to do with the
strength of the individual bonds and much more to do with
the high degree of polymerization (75%) in the tetrahe-
dral sheet. This is obvious from comparisons between the
phyllosilicates and less polymerized silicate minerals such as
willemite and hemimorphite. Both of these dissolve com-
pletely in dilute sulfuric acid without forming silica gels
(Matthew and Elsner, 1976). The Si-O bond strength is
the same what differs is that the soluble silicates have a
lower degree of polymerization, with at most 25% of the
silica tetrahedra bonded to other silica tetrahedra through
bridging oxide ions. This lack of polymerization effectively
puts the silica, though itself strongly bonded, into digest-
ible units, which exist in the solution as silicic acid species
with unbroken Si-O bonds (Terry, 1983b). Simply put, tet-
rahedrally coordinated silica in silicate minerals only enters
solution if everything else dissolves from around the tetra-
hedron. This happens in low-polymerized silicates, where
the Si-O units are attached to weaker-bonded species, but
not in phyllosilicates, where tetrahedra are mostly attached
to each other and the high average bond strength protects
the entire layer.
Metallurgical Implications
Applying these two crystal-chemical parameters – bond
strength and silica polymerization – to the different types
of phyllosilicates explains their variability in acid consump-
tion and silica gel formation.
The main control on acid consumption is the extent
of dissolution in the octahedral layer. Under leaching
conditions, interlayers are exchanged and alkali layers
dissolve more or less equally from all phyllosilicates that
possess them, but phyllosilicates whose octahedral layers
are Al-dominated tend to have far lower acid consump-
tion than those with more Mg and Fe. Even though the
total amount of charge in the layer is equal (6+ per for-
mula unit), the higher average bond strength and conse-
quent lower dissolution of Al means that less H+ has to be
exchanged. This also explains the higher acid consumption
of chlorite compared to a biotite with equivalent composi-
tion: where biotite is a T-O-T phyllosilicate, chlorite has a
T-O structure with twice as many octahedral layers per unit
(Baum, 1999). If these are dominated by Mg, Fe, and other
divalent cations, their dissolution leads to extreme molar
acid consumption over the long term.
The tendency to form a silica gel is a function of phyl-
losilicate polymerization, which is the same 75% no matter
the type of phyllosilicate. If that were the only major influ-
ence, all phyllosilicates would form gels during leaching.
Here again, the observed variable behavior can be traced
back to the octahedral layer. If it stays more or less intact
during leaching, the phyllosilicate’s crystal structure is pre-
served and the tetrahedral layer does not rearrange into
a gel (Figure 3). If the octahedral layer dissolves to any
significant extent, the resulting underbonding makes it
impossible for the tetrahedral layer to maintain both charge
balance and its layered structure. It therefore crumples and
forms a hydrous gel with Si-O bonds still intact, but rear-
ranged into a noncrystalline framework.
PHYLLOSILICATE DISSOLUTION AND
METAL EXTRACTION
For zinc, nickel, vanadium, and some other base metals,
phyllosilicates are an important ore type. Because of their
refractory nature, it is common to treat them via roasting,
vat leaching, or other process options more powerful than
standard room-temperature, relatively dilute leaching. Even
with these options, though, low recoveries from phyllosili-
cate ores are common. This is partly due to unrecognized
mineralogical complexity: not all metals in phyllosilicates
actually occur on the phyllosilicate lattice proper (Barton
et al., 2023 Drexler et al., 2023). For those that do, it is
worth exploring the implications of phyllosilicate dissolu-
tion as discussed above.
The same crystal-chemical factors that determine
leaching behavior control metal recovery from phyllosili-
cates. The strength of the individual metal bond determines
whether or not it will dissolve during leaching, and the
average bond strength of the layer that hosts it determines
whether or not a strongly bonded species will be forced into
solution by the dissolution of its surroundings. Thus recov-
ery of a divalent metal in octahedral coordination (e.g.,
zinc) is typically high, while higher-valent (and thus more
strongly bonded) metals such as vanadium pose more of a
problem. If they are trivalent and hosted in an otherwise
divalent octahedral layer, they will probably leach as every-
thing around them dissolves. If they are trivalent and so is
the rest of the layer, leaching is unlikely to recover much
of them. At least in the case of vanadium, the target metal
can exist in a mix of valences and coordination states in
phyllosilicates, leading to highly variable recovery (Drexler
et al., 2023).
UNANSWERED QUESTIONS AND
FUTURE RESEARCH
The dissolution of phyllosilicates is seldom studied in detail
in practical leaching contexts. For that reason, much of the
above is conjectural: based on the known crystal chemis-
try and observed variations in e.g., acid consumption, but
not actually demonstrated at a mechanistic level. There