Minerals Engineering 129 (2018) 102–105

Contents lists available at ScienceDirect

Minerals Engineering
journal homepage: www.elsevier.com/locate/mineng

Short communication

Use of molecular dynamics to study the conformation of an anionic T

polyelectrolyte in saline medium and its adsorption on a quartz surface
⁎
Gonzalo R. Quezadaa,d, , Ricardo I. Jeldresb, Phillip D. Fawellc, Pedro G. Toledoa,d
a
Water Research Center for Agriculture and Mining (CRHIAM), Concepción, Chile
b
Departamento de Ingeniería Química y Procesos de Minerales, Facultad de Ingeniería, Universidad de Antofagasta, PO Box 170, Antofagasta, Chile
c
CSIRO Mineral Resources, Waterford, Western Australia, Australia
d
Chemical Engineering Department and Surface Analysis Laboratory (ASIF), Universidad de Concepción, PO Box 160-C, Correo 3, Concepción, Chile

A R T I C LE I N FO A B S T R A C T

Keywords: In seeking to gain insights on the performance of long chain water-soluble polymer flocculants when conducting
Adsorption mineral processing in highly saline liquors, molecular dynamic simulation was used to study the effect of NaCl
Dynamic simulation concentration on the conformation of an anionic polyelectrolyte and its adsorption on a quartz surface at pH
Flocculation values above its isoelectric point. Increased salinity led to two competing effects on the overall flocculation
Polyelectrolyte
process: (i) increased polyelectrolyte adsorption on the quartz surface favours the initial aggregate-forming
Salinity effect
collisions, but (ii) the radius-of-gyration of the polyelectrolyte is reduced, which then may limit the aggregate
size achieved. Such results support reported experimental trends and show the advantages of computational
methodologies to describe complex systems.

1. Introduction surface is weak, but adsorption of multiple functionalities from one

Mineral processing to separate and concentrate valuable phases
involves large volumes of water and the generation of tailings, i.e. fine
gangue phases in dilute suspensions. The tailings are treated in gravity
thickeners, large settling tanks in which water is recovered for upstream
reuse and solid phases are concentrated for deposition or disposal,
ideally over a low area footprint.
High volumetric throughputs require particles to settle at rates well
in excess of what they would on the basis of their size. This is achieved
by adding flocculants - high molecular weight water-soluble polyelec-
trolytes that adhere to specific active surface sites, bridging particles to
form larger structures (aggregates) and accelerating settling by up to
several orders of magnitude. Decisions on reagent management, such as
polymer and dosage selection, are made empirically and often with
little fundamental understanding, despite flocculation being highly
sensitive to a variety of variables.
The physical characteristics of aggregates on a microscopic scale
govern flocculation responses, which can in part be manipulated
through knowledge of the flocculant’s solution and adsorption proper-
ties. Mineral processing flocculants are based on hydrolysed poly-
acrylamide (HPAM), typically acrylamide and acrylate copolymerised
at varying ratios, the latter’s charged functionality favouring chain
extension in aqueous solutions. Adsorption of a single functionality to a
chain (a “train”) is effectively irreversible, with unadsorbed “tails” then
able to offer bridging.
While tailings composition varies widely in mineralogy and size, in
general the fines fraction (< 20 µm) determining solid-liquid separa-
tion and flocculation behaviour is dominated by clays and fine quartz
phases. Quartz surfaces consist of silanol groups (Si-OH), but clay sur-
face chemistry is complicated by their plate-like structures. The major
basal faces consist of silanol groups which are negatively charged over
most pH values and effectively control the bulk particle charge, while
the minor edge faces have a higher proportion of aluminol groups (Al-
OH) with a greater prospect of providing neutral sites for flocculant
adsorption at most slurry pH values (Hocking et al., 1999).
There is strong interest (and a need in some cases) to use highly
saline liquors and even seawater in mineral processing (Cisternas and
Moreno, 2014), which influences the flocculation process. At pH va-
lues > 3, minimal flocculant adsorbs on quartz surfaces in the absence
of salt (Somasundaran and Zhang, 2000). Adsorption may then be
through salt bridges, where cations link between the hydrolysed silanol
sites and anionic HPAM groups. However, the measured flocculation
trends in saline liquors can be complex and contrasting (Witham et al.,
2012; Jeldres et al., 2017; Costine et al., 2018), making ultimate effects
on separation processes difficult to predict.
Techniques such as viscometry, carbon analysis, x-ray spectroscopy

⁎
Corresponding author.
E-mail address: gonzaloqe@gmail.com (G.R. Quezada).

https://doi.org/10.1016/j.mineng.2018.09.025
Received 30 May 2018; Received in revised form 27 September 2018; Accepted 28 September 2018
0892-6875/ © 2018 Elsevier Ltd. All rights reserved.
G.R. Quezada et al. Minerals Engineering 129 (2018) 102–105

and atomic force microscopy have been applied to probe the state of
flocculants both in solution and adsorbed on surfaces (Girod et al.,
1988, Lecourtier et al., 1990; Allen et al., 1998; Dalton et al., 2002),
although most approaches give indirect information or require drying,
the latter likely to induce changes. Molecular dynamics (MD) simula-
tions offer the potential to describe how systems of interest behave on a
microscopic scale. The quartz-water interface has been extensively si-
mulated - for example, Quezada et al. (2017) showed ion adsorption
obeys the concept of “like absorbs like”, where weakly hydrated ions
are better adsorbed on low isoelectric point quartz surfaces that are
weakly hydrated. On the other hand, Chen et al. (2012) showed
through MD that Na+ ions reduce the HPAM size by decreasing the
repulsion of its charged groups. Abdel-Azeim and Kanj (2018) predicted
this effect to be even greater in the presence of divalent cations.
The interaction of biological molecules with mineral surfaces has
been studied by MD (e.g. Emami et al., 2014). Simulations of the effect
of salt type on HPAM adsorption to quartz at different pH have also
shown better adhesion with cations of higher electrical density, such as
Li+ and Na+, than with Cs+ (Quezada et al., 2018). In this work, the
effect of salinity (NaCl) on the phenomena involved in HPAM adsorp-
tion on a quartz mineral (1 0 1) surface is analysed, with particular
emphasis on the density of adsorption and the polyelectrolyte con-
formation.

2. Methodology

Molecular dynamics simulations were carried out for: (i) a model

polymer in the presence of salts for prediction of the radius-of-gyration
in solution and (ii) a quartz surface in the presence of the polymer and
salts for prediction of subsequent adsorption. The HPAM flocculant is
represented by a 25% anionic polymer formed from 54 acrylamide and
18 acrylate evenly distributed monomers, the functionalities of the
latter in deprotonated form (eCOO−) at pH > 6. The surface is crys-
talline quartz with an area of 5.8 × 5.5 nm2 and the face (1 0 1) ex-
posed to the medium having explicit silanol groups; this face represents
one the most common cleavages found in nature (de Leeuw et al.,
1999). The ions are spheres of Na+ and Cl−, while the water used is the
SPC/E model and is represented by three explicit site atoms. Computer
simulations were carried out in the package Gromacs 5.1.2. The inter-
action potentials used were AMBER99SB for HPAM, CLAYFF for quartz
and the potential shown in Li et al. (2015) for ions. The similarity of
these functions facilitates their use together. For a correct representa-
tion of the electronic density of the quartz surface to incorporate pH
effects through charge density, first principles calculations for partial
charges of the HPAM and the surface deprotonation, respectively, were
performed in Gaussian 09 using the DFT B3LYP method with the base
functions 3-21G for the non-oxygen atoms and 6-31G (d,p) for the
oxygens (Kroutil et al., 2015). An initial energy minimisation step was
made by steepest descent, followed by (i) a constant particle, volume
and temperature (NVT) simulation of 1 ns only as free water molecules
to generate hydration layers, and (ii) a constant particle, pressure and
Fig. 1. Flocculant response to changes in salinity: (a) radius of gyration for
temperature (NPT) step of 2 ns, initially using annealing with an in- pH > 6, (b) adsorption on a quartz surface, (c) flocculant conformation on a
crease to 430 K per 100 ps and then lower to 300 K, thus eliminating quartz surface.
metastable HPAM configurations. Finally, an NVT simulation of 80 ns is
used to generate results. The step of integration is 2 fs, and the con-
strength, but such repulsions are reduced by shielding from increasing
stants of the thermostat and barostat of Berendsen are 0.1 and 2.0 ps,
the counter-ion concentration (Na+), and the polyelectrolyte’s effective
respectively.
size decreases. Fig. 2a is a representation from the simulations of this
behaviour, clearly showing the change in chain extension.
3. Results
Simulation of the polyelectrolyte interactions with a quartz surface
(Fig. 1b) indicated adsorption density was strongly influenced by higher
The predicted radius-of-gyration (rg) of the HPAM polymer in re-
salt concentration, increasing by more than two orders of magnitude on
sponse to increasing salinity (Fig. 1a) clearly shows a contraction in
going from 0.006 M to 0.6 M NaCl, with pH having no discernible ef-
conformation. For context, rg for the polymer in the absence of salt was
fect. The enhancement of adsorption is captured by the simulations in
1.85 nm, and hence the presence of 0.6 M NaCl represents a reduction
Fig. 2b, where the Na+ cations that shield the eCOO− groups of HPAM
in size of one-third. Electrostatic repulsions between the eCOO−
and also adsorb on the quartz surface combine to give closer approach
groups serve to keep the polyelectrolyte chain stretched at low ionic

103
G.R. Quezada et al.
CNaCl = 0.006 M
0.006 M
Fig. 2. Simulation of HPAM response to changes in salinity: (a) flocculant in solution for pH > 6, (b) interaction with quartz surface (red represents oxygen, blue
spheres are sodium cations). Water molecules are not shown for better clarity on the flocculant-surface interaction. (For interpretation of the references to color in
this figure legend, the reader is referred to the web version of this article.)
to the surface by a greater proportion of HPAM functionalities, fa-
vouring interactions between the polyelectrolyte and quartz by hy-
drogen bonds and cationic bridging.
Fig. 1c provides estimates of the fractions of adsorbed trains and
unadsorbed loops and tails for the simulated polyelectrolyte. Given the
quite short chain length, the proportions of trains and (in particular)
loops were expected to be low, but the critical prediction was the in-
crease in trains with ionic strength. Adsorption of individual polymer
functionalities is weak, but having multiple in an adsorbed train leads
to an effectively irreversible bond (Lyklema, 1988), and therefore to the
potential for bridging.
4. Implications
The MD simulation results are consistent with published experi-
mental observations (e.g. Witham et al., 2012; Ji et al., 2013; Jeldres
et al., 2017) that highlight the competing effects of salinity on the
flocculation process, i.e. higher salinity favours flocculant adsorption
due to the counter-ion effects on surface interactions, but at same time
the flocculant’s reduced rg limits the aggregate size that may then form.
Added to this is the fact that while enhanced adsorption density is in-
itially beneficial to aggregate formation, the process becomes effective
at quite moderate surface coverage, and so returns from higher cov-
erage may be small (Owen et al., 2008). In practical terms, while
flocculation may be effective and even promoted by high salinity in low
settling rate applications, it may have a detrimental effect when higher
settling rates are required.
The polyelectrolye simulated is clearly much smaller in length and
solution size than the flocculants used industrially, for which the re-
lative proportion of adsorbed trains and bridging tails may differ. It is
therefore not expected that MD simulations will approach quantitative
predictions that represent full-scale behaviour. However, such simula-
tions can still offer invaluable insights on how variations in monomer
Minerals Engineering 129 (2018) 102–105
0.060 M 0.600 M
a)
0.060 M 0.600 M
b)
chemistry, liquor composition and surface structures impact on par-
ticle-particle collision efficiencies and even aggregate structures
(Costine et al., 2018), which may then influence flocculant selection.
5. Conclusions
The potential value of MD simulations in gaining insights on how
salinity promotes the HPAM flocculant adsorption process on a quartz
surface has been demonstrated. The simulated polyelectrolyte consisted
of 72 monomers (a mixture of acrylamide and acrylate), some three
orders of magnitude smaller than flocculants used in mineral proces-
sing, but the interactions with surfaces are expected to be the same.
While simulations of how increasing salt concentrations affect poly-
electrolytes in solution and quartz surfaces are known, we believe this is
the first study to capture in detail how adsorption of a bridging floc-
culant’s surrogate on such surfaces will respond under these conditions.
Increased salinity favoured the polyelectrolyte’s adsorption, but this is
balanced by a reduction in its solution dimensions, which may limit
bridging capacity.
Acknowledgements
RIJ acknowledges support CONICYT Fondecyt n° 11171036. GRQ
and PGT thank Centro CRHIAM through Project Conicyt/Fondap/
15130015 for financial support and The Southern GPU-cluster (SGPU-
C) UDEC funded by FONDEQUIP EQM150134 for computational sup-
port.
References
Abdel-Azeim, S., Kanj, M.Y., 2018. Dynamics, aggregation, and interfacial properties of
the partially hydrolyzed polyacrylamide polymer for enhanced oil recovery appli-
cations: insights from molecular dynamics simulations. Energy Fuels 32, 3335–3343.
Allen, G.C., Hallam, K.R., Eastman, J.R., Graveling, G.J., Ragnarsdottir, V.K., Skuse, D.R.,
104
G.R. Quezada et al.
1998. XPS analysis of polyacrylamide adsorption to kaolinite, quartz and feldspar.
Surf. Interface Anal. 26 (7), 518–523.
Chen, P., Yao, L., Liu, Y., Luo, J., Zhou, G., Jiang, B., 2012. Experimental and theoretical
study of dilute polyacrylamide solutions: effect of salt concentration. J. Mol. Model.
18, 3153–3160.
Costine, A., Cox, J., Travaglini, S., Lubansky, A., Fawell, P., Misslitz, H., 2018. Variations
in the molecular weight response of anionic polyacrylamides under different floc-
culation conditions. Chem. Eng. Sci. 176, 127–138.
Cisternas, L., Moreno, L., 2014. Sea Water in Mining: Fundamentals and Applications (in
Spanish). Master Rill, Santiago, Chile.
Dalton, J.S., Allen, G.C., Hallam, K.R., Elton, N.J., Hooper, J.J., Skuse, D.R., 2002. An
AFM and XPS investigation of the selective flocculation of kaolinite from a mineral
mixture. Spec. Publ. - R. Soc. Chem. 282, 71–80.
de Leeuw, N.H., Higgins, F.M., Parker, S.C., 1999. Modeling the surface structure and
stability of α-quartz. J. Phys. Chem. B 103, 1270–1277.
Emami, F.S., Puddu, V., Berry, R.J., Varshney, V., Patwardhan, S.V., Perry, C.C., Heinz,
H., 2014. Force field and a surface model database for silica to simulate interfacial
properties in atomic resolution. Chem. Mater. 26, 2647–2658.
Girod, G., Lamarche, J.M., Foissy, A., 1988. Adsorption of partially hydrolyzed poly-
acrylamides on titanium dioxide. J. Colloid Interface Sci. 121 (1), 265–272.
Hocking, M.B., Klimchuk, K.A., Lowen, S., 1999. Polymeric flocculants and flocculation.
J. Macromol. Sci. Rev. Macromol. Chem. Phys. C39, 177–203.
Jeldres, R.I., Piceros, E.C., Leiva, W.H., Toledo, P.G., Herrera, N., 2017. Viscoelasticity
and yielding properties of flocculated kaolinite sediments in saline water. Colloids
Surf. A: Physicochem. Eng. Asp. 529, 1009–1015.
Ji, Y., Lu, Q., Liu, Q., Zeng, H., 2013. Effect of solution salinity on settling of mineral
105
Minerals Engineering 129 (2018) 102–105
tailings by polymer flocculants. Colloids Surf. A: Physicochem. Eng. Asp. 430, 29–38.
Kroutil, O., Chval, Z., Skelton, A.A., Predota, M., 2015. Computer simulations of quartz
(101)–water interface over a range of pH values. J. Phys. Chem. C 119, 9274–9286.
Lecourtier, J., Lee, L.T., Chauveteau, G., 1990. Adsorption of polyacrylamides on siliceous
minerals. Colloids Surf. 47C, 219–231.
Li, P., Song, L.F., Merz Jr, K.M., 2015. Systematic parameterization of monovalent ions
employing the nonbonded model. J. Chem. Theory Comput. 11, 1645–1657.
Lyklema, J., 1988. The colloidal background of flocculation and dewatering. In: Moudgil,
B.M., Scheiner, B.J. (Eds.), Flocculation and Dewatering. Engineering Foundation,
Palm Coast, Florida, pp. 1–20.
Owen, A.T., Fawell, P.D., Swift, J.D., Labbett, D.M., Benn, F.A., Farrow, J.B., 2008. Using
turbulent pipe flow to study the factors affecting polymer-bridging flocculation of
mineral systems. Int. J. Miner. Process. 87, 90–99.
Quezada, G.R., Rozas, R.E., Toledo, P.G., 2017. Molecular dynamics simulations of quartz
(101)–water and corundum (001)–water interfaces: effect of surface charge and ions
on cation adsorption, water orientation, and surface charge reversal. J. Phys. Chem. C
121, 25271–25282.
Quezada, G.R., Rozas, R.E., Toledo, P.G., 2018. Molecular dynamics simulations of par-
tially hydrolyzed polyacrylamide adsorption on (101) quartz surfaces in saltwater for
a range of pH values: adsorption bridges and lifetimes (submitted for publication).
Somasundaran, P., Zhang, L., 2000. Modification of silica-water interfacial behavior by
adsorption of surfactants, polymers, and their mixtures. Surf. Sci. Ser. 441–462.
Witham, M.I., Grabsch, A.F., Owen, A.T., Fawell, P.D., 2012. The effect of cations on the
activity of anionic polyacrylamide flocculant solutions. Int. J. Miner. Process.
114–117, 51–62.