Applied Surface Science 466 (2019) 607–614

Contents lists available at ScienceDirect

Applied Surface Science

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

Full Length Article

Interaction between polysaccharide monomer and SiO2/Al2O3/CaCO3 T

surfaces: A DFT theoretical study
⁎
Hui Zhao, Na Qi, Ying Li
Key Laboratory of Colloid and Interface Chemistry of State Education Ministry, Shandong University, 27 South Road of ShanDa, Jinan, Shandong 250100, PR China

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

Keywords: In the present study, the interaction between a model polysaccharides monomer β-D-glucopyranose with dif-
Polysaccharides monomer-Al2O3/CaCO3/SiO2 ferent mineral solid surfaces, including hydroxylated (0 0 1) surface of SiO2, (0 0 0 1) surface of Al2O3 and (1 0 4)
surfaces interaction surface of CaCO3, were explored by theoretical calculations based on Density Functional Theory (DFT) under
DFT calculation periodic boundary conditions. The adsorption geometry of the β-D-glucopyranose monomer (GM) at different
Interaction mechanism
solid surface and the interaction energies were analyzed in detail, and the interaction mechanism was de-
termined via electron density differences iso-surface, Mulliken charge, as well as projected density states (PDOS)
analysis. Very strong interaction was found between GM and Al2O3 (0 0 0 1) surface, which was mainly attrib-
uted to the formation of bridging AleO bonds and H-bonds. There was the bonding between H-1s and O-2p
orbitals for the formed hydrogen bonds and hybridization between O-p states and Al-d states for the bridging
AleO bonds. The interaction between GM and CaCO3 (1 0 4) surface was mainly attributed to the electrostatic
interaction and hydrogen bonded interaction, and was fairly strong, too. But there was only hydrogen bonds
between GM and SiO2 surface, which was relative weak. In all the three adsorbed systems, significant charge
redistribution was observed which contributed to the interaction between them, though little electron trans-
ferred between them. The finding is very meaningful for providing theoretical direction in proceeding sufficient
application of diverse kind of natural sugar-based functional substances in many fields.

1. Introduction
As for the increasing social concern about the serious environmental
problems, environmental friendly technologies and products have
gained more and more attention, and the application of the green
functional substances derived from natural raw materials is on rise [1]
in various industrial process. Polysaccharides and modified chemicals
based on sugar are widely concerned because of their non-toxic and
non-hazardous nature which have broad applications in many in-
dustrial processes, ranging from enhanced oil recovery [2–5], disper-
sion [6], ceramic industry [7,8] to biomineralization [9,10], mineral
scaling [11] and so on. In these applications, the adsorption of these
functional substances upon mineral surfaces is a key scientific question,
which has aroused researchers’ interest [12–14]. For example, in En-
hanced Oil Recovery (EOR) process, the adsorption loss often occurs
due to the interaction between polysaccharides polymer or sugar based
surfactants and the mineral surfaces such as SiO2 and CaCO3, which
needs to be controlled [15,16]. While in some other cases, the strong
adsorption of polysaccharides upon the mineral surfaces is favorable.
For instance, polysaccharides could control the biomineralization
process which attributes to the strong interaction with CaCO3 surfaces
[9,10,17,18] and be proposed for the aqueous gelcasting of ceramic and
Al2O3 powers in ceramic industry [7,8].
The adsorption of substances on solid rock surface is a complex
process, which is affected by the environment such as pH, salinity as
well as additives [13,14]. Thus, it is meaningful and helpful to elucidate
how the functional molecules interact with the mineral surfaces exactly
from microscopic view for better understand the corresponding ad-
sorption behavior, while it is difficult to be analyzed at atomic scale by
experimental techniques.
Molecular dynamic simulations have been recognized as a powerful
method to study the adsorption behaviors of small molecules or mac-
romolecules at mineral surfaces, as well as the interaction between
them [19–22]. But some detailed information at atomic scale is ignored
in these calculations, such as the change in the surface properties after
adsorption, the electronic structure and the charge transfer and so on.
Theoretical calculation based on the ab initio method, Density Func-
tional Theory (DFT) have proved to be an appropriate method to un-
ravel the interaction mechanism of organic and inorganic molecules
with solid surfaces, as they could give out the more specific details

⁎
Corresponding author at: School of Chemistry and Chemical Engineering, Shandong University, 27 South Road of ShanDa, Jinan, Shandong 250100, PR China.
E-mail address: yingli@sdu.edu.cn (Y. Li).

https://doi.org/10.1016/j.apsusc.2018.10.085
Received 21 July 2018; Received in revised form 22 September 2018; Accepted 9 October 2018
Available online 10 October 2018
0169-4332/ © 2018 Elsevier B.V. All rights reserved.
H. Zhao et al.
about the adsorption geometry, identify charge transfer and electronic
effects at the atomic scale [23–32].
In the present work, to explore the interaction mechanism between
polysaccharides or sugar based surfactants and the mineral surfaces
essentially, a model polysaccharides monomer β-D-glucopyranose was
selected in the calculations, which represented the constituent re-
peating unit of polysaccharides. The interaction mechanism between
the β-D-glucopyranose monomer (GM) and SiO2 (0 0 1), Al2O3 (0 0 0 1),
CaCO3 (1 0 4) surfaces at atomic scale was determined using DFT cal-
culations. The adsorption geometry, interaction energy, charge transfer
as well as the projected density states were analyzed. The results of this
study provide important theoretical knowledge which is very helpful
for better understand about the adsorption behavior of the poly-
saccharides and sugar based additives at the solid surfaces at the atomic
level, which is one of the key issues in their efficient usage in various
applications.
2. Modeling and DFT computational details
The primitive unit cells of SiO2, Al2O3 and CaCO3 were optimized
using first principle Density Functional Theory (DFT) implemented in
Dmol3 code (Dmol3 is available as part of Material Studio) [33,34]. The
generalized gradient approximation (GGA) for
Perdew − Burke − Ernzerhof (PBE) exchange–correlation was used in
the calculations [35–37]. To describe the interaction more appro-
priately, the DFT-D2 [38] approach of Grimme was used to account for
the London dispersion force [39], which has been shown to perform
well for the adsorption properties of molecules on these three solid
surfaces [31,40–48]. The all-electron Konhn-Sham wave functions and
DNP numerical basis set with a polarization d function and a polar-
ization p function on hydrogen were used. The atom-centered grids
were used for the numerical integration with a real-space cutoff of
4.2 Å. The k-space integrations were performed using a Mon-
khorst–Pack [49] grid and the k-point mesh was 3 × 3 × 4 for the
optimization of SiO2, Al2O3 and CaCO3 unit cells. The self-consistent-
field (SCF) convergence criterion was set as 10−5 ev, and the con-
vergence criteria were 5.44 × 10−4 eV for energy, 0.004 Ha/Å for
force, 0.005 Å for maximum displacement.
Once the optimized unit cells of SiO2, Al2O3 and CaCO3 were ob-
tained, the periodic supercells: (4 × 4 × 1) for Al2O3 and (4 × 4 × 2)
for SiO2 with a vacuum thickness of 17 Å were constructed for GM
adsorption. The chemical structure of β-D-glucopyranose was shown in
Fig. 1. The (0 0 1) surface of SiO2, (0 0 0 1) surface of Al2O3 and (1 0 4)
surface of CaCO3 were considered in the calculations. When optimiza-
tion, half of the “sublayers” atoms were fixed and the constructed initial
adsorption models of GM on SiO2, Al2O3 and CaCO3 surfaces were
shown in Fig. 2. The optimization of GM was performed in a
(19 Å × 19 Å × 27 Å) cell which had the same size with SiO2 and Al2O3
systems, and in a (25 Å × 21 Å × 31 Å) cell which was same as the
CaCO3 adsorbed system. All the calculations were performed with PBE/
DNP and Γ-point, which were shown to be sufficient to achieve con-
vergence for the adsorption energies [41].
The interaction strength between GM and the solid surfaces could be
represented by the interaction energy (ΔEint ), which was defined as
following:
Fig. 1. Chemical structure of β-D-glucopyranose in the DFT calculations.
608
Applied Surface Science 466 (2019) 607–614
ΔEint = EGM + slab−(EGM + Eslab)
where EGM + slab , EGM and Eslab represented the energy of β-D-gluco-
pyranose adsorbed on the mineral surfaces, the energy of β-D-gluco-
pyranose in the same cell with the adsorbed systems, the energy of the
bare surfaces, respectively. Usually, the more negative was the inter-
action energy, the stronger was the interaction strength between the
adsorbate and the surface [50]. Finally, to further consider the effect of
water, the COSMO solvation model [51–53] was taken to optimize the
adsorption of GM at the mineral surfaces.
3. Results and discussion
3.1. DFT calculations
3.1.1. Geometry optimization of SiO2/Al2O3/CaCO3 unit structure
Our optimized primitive unit cells were characterized by the para-
meters: a = b = 4.868 Å, c = 5.404 Å for SiO2; a = b = 4.800 Å,
c = 13.010 Å for Al2O3, a = b = 5.023 Å, c = 17.409 Å for CaCO3,
closely coincided with the experimental values [54–56] and the pre-
viously published DFT calculated results [56–59] (see Table 1). Like-
wise, it could be seen from Table 1 that the bond length of Si-O of SiO2,
AleO of Al2O3 and CeO of CaCO3 calculated here generally agreed with
the experimental and DFT calculated values.
3.1.2. Interaction energies between GM and SiO2/Al2O3/CaCO3 surfaces
In the calculations, two different adsorption geometries: a parallel
geometry where GM was lying on SiO2, Al2O3 and CaCO3 surfaces and a
vertical geometry where GM was almost perpendicular to the surfaces
were considered, as shown in Fig. 2. The calculated interaction energies
between GM and SiO2 (0 0 1), Al2O3 (0 0 0 1), CaCO3 (1 0 4) surfaces
were shown in Table 2. It was found that the interaction between GM
and Al2O3, CaCO3 surfaces was quite strong while it was rather weak
with SiO2 surface.
3.1.3. Interaction geometries between GM and SiO2/Al2O3/CaCO3 surfaces
To determine the interaction mechanism between GM and SiO2,
Al2O3, CaCO3 surfaces, the optimized configurations were analyzed. By
analysis of the parallel adsorption geometry at SiO2 (0 0 1) surface, we
observed only small elongations in the O5eC1, O5eC5 and O1eH1
bonds and the C1eO5eC5 and C1eO1eH1 angles (Table 3, the atom
label was shown in Fig. 1) compared to the structural data of the free β-
D-glucopyranose molecule, which may be attributed to the hydrogen-
bonded interactions with SiO2 (0 0 1) surface. As was shown in Fig. 3,
two hydrogen bonds were formed between GM and SiO2 surface in
parallel and vertical geometry. The corresponding hydrogen bond
length and angle were listed in Table 4. In the hydrogen bond studies,
the hydrogen bonds were classified into three types based on a distance
for H⋯Y and an angle for XeH⋯Y [60], such as strong (1.2 Å <
H⋯Y < 1.5 Å, 170° < X-H⋯Y < 180°), moderate (1.5 Å <
H⋯Y < 2.2 Å, X-H⋯Y > 130°) and weak (H⋯Y > 3.2 Å, X-
H⋯Y > 90°) hydrogen bonds [61]. From Table 4, we observed the
strength of the most hydrogen bonds formed between GM and SiO2
surface was moderate. This suggested that hydrogen-bonded interac-
tions contributed significantly to the adsorption of GM at SiO2 (0 0 1)
surface, which could explain why the interaction between GM and SiO2
surface was relative weak.
The adsorption configuration of GM at Al2O3 (0 0 0 1) surface was
displayed in Fig. 4, wherein GM interacted with Al and O atoms of
Al2O3 surface via its O atom and H atom, respectively. Before GM ad-
sorption, all the Al atoms were flat at Al2O3 surface (Fig. 2c and d). But
when interacted with O atoms of GM, Al atoms were raised and the
AleO bond length at Al2O3 surface increased from 1.705 Å to 1.733 Å.
In the parallel adsorbed configuration (Fig. 4a), we have observed
elongations in O5eC1, O5eC5 and O3eC3 bonds as well as decrease in
C5eO5eC1 and C3eO3eH3 angles (Table 5), compared to the
H. Zhao et al. Applied Surface Science 466 (2019) 607–614

Fig. 2. The initial adsorbed conformation of GM on (a), (b) SiO2 (0 0 1) surface, (c), (d) Al2O3 (0 0 0 1) surface and (e) (f) CaCO3 (1 0 4) surface with parallel and
vertical geometries. The C, O, H, Al, Si Ca atoms were drawn in green, red, white, pink and yellow, blue respectively.

Table 1
Comparison of experimental values and previously published results with our calculated lattice parameters (in Å) of SiO2, Al2O3 and CaCO3.
SiO2 Al2O3 CaCO3

a b c Si-O a b c AleO a b c CeO

Experimental value 4.916 4.916 5.405 1.605 4.759 4.759 12.993 1.853 5.004 5.004 16.969 1.300
[54] [55] [56]
Published DFT value 5.052 5.052 5.547 1.630 4.800 4.800 13.114 1.872 5.100 5.100 17.160 1.100
[57] [58] [59]
This work 4.868 4.868 5.404 1.632 4.800 4.800 13.010 1.866 5.023 5.023 17.409 1.296

Table 2 structural data of the unabsorbed β-D-glucopyranose molecule, which

Interaction energies (in eV) of GM adsorbed at SiO2/Al2O3/CaCO3 surfaces with could attribute to the direct interaction between O atoms in GM and Al
different geometries. atoms in Al2O3 surface. The formed bridging OeAl distances between
SiO2 surface Al2O3 surface CaCO3 surface
GM and Al2O3 surface were calculated at 1.993–1.995 Å, as can be seen
in Fig. 4. Besides, O1eH1, O3eH3 and O6eH6 bonds elongations were
Eparal −0.90 −3.71 −2.35 also observed (Table 5) of β-D-glucopyranose molecule, which may be
Evert −0.81 −2.96 −1.03 attributed to the presence of hydrogen-bonded interactions between the
H atoms of GM and O atoms of Al2O3 surface as reported in Fig. 4. The
analysis of bond length and angles (Table 4) of the formed H-bonds
Table 3
between GM and Al2O3 surface told us that the strength of some hy-
Interatomic distance and angle of β-D-glucopyranose molecule before and after
adsorption at SiO2 (0 0 1) surface in parallel geometry. The atom label was drogen bonds formed at Al2O3 surface was weak. But the interactions
shown in Fig. 1. between them were quite strong (−3.71 eV), which therefore suggested
that the formed bridging AleO bonds contributed significantly to the
Before adsorption After adsorption
strong adsorption of GM at Al2O3 (0 0 0 1) surface.
Bond length/Å d(O5eC1) 1.440 1.479 To ensure the important role of the bridging AleO bonds on the
d(O5eC5) 1.429 1.439 strong interaction between β-D-glucopyranose and Al2O3 surface, cal-
d(O1eH1) 0.973 0.993 culations of cellobiose which consistes of two β-D-glucopyranose con-
Angle/° ∠(C1eO5eC5) 112.149 112.784 nected through the 1 → 4 β-glucosidic linkage at Al2O3 and SiO2 surface
∠(C1eO1eH1) 108.571 109.730 were performed. The optimized configurations and the corresponding
interaction energies were shown in Fig. 5a and b. As can be seen, one
more bridging AleO bond was observed between cellobiose and Al2O3
surface, and the interaction energy was about 3 eV lower than that with
GM. This confirmed the important contribution of the bridging AleO

609
H. Zhao et al. Applied Surface Science 466 (2019) 607–614

Fig. 3. Optimized configurations of GM adsorbed on SiO2 (0 0 1) surface. (a) Side view of the parallel geometry (b) side view of the vertical geometry. 1a, 1b were the
hydrogen bonds formed between them and the corresponding bond length. Atom colors: C, green; O, red; Si, yellow; H, white.

Table 4
Bond length, in angstroms (Å) and angles in degrees (°) of hydrogen bonds formed by GM and SiO2/Al2O3/CacO3 surfaces. The hydrogen labels of 1a, 1b, 2a⋯2d, 3a,
3b were shown in Figs. 3, 4 and 6.
Model SiO2 surface Al2O3 surface CaCO3 surface

Bond Length Angle Bond Length Angle Bond length Angle

GM-paral* 1a 1.642 170.81 2a 1.985 160.79 3a 1.869 162.10

1b 1.727 174.90 2b 2.389 110.96 3b 1.716 161.04
2c 1.715 168.11

GM-vert* 1a 1.620 164.89 2a 1.773 156.84 3a 1.789 157.99

1b 1.754 156.13 2b 1.402 172.44 3b 1.768 145.21
2c 2.454 92.79
2d 2.183 102.45

* GM-paral and GM-vert represented β-D-glucopyranose adsorbed at SiO2, Al2O3 and CaCO3 surfaces with a parallel and vertical geometry, respectively.

Fig. 4. Optimized configurations of GM ad-

sorbed on Al2O3 (0 0 0 1) surface. (a) Side view
of the parallel geometry, (b) side view of the
vertical geometry. 2a, 2b⋯2d were the hydrogen
bonds formed between GM and Al2O3 surfaces.
The number in green represented the bridging
AleO bond length. Atom colors: C, green; O, red;
Al, pink; H, white.

Table 5 hydrogen bonds were formed which induced relative stronger interac-
Interatomic distance and angle of GM before and after adsorption at Al2O3 tion strength (−1.81 eV).
(0 0 0 1) surface in parallel geometry. The atom label was shown in Fig. 1. For the adsorption of GM at CaCO3 (1 0 4) surface, the optimized
Before adsorption After adsorption adsorption geometry was shown in Fig. 6. As O atom of GM was elec-
tronegative, the Ca atom on CaCO3 surface was positive charged,
Bond length/Å d(O5eC1) 1.445 1.535 electrostatic interactions contributed to the adsorption energy. As can
d(O5eC5) 1.430 1.477
be seen, the calculated Ca (CaCO3 surface)-O (GM) distance was
d(O3eC3) 1.432 1.490
d(O3eH3) 0.974 0.983 2.357–2.597 Å, confirming the electrostatic interactions between them.
d(O1eH1) 0.974 1.011 As was displayed in Table 6, the electrostatic interactions induced small
d(O6eH6) 0.970 0.990 elongations in the OeC (C2eO2, C3eO3, C6eO6) bonds and obvious
Angle/° ∠(C5eO5eC1) 112.640 109.563 increase in the CeOeH (C1eO1eH1, C6eO6eH6) angles, compared to
∠(C3eO3eH3) 106.134 104.571 the structural data of the free β-D-glucopyranose molecule. Analysis of
∠(C1eO1eH1) 107.963 110.032 the bond length and angle of the hydrogen bond (Table 4) between GM
∠(C6eO6eH6) 107.977 106.150
and CaCO3 surface showed that moderate hydrogen bonds were formed
between them, via H atom of GM and O atom of CaCO3 surface. Briefly,
there were two main contributions to the adsorption energy of GM at
bonds for the strong interaction with Al2O3 surfaces. From Fig. 5a, it
CaCO3 (1 0 4) surface: electrostatic interaction and hydrogen-bonded
could be seen that cellobiose interacted with SiO2 surface was mainly
interaction. Similar adsorption configuration and interaction me-
through hydrogen bonding interaction, similar with GM. But two more
chanism of the cellobiose on CaCO3 surface were obtained as was

610
H. Zhao et al. Applied Surface Science 466 (2019) 607–614

Fig. 5. Optimized configurations of cellobiose adsorbed on (a) SiO2 surface and (b) Al2O3 surface (c) CaCO3 surface. 1a⋯3d were the hydrogen bonds. The number in
green and blue represented the bridging AleO bond length and the distance between O atom of cellobiose and Ca atom of CaCO3, respectively. Atom colors: C, green;
O, red; Al, pink; Si, yellow; Ca, blue; H, white.

Fig. 6. Optimized configurations of GM ad-

sorbed on CaCO3 surface. (a) Side view of
the parallel geometry (b) side view of the
vertical geometry. 3a, 3b were the hydrogen
bonds formed between them. The number in
blue represented the distance between O
atom and Ca atom. The number in purple
was the charge of O atom. Atom colors: C,
green; O, red; Ca, blue; H, white.

Table 6 and depletion of electrons, respectively. An inspection of the iso-sur-

Interatomic distance and angle of GM before and after adsorption at CaCO3 faces revealed electron density rearrangement within the bonding re-
(1 0 4) surface in parallel geometry. The atom label was shown in Fig. 1. gions between GM and SiO2 (0 0 1), Al2O3 (0 0 0 1) and CaCO3 (1 0 4)
Before adsorption After adsorption surfaces, which was consistent with the formation of new bonds.
In the case of the GM-SiO2 (0 0 1) complex (Fig. 7a), we could see
Bond length/Å d(C2eO2) 1.421 1.448 electron density redistribution between H and O atoms indicative of
d(O2eH2) 0.974 0.977
hydrogen-bonded interaction. But the extent of electron density re-
d(O3eC3) 1.423 1.445
d(O3eH3) 0.974 0.980
arrangement was relative small and no electron transferred between
d(C6eO6) 1.434 1.435 GM and SiO2 (0 0 1) surface, as revealed from the Mulliken charge
d(O6eH6) 0.970 1.002 analysis (Table 7). Therefore, interaction between GM and SiO2 (0 0 1)
d(O1eH1) 0.975 1.006 surface was rather weak. In the case of the GM-Al2O3 (0 0 0 1) complex
d(O4eH4) 0.969 0.994
(Fig. 7b), there were significant electron density accumulation around
Angle/° ∠(C2eO2eH2) 107.803 106.317 O atom of GM and electron density depletion around Al atom of Al2O3
∠(C3eO3eH3) 105.369 103.470
surface, which was consistent with the formation of the bridging AleO
∠(C6eO6eH6) 107.773 111.485
∠(C1eO1eH1) 107.791 114.382 bonds. And the O atom of the surface accumulated electron density
∠(C4eO4eH4) 107.721 106.603 from the H atom of GM, corresponding to the formed hydrogen bonds.
Despite the strong electron density redistribution within the β-D-glu-
copyranose-Al2O3 complex, only little charge transfer occurred from
shown in Fig. 5c. GM to Al2O3 (0 0 0 1) surface. The charge gained by Al2O3 (0 0 0 1)
surface was only 0.08 e−, as shown in Table 7. Similar result was ob-
3.1.4. Electronic structure tained for GM-CaCO3 (1 0 4) system (Fig. 7c). There were electron
To further clarify the mechanism of GM interacted with SiO2 (0 0 1), density accumulation around O atom of GM and electron density de-
Al2O3 (0 0 0 1) and CaCO3 (1 0 4) surfaces, the electron density redis- pletion around Ca atom, corresponding to the electrostatic interaction
tribution within the GM-substrate systems was determined by analyzing between them. And the O atom of CaCO3 surface accumulated electron
the iso-surface of the differential charge density. The electron density density from H atom of GM, which was consistent with the formation of
difference was obtained by subtracting from the charge density of the hydrogen bonds. Even though the relative greater extent of the electron
GM-substrate systems, the sum of the charge densities of the GM and density redistribution between GM and CaCO3 surface, only 0.1 elec-
the substrate surfaces. The electron density difference of GM interacted tron transferred from surface to β-D-glucopyranose molecule. In a word,
with SiO2 (0 0 1), Al2O3 (0 0 0 1) and CaCO3 (1 0 4) surfaces was shown although Al2O3 and CaCO3 surfaces had little charge transfer when
in Fig. 7. The blue and yellow regions represented the accumulation

611
H. Zhao et al.
Fig. 7. Charge density difference iso-surface of (a) GM-SiO2 (b) GM-Al2O3 (c) GM-CaCO3 systems. The blue and yellow contours indicated electron accumulation and
electron depletion by 0.02 e/Å, respectively.
Table 7
Variation of the Mulliken charge of the mineral surfaces after GM adsorption.
Surface SiO2 (0 0 1) surface Al2O3 (0 0 0 1) surface CaCO3 (1 0 4) surface
Δq/e- 0.00 −0.08 0.1
interacted with GM, the greater extent of electron density redistribution
between Al2O3 (0 0 0 1) and CaCO3 (1 0 4) surface and β-D-glucopyr-
anose molecule contributed to the stronger interaction between them.
To gain insight into the nature of the interactions between GM and
different mineral surface, we carried out projected density of states
(PDOS) analysis of the free β-D-glucopyranose molecule (Fig. 8a) as
well as the interaction between H atom of GM and O atom of surfaces,
shown in Fig. 9. In the free β-D-glucopyranose PDOS, it was observed
that the states around the Fermi level were dominated by p-states of O
atoms, which was associated with the lone pair electron density of O
atoms, as shown in the highest occupied molecular orbital (HOMO)
(Fig. 8b). The p-orbitals of O atoms might interact with the orbitals of
the surface atoms, during the adsorption process of GM at the mineral
surfaces. Actually, when interacted with Al2O3 surface, reduction of the
O5-p states of GM around the Fermi level was observed in Fig. 9a, due
to the hybridization with the Al-d states. In addition, it was observed
that O-p states shifted to lower energy levels, suggested the stabilization
of GM molecule after interacted with Al2O3 surface.
The DOS results for the interaction between H atom of GM (HGM)
and O atom of mineral surfaces (Osurf) were shown in Fig. 9b–d. It was
observed that, the interaction between HGM and Osurf atoms was via the
hybridization of Osurf-2p and HGM-1s states around −7.0 eV to −6.8 eV
and −4.0 eV to −2.4 eV at Al2O3 surface, −8.2 eV to −5.0 eV at SiO2
Fig. 8. (a) DOS and (b) highest occupied molecular orbital (HOMO) for β-D-glucopyranose molecule in the free state.
612
Applied Surface Science 466 (2019) 607–614
surface, −3.6 eV to −2.8 eV at CaCO3 surface.
3.1.5. Interactions between GM and SiO2/Al2O3/CaCO3 surfaces with
COSMO solvation model
In the above text, the interaction mechanism between GM and SiO2/
Al2O3/CaCO3 surfaces was investigated in the gas phase. Considering
the water environment during the adsorption process, the GM and the
mineral surfaces would be hydrated by water molecules which might
influence the interactions between them. Therefore, effect of water
molecules should not be ignored. Thus, in the calculations implicit
COSMO solvation model was used to examine solvation effect on the
interaction between GM and mineral surfaces. The interaction energies
using COSMO solvation model were shown in Table 8. As can be seen,
interaction energies between GM and SiO2/Al2O3/CaCO3 surfaces got
be higher when considering the implicit water solvent effect. This could
be concluded that, in the water phase, the interaction strength between
GM and these mineral surfaces would be weakened.
4. Conclusion
In this work, theoretical study of the interaction between a model
polysaccharides monomer β-D-glucopyranose and different mineral
solid surfaces, modeled through the (0 0 1) surface of SiO2, (0 0 0 1)
surface of Al2O3 and (1 0 4) surface of CaCO3, has been performed
through DFT periodic calculations to determine the interaction me-
chanism at atomistic scale.
It was found that only moderate hydrogen bonds were formed be-
tween GM and SiO2 surface, which was relatively weak. While at CaCO3
surface, both hydrogen bonding interaction and electrostatic interac-
tion contributed to the strong interaction between GM and the solid
H. Zhao et al.
Fig. 9. DOS of (a) atom O5 of GM which directly interacted with surface Al atom. DOS of interaction of H atom of GM with surface O atom: (b) Al2O3 (0 0 0 1) surface
(c) SiO2 (0 0 1) surface (d) CaCO3 (1 0 4) surface.
Table 8
Interaction energies (eV) between GM and SiO2/Al2O3/CaCO3 surfaces in par-
allel adsorption geometries using COSMO solvation model and without COSMO
solvation model.
GM Without COSMO model With COSMO model
SiO2 −0.90 −0.61
Al2O3 −3.71 −3.23
CaCO3 −2.35 −1.49
surface. Quite strong interaction between GM and Al2O3 surface mainly
attributed to the formed bridging AleO bonds and hydrogen bonds.
Detailed information about the interaction mechanism was obtained by
electron density difference iso-surface and projected density of states
(PDOS) analyses. Only small electron transferred between GM and the
mineral surfaces, but obvious charge redistribution between them was
found, especially when interacted with Al2O3 surface. The bridging
AleO bonds formed between GM and Al2O3 (0 0 0 1) surface was
mainly through hybridization between O-p states and Al-d states. The
H-bonding interaction between GM and the three solid surfaces was
mainly through hybridization between O-2p states and H-1s states.
Finally, solvation effect on the interactions with the three solids sur-
faces was investigated by COSMO solvation model. The above knowl-
edge is very meaningful for better understanding the adsorption of
sugar based functional substances, which is an important issue in pro-
ceeding sufficient application in many fields.
613
Applied Surface Science 466 (2019) 607–614
Acknowledgements
The funding from the National Science Fund of China (Nos.
21473103 and 21872084) is gratefully acknowledged.
References
[1] F.W. Lichtenthaler, S. Peters, Carbohydrates as green raw materials for the chemical
industry, Comptes Rendus Chimie 7 (2004) 65–90.
[2] D.A.Z. Wever, F. Picchioni, A.A. Broekhuis, Polymers for enhanced oil recovery: a
paradigm for structure–property relationship in aqueous solution, Prog. Polym. Sci.
36 (2011) 1558–1628.
[3] V. Alvarado, E. Manrique, Enhanced oil recovery: an update review, Energies 3
(2010) 1529–1575.
[4] H. Chen, A.Z. Panagiotopoulos, E.P. Giannelis, Atomistic molecular dynamics si-
mulations of carbohydrate-calcite interactions in concentrated brine, Langmuir 31
(2015) 2407–2413.
[5] R. Liu, W. Pu, L. Wang, Q. Chen, Z. Li, Y. Li, B. Li, Solution properties and phase
behavior of a combination flooding system consisting of hydrophobically ampho-
teric polyacrylamide, alkyl polyglycoside and n-alcohol at high salinities, RSC Adv.
5 (2015) 69980–69989.
[6] S. Li, Y. Zhang, X. Xu, L. Zhang, Triple helical polysaccharide-induced good dis-
persion of silver nanoparticles in water, Biomacromolecules 12 (2011) 2864–2871.
[7] A.J. Millán, R. Moreno, M.a.I. Nieto, Thermogelling polysaccharides for aqueous
gelcasting—part I: a comparative study of gelling additives, J. Eur. Ceram. Soc. 22
(2002) 2209–2215.
[8] I. Santacruz, C. Baudı́n, M.I. Nieto, R. Moreno, Improved green properties of gelcast
alumina through multiple synergistic interaction of polysaccharides, J. Eur. Ceram.
Soc. 23 (2003) 1785–1793.
[9] J.L. Arias, M.S. Fernández, Polysaccharides and proteoglycans in calcium carbo-
nate-based biomineralization, Chem. Rev. 108 (2008) 4475–4482.
[10] J.W. Nielsen, K.K. Sand, C.S. Pedersen, L.Z. Lakshtanov, J.R. Winther,
M. Willemoës, S.L.S. Stipp, Polysaccharide effects on calcite growth: the influence
H. Zhao et al.
of composition and branching, Cryst. Growth Des. 12 (2012) 4906–4910.
[11] H. Ono, Y. Deng, Flocculation and retention of precipitated calcium carbonate by
cationic polymeric microparticle flocculants, J. Colloid Interface Sci. 188 (1997)
183–192.
[12] K. Zhao, W. Yan, X. Wang, B. Hui, G. Gu, H. Wang, The flotation separation of pyrite
from pyrophyllite using oxidized guar gum as depressant, Int. J. Miner. Process. 161
(2017) 78–82.
[13] K. Shrimali, V. Atluri, X. Wang, J.D. Miller, Adsorption of corn starch molecules at
hydrophobic mineral surfaces, Colloids Surf. A 546 (2018) 194–202.
[14] S. Jin, Q. Shi, Q. Li, L. Ou, K. Ouyang, Effect of calcium ionic concentrations on the
adsorption of carboxymethyl cellulose onto talc surface: flotation, adsorption and
AFM imaging study, Powder Technol. 331 (2018) 155–161.
[15] M.S. Kamal, A.S. Sultan, U.A. Al-Mubaiyedh, I.A. Hussein, Review on polymer
flooding: rheology, adsorption, stability, and field applications of various polymer
systems, Polym. Rev. 55 (2015) 491–530.
[16] W. Pu, C. Shen, B. Wei, Y. Yang, Y. Li, A comprehensive review of polysaccharide
biopolymers for enhanced oil recovery (EOR) from flask to field, J. Ind. Eng. Chem.
61 (2018) 1–11.
[17] V.V. Hardikar, E. Matijević, Influence of ionic and nonionic dextrans on the for-
mation of calcium hydroxide and calcium carbonate particles, Colloid Surf. A 186
(2001) 23–31.
[18] K.K. Sand, C.S. Pedersen, S. Sjöberg, J.W. Nielsen, E. Makovicky, S.L.S. Stipp,
Biomineralization: long-term effectiveness of polysaccharides on the growth and
dissolution of calcite, Cryst. Growth Des. 14 (2014) 5486–5494.
[19] Z. Benkova, M.N. Cordeiro, Molecular dynamics simulations of poly(ethylene oxide)
grafted onto silica immersed in melt of homopolymers, Langmuir 31 (2015)
10254–10264.
[20] A.N. Rissanou, H. Papananou, V.S. Petrakis, M. Doxastakis, K.S. Andrikopoulos,
G.A. Voyiatzis, K. Chrissopoulou, V. Harmandaris, S.H. Anastasiadis, Structural and
conformational properties of poly(ethylene oxide)/silica nanocomposites: effect of
confinement, Macromolecules 50 (2017) 6273–6284.
[21] N.R. Tummala, L. Shi, A. Striolo, Molecular dynamics simulations of surfactants at
the silica-water interface: anionic vs nonionic headgroups, J. Colloid Interface Sci.
362 (2011) 135–143.
[22] C.P. Kong, E.A. Peters, G. de With, H.X. Zhang, Molecular dynamics simulation of a
DOPA/ST monolayer on the Au(111) surface, Phys. Chem. Chem. Phys. 15 (2013)
15426–15433.
[23] A.J.A. Aquino, D. Tunega, G. Haberhauer, M.H. Gerzabek, H. Lischka, Quantum
chemical adsorption studies on the (110) surface of the mineral goethite, J. Phys.
Chem. C 111 (2007) 877–885.
[24] J. Chen, X. Long, Y. Chen, Comparison of multilayer water adsorption on the hy-
drophobic galena (PbS) and hydrophilic pyrite (FeS2) surfaces: a DFT study, J.
Phys. Chem. C 118 (2014) 11657–11665.
[25] G. Zhang, X. Xiang, F. Yang, L. Liu, T. Tang, Y. Shi, X. Wang, First principles in-
vestigation of helium physisorption on an alpha-Al2O3(0 0 0 1) surface, Phys. Chem.
Chem. Phys. 18 (2016) 15711–15718.
[26] A. Torres, J. Amaya Suárez, E.R. Remesal, A.M. Márquez, J. Fernández Sanz,
C. Rincón Cañibano, Adsorption of prototypical asphaltenes on silica: first-princi-
ples DFT simulations including dispersion corrections, J. Phys. Chem. B (2017).
[27] S. Sheng, M. Miller, J. Wu, Molecular theory of hydration at different temperatures,
J. Phys. Chem. B 121 (2017) 6898–6908.
[28] D.L. Geatches, A. Jacquet, S.J. Clark, H.C. Greenwell, Monomer adsorption on
kaolinite: modeling the essential ingredients, J. Phys. Chem. C 116 (2012)
22365–22374.
[29] J. Wang, S. Xia, L. Yu, Adsorption of Pb(II) on the kaolinite(001) surface in aqueous
system: a DFT approach, Appl. Surf. Sci. 339 (2015) 28–35.
[30] J. Chen, F.-F. Min, L. Liu, C. Liu, F. Lu, Experimental investigation and DFT cal-
culation of different amine/ammonium salts adsorption on kaolinite, Appl. Surf.
Sci. 419 (2017) 241–251.
[31] A. Budi, S.L.S. Stipp, M.P. Andersson, The effect of solvation and temperature on the
adsorption of small organic molecules on calcite, Phys. Chem. Chem. Phys. 20
(2018) 7140–7147.
[32] M. Ruan, H. Hou, W. Li, B. Wang, Theoretical study of the adsorption/dissociation
reactions of formic acid on the α-Al2O3(0 0 0 1) surface, J. Phys. Chem. C 118
(2014) 20889–20898.
[33] B. Delley, An all-electron numerical method for solving the local density functional
for polyatomic molecules, J. Chem. Phys. 92 (1990) 508–517.
[34] B. Delley, From molecules to solids with the DMol3 approach, J. Chem. Phys. 113
(2000) 7756–7764.
[35] J.P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approximation made
simple, Phys. Rev. Lett. 77 (1996) 3865.
614
Applied Surface Science 466 (2019) 607–614
[36] K. Müller, D. Lu, S.D. Senanayake, D.E. Starr, Monoethanolamine adsorption on
TiO2(110): bonding, structure, and implications for use as a model solid-supported
CO2 capture material, J. Phys. Chem. C 118 (2014) 1576–1586.
[37] A. Gouron, J. Kittel, T. de Bruin, B. Diawara, Density functional theory study of
monoethanolamine adsorption on hydroxylated Cr2O3 surfaces, J. Phys. Chem. C
119 (2015) 22889–22898.
[38] S. Grimme, S. Ehrlich, L. Goerigk, Effect of the damping function in dispersion
corrected density functional theory, J. Comput. Chem. 32 (2011) 1456–1465.
[39] D. Tunega, T. Bucko, A. Zaoui, Assessment of ten DFT methods in predicting
structures of sheet silicates: importance of dispersion corrections, J. Chem. Phys.
137 (2012) 114105.
[40] F. Chiter, C. Lacaze-Dufaure, H. Tang, N. Pebere, DFT studies of the bonding me-
chanism of 8-hydroxyquinoline and derivatives on the (111) aluminum surface,
Phys. Chem. Chem. Phys. 17 (2015) 22243–22258.
[41] E. Ataman, M.P. Andersson, M. Ceccato, N. Bovet, S.L.S. Stipp, Functional group
adsorption on calcite: I. Oxygen containing and nonpolar organic molecules, J.
Phys. Chem. C 120 (2016) 16586–16596.
[42] M.A.O. Lourenco, P. Ferreira, J.R.B. Gomes, Flue gas adsorption on periodic me-
soporous phenylene-silica: a DFT approach, Phys. Chem. Chem. Phys. 20 (2018)
16686–16694.
[43] D. Addari, A. Satta, Influence of HCOO– on calcite growth from first-principles, J.
Phys. Chem. C 119 (2015) 19780–19788.
[44] E. Ataman, M.P. Andersson, M. Ceccato, N. Bovet, S.L.S. Stipp, Functional group
adsorption on calcite: II. Nitrogen and sulfur containing organic molecules, J. Phys.
Chem. C 120 (2016) 16597–16607.
[45] M.P. Andersson, K. Dideriksen, H. Sakuma, S.L. Stipp, Modelling how incorporation
of divalent cations affects calcite wettability-implications for biomineralisation and
oil recovery, Sci. Reports 6 (2016) 28854.
[46] D. Costa, T. Ribeiro, P. Cornette, P. Marcus, DFT modeling of corrosion inhibition
by organic molecules: carboxylates as inhibitors of aluminum corrosion, J. Phys.
Chem. C 120 (2016) 28607–28616.
[47] S. Heiden, Y. Yue, H. Kirsch, J. Wirth, P. Saalfrank, R.K. Campen, Water dissociative
adsorption on α-Al2O3(112̅0) is controlled by surface site undercoordination, den-
sity, and topology, J. Phys. Chem. C 122 (2018) 6573–6584.
[48] M.A.O. Lourenço, C. Siquet, M. Sardo, L. Mafra, J. Pires, M. Jorge, M.L. Pinto,
P. Ferreira, J.R.B. Gomes, Interaction of CO2 and CH4 with functionalized periodic
mesoporous phenylene–silica: periodic DFT calculations and gas adsorption mea-
surements, J. Phys. Chem. C 120 (2016) 3863–3875.
[49] H.J. Monkhorst, J.D. Pack, Special points for Brillouin-zone integrations, Phys. Rev.
B 13 (1976) 5188–5192.
[50] Y. Han, W. Liu, J. Zhou, J. Chen, Interactions between kaolinite Al OH surface and
sodium hexametaphosphate, Appl. Surf. Sci. 387 (2016) 759–765.
[51] J. Tomasi, B. Mennucci, R. Cammi, Quantum mechanical continuum solvation
models, Chem. Rev. 105 (2005) 2999–3094.
[52] A. Klamt, The COSMO and COSMO-RS solvation models, Wires Comput. Mol. Sci. 1
(2011) 699–709.
[53] J. Andzelm, C. Kölmel, A. Klamt, Incorporation of solvent effects into density
functional calculations of molecular energies and geometries, J. Chem. Phys. 103
(1995) 9312–9320.
[54] L. Levien, C.T. Prewitt, D.J. Weidner, Structure and elastic properties of quartz at
pressure, Am. Mineral. 65 (1980) 920–930.
[55] D. Többens, N. Stüßer, K. Knorr, H. Mayer, G. Lampert, E9: the new high-resolution
neutron powder diffractometer at the Berlin neutron scattering center, Mater. Sci.
Forum, Trans. Tech. Publ. (2001) 288–293.
[56] R. Demichelis, P. Raiteri, J.D. Gale, R. Dovesi, Examining the accuracy of density
functional theory for predicting the thermodynamics of water incorporation into
minerals: the hydrates of calcium carbonate, J. Phys. Chem. C 117 (2013)
17814–17823.
[57] T.P. Goumans, A. Wander, W.A. Brown, C.R. Catlow, Structure and stability of the
(001) alpha-quartz surface, Phys. Chem. Chem. Phys. 9 (2007) 2146–2152.
[58] C. Arrouvel, B. Diawara, D. Costa, P. Marcus, DFT periodic study of the adsorption
of glycine on the anhydrous and hydroxylated (0 0 0 1) surfaces of α-alumina, J.
Phys. Chem. C 111 (2007) 18164–18173.
[59] R.S. Alvim, F.C.D.A. Lima, V.M. Sánchez, T.F. Headen, E.S. Boek, C.R. Miranda,
Adsorption of asphaltenes on the calcite (10.4) surface by first-principles calcula-
tions, RSC Adv. 6 (2016) 95328–95336.
[60] X.-J. Hou, H. Li, S. Li, P. He, Theoretical study of the intercalation behavior of
ethylene glycol on kaolinite, J. Phys. Chem. C 118 (2014) 26017–26026.
[61] G.A. Jeffrey, G.A. Jeffrey, An Introduction to Hydrogen Bonding, Oxford University
Press, New York, 1997.