Journal of Colloid and Interface Science 527 (2018) 195–201

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science

journal homepage: www.elsevier.com/locate/jcis

Regular Article

Effect of Ca2+ ion concentration on adsorption of poly(carboxylate

ether)-based (PCE) superplasticizer on mica
Bo Wu a,b, Byong-Wa Chun c, Le Gu a, Tonya L. Kuhl b,⇑
a
School of Mechatronics Engineering, Harbin Institute of Technology, Harbin 150001, China
b
Department of Chemical Engineering and Materials Science, University of California at Davis, Davis, CA 95616, United States
c
GCP Applied Technologies, Cambridge, MA 02140, United States

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o a b s t r a c t

Article history: Hypothesis: Poly(carboxylate ether)-based (PCE) superplasticizers consist of a carboxylic acid backbone
Received 19 March 2018 and grafted poly(ethylene glycol) (PEG) side chains. Ca2+ ion bridging mechanism is commonly purported
Revised 8 May 2018 to control PCE’s adsorption on negatively charged cement particle surfaces in cement suspension, thus
Accepted 8 May 2018
PCE was expected to adsorb on negatively charged surfaces in synthetic pore solutions via Ca2+/ACOO

Available online 9 May 2018
interactions.
Experiments: Adsorption behaviors of a commercial PCE on negatively charged mica were studied in
Keywords:
aqueous electrolyte solutions by a surface forces apparatus.
Surface forces apparatus (SFA)
Poly(carboxylate ether) (PCE)
Findings: Direct force measurements indicated that the PCE adsorbed onto mica from 0.1 M K2SO4 due to
Superplasticizer K+ ion chelation by the ether oxygen units ACH2CH2OA on the PEG chains, but surprisingly did not adsorb
Poly(ethylene glycol) (PEG) from either 0.1 M K2SO4 with saturated Ca(OH)2 or 0.1 M Ca(NO3)2. The adsorption in K2SO4 was weak,
Polymer adsorption enabling the adsorbed PCE layers to be squeezed out under modest compression. Upon separating the
Steric force surfaces, the PCE immediately achieved an identical re-adsorption. In high-calcium conditions, the PCE
Ion chelation was highly positively charged due to Ca2+ ion chelation by PEG chains and backbone carboxylic groups
Cement ACOO
, and mica also underwent charge reversal due to electrostatic adsorption/binding of Ca2+ ions.
Suspension
Consequently, the interaction between mica and PCE was electrostatically repulsive and no PCE adsorp-
Dispersant
tion occurred. These findings can be explained by the complex interplay of ion chelation by PEG chains,
electrostatic binding and screening interactions with charged surfaces in the presence of monovalent and
divalent counterions, and ultimately charge reversal of both the charged surfaces and polyelectrolyte in
high divalent ion conditions.
Ó 2018 Elsevier Inc. All rights reserved.

⇑ Corresponding author.
E-mail address: tlkuhl@ucdavis.edu (T.L. Kuhl).

https://doi.org/10.1016/j.jcis.2018.05.016
0021-9797/Ó 2018 Elsevier Inc. All rights reserved.
196 B. Wu et al. / Journal of Colloid and Interface Science 527 (2018) 195–201
1. Introduction
Polyelectrolytes are extensively used as dispersing agents to
improve the rheological properties of colloidal suspensions in a
great variety of industries including paints, ceramic slurry process-
ing, cement and concrete placement, cosmetics and pharmaceuti-
cal formulations [1,2]. In the case of cement, polyelectrolyte
additives enable improved flowability at reduced water to cement
ratios [3–5]. Lower water content in cement suspension con-
tributes to higher mechanical strength and chemical resistance
[6–9]. A variety of water-reducing admixtures, also called super-
plasticizers, are incorporated into high performance cement as
essential components to improve effective flowability of hydrating
cement paste. The mechanism for improved workability is thought
to rely on adsorption of superplasticizers onto cement particle sur-
faces through their charged backbones. The adsorbed polymer lay-
ers prevent the agglomeration of cement particles by electrostatic
and/or steric repulsions and release entrapped water to reduce the
viscosity of cement suspension [10].
Comb-like (or bottle-brush) polymers have a linear backbone
with grafted side chains [11–13]. Steric repulsion between bulky
side chains can cause both the backbone and side chains to highly
extend [14,15]. This unusual molecular configuration makes this
class of polymers attractive for a wide range of applications such
as cement dispersion [16], boundary friction modification [17],
biomimetic lubrication [18] and protein adsorption [19]. Comb-
like poly(carboxylate ether)-based (PCE) superplasticizers are the
most common type of superplasticizer in aqueous systems [16].
They consist of a negatively charged carboxylic backbone and
grafted side chains mainly composed of poly(ethylene glycol)
(PEO or PEG), as shown in Fig. 1. PEG side chains are used because
of their high osmotic repulsion in water and insensitivity to ionic
strength [20]. In a hydrating cement system, the anionic backbones
are electrostatically attracted onto positively charged cement par-
ticle surfaces (e.g., tricalcium aluminate ðC 3 AÞ and ettringite ðAFtÞ)
and are also thought to be bridged to negatively charged particle
surfaces (e.g., tricalcium silicate ðC 3 SÞ and calcium silicate hydrate
ðC
S
HÞ) through electrostatic and entropic Ca2þ =
COO
inter-
actions [16,21], with the side chains extending away from the par-
ticle surfaces into the cement pore solution. Consequently, the
cement particles are dispersed primarily by the steric hindrance
produced by the side chains. PCE polymers exhibit superior dis-
persing ability compared to conventional dispersants such as lig-
nosulfonate and naphthalene or melamine formaldehyde
condensates based superplasticizers [10]. Large bodies of research
on zeta potential [5,6,8,10], adsorption isotherm [4–10] and rheo-
logical properties [4,5,10] of model cement suspensions and/or
actual cement pastes have indicated the significance of added
PCE dispersants, where enhanced flow properties were ascribed
to strong steric repulsion of the side chains.
Atomic force microscope (AFM) [22,23] and surface forces appa-
ratus (SFA) [24], two techniques for direct force measurement,
Fig. 1. Molecular structural schematic of the PCE polymer.
have been widely employed to measure the interaction forces
between polyelectrolyte adsorption layers in aqueous solutions
[25–29]. Uchikawa et al. [30] used AFM for the first time to demon-
strate the existence of steric repulsive force from the adsorbed
superplasticizer on cement particle surfaces. Since this pioneering
work, measurements of the steric force between PCE adsorption
layers have been carried out by AFM and SFA. Ferrari et al. [31]
used an AFM silicon tip to detect the interaction force with calcium
silicate and ettringite surfaces in Milli-Q water with PCE. The mea-
sured force was very small, suggesting little PCE’s adsorption under
these conditions. In contrast, significant repulsion between mica
surfaces was measured in aqueous PCE solution under both acidic
and alkaline conditions by SFA [32]. Adhesion was observed upon
separation in low PCE concentration conditions, which was
explained by the bridging of the PCE molecules between the two
mica surfaces through hydrogen bonding. Kauppi et al. [1] mea-
sured the interaction forces between MgO surfaces in KCl solution
with anionic acrylic ester-ethylene oxide (AAE-EO) copolymer at
pH 10 using AFM. Analysis of the force-distance curves revealed
that the measured force had a long-range electrostatic component
and a short-range steric component, suggesting that the acrylic
backbone adsorbed onto the MgO surface with the PEG chains pro-
truding into the solution in a coiled conformation. Ferrari and
coworkers [8] also employed AFM to study the interactions
between a silicon nitride tip and multiple substrates (calcite,
quartz, mica and MgO) in Milli-Q water, K2SO4 and KOH solutions
with PCE, respectively. Their force measurements showed that the
addition of PCE eliminated attractive jump-into-contact behavior
and induced repulsive force between the tip and substrate in all
cases. Along with zeta potential measurements and adsorption iso-
therms, they concluded that the PCE did not adsorb on negatively
charged substrates (calcite, quartz and mica) but adsorbed on the
positively charged silicon nitride tip and MgO in these aqueous
monovalent electrolyte conditions. However, reports on establish-
ing PCE’s adsorption through divalent cation bridging are very lim-
ited in the literature. Flatt and coworkers [2] investigated the
interactions between C
S
H surfaces in 5 mM Ca(OH)2 solution
with PCE at different ionic strengths (adding 10 or 100 mM NaCl)
by AFM, but the adsorption of PCE on CASAH could not be con-
firmed from the measured interaction forces. Similarly, direct mea-
surements of adsorption of polyelectrolyte or comb-like polymer
due to divalent counterion bridging are very sparse, though this
mechanism is purported to explain phenomenological changes in
such systems.
The goal of the present study was to determine the adsorption
behavior of a commercial PCE on negatively charged mica in syn-


thetic pore solution containing abundant Ca2+, K+, SO2
4 and OH
ions that mimics actual cement suspension conditions. Direct force
measurements at the molecular level by SFA showed that the PCE
polymer adsorbed onto mica from 0.1 M K2SO4 solution but did not
adsorb from either the synthetic pore solution or 0.1 M Ca(NO3)2
solution. This work is relevant to understanding and improving
PCEs as cement/concrete admixtures, not only in terms of their
action as dispersing agents, but also their interaction with aggre-
gate minerals such as alminosilicate clays. More broadly, the work
increases our understanding of the properties of adsorbed poly-
electrolyte in high-salt conditions in the absence and presence of
divalent counterions.
2. Experimental
2.1. Materials
A commercial cement dispersant PCE polymer was utilized as
received from GCP Applied Technologies (Cambridge, USA): 55 wt
 B. Wu et al. / Journal of Colloid and Interface Science 527 (2018) 195–201
% aqueous solution, pH 6–7. Its molecular configuration was comb-
like, consisting of a carboxylic backbone and grafted PEG side
chains (Fig. 1). On the backbone, the ratio n/m of carboxylic groups
(
COO
) to PEG chains was approximately 5:1. The side chain PEG
had about 45 repeating ether oxygen units
CH2 CH2 O
. The total
molecular weight (MW) was MW  13,361 g/mol (Mn  4,878
g/mol), yielding a polydispersity index (PDI) of 2.7. Potassium sul-
fate (K2SO4, 99.99% trace metals basis) and calcium hydroxide (Ca
(OH)2, 99.995% trace metals basis) were purchased from SIGMA-
ALDRICH (MO, USA). Calcium nitrate hydrate (Ca(NO3)24H2O,
99.995% metals basis) was purchased from Alfa Aesar (MA, USA).
2.2. Preparation of aqueous PCE polymer solutions
All aqueous PCE polymer solutions were prepared at room tem-
perature. Aqueous 0.1 M K2SO4 and 0.1 M Ca(NO3)2 solutions were
directly prepared with Milli-Q water (17.8 MX cm). A low-
calcium K2SO4 solution was prepared by adding a small quantity
of Ca(OH)2 (about 4 mg) into 20 mL of 0.1 M K2SO4 solution, fol-
lowed by sonication for at least 1 h. After sonication, this low-
calcium K2SO4 solution was filtered through a 0.02 lm aluminum
matrix anotop filter (Whatman, England). 0.1 M K2SO4 solution
with saturated Ca(OH)2 was prepared by the similar procedure:
adding about 20 mg Ca(OH)2 into 20 mL of 0.1 M K2SO4 solution,
followed by sonication for more than 1 h. The filtered synthetic
pore solution pH was 12. All the aqueous PCE electrolyte solu-
tions with desired PCE concentration (0.1 wt% or 1 wt%) were
obtained by direct addition of the PCE polymer to the electrolyte
solution without any further processing. In particular, these fresh
aqueous PCE electrolyte solutions were clear, with no precipitation
observed.
2.3. Force-distance measurements
A SFA was used to measure the interaction forces between mica
surfaces in aqueous electrolytes in the absence and presence of the
PCE polymer. The SFA technique directly measures the interaction
forces between two macroscopic surfaces at the molecular level
and has been extensively described elsewhere [24,33,34]. Briefly,
back silvered (550 Å) mica sheets of uniform thickness were glued
onto curved cylindrical silica disks and then mounted in a cross-
cylinder configuration in the SFA chamber. This configuration
was locally equivalent to a sphere-on-flat geometry as the surface
separation D was far smaller than the surface curvature. The abso-
lute separation D between the two mica surfaces was directly
determined by multiple beam interferometry (MBI) based on inter-
ference fringes of equal chromatic order (FECO) [35], which were
produced by passing white light through the opposing surfaces.
The FECO fringe wavelengths were measured using a spectrometer,
enabling the distance D to be determined with a resolution up to 1
Å. ‘‘Zero” distance was defined as the contact of two bare mica sur-
faces in air. The interaction force F ðDÞ between the surfaces was
measured from the deflection of a double cantilever spring sup-
porting the lower surface. To enable quantitative comparison
between different experiments, the force F ðDÞ was normalized by
the geometric mean radius R of the contact position according to
the Derjaguin approximation:
F ðDÞ ¼ 2pREðDÞ
whereEðDÞ is the interaction energy per unit area between two flat
surfaces. The measured radius R was about 1.5 cm in the reported
experiments. The Derjaguin approximation is valid when D  R.
When @FðDÞ
@D
is greater than the stiffness of the spring, there will be
a mechanical instability that can induce the lower surface to jump
either toward or away from the upper surface during approach
197
and separation, respectively [33]. Jump-into-contact and jump-out
behaviors were frequently observed in the absence of PCE’s adsorp-
tion. All the SFA experiments were conducted at 25 °C with droplets
of the various solutions injected between the surfaces. To minimize
evaporation, a small volume of water was placed at the bottom of
the SFA chamber. In most cases, experiments were done by first
measuring the interaction in pure electrolyte solution and then
exchanging the solution with the desired PCE solution.
3. Results
3.1. Interaction between mica surfaces in K2SO4 with PCE
The interaction force as a function of distance between mica
surfaces in 0.1 M K2SO4 solution is shown in Fig. 2a. No long-
range electrostatic repulsion was observed due to the high elec-
trolyte concentration. The Debye length j
1 was only 5.6 Å under
this condition, calculated by:
j
1 ¼ 0:176½K 2 SO4 
0:5 ð2Þ
where [K 2 SO4 ] is the concentration in mol/L [33]. Upon close
approach, the surfaces jumped into hard contact from about 40 Å
due to the attractive van der Waals force of the mica substrates.
Upon separation, the surfaces jumped out of adhesive contact with
F ad

3mN=m.
R
With addition of 0.1 wt% PCE polymer, a repulsive force was
detected at a separation of about 100 Å. With further approach,
the repulsion increased exponentially, as shown in the semi-
logarithmic graph in Fig. 2b. The characteristic decay length was
about 20 Å, much larger than the Debye length (5.6 Å). These mea-
surements clearly demonstrate that the PCE adsorbed onto the
negatively charged mica from 0.1 M K2SO4 solution. The repulsion
was not electrostatic but due to osmotic pressure from the
adsorbed PCE layers. The onset of steric force at 100 Å indicates
that the two adsorption layers began to interact at this separation.
Assuming that the polymer layers were symmetric, the thickness L
of one uncompressed PCE adsorption layer was about 50 Å. With
further compression, the surfaces jumped into the same hard wall
contact as in the absence of PCE, but from a smaller separation of
approximately 25 Å. These force profiles demonstrate that the
adsorbed PCE layers were squeezed out from the gap at a relatively
F
small compression Rsq  9 mN : At lower compressions, the adsorp-
m
tion layers were not squeezed out, but hysteresis was observed.
Hysteresis between the approach and separation of the surfaces
was due to structural rearrangement and slow relaxation of the
adsorbed polymer films during compression/decompression.
3.2. Interactions between mica surfaces in K2SO4 with PCE at different
calcium concentrations
The series of profiles in Fig. 3 show the interaction forces
between mica surfaces in 0.1 M K2SO4 solution with different
amounts of Ca(OH)2 in the absence and presence of PCE polymer.
With addition of a small quantity of Ca(OH)2, as shown in Fig. 3a,
the interaction was similar to that in pure 0.1 M K2SO4 solution:
van der Waals attraction at small distances with jumps into and
out of adhesive contact. The addition of 1 wt% PCE caused a similar
ð1Þ repulsion due to PCE’s adsorption as in the absence of calcium. The
repulsion was exponential with almost the same characteristic
decay length, as indicated by the corresponding semi-logarithmic
graph in Fig. 3b. The adsorption layers also exhibited hysteresis
and squeeze-out behaviors at low and high compressions, respec-
tively. By comparing these force profiles with those obtained in 0.1
M K2SO4 with 0.1 wt% PCE, the repulsion due to the adsorbed PCE
was significantly reduced, though the onset of repulsion arose at
198 B. Wu et al. / Journal of Colloid and Interface Science 527 (2018) 195–201

Fig. 2. Interaction force profiles between mica surfaces in 0.1 M K2SO4 in the absence and presence of 0.1 wt% PCE: (a) linear-linear plot; (b) linear-log plot of (a). Solid arrows
designate jump-into or jump-out-of contact. Dashed arrows indicate direction of approach or separation.

Fig. 3. Interaction force profiles between mica surfaces in 0.1 M K2SO4 with different amounts of Ca(OH)2 in the absence and presence of 1 wt% PCE: (a) with a small quantity
of Ca(OH)2, linear-linear plot; (b) linear-log plot of (a); (c) with saturated Ca(OH)2. Solid arrows designate jump-into or jump-out-of contact. Dashed arrows indicate direction
of approach or separation.
 B. Wu et al. / Journal of Colloid and Interface Science 527 (2018) 195–201
almost the same separation (around 100 Å). In particular, the
F
repulsive force at the critical squeeze-out state, Rsq , was about half.
Additionally, the critical squeeze-out separation was slightly
shifted to larger separation. The decrease in repulsion and increase
in squeeze-out/jump-into-contact separation imply a higher com-
pressibility and a lower adsorption strength. The addition of a
small amount of Ca(OH)2, therefore, weakened the adsorption
strength of PCE on mica compared to the pure K2SO4 solution.
Overall, the adsorption layers established in only K2SO4 solution
and with low calcium shared the same interaction characteristics,
suggesting that the adsorption mechanism was the same.
To mimic cement pore solution conditions, force measurements
were further performed in 0.1 M K2SO4 solution with saturated Ca
(OH)2 in the absence and presence of PCE. Fig. 3c shows that the
force-distance profiles obtained in the synthetic pore solution with
1 wt% PCE almost overlap those in the same solution without PCE.
Thus, almost no PCE adsorbed onto mica from the calcium-
saturated K2SO4 solution. The differences in adhesion were due
to changes in the effective charge density of mica surface under
these electrolyte solution conditions. In addition, differences in
the orientation of the mica substrates (alignment of the crystallo-
graphic planes) can also affect the measured adhesion [36]. Com-
parison of the force profiles in Fig. 3a and c clearly shows that
increasing the concentration of Ca(OH)2 weakened and decreased
PCE’s adsorption on mica in 0.1 M K2SO4 solution.
3.3. Interaction between mica surfaces in Ca(NO3)2 with PCE
It was very surprising that PCE adsorbed onto mica from 0.1 M
K2SO4 solution but did not after adding saturated Ca(OH)2. To con-
firm that calcium inhibited PCE’s adsorption, rather than the previ-
ously suggested bridging mechanism that induces PCE’s adsorption
on negatively charged surfaces via Ca2+/ACOO
interactions,
experiments were conducted in 0.1 M Ca(NO3)2 solution without
and with 1 wt% PCE, respectively. The measured force profiles are
shown in Fig. 4. In the pure Ca(NO3)2 solution, a significant,
short-range repulsion was observed before the surfaces jumped
into contact from about 30 Å. This repulsion was consistent with
Fig. 4. Interaction force profiles between mica surfaces in 0.1 M Ca(NO3)2 in the
absence and presence of 1 wt% PCE. Solid arrows designate jump-into or jump-out-
of contact. Dashed arrow indicates direction of approach.
199
hydration repulsion due to strong electrostatic binding/adsorption
of Ca2+ ions on mica [37]. In particular, the negatively charged mica
substrates underwent charge reversal in 0.1 M Ca(NO3)2 solution
due to calcium condensation at this concentration [38,39]. Thus,
negatively charged PCE would electrostatically adsorb onto the
now positively charged mica through its carboxylic groups ACOO
.
However, the force profiles without and with 1 wt% PCE almost
overlap, indicating that very little PCE adsorbed onto mica from
this calcium-ion-rich solution.
4. Discussion
4.1. PCE’s adsorption on mica from 0.1 M K2SO4
PEG is sometimes called ‘‘a poor chemist’s crown” for its ability
to chelate alkali metal ions [40,41]. The chelation mechanism is a
complexation interaction between the alkali metal ion and the lone
pair oxygen electrons in the ether oxygen unit
CH2 CH2 O
[42,43].
It has been reported that six ether oxygen atoms chelate one K+ ion
to form a crown ether ring [44]. For example, Zhivkova et al. [45]
found that PEO behaved as a positively charged polyelectrolyte in
aqueous solution due to cation chelation. Thus, the PCE in our stud-
ies behaved as a bifunctional polyelectrolyte with negatively
charged carboxylic groups ACOO
along the backbone and posi-
tively charged PEG arms due to K þ ion chelation. The high elec-
trolyte concentration ensured a high level of K+ ion chelation by
the PEG chains. As a result, the PCE was sufficiently positively
charged to adsorb onto negatively charged mica surfaces. This is
further supported by the thickness of the adsorbed PCE layer.
The Flory radius RF of the PEG side chain was about 35 Å, calculated
by:
RF ¼ aðMw =M w;monomer Þ0:6 ð3Þ
where a is the monomer (
CH2 CH2 O
) size (3.5 Å), Mw is the
molecular weight (2000 g/mol) of a PEG side chain and
Mw;monomer is the molecular weight (44 g/mol) of a monomer
(
CH2 CH2 O
). The thickness L of the adsorbed PCE layer was about
50 Å, suggesting that some PEG chains adsorbed on mica while
others extended into solution. A range betweenRF and 2RF is
expected for a weakly adsorbed polymer [33].
The surprising squeeze-out behavior at a relatively low com-
pression (9 mN/m) indicates that PCE’s adsorption on mica due
to K þ ion chelation was weak. An earlier study by Chai et al. [43]
showed that the adsorption layers of PEO (M w ¼ 170; 000g=mol,
Mw
Mn
¼ 1:02, concentration = 150 lg/mL) on mica in 0.1 M KNO3 were
not squeezed out. Compared to this study with high molecular
weight PEO, the PCE here has 5–6 relatively short PEG side chains
(M w  2; 000g=mol per chain). The adsorption strength scales with
the molecular weight and the molecular weight also dictates the
total positive charge chelation of the polymer [46]. Further, the
negatively charged carboxylic backbone should also reduce the
binding strength to mica because of electrostatic repulsion. As a
result, the PCE had a much lower binding strength and could be
squeezed out at modest compression. Such squeeze-out phe-
nomenon is relatively rare, but has been observed with low MW
PEG chains adsorbing to lipid bilayers in KNO3 solutions [44,47].
Similarly, it has also been reported that hyaluronic acid layers
adsorbing on mica can be squeezed out with compression in CaCl2
Hepes buffer solution [48]. In this case, the weak adsorption was
attributed to the mediation of a water molecule, which decreased
the strength of the Ca2+ ion bridging interaction between mica
and
COO
groups in hyaluraonic acid. Additionally, shear was also
reported to induce abrupt removal of adsorbed polyelectrolyte and
neutral polymer brushes from contacting mica surfaces [49,50].
200 B. Wu et al. / Journal of Colloid and Interface Science 527 (2018) 195–201

Another unusual phenomenon observed here was that force- was abolished due to electrostatic repulsion between both the
distance profiles of multiple successive approach-separation runs effectively positively charged mica surface and PCE.
mostly overlap. When the PCE polymer was excluded from the The situation in practical cement suspensions is much more
gap at high compression, attractive van der Waals force brought complicated. High concentrations of ions including Ca2+ are
the surfaces into direct contact. Upon separating the surfaces sev- released upon cement hydration. Both positively and negatively
eral hundred angstroms, the polymer immediately achieved an charged particles are present and PCE may adsorb onto both types
almost identical re-adsorption. In addition, the reproducibility of of surfaces depending on the charge density of the surface, rate of
the force profiles showing compression-decompression hysteresis hydration and local effective salt solution conditions. Given the
also indicates that the PCE adsorption layers almost immediately effort to keep the water to cement ratio as low as possible, non-
recovered to their uncompressed state once the surfaces were sep- equilibrium effects that limit the rate of dissociation may also be
arated more than 2L. Significant time-dependent effects in succes- very important in real applications. Additionally, there is a
sive compression-decompression cycles between monodisperse dynamic balance between two driving forces of PCE’s adsorption
PEO (M w ¼ 160; 000g=mol) adsorption layers on mica in 0.1 M to cement particles: enthalpic interactions of carboxylic acid with
KNO3 were observed by Klein and Luckham [51]. Pettersson et al. calcium and entropic contributions due to the release of water
[52] found that the molar percentage of charged backbone seg- and counterions from carboxylic acid and calcium on the surfaces.
ments had an important influence on the hysteresis between com- The current study implies that cement dispersants may also func-
pression and decompression of PEG bottle brush polymers. The tion as a lubricant between the aluminosilicate clay surfaces in the
reproducibility of the profiles in this work suggests that the pore solution under high calcium concentration.
adsorbed PCE molecules were retained in the gap between the sur-
faces during compression until the compression reached the critical
5. Conclusions
squeeze-out level. A structurally similar polyelectrolyte PLL–g–PEG
also exhibited this retention characteristic during solvent exchange
Adsorption behaviors of a commercial PCE on negatively
and rinsing [53,54]. Although the squeezing out and relatively mod-
charged mica in 0.1 M K2SO4 solution with different concentrations
est adsorption strength of PCE to negatively charged surfaces are
of Ca(OH)2 were directly investigated by SFA. The PCE weakly
undesirable behaviors when considering PCE as a long-term
adsorbed onto mica from the K2SO4 solution due to K+ ion chelation
lubricant and dispersion modifier, the rapid re-adsorption likely
by PEG chain segments (ether oxygen units ACH2CH2OA) and the
mitigates this effect in rheological flow conditions, which is of pri-
adsorption layers were squeezed out with modest compression.
mary importance in cement placement and consolidation.
Adding a small quantity of Ca(OH)2 further weakened the adsorp-
tion strength and increasing Ca(OH)2 to saturated solution condi-
tions abolished the adsorption. The squeeze-out phenomenon in
4.2. Ca2+ ion effect on PCE’s adsorption in 0.1 M K2SO4
K2SO4 solution is relatively rare, which increases our understanding
of polyelectrolyte adsorption behaviors in high-salt conditions. Ca2+
One adsorption mechanism of PCE is thought to rely on Ca2+ ion
ions are commonly thought to anchor the backbone carboxylic
bridging between the backbone carboxylic groups
COO
and
groups ACOO
to negatively charged cement particle surfaces to
negatively charged particle surfaces [16,21]. Ca2+ ion bridging has
form robust PCE adsorption layers for cement suspension stabiliza-
also been suggested as the mechanism for the adsorption of
tion [16,21]. Therefore, PCE was expected to adsorb on mica in syn-
ACOO
-containing sodium polyacrylate [55] and hyaluronic acid
thetic pore solution conditions through Ca2+/ACOO
interactions.
[48] on negatively charged mica. In contrast, the high-resolution
Surprisingly, however, the effect of Ca2+ ion on the adsorption
force measurements by SFA here clearly reveal that the expected
was inverse to the hypothesized bridging mechanism. Compared
adsorption of PCE on mica by Ca2+ ion bridging did not occur and
to previous force-distance measurements in Milli-Q water [8,31],
that PCE’s adsorption in 0.1 M K2SO4 was significantly weakened
monovalent electrolyte aqueous solutions [1,8,30,32] and low-
after adding a small quantity of Ca(OH)2.
calcium aqueous solutions [2] by AFM and SFA, the present mea-
As with monovalent K+ ions, divalent Ca2+ ions can be also che-
surements give us new understanding of the interactions between
lated by the PEG chain segments (ether oxygen units ACH2CH2OA)
PCE and negatively charged surfaces in high-calcium aqueous solu-
[40,56,57]. Thus, the Ca2+/ACOO
and Ca2+/ACH2CH2OA complexa-
tions at high ionic strength and pH, which have not been previously
tion interactions could potentially induce PCE aggregation, thereby
explored. It is precisely these conditions where our understanding
reducing the effective amount of PCE available to adsorb. However,
of polyelectrolyte absorption behavior is limited. The surprising
the hydrodynamic diameter determined by dynamic light scatter-
effect of Ca2+ ions can be explained by charge reversal that hap-
ing (Zetasizer Nano (UK) at 25 °C) at the concentration of 1 wt% in
pened to both the mica surface due to strong electrostatic adsorp-
Milli-Q water was 9.3 nm and was only slightly decreased to 7.3
tion/binding of Ca2+ ions and the PCE due to Ca2+ ion chelation by
nm and 6.9 nm in 0.1 M K2SO4 solution and 0.1 M Ca(NO3)2 solution
the ether oxygen units ACH2CH2OA on the PEG side chains and
due to charge screening, respectively. Therefore, PCE molecules did
backbone carboxylic groups ACOO
. What is less appreciated in
not aggregate in the experimental solution conditions.
the literature is that divalent ion bridging mechanisms should only
Nevertheless, the presence of calcium changed the charging
function over a finite concentration range and will depend on the
behaviors of both PCE and the mica surface. With addition of a
charge density of the particles, architecture of the polymer disper-
small amount of Ca(OH)2, the electrostatic adsorption/binding of
sant and solution conditions. Correspondingly, future studies of
Ca2+ ions can partially neutralize some of the negatively charged
PCE’s adsorption behaviors as a function of calcium concentration
sites on the mica surface as well as the negatively charged car-
will further inform their use as dispersants under high ionic
boxylic groups along the PCE backbone. As a result, the electro-
strength conditions.
static attraction between PCE and mica was weakened in the
low-calcium conditions, thereby decreasing the adsorption
strength. At high calcium concentrations, the PCE became more Acknowledgements
positively charged due to Ca2+ ion chelation by PEG chain units
ACH2CH2OA and carboxylic groups ACOO
. Meanwhile, the mica This work was supported by GCP Applied Technologies,
was subjected to charge reversal due to strong electrostatic Cambridge, Massachusetts, USA and the National Natural Science
adsorption/binding of Ca2+ ions [38,39]. Consequently, adsorption Foundation of China (51675120, U1537214 and U1637206). The
 B. Wu et al. / Journal of Colloid and Interface Science 527 (2018) 195–201
overseas study of author Bo Wu at the University of California at
Davis was sponsored by the non-profit China Scholarship Council.
References
[1] A. Kauppi, K.M. Andersson, L. Bergström, Probing the effect of superplasticizer
adsorption on the surface forces using the colloidal probe AFM technique, Cem.
Concr. Res. 35 (2005) 133–140.
[2] R.J. Flatt, I. Schober, E. Raphael, C. Plassard, E. Lesniewska, Conformation of
adsorbed comb copolymer dispersants, Langmuir 25 (2008) 845–855.
[3] P.-C. Aitcin, C. Jolicoeur, J.G. MacGregor, Superplasticizers: how they work and
why they occasionally don’t, Concr. Int. 16 (1994) 45–52.
[4] E. Sakai, A. Ishida, A. Ohta, New trends in the development of chemical
admixtures in Japan, J. Adv. Concr. Technol. 4 (2006) 211–223.
[5] Y. Li, C. Yang, Y. Zhang, J. Zheng, H. Guo, M. Lu, Study on dispersion, adsorption
and flow retaining behaviors of cement mortars with TPEG-type polyether
kind polycarboxylate superplasticizers, Constr. Build. Mater. 64 (2014) 324–
332.
[6] K. Yoshioka, E.-I. Tazawa, K. Kawai, T. Enohata, Adsorption characteristics of
superplasticizers on cement component minerals, Cem. Concr. Res. 32 (2002)
1507–1513.
[7] H. Hommer, Interaction of polycarboxylate ether with silica fume, J. Eur.
Ceram. Soc. 29 (2009) 1847–1853.
[8] L. Ferrari, J. Kaufmann, F. Winnefeld, J. Plank, Interaction of cement model
systems with superplasticizers investigated by atomic force microscopy, zeta
potential, and adsorption measurements, J. Colloid Interface Sci. 347 (2010)
15–24.
[9] T. Sowoidnich, T. Rachowski, C. Rößler, A. Völkel, H.-M. Ludwig, Calcium
complexation and cluster formation as principal modes of action of polymers
used as superplasticizer in cement systems, Cem. Concr. Res. 73 (2015) 42–50.
[10] O. Akhlaghi, Y.Z. Menceloglu, O. Akbulut, Poly(carboxylate ether)-based
superplasticizer achieves workability retention in calcium aluminate cement,
Sci. Rep., 7, 2017.
[11] S.S. Sheiko, B.S. Sumerlin, K. Matyjaszewski, Cylindrical molecular brushes:
synthesis, characterization, and properties, Prog. Polym. Sci. 33 (2008) 759–
785.
[12] J. Plank, E. Sakai, C. Miao, C. Yu, J. Hong, Chemical admixtures—chemistry,
applications and their impact on concrete microstructure and durability, Cem.
Concr. Res. 78 (2015) 81–99.
[13] C.M. Bates, F.S. Bates, 50th anniversary perspective: block polymers pure
potential, Macromolecules 50 (2016) 3–22.
[14] S. Rathgeber, T. Pakula, A. Wilk, K. Matyjaszewski, K.L. Beers, On the shape of
bottle-brush macromolecules: systematic variation of architectural
parameters, J. Chem. Phys. 122 (2005) 124904.
[15] B. Zhang, F. Gröhn, J.S. Pedersen, K. Fischer, M. Schmidt, Conformation of
cylindrical brushes in solution: effect of side chain length, Macromolecules 39
(2006) 8440–8450.
[16] E. Sakai, K. Yamada, A. Ohta, Molecular structure and dispersion-adsorption
mechanisms of comb-type superplasticizers used in Japan, J. Adv. Concr.
Technol. 1 (2003) 16–25.
[17] S. Lee, M. Müller, M. Ratoi-Salagean, J. Vörös, S. Pasche, S.M. De Paul, H.A.
Spikes, M. Textor, N.D. Spencer, Boundary lubrication of oxide surfaces by poly
(L
lysine)
g
poly(ethylene glycol) (PLL
g
PEG) in aqueous media, Tribo.
Lett. 15 (2003) 231–239.
[18] X. Banquy, J. Burdyńska, D.W. Lee, K. Matyjaszewski, J. Israelachvili,
Bioinspired bottle-brush polymer exhibits low friction and amontons-like
behavior, J. Am. Chem. Soc. 136 (2014) 6199–6202.
[19] Z. Zhang, H. Ma, D.B. Hausner, A. Chilkoti, T.P. Beebe, Pretreatment of
amphiphilic comb polymer surfaces dramatically affects protein adsorption,
Biomacromolecules 6 (2005) 3388–3396.
[20] S. Pasche, S.M. De Paul, J. Vörös, N.D. Spencer, M. Textor, Poly(l
lysine)

graft
poly(ethylene glycol) assembled monolayers on niobium oxide
surfaces: a quantitative study of the influence of polymer interfacial
architecture on resistance to protein adsorption by ToF-SIMS and in situ
OWLS, Langmuir 19 (2003) 9216–9225.
[21] J. Plank, B. Sachsenhauser, J. De Reese, Experimental determination of the
thermodynamic parameters affecting the adsorption behavior and dispersion
effectiveness of PCE superplasticizers, Cem. Concr. Res. 40 (2010) 699–709.
[22] G. Binnig, C.F. Quate, C. Gerber, Atomic force microscope, Phys. Rev. Lett. 56
(1986) 930–933.
[23] W.A. Ducker, T.J. Senden, R.M. Pashley, Direct measurement of colloidal forces
using an atomic force microscope, Nature 353 (1991) 239–241.
[24] J. Israelachvili, Y. Min, M. Akbulut, A. Alig, G. Carver, W. Greene, K. Kristiansen,
E. Meyer, N. Pesika, K. Rosenberg, Recent advances in the surface forces
apparatus (SFA) technique, Rep. Prog. Phys. 73 (2010) 036601.
[25] E. Spruijt, M.A. Cohen Stuart, J. van der Gucht, Dynamic force spectroscopy of
oppositely charged polyelectrolyte brushes, Macromolecules 43 (2010) 1543–
1550.
[26] Q. Lu, D.S. Hwang, Y. Liu, H. Zeng, Molecular interactions of mussel protective
coating protein, mcfp-1, from Mytilus californianus, Biomaterials 33 (2012)
1903–1911.
201
[27] J. Yu, J. Mao, G. Yuan, S. Satija, Z. Jiang, W. Chen, M. Tirrell, Structure of
polyelectrolyte brushes in the presence of multivalent counterions,
Macromolecules 49 (2016) 5609–5617.
[28] J. Faivre, B.R. Shrestha, J. Burdynska, G. Xie, F. Moldovan, T. Delair, S. Benayoun,
L. David, K. Matyjaszewski, X. Banquy, Wear protection without surface
modification using a synergistic mixture of molecular brushes and linear
polymers, ACS Nano 11 (2017) 1762–1769.
[29] U. Raviv, S. Glasson, N. Kampf, J.-F. Gohy, Lubrication by charged polymers,
Nature 425 (2003) 163–165.
[30] H. Uchikawa, S. Hanehara, D. Sawaki, The role of steric repulsive force in the
dispersion of cement particles in fresh paste prepared with organic admixture,
Cem. Concr. Res. 27 (1997) 37–50.
[31] L. Ferrari, L. Bernard, F. Deschner, J. Kaufmann, F. Winnefeld, J. Plank,
Characterization of polycarboxylate-ether based superplasticizer on cement
clinker surfaces, J. Am. Ceram. Soc. 95 (2012) 2189–2195.
[32] L. Zhang, Q. Lu, Z. Xu, Q. Liu, H. Zeng, Effect of polycarboxylate ether comb-type
polymer on viscosity and interfacial properties of kaolinite clay suspensions, J.
Colloid Interface Sci. 378 (2012) 222–231.
[33] J.N. Israelachvili, Intermolecular and Surface Forces, third ed., Academic press,
London, 2010.
[34] J. Israelachvili, Direct measurements of forces between surfaces in liquids at
the molecular level, Proc. Natl. Acad. Sci. 84 (1987) 4722–4724.
[35] J. Israelachvili, Thin film studies using multiple-beam interferometry, J. Colloid
Interface Sci. 44 (1973) 259–272.
[36] P.M. Mcguiggan, J.N. Israelachvili, Measurements of the effect of angular lattice
mismatch on the adhesion energy between 2 mica surfaces in water, Mater.
Res. Soc. 138 (1989) 349–360.
[37] R. Kjellander, S. Marčelja, R. Pashley, J. Quirk, A theoretical and experimental
study of forces between charged mica surfaces in aqueous CaCl2 solutions, J.
Chem. Phys. 92 (1990) 4399–4407.
[38] D.F. Parsons, B.W. Ninham, Charge reversal of surfaces in divalent electrolytes:
the role of ionic dispersion interactions, Langmuir 26 (2010) 6430–6436.
[39] P. Kékicheff, S. Marc̆elja, T. Senden, V. Shubin, Charge reversal seen in electrical
double layer interaction of surfaces immersed in 2:1 calcium electrolyte, J.
Chem. Phys. 99 (1993) 6098–6113.
[40] F. Bailey, R. Callard, Some properties of poly(ethylene oxide)1 in aqueous
solution, J. Appl. Polym. Sci. 1 (1959) 56–62.
[41] D. Balasubramanian, B. Chandani, Poly(ethylene glycol): a poor chemist’s
crown, J. Chem. Educ. 60 (1983) 77–78.
[42] L. Chai, J. Klein, Role of ion ligands in the attachment of poly(ethylene oxide) to
a charged surface, J. Am. Chem. Soc. 127 (2005) 1104–1105.
[43] L. Chai, R. Goldberg, N. Kampf, J. Klein, Selective adsorption of poly(ethylene
oxide) onto a charged surface mediated by alkali metal ions, Langmuir 24
(2008) 1570–1576.
[44] T.L. Kuhl, A.D. Berman, S.W. Hui, J.N. Israelachvili, Part 2. Crossover from
depletion attraction to adsorption: polyethylene glycol induced electrostatic
repulsion between lipid bilayers, Macromolecules 31 (1998) 8258–8263.
[45] I. Zhivkova, A. Zhivkov, D. Stoychev, Electrostatic behavior of polyethylene
oxide, Eur. Polym. J. 34 (1998) 531–538.
[46] P.G. Degennes, Scaling theory of polymer adsorption, Journal de Physique 37
(1976) 1445–1452.
[47] T.L. Kuhl, A.D. Berman, S.W. Hui, J.N. Israelachvili, Part 1. Direct measurement
of depletion attraction and thin film viscosity between lipid bilayers in
aqueous polyethylene glycol solutions, Macromolecules 31 (1998) 8250–8257.
[48] R. Tadmor, N. Chen, J. Israelachvili, Normal and shear forces between mica and
model membrane surfaces with adsorbed hyaluronan, Macromolecules 36
(2003) 9519–9526.
[49] U. Raviv, S. Giasson, N. Kampf, J.-F. Gohy, R. Jérôme, J. Klein, Normal and
frictional forces between surfaces bearing polyelectrolyte brushes, Langmuir
24 (2008) 8678–8687.
[50] J. Klein, E. Kumacheva, D. Perahia, L. Fetters, Shear forces between sliding
surfaces coated with polymer brushes: the high friction regime, Acta
Polymerica 49 (1998) 617–625.
[51] J. Klein, P.F. Luckham, Forces between two adsorbed poly(ethylene oxide)
layers in a good aqueous solvent in the range 0–150 nm, Macromolecules 17
(1984) 1041–1048.
[52] T. Pettersson, A. Naderi, R. Makuška, P.M. Claesson, Lubrication properties of
bottle-brush polyelectrolytes: an AFM study on the effect of side chain and
charge density, Langmuir 24 (2008) 3336–3347.
[53] M.T. Müller, X. Yan, S. Lee, S.S. Perry, N.D. Spencer, Lubrication properties of a
brushlike copolymer as a function of the amount of solvent absorbed within
the brush, Macromolecules 38 (2005) 5706–5713.
[54] E. Guzman, F. Ortega, M.G. Prolongo, V.M. Starov, R.G. Rubio, Influence of the
molecular architecture on the adsorption onto solid surfaces: comb-like
polymers, Phys. Chem. Chem. Phys. 13 (2011) 16416–16423.
[55] J.M. Berg, P.M. Claesson, R.D. Neuman, Interactions between mica surfaces in
sodium polyacrylate solutions containing calcium ions, J. Colloid Interface Sci.
161 (1993) 182–189.
[56] M.J. Reddy, P.P. Chu, Effect of Mg2+ on PEO morphology and conductivity, Solid
State Ionics 149 (2002) 115–123.
[57] N. Rakkapao, V. Vao-soongnern, Y. Masubuchi, H. Watanabe, Miscibility of
chitosan/poly(ethylene oxide) blends and effect of doping alkali and alkali
earth metal ions on chitosan/PEO interaction, Polymer 52 (2011) 2618–2627.