scispace - formally typeset
Open AccessJournal ArticleDOI

A computational study of carbon dioxide adsorption on solid boron

TLDR
The theoretical results demonstrate that the electron deficient boron materials, such as α-B12 and γ-B28, can bond strongly with CO2 due to Lewis acid-base interactions because the electron density is higher on their surfaces and are predicted to be good candidates for CO2 capture.
Abstract
Capturing and sequestering carbon dioxide (CO2) can provide a route to partial mitigation of climate change associated with anthropogenic CO2 emissions. Here we report a comprehensive theoretical study of CO2 adsorption on two phases of boron, α-B12 and γ-B28. The theoretical results demonstrate that the electron deficient boron materials, such as α-B12 and γ-B28, can bond strongly with CO2 due to Lewis acid–base interactions because the electron density is higher on their surfaces. In order to evaluate the capacity of these boron materials for CO2 capture, we also performed calculations with various degrees of CO2 coverage. The computational results indicate CO2 capture on the boron phases is a kinetically and thermodynamically feasible process, and therefore from this perspective these boron materials are predicted to be good candidates for CO2 capture.

read more

Content maybe subject to copyright    Report

(83@/<=3>C90)966981981(83@/<=3>C90)966981981
%/=/+<-2"8638/%/=/+<-2"8638/
?=><+63+88=>3>?>/09<889@+>3@/!+>/<3+6=
#+:/<=
?=><+63+88=>3>?>/09<889@+>3@/!+>/<3+6=

-97:?>+>398+6=>?.C90-+<,98.39B3./+.=9<:>39898=963.,9<98-97:?>+>398+6=>?.C90-+<,98.39B3./+.=9<:>39898=963.,9<98
$3+9&?8
(83@/<=3>C90$?//8=6+8.
!/81)+81
(83@/<=3>C90$?//8=6+8.
7A?9A7+36/.?+?
*2/8 3
(83@/<=3>C90)966981981
D2/86?9A/.?+?
34?8?
$?//8=6+8.(83@/<=3>C90'/-289691C
/,<+&/+<6/=
(83@/<=3>C90$?//8=6+8.
9669A>23=+8.+..3>398+6A9<5=+>2>>:=<9?9A/.?+?+337:+:/<=
#+<>90>2/8138//<38197798=+8.>2/#2C=3-+6&-3/8-/=+8.!+>2/7+>3-=97798=
%/-977/8./.3>+>398%/-977/8./.3>+>398
&?8$3+9)+81!/81 3*2/8?34?8+8.&/+<6/=/,<+-97:?>+>398+6=>?.C90-+<,98
.39B3./+.=9<:>39898=963.,9<98
?=><+63+88=>3>?>/09<889@+>3@/!+>/<3+6=#+:/<=

2>>:=<9?9A/.?+?+337:+:/<=
%/=/+<-2"8638/3=>2/9:/8+--/==38=>3>?>398+6</:9=3>9<C09<>2/(83@/<=3>C90)9669819819<0?<>2/<3809<7+>398
-98>+->>2/(") 3,<+<C</=/+<-2:?,=?9A/.?+?

-97:?>+>398+6=>?.C90-+<,98.39B3./+.=9<:>39898=963.,9<98-97:?>+>398+6=>?.C90-+<,98.39B3./+.=9<:>39898=963.,9<98
,=><+->,=><+->
+:>?<381+8.=/;?/=>/<381-+<,98.39B3./"-+8:<9@3./+<9?>/>9:+<>3+673>31+>39890-637+>/
-2+81/+==9-3+>/.A3>2+8>2<9:91/83-"/73==398=/</A/</:9<>+-97:</2/8=3@/>2/9</>3-+6=>?.C
90"+.=9<:>39898>A9:2+=/=90,9<98α+8.γ'2/>2/9</>3-+6</=?6>=./798=><+>/>2+>>2/
/6/-><98./F-3/8>,9<987+>/<3+6==?-2+=α+8.γ-+8,98.=><9816CA3>2".?/>9 /A3=
+-3.,+=/38>/<+->398=,/-+?=/>2//6/-><98./8=3>C3=2312/<98>2/3<=?<0+-/=89<./<>9/@+6?+>/>2/
-+:+-3>C90>2/=/,9<987+>/<3+6=09<"-+:>?</A/+6=9:/<09<7/.-+6-?6+>398=A3>2@+<39?=./1<//=
90"-9@/<+1/'2/-97:?>+>398+6</=?6>=38.3-+>/"-+:>?</98>2/,9<98:2+=/=3=+538/>3-+66C+8.
>2/<79.C8+73-+66C0/+=3,6/:<9-/==+8.>2/</09</0<97>23=:/<=:/->3@/>2/=/,9<987+>/<3+6=+</
:</.3->/.>9,/199.-+8.3.+>/=09<"-+:>?</
/CA9<.=/CA9<.=
.39B3./+.=9<:>398=963.,9<98=>?.C-97:?>+>398+6-+<,98
3=-3:638/=3=-3:638/=
8138//<381E#2C=3-+6&-3/8-/=+8.!+>2/7+>3-=
#?,63-+>398/>+36=#?,63-+>398/>+36=
&?8$)+81! 3*?&/+<6/=-97:?>+>398+6=>?.C90-+<,98.39B3./+.=9<:>398
98=963.,9<98#2C=3-+62/73=><C2/73-+6#2C=3-=
'23=49?<8+6+<>3-6/3=+@+36+,6/+>%/=/+<-2"8638/2>>:=<9?9A/.?+?+337:+:/<=

A Computational Study of Carbon Dioxide
Adsorption on Solid Boron
Qiao Sun
*,a
, Meng Wang
a,b
, Zhen Li
*,
c
, Aijun Du
d
and Debra J. Searles
*,a,e
Capturing and sequestering carbon dioxide (CO
2
) can provide a route to partial mitigation of
climate change associated with anthropogenic CO
2
emissions. Here we report a comprehensive
theoretical study of CO
2
adsorption on two phases of boron, α-B
12
and γ-B
28.
The theoretical
results demonstrate that the electron deficient boron materials, such as α-B
12
and γ-B
28
, can
bond strongly with CO
2
due to Lewis acid-base interactions because the electron density is
higher on their surfaces. In order to evaluate the capacity of these boron materials for CO
2
capture, we also performed calculations with various degrees of CO
2
coverage. The
computational results indicate CO
2
capture on the boron phases is a kinetically and
thermodynamically feasible process, and therefore from this perspective these boron materials
are predicted to be good candidates for CO
2
capture.
1 Introduction
Carbon dioxide, CO
2
, is a greenhouse gas whose concentration in the
atmosphere has been increasing since the industrial revolution, and
this increase is largely caused by the burning of fossil fuels.
Therefore it is essential that this trend be halted, and new
technologies aimed at removing CO
2
from combustion products to
reduce its concentration in the atmosphere are being developed.
1,2
An alternative approach to reduction of CO
2
from combustion is to
select fuels that produce less CO
2
on combustion than conventional
fossil fuels (e.g. natural gas), or no CO
2
(H
2
). These fuels often
contain contaminants including CO
2
, so CO
2
capture from these
resources before combustion is necessary and is a well-established
process in industry. The most common processes involve treatment
with amine solutions or chilled ammonia, however these processes
have some problems in that they are not energy efficient, the amine
is toxic and solvent can easily be lost.
2,3
In both these approaches to
reducing CO
2
in the atmosphere, it is important that new
environmentally friendly and economically feasible processes for
CO
2
capture be developed.
____________________________________________
a
Centre for Theoretical and Computational Molecular Science,
Australian Institute for Bioengineering and Nanotechnology, The
University of Queensland, QLD 4072, Brisbane, Australia. Email:
q.sun@uq.edu.au; d.bernhardt@uq.edu.au
b
Center for Bioengineering and Biotechnology, China University of
Petroleum (East China), Qingdao 266555, China.
c
Institute of Superconducting & Electronic Materials, The University
of Wollongong, NSW 2500, Australia. Email: zhenl@uow.edu.au
d
School of Chemistry, Physics and Mechanical Engineering,
Queensland University of Technology, Brisbane, QLD 4001,
Australia.
e
School of Chemistry and Molecular Biosciences, The University of
Queensland, QLD 4072, Brisbane, Australia.
Solid adsorbents are anticipated to play a key role in new
technologies, and there have been many materials considered in
recent years.
4
To be of practical use the materials must be able to
strongly adsorb CO
2
and have large surface areas, however many
materials only weakly bind CO
2
. In this work we find that solid
boron materials such as α-B
12
and γ-B
28
can adsorb CO
2
strongly and
therefore may be useful materials for CO
2
capture.
Boron readily bonds with other boron atoms, forming a variety of
different structures with complex features such as three-center two-
electron bonds or electron deficient bonds. In the pure boron solids,
the B
12
icosahedron is the basic structural unit which can be flexibly
linked or fused to form rigid structures,
5-10
and the existence of this
unit and it connectivity is associated with the electron deficiency, or
hypovalency, of boron. Four reported boron phases correspond to the
pure element: rhombohedral α-B
12
6,9,11,12
and β-B
106
5
(with 12 and
106 atoms in the unit cell, respectively), tetragonal T-192
7
(with
190192 atoms per unit cell) and γ-B
28
8,13,14
(with 28 atoms in the
unit cell), whereas there is a large variety of boron-rich compounds.
Much work on boron rich compounds has been carried out due to
their physical and chemical properties which have resulted in
research for their suitability in applications from nuclear reactors to
super-hard, thermoelectric and high-energy materials as well as
hydrogen storage materials.
8,12,15-18
In this study, we will investigate
CO
2
capture using α-B
12
and γ-B
28
. The α-B
12
phase consists of one
B
12
icosahedron per unit cell and γ-B
28
consists of icosahedral B
12
clusters and B
2
pairs in a NaCl-type arrangement.
8
It has been shown
that the electronic properties of the B
2
pairs and B
12
clusters in γ-B
28
,
are different, and this results in charge transfer between them.
8
For
α-B
12
and γ-B
28
, our study found that there is charge transfer among
the atoms on their surfaces and the internal atoms, indicating that the
α-B
12
and γ-B
28
slabs are amphoteric with acidic and basic sites. The
regioselectivity of α-B
12
and γ-B
28
indicate CO
2
(Lewis acid) might
form strong interactions with the basic sites of their B materials due
to Lewis acid-base interactions. In order to test this hypothesis, we
have carried out DFT calculations of CO
2
capture on α-B
12
and γ-B
28
surfaces. The primary motivation of the theoretical study is to
stimulate further experiments to verify our prediction that CO
2
can
be captured by these boron materials.
2 Computational methods
Ab initio DFT calculations are performed with the DMol3 module in
Material Studio.
19,20
The configurations of CO
2
adsorbed on the
boron phases are fully optimized using the generalized gradient
approximation
21
treated by the Perdew-Burke-Ernzerhof exchange-
correlation potential with long rang dispersion correction carried out
using the Grimme’s scheme
22
. This method has been used to
successfully determine the geometrical, energetic and electronic
structural properties of boron clusters, boron phases and boron

containing materials.
23-29
The basis set employed is an all-electron
double-numerical atomic orbital basis set, augmented by d-
polarization functions (DNP). The cell parameters of α-B
12
and γ-B
28
in the calculations are all optimized and are consistent with the
experimental values.
8,11
Detailed information on the cell parameters
is listed in our previous publication.
27
The slab thicknesses of α-B
12
and γ-B
28
are 8.012 Å and 6.914 Å (corresponding to two layers of
B
12
or B
14
clusters), respectively. The 2 × 2 α-boron (001) and γ-
boron (001) surfaces were chosen with a 15 Å vacuum above the
slab in order to avoid interactions between their periodic images, and
the Brillouin zone is sampled by 6 × 6 × 1 k-points using the
Monkhorst-Pack scheme.
The calculations of CO
2
adsorption on α-B
12
(001) and γ-B
28
(001)
surfaces are based on the fully optimized boron surfaces.
27
The
adsorption energy of CO
2
on α-B
12
and γ-B
28
are calculated from Eq.
1:
E
ads
= E
(boron phaseCO
2
)
- E
CO
2
- E
(boron phase)
(1)
where E
(boron phaseCO
2
)
is the total energy of the boron surface with a
CO
2
molecule adsorbed. In order to better clarify the adsorption and
the nature of the interaction of CO
2
with α-B
12
and γ-B
28
, atoms in
molecules (AIM) theory has been employed in the study. Based on
the optimized structures at the DFT-D level, we calculate the
wavefunctions at the B3LYP/6-31G(d) level of theory,
21
we then use
AIM theory, which has been used to successfully determine
intermolecular interactions of different systems.
27,30,31
In the AIM
analyses,
32
the existence of an interaction is indicated by the
presence of a bond critical point (BCP), and the strength of the bond
can be estimated from the magnitude of the electron density (
ρ
bcp
) at
the BCP. Similarly, the ring or cage structures are characterized by
the existence of a ring critical point (RCP) or cage critical point
(CCP). Furthermore, the nature of the molecular interaction can be
predicted from the topological parameters at the BCP, such as the the
acian of electron density (
2
ρ
bcp
) and energy density (H
bcp
). The
topological analysis of the system was carried out via the AIMALL
program.
32
The transition states between chemisorbed and physisorbed CO
2
have been investigated using the complete LST (linear synchronous
transit)/QST (quadratic synchronous transit) method
33
implemented
in the DMol3 code. The reactants and products correspond to the
optimized structures of CO
2
physisorbed and chemisorbed on α-B
12
and γ-B
28
surfaces, respectively. Electron distributions and transfer
mechanisms are determined with the Mulliken method.
34
3 Results and discussion
The α-B
12
and γ-B
28
slabs were optimized and the fully relaxed α-B
12
(001) and γ-B
28
(001) surfaces with cell vectors are shown in Fig.
1(a) and (b), respectively. We firstly investigated the Mulliken
atomic charge distributions on the α-B
12
(001) and γ-B
28
(001)
surfaces and then carried out a study of the mechanism of CO
2
adsorption on the surfaces of the two boron materials. All the
possible sites for CO
2
adsorption on α-B
12
and γ-B
28
surfaces have
been considered. In this manuscript we will discuss the strongest
adsorption sites which we classify as having two different types of
interaction: type A and B. In type A interactions, the carbon atom
and one oxygen atom in the CO
2
molecule directly connect with two
boron atoms, and in type B interactions, two oxygen atoms of the
CO
2
molecule directly connect with the boron atoms of the α-B
12
and
γ-B
28
surfaces.
(a) (b)
Fig. 1 The top view of the surfaces of fully relaxed α-B
12
(001) and
γ-B
28
(001) slabs with cell vectors shown. Atom color code: pink,
boron.
3.1 Atomic charge distributions of α-B
12
and γ-B
28
It is well known that boron phases, such as α-B
12
and γ-B
28
, are
electron deficient materials because there is only one electron on the
p orbital of each boron atom, while CO
2
is a Lewis acid and it would
prefer to gain electrons in reactions. Because of this the electron
deficient boron materials might not be expected to be good
adsorbents for CO
2
capture. However, from the Mulliken atomic
charge distributions of the α-B
12
and γ-B
28
slabs that
we modeled we
can see that there is electron transfer from the atoms within the slabs
to the surface atoms of α-B
12
and γ-B
28
. Fig. 2 shows the optimized
α-B
12
and γ-B
28
structures and the Mulliken atomic charges for the
atoms labeled in Fig. 2 are listed in Table 1.
(a)
(b)
Fig. 2 The side view of the optimized α-B
12
(001) and γ-B
28
(001)
slabs. Mulliken atomic charges of these atoms (with atom numbers
in Fig. 2) are listed in Table 1.
The
charges on the surface atoms of α-B
12
are -0.366e (atom B1), -
0.321e
(atom B2), -0.008e (atom B3 and B4) according to the
Mulliken atomic charge analysis, and the atoms in the internal part
of the slab have positive charges. The Mulliken atomic charge
distributions of γ-B
28
are similar to those of the α-B
12
slab: the
surface atoms B1 and B2 have negative charges of -0.311e, B3 and
B4 have negative charges of -0.098e, and the internal atoms either
have positive charges (atoms B9-B14) or are slightly negative

(atoms B7 and B8). The charge difference between the atoms on the
surfaces and the internal atoms mean that the α-B
12
and γ-B
28
slabs
are regioselective and amphoteric, with both acidic and basic sites.
Similar behaviour in boron materials, such as B
80
, has been
investigated preciously by Muya et al., who found the cap atoms of
B
80
act as acid sites and the frame atoms act as basic sites in this
regioselective molecule.
35
From the above analysis we have shown
that, in principle, CO
2
(Lewis acid) will form strong interactions
with the basic sites of α-B
12
and γ-B
28
due to Lewis acid-base
interactions. In the following section, we will investigate CO
2
the
mechanism of CO
2
adsorption on the α-B
12
and γ-B
28
surfaces.
Table 1 Atom number and Mulliken atomic charges (e) of the
optimized α-B
12
and γ-B
28
slabs.
α-B
12
γ-B
28
Atom number
Atomic charge
Atom number
Atomic charge
B1
-0.366
B1
-0.311
B2
-0.321
B2
-0.311
B3
-0.008
B3
-0.098
B4
-0.008
B4
-0.098
B5
0.139
B5
-0.074
B6
0.139
B6
-0.074
B7
0.119
B7
-0.025
B8
0.119
B8
-0.025
B9
0.033
B9
0.133
B10
0.033
B10
0.133
B11
0.066
B11
0.171
B12
0.055
B12
0.171
B13
0.204
B14
0.204
3.2 CO
2
adsorption on the α-B
12
surface
Currently 16 allotropes of elemental boron have been reported, with
the α-boron structure being the simplest.
18
It contains 12 atoms in a
rhombohedral unit cell, forming a slightly distorted icosahedral B
12
.
In this section, we will investigate the reaction mechanism of CO
2
adsorption on α-B
12
. We have identified two ways in which CO
2
adsorbs on α-B
12
, labelled type A and type B. In type A one O
atom
and the C atom of CO
2
bond with one BB bond of α-B
12
; and in
type B two O atoms of CO
2
interact with one BB bond of α-B
12
.
Fig. 3 (a) shows the optimized structures of the two possible
minimum-energy type A configurations, and the transition states for
the process of CO
2
adsorption on the α-B
12
surface. Corresponding
results for type B interactions have been listed in the supporting
information, see Fig. S1 (a). The important structural properties,
relative energies and the electron transfers from the α-B
12
to CO
2
are
listed in Table 2 and Table S1 in supporting information. For the two
types of interaction of CO
2
with α-B
12
, we can see that the
chemisorbed type A configuration is the most stable, so the
discussion will focus on the adsorption through interactions of type
A. For CO
2
capture on α-B
12
through type A interactions, we
identified two stationary states, corresponding to physisorption and
chemisorption. In the physisorbed configuration, CO
2
interacts with
one BB bond of α- B
12
and lies parallel to it. The O–C–O angle is
179.9° and BO distance is approximately 3.1 Å, the long distance
indicates the interaction is very weak and the adsorption is mainly
due to the van der Waals interaction of CO
2
and α-B
12
.
Table 2 Adsorption energy in kcal/mol, bond distance (r) in Å, bond
angle (α) in deg and charge transfer (CT) in electrons for the type A
of CO
2
adsorption on α-B
12
and γ-B
28
surfaces.
Chem
Phys
TS
Chem
α-B
12
surface
-4.95
-0.46
-47.76
3.042
1.967
1.477
3.107
2.648
1.647
1.176
1.208
1.442
1.176
1.170
1.203
1.597
1.636
1.694
179.9
170.5
122.7
0.004
0.05
-0.584
γ-B
28
Surface
-4.84
-2.50
-29.18
2.896
1.738
1.482
3.304
2.616
1.649
1.178
1.216
1.413
1.175
1.166
1.206
1.671
1.672
1.740
179.7
168.8
123.4
0.007
0.077
-0.528
(a)
(b)
Fig. 3 Computed minimum energy configurations and transition
state for CO
2
adsorption on the α-B
12
(a) and γ-B
28
(b) surfaces
involving type A interactions (top and side view of the two
minimum energy configurations and transition state).
Table 3 Atom number and Mulliken atomic charges (e) and the sum
of charge (e) of CO
2
, the boron cell of α-B
12
_CO
2
and γ-B
28
_CO
2
complexes (chemisorbed configurations) as well as the charge
transfer of α-B
12
_CO
2
(CT1) and γ-B
28
_CO
2
(CT2) complexes
comparing with those of the isolated forms.
Atom type/Sum
CO
2
α-B
12
CT1
γ-B
28
CT2

Figures
Citations
More filters
Journal Article

Charge-controlled switchable CO2 capture on boron nitride nanomaterials

TL;DR: In this paper, the adsorption of CO2, CH4, and H2 on boron nitride (BN) nanosheets and nanotubes (NTs) with different charge states was investigated.
Journal ArticleDOI

Electric field controlled CO2 capture and CO2/N2 separation on MoS2 monolayers.

TL;DR: In this article, the authors investigated the effect of external electric fields on CO2 capture and regeneration on MoS2 monolayers controlled by turning on/off external electric field.
Journal ArticleDOI

Novel two-dimensional MOF as a promising single-atom electrocatalyst for CO2 reduction: A theoretical study

TL;DR: In this article, a family of emerging 2D metal-organic frameworks (MOFs) utilizing density functional theory (DFT) calculations were investigated for CO2 activation and reduction under ambient conditions.
Journal ArticleDOI

High capacity and reversible hydrogen storage on two dimensional C2N monolayer membrane

TL;DR: In this paper, the authors explored the potential application of C2N monolayer using as a promising material for hydrogen storage through a comprehensive density functional theory (DFT) investigation.
Journal ArticleDOI

First-Principles Study of Electrocatalytically Reversible CO2 Capture on Graphene-like C3N

TL;DR: It is demonstrated that the CO2 capture, regeneration and separation on C3 N monolayer can be controllable with the method of switching on/off the charge state or electric field during the adsorption.
References
More filters
Journal ArticleDOI

Generalized Gradient Approximation Made Simple

TL;DR: A simple derivation of a simple GGA is presented, in which all parameters (other than those in LSD) are fundamental constants, and only general features of the detailed construction underlying the Perdew-Wang 1991 (PW91) GGA are invoked.
Journal ArticleDOI

Semiempirical GGA-type density functional constructed with a long-range dispersion correction.

TL;DR: A new density functional of the generalized gradient approximation (GGA) type for general chemistry applications termed B97‐D is proposed, based on Becke's power‐series ansatz from 1997, and is explicitly parameterized by including damped atom‐pairwise dispersion corrections of the form C6 · R−6.
Book

Atoms in molecules : a quantum theory

TL;DR: In this article, the quantum atom and the topology of the charge desnity of a quantum atom are discussed, as well as the mechanics of an atom in a molecule.
Journal ArticleDOI

Electronic Population Analysis on LCAO–MO Molecular Wave Functions. I

TL;DR: In this paper, an analysis in quantitative form is given in terms of breakdowns of the electronic population into partial and total ''gross atomic populations'' and ''overlap populations'' for molecules.
Journal ArticleDOI

An all‐electron numerical method for solving the local density functional for polyatomic molecules

TL;DR: In this paper, a method for accurate and efficient local density functional calculations (LDF) on molecules is described and presented with results using fast convergent threedimensional numerical integrations to calculate the matrix elements occurring in the Ritz variation method.
Frequently Asked Questions (15)
Q1. What have the authors contributed in "A computational study of carbon dioxide adsorption on solid boron" ?

Here the authors report a comprehensive theoretical study of CO2 adsorption on two phases of boron, α-B12 and γ-B28. In order to evaluate the capacity of these boron materials for CO2 capture, the authors also performed calculations with various degrees of CO2 coverage. 

According to molecular orbital theory, the frontier orbits and nearby molecular orbits are the most important factors determining the stability of molecules. 

Much work on boron rich compounds has been carried out due to their physical and chemical properties which have resulted in research for their suitability in applications from nuclear reactors to super-hard, thermoelectric and high-energy materials as well as hydrogen storage materials. 

the nature of the molecular interaction can be predicted from the topological parameters at the BCP, such as the the acian of electron density (∇2ρbcp) and energy density (Hbcp). 

The energy gaps of ∆E between their highest occupied molecular orbitals (HOMO) and their lowest unoccupied molecular orbitals (LUMO) (∆E = ELUMO − EHOMO) for α-B12 and γ-B28 surfaces are 0.046 and 0.854 eV, respectively. 

The adsorption energy of the physisorbed CO2 is calculated to be 4.84 kcal/mol so the interaction between CO2 and γ-B28 is very weak. 

Therefore pure solid boron could be used a material for permanent capture of CO2; or alternatively the solid boron functionalised with CO2 could be used as a material for capture of further CO2 and release of the additional CO2.4 ConclusionsUsing DFT calculations the authors have investigated the reaction mechanisms for CO2 capture on α-B12 and γ-B28 surfaces. 

The transition states between chemisorbed and physisorbed CO2 have been investigated using the complete LST (linear synchronous transit)/QST (quadratic synchronous transit) method33 implemented in the DMol3 code. 

In detail, the adsorption energies for CO2 molecules adsorbed on 2 × 2 α-B12 reduce from -47.76 to -33.79 kcal/mol as the number of CO2 molecules increases from one to four, with the energy required to remove one CO2 molecule from the system with four adsorbed being -20.89 kcal/mol. 

This results in ΔG increasing almost linearly with an increase in temperature for CO2 capture on γ-B28, as the free CO2 is chemisorbed on the surface. 

The authors find that the α-B12 and γ-B28 surfaces can effectively capture up to four CO2 molecules through chemisorption with configurations which are similar to those when one CO2 isadsorbed on these surfaces. 

For the two types of interaction of CO2 with α-B12, the authors can see that the chemisorbed type A configuration is the most stable, so the discussion will focus on the adsorption through interactions of type A. For CO2 capture on α-B12 through type A interactions, the authors identified two stationary states, corresponding to physisorption and chemisorption. 

In summary, the low barrier and the negative ΔG within the temperature range demonstrate that CO2 adsorption on γ-B28 is a kinetically and thermodynamically favorable process. 

Here the authors note that the interaction between CO2 and α-B12 is the strongest value for CO2 adsorption on substrates so far, as far as the authors are aware. 

The short distance indicates a strong interaction between the CO2 molecule and α-B12, with the calculated adsorption binding energy is 47.76 kcal/mol.