Screening and Characterization of Ternary Oxides for High-Temperature Carbon Capture

Article abstract of DOI:10.1021/acs.chemmater.7b04679

Carbon capture and storage (CCS) is increasingly being accepted as a necessary component of any effort to mitigate the impact of anthropogenic climate change, as it is both a relatively mature and easily implemented technology. High-temperature CO₂ absorption looping is a promising process that offers a much lower energy penalty than the current state of the art amine scrubbing techniques, but more effective materials are required for widespread implementation. This work describes the experimental characterization and CO₂ absorption properties of several new ternary transition metal oxides predicted by high-throughput DFT screening. One material reported here, Li₅SbO₅, displays reversible CO₂ sorption, and maintains ～72% of its theoretical capacity out to 25 cycles. The results in this work are used to discuss major influences on CO₂ absorption capacity and rate, including the role of the crystal structure, the transition metal, the alkali or alkaline earth metal, and the competing roles of thermodynamics and kinetics. Notably, this work shows the extent and rate to which ternary metal oxides carbonate are driven primarily by the identity of the alkali or alkaline earth ion and the nature of the crystal structure, whereas the identity of the transition ion carries little influence in the systems studied here.

Full text of DOI:10.1021/acs.chemmater.7b04679

Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES
Screening and characterization of ternary
oxides for high temperature carbon capture
Michael W. Gaultois, Matthew T. Dunstan, Adam W. Bateson, Martin S. C. Chan, and Clare P. Grey
Chem. Mater., Just Accepted Manuscript • DOI: 10.1021/acs.chemmater.7b04679 • Publication Date (Web): 29 Mar 2018
Downloaded from http://pubs.acs.org on March 29, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 11
1
Chemistry of Materials
Screening and characterization of ternary oxides for high temperature carbon capture
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
, a)
1, a)
1
2
1, b)
Michael W. Gaultois,
Matthew T. Dunstan,
Adam W. Bateson, Martin S. C. Chan, and Clare P. Grey
1
2
)
)
Department of Chemistry, University of Cambridge, Cambridge, CB2 1EW, United Kingdom
Department of Chemical Engineering and Biotechnology, University of Cambridge, Cambridge, CB2 3RA,
United Kingdom
Carbon capture and storage (CCS) is increasingly being accepted as a necessary component of any effort to
mitigate the impact of anthropogenic climate change, as it is both a relatively mature and easily implemented
technology. High-temperature CO absorption looping is a promising process that offers a much lower energy
2
penalty than the current state of the art amine scrubbing techniques, but more effective materials are required
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
for widespread implementation. This work describes the experimental characterisation and CO absorption
2
properties of several new ternary transition metal oxides predicted by high-throughput DFT screening. One
material reported here, Li SbO , displays reversible CO sorption, and maintains ∼72 % of its theoretical
5
5
2
capacity out to 25 cycles. The results in this work are used to discuss major influences on CO absorption
2
capacity and rate, including the role of the crystal structure, the transition metal, the alkali or alkaline earth
metal, and the competing roles of thermodynamics and kinetics. Notably, this work shows the extent and rate
to which ternary metal oxides carbonate is driven primarily by the identity of the alkali or alkaline earth ion
and the nature of the crystal structure, whereas the identity of the e transition ion carries little influence in
the systems studied here.
I. INTRODUCTION
screening method. While these experiments confirmed
the relative accuracy of the screening methodology in
predicting the reaction thermodynamics of the studied
materials, they did not account for variations in kinetics
or cycle stability, with many promising candidates such
The most expensive process in all carbon capture and
sequestration strategies is the separation of CO from
waste streams, making the discovery of suitable mate-
rials for separation processes a critical step to achiev-
ing a neutral carbon footprint. One technology that
has shown recent promise is high-temperature CO ab-
sorption looping, where a material first chemically re-
2
as Mg MnO , Ca Nb O and Na SbO failing on one or
6
8
4
2
9
3
4
1
both of these requirements. Furthermore, while many
promising candidates were obtained from the screening,
most were not tested in this initial wave of experimental
exploration and validation.
2
acts with CO to form a solid carbonate phase which is
2
then heated in a second reactor to decompose and re-
In this work, we used the results of the high-
throughput DFT screening as the starting point for a
more extensive experimental screening. 9 ternary metal
oxide materials were prepared to investigate the role of
crystal structure and chemical composition on the rate
lease pure CO and regenerate the starting material. Ca-
2
based sorbents, such as limestone, are the most widely
2
,3
tested and used materials for this purpose, and while
being abundant and cheap, their CO absorption capacity
2
rapidly decays with use because of undesirable changes
and reversibility of CO absorption reactions. The pre-
2
4
to the microstructure of the cycled CaCO particles.
dicted carbonation behaviour, CO capacity and reaction
3
2
The deficiencies in the CaO-CaCO system have mo-
are summarised in Table I. The entry for Li WO was cal-
3
6
6
tivated the exploration of various alternative ternary
culated directly from the Materials Project, as the initial
screening omitted it due to the phase being unstable with
respect to other phases in the phase diagram. The entry
for Ca WO was directly calculated in a similar fashion
5,6
7,8
metal oxide phases, including Li ZrO ,
Na ZrO ,
2
3
2 3
9,10
11,12
13
Li SiO ,
Li AlO
and Li FeO . In particular, the
4
4
5
4
5 4
perovskite material Ba Sb O was found to be able to
4
2
9
3
6
carbonate reversibly with negligible capacity loss even
after 100 cycles, underlining the importance of search-
due to it being added to the database after the original
screening.
14
ing for new materials with optimal CO absorption prop-
The materials chosen here were selected for several
reasons. Li MnO and Li ZnO have similarly high theo-
2
erties.
6
4
6
4
Towards this goal, a previous study performed a
large scale screening of the Materials Project database
retical gravimetric CO absorption capacities as Li AlO
2 5 4
and Li FeO and can be regenerated at lower temper-
5
4
1
5
(www.materialsproject.org), which contains the struc-
tural and theoretical ground state energies of over
67 000 solid state materials. After calculating the car-
bonation enthalpies of over 432 ternary oxide phases, a
number of candidates with desirable predicted proper-
ties were synthesised and characterised to validate the
atures than the formerly studied materials, possibly al-
lowing more efficient cycling. Li WO , Li WO , and
2
4
4
5
Li WO all lie in the complex Li–W–O phase diagram,
6
6
where the existence of small energy differences between
multiple stable phases may allow rapid and stable cy-
cling between different phases during carbonation, much
like the Li–Si–O system.16 The materials from the Ca–W–
O phase diagram, CaWO and Ca WO , were chosen to
determine the effect of changing the alkali element and
4
3
6
a)
these authors contributed equally
Electronic mail: cpg27@cam.ac.uk
b)
how W substitution influences the CaO-CaCO carbona-
3
ACS Paragon Plus Environment

Chemistry of Materials
Page 2 of 11
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
TABLE I. Carbonation reactions and carbonation temperatures under pCO2 = 1×10 Pa of materials studied in this work, compared
to the prototypical CaO sorbent. Carbonation temperatures were derived from ∆H and ∆S values largely determined by DFT
calculations performed in a previous large scale screening of the Materials Project database15 (values for Li WO and Ca WO
6
6
3
6
were calculated as part of this study).
Compound Tcarbonation CO capacity
Reaction
2
(K)
(gCO2 /gsorbent)
CaO
704
532
781
903
846
821
817
305
503
1005
323
949
0.785
0.213
0.186
0.165
0.822
0.450
0.772
0.168
0.151
0.137
0.153
0.330
CaO + CO −−→ CaCO
2
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Li SbO4
Li SbO + CO −−→ LiSbO + Li CO
3
3
4
2
3
2
3
Li SbO5
Li SbO + CO −−→ Li SbO + Li CO
5
5
5
2
3
4
2
3
3
Li SbO6
Li SbO + CO −−→ Li SbO + Li CO
7
7
6
2
5
1
3
5
2
1
3
Li MnO4
Li MnO + CO −−→ MnO + Li CO
6
6
4
2
2
3
4
7
1
7
Li ZnO4
Li ZnO + CO −−→ Li Zn O + Li CO
6
6
1
3
4
2
10
1
3
4
9
2
3
Li ZnO4
Li ZnO + CO −−→ ZnO + Li CO3
6
6
4
2
2
Li WO4
2
Li WO + CO −−→ WO + Li CO
2
4
2
3
2
3
4
Li WO5
Li WO + CO −−→ Li WO + Li CO
4
5
2
2
4
2
3
3
6
Li WO6
Li WO + CO −−→ Li WO + Li CO
6
6
2
4
5
2
CaWO4
CaWO + CO −−→ WO + CaCO
4
2
3
3
1
3
1
3
Ca WO6
Ca WO + CO −−→ WO + CaCO
3
3
6
2
3
3
tion reaction. Li SbO has a reasonable gravimetric ca-
PanAlytical Empyrean diffractometer in Bragg-Brentano
geometry using Cu Kα radiation. The instrument con-
tributions to the peak shape were determined using a
diffraction pattern of NIST SRM Si 640c collected in a
separate measurement. Rietveld and Pawley refinement
were performed using the Topas software package. In
situ XRD measurements above room temperature were
performed using an Anton-Paar XRK 900 furnace, with a
5
5
pacity without requiring high temperatures for regenera-
tion, and provides another point of reference for the role
that Sb plays in CCS materials in light of previous work
14
on Sb-containing ternary oxides, such as Ba Sb O .
4
2
9
1
7
Taken as a whole, the selection of materials studied
here allows us to study the impact of chemical, struc-
tural and thermodynamic factors on CCS performance,
working towards the twin goals of materials discovery
and increased fundamental understanding of structural
evolution during carbonation. It also allows further vali-
dation of the original screening methodology to improve
future studies.
−
1
total gas flow rate of 100 mL min
.
The carbonation and regeneration reactions of the
screened materials were investigated with thermogravi-
metric analysis (TGA) (TGA/DSC 1, Mettler Toledo) op-
erating at atmospheric pressure. In each experiment,
a sample of ∼20 mg of powder was placed in a 70 µL
Al O crucible, supported on a cantilever-type balance.
The reaction chamber was heated by a tube furnace sur-
rounding the balance. Both the protective gas and the
purge gas were N , and were fed to the TGA reaction
2
3
II. METHODS
Materials were prepared using ceramic methods; sto-
ichiometric amounts of solid precursor powders were
ground together in an agate mortar and pestle, then re-
acted at high temperatures under air or argon. Owing to
the wide range of materials, the same processing condi-
tions could not be used for every materials; the precur-
sors for each material, along with the processing sched-
ule and atmosphere, are summarised in Table II. The re-
actions involving LiOH were pretreated at 423 K for 1 h
in air to remove volatile components. Most complica-
tions in preparation or carbonation arose from melting
of samples.
2
chamber from a common source with a total flow rate of
−1
100 mL min . The reactive gas was either N or CO ,
2
2
fed by a capillary so that the gas could flow over the
top of the crucible. The partial pressure of CO at the
surface of the solid sample was adjusted by varying the
mix of N and CO in the reactive gas, while keeping a
constant overall flow rate of 100 mL min . The supply
of reactive gas was controlled by solenoid valves, allow-
ing automatic changing of reactive gas at different points
within the TGA program. The actual CO concentra-
tion at the gas-solid interface was calibrated against the
well-understood thermodynamic CaO/CaCO3 carbona-
tion equilibrium. Over a range of different pCO , the
temperature of the onset of CaCO decomposition was
2
2
2
−
1
2
To analyse structural changes upon carbonation, a por-
tion of the prepared material was subsequently annealed
2
under a flow of pure CO in a tube furnace at 973 K for
2
3
2
h (723 K for Li WO ). Both the as-prepared and car-
recorded and the corresponding CO partial pressure in
contact with the solid phase was determined from the
phase diagram of the CaO–CaCO –CO system.
4
5
2
bonated samples were characterised using room temper-
ature powder X-ray diffraction (XRD), performed on a
3
2
ACS Paragon Plus Environment

Page 3 of 11
1
Chemistry of Materials
3
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
TABLE II. List of precursor materials and processing schedules used during the solid-state preparation of the screened CCS materials.
CCS material Precursor materials
Reaction programme
Atmosphere
Li SbO5
LiOH and Sb O5
973 K for 24 h, 1073 K for 12 h
1223 K for 12 h
argon
argon
argon
argon
argon
air
5
2
Li MnO4
Li O and MnO
6
2
Li ZnO4
LiOH and ZnO 973 K for 24 h, 1073 K for 12 h, 1173 K for 24 h
LiOH and WO3
6
Li WO4
973 K for 12 h
973 K for 12 h, 1073 K for 12 h
973 K for 48 h
2
Li WO5
Li O and WO3
4
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Li WO6
6
Li O and WO3
2
CaWO4
CaCO and WO3
1273 K for 12 h
air
3
Ca WO6
CaCO and WO3
1273 K for 48 h
air
3
3
For single cycle carbonation and regeneration TGA ex-
periments, the material was heated from 323 K to 973 K
duced without the persistence of significant amounts of
MnO and Li CO (see Supporting Information). The ma-
2
3
−
1
at a rate of 10 K min , and then held at 973 K for
20 min. In the case of Li ZnO this temperature was
terial was still included in the initial thermogravimetric
experiments, as Li MnO and the impurities are expected
1
6
4
6
4
held for only 10 min. In the case of Li WO , the max-
to participate in the carbonation reactions.
The powder patterns of all materials and mixtures
were well-fit by the structural models, except Li WO .
4
5
imum temperature used was 732 K. In each case CO2
was supplied as the reactive gas for the duration of the
experiment. For the multiple cycling experiments of
Li SbO , Li WO and Li WO , the material was heated
6
6
The average crystal structure of Li WO (Immm) is de-
6
6
scribed by half-filled edge-sharing WO6 octahedra ex-
tending along the b direction, and corner-sharing along
the c direction (Figure 1). The measured diffraction pat-
tern of Li WO contains sharp reflections well-modelled
5
5
6
6
4
5
isothermally at 973 K (723 K for Li WO ) for 60 min un-
4
5
5
der pCO = 1.05 ± 0.12 ×10 Pa, then heated to 1223 K
2
(
4
823 K for Li WO ) and held at that temperature for
4 5
6
6
0 min under pure N , before being cooled to 973 K
by the Immm spacegroup, along with several additional
broad reflections with Warren-type lineshapes, beyond
the expected reflections (Figure 1). This is consistent
with the original report by Hauck, where extra reflec-
2
(
723 K for Li WO ) and held for 20 min under N . This
4 5 2
sequence was repeated for 25 cycles of carbonation and
regeneration. The samples obtained at the end of these
cycles were characterised using room-temperature XRD
as described above, along with other samples cycled in a
tube furnace for 3 cycles, with a program of carbonation
6+
19
tions were attributed to partial ordering of W ions in
the formally half-filled 2a site. This is sensible in light of
Pauling’s rules, where structures will seek to maximize
the distance between highly charged centres to mini-
mize the Coulombic energy penalty. In this case, given
at 973 K for 2 h under pure CO followed by regeneration
2
at 1223 K for 2 h under pure N (temperatures of 723 K
2
6
+
and 823 K were used for Li WO ).
the high charge-radius ratio of W , it is energetically-
4
5
Backscattered electron micrographs were collected us-
ing a field emission gun scanning electron microscope
unfavourable to form edge-sharing pairs of WO octahe-
dra, and W orders on the 2a site (at least locally) to
6
6
+
(
Camscan MX2600) operating at an accelerating voltage
minimize the system energy.
The surface area of the candidate materials was char-
acterised using BET (see Supporting Information); all
materials exhibit low surface areas (∼1 m g ), which
is expected as the ceramic methods used in the material
preparation involve annealing at high temperatures for
long dwell times. Owing to the similar, low surface ar-
eas, we do not expect sample morphology or surface area
to be a factor in the relative performance of the pristine
materials studied here.
of 10 kV. 10 nm of Pd was sputter-deposited on speci-
mens to minimize charging.
2
−1
The porosity and specific surface area (SSA) of the
candidate materials were determined using volumet-
ric sorption measurements (TriStar3000 analyzer, Mi-
cromeritics) in N2 at 77 K. The SSA was calculated
using Brunauer-Emmett-Teller (BET) analysis using N2
adsorption.18
Analysis of the materials after carbonation using room-
temperature XRD confirm the following reactions take
place (diffractograms shown in Supporting Information):
III. RESULTS AND DISCUSSION
A. Characterisation of screened materials
Li SbO + CO −−→ Li SbO + Li CO
(1)
(2)
5
5
2
3
4
2
3
XRD was used to characterise the identity and purity
of all materials, which were single-phase after prepara-
tion with the exception of Li MnO . Despite multiple dif-
6
4
1
3
1
Li MnO + CO −−→ MnO + Li CO3
ferent synthetic approaches, Li MnO could not be pro-
6
4
2
2
6
4
3
ACS Paragon Plus Environment

Chemistry of Materials
Page 4 of 11
4
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
3
1
Li ZnO + CO −−→ ZnO + Li CO3
(3)
6
4
2
2
3
Li WO + CO −−→ Li WO + Li CO
(4)
(5)
4
5
2
2
4
2
3
Li WO + CO −−→ Li WO + Li CO
6
6
2
4
5
2
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
These results agree with the predicted carbonation
reactions obtained from the Materials Project screen-
ing (Table I), with several exceptions. First, CaWO4,
Ca WO and Li WO showed no evidence of carbona-
tion in their diffractograms following exposure to flow-
ing CO in a tube furnace, and were thus not subjected
to any further experiments. Second, we observe that
Li ZnO carbonates to form ZnO and Li CO , rather than
the products predicted by the screening (Li Zn O and
3
6
2
4
2
6
4
2
3
10
4
9
Li CO ). This was more carefully confirmed by perform-
2
3
ing in situ XRD at 10 minute intervals during carbona-
tion of Li ZnO in atmospheres of 100% CO and 50%
6
4
2
CO (N ), at 50 K intervals between 773 K and 973 K;
there was no evidence for the formation of Li Zn O
nor other Zn species. Further, XRD measurements on
Li ZnO4 powder exposed to ambient lab atmosphere
for prolonged periods revealed the presence of only
Li ZnO , ZnO, and Li CO .
For each material in the Materials Project, our screen-
ing metholodogy finds several different possible carbon-
ation reactions, and ranks them based on the calculated
carbonation enthalpy of each reaction. We select the re-
action with the lowest (most negative) carbonation en-
thalpy and assume that it will be the only reaction that is
activated.
Selecting the reaction with the lowest carbonation en-
thalpy works well when there is a reaction that is clearly
more favourable than the other possibilities, but in this
case the two reactions have extremely similar carbona-
tion enthalpies, well within the errors of the underlying
DFT metholodogy:
2
2
1
0
4
9
6
6
4
2
3
1
3
1
Li ZnO + CO −−→ ZnO + Li CO3
6
4
2
2
3
(6)
(7)
FIG. 1. (a) XRD pattern and Rietveld refinement of Li WO
6
6
−
1
in the Immm spacegroup (λ = 1.5405 A˚ ). Additional peaks
∆
H = −2.25 eV(−217 kJ mol
)
with Warren-type broadening (*) are present due to ordering
6
+
of W ions in the half-filled 2a site. It is energetically un-
favourable to have edge-sharing octahedra, which would bring
4
7
1
6+
Li ZnO + CO −−→ Li Zn O + Li CO
together highly-charged W ions. Ordering of W to prevent
6
4
2
10
4
9
2
3
7
edge-sharing WO pairs in the local structure lowers the system
6
−1
∆
H = −2.26 eV(−217 kJ mol
)
energy, and leads to the broadened extra reflections observed
in the XRD pattern. (b) The average crystal structure of Li WO
6
6
(
Immm) is described by half-filled edge-sharing WO octahe-
Selecting the reaction with the lowest carbonation en-
thalpy also neglects kinetic limitations, which are likely
responsible for the inert behaviour of CaWO , Ca WO
6
dra extending along the b direction, and corner-sharing along
the c direction.
4
3
6
and Li WO4 seen here. These kinetic considerations
2
cannot be easily treated from thermodynamic quanti-
ties such as ∆H and ∆S calculated from the Materials
Project.
ACS Paragon Plus Environment

Page 5 of 11
1
Chemistry of Materials
5
B. Initial carbonation reaction thermogravimetry
CO in the high-temperature regime is ascribed to bulk
2
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
solid-state reaction, enabled by a sufficiently high rate of
ionic transport activated at higher temperatures.
To investigate the initial carbonation capacity and ki-
netics of the candidate materials, ramp TGA experiments
were carried out at a rate of 10 K min (Figure 2). The
experimental capacities and reaction temperatures for
the initial carbonation step compared to the theoreti-
cal values from the Materials Project screening are sum-
marised in Table III.
−
1
C. Cycling experiments
Of the materials that could initially carbonate, only
Li SbO , Li WO and Li WO could be regenerated and
5
5
4
5
6
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
cycled in the TGA. Li ZnO and Li MnO both show evi-
6
4
6
4
dence of melting at the higher temperatures required for
regeneration, most likely due to the formation of Li CO
2
3
(
with a melting point of 996 K), making them unsuitable
for further use.
The materials that could be regenerated were cycled
5 times in the TGA; mass–time curves for the 1st, 2nd,
0th and 25th cycles are shown in Figure 3. The differ-
2
1
ence in mass after 60 minutes of exposure to CO at reac-
2
tion temperature was used to determine the mass of CO2
absorbed, which was then used calculate the percentage
of theoretical capacity according to the appropriate reac-
tion stoichiometry shown earlier in Equations 1–5. The
CO sorption capacity as a percentage of theoretical ca-
2
pacity for the material is plotted as a function of cycle
number, along with the corresponding performance of
CaO for comparison (Figure 4).
Li SbO displays an initial carbonation capacity >90%
5
5
of the theoretical capacity, and while this quickly de-
creases to ∼75% by the 3rd CO sorption/desorption
2
cycle, Li SbO displays excellent capacity retention for
5
5
the duration of the cycling experiment (Figure 4), with
∼72% of the theoretical capacity after 25 cycles. Fur-
thermore, in addition to the small decrease in capacity
with cycling, the material shows much slower carbona-
tion kinetics. In the first cycle, Li SbO absorbs 70% of
FIG. 2. TGA traces for selected candidate materials heated un-
4
der CO2 (pCO2 = 1 × 10 Pa). Most materials studied here
exhibit similar carbonation profiles, with the onset of a small
initial mass gain at low temperature around ∼600 K followed
by a sharp onset and substantial mass gain initiated between
850 K and 900 K. Traces from previous studies for Li FeO and
5
5
the theoretical capacity of CO within 5 minutes when
2
5
4
exposed to CO , whereas achieving the same sorption af-
15
2
Li CoO are added for comparison
.
6
4
ter 10 cycles takes 60 minutes (Figure 3).
Li WO displays an initial carbonation capacity ∼70%
6
6
The initial TGA experiments are consistent with the
of the theoretical capacity, and the sorption capacity de-
creases steadily as the material is cycled (Figure 4). By
T
values obtained from the screening; the ma-
carbonation
jority of materials lie within ∼100 K of their experimental
values. This corresponds to an error of ∼10 meV, and is
well within the limits of the DFT methodology. However,
the experimental carbonation capacities are all lower
than the corresponding theoretical limits, with some ma-
terials showing significant deviations (e.g. Li MnO ,
the 25th cycle, the CO sorption capacity is ∼35% of the
2
theoretical capacity. The first cycle has an anomalous
sorption capacity because some of the material was likely
partially carbonated prior to the experiment, as Li WO
6
6
spontaneously carbonates in ambient conditions. Despite
the steady degradation in sorption capacity as the mate-
rial is cycled, Li WO demonstrates impressive carbona-
6
4
Li WO , and Li WO ).
4
5
6
6
6
6
It is clear from the TGA ramp experiments that
tion kinetics; the material reaches maximum capacity in
less than 1 minute, with no change upon repeated cy-
cling.
Li MnO and Li ZnO exhibit a two-stage carbonation
6
4
6
4
behaviour, commonly observed in other materials with
a similar chemical composition and structure such as
Li WO displays poor CO sorption capacity compared
4
5
2
1
1,13,15
Li AlO , Li FeO , and Li CoO .
This two-stage car-
to its theoretical capacity, but exhibits the most stable
gravimetric capacity of all the materials studied here
5
4
5
4
6
4
bonation features the onset of a small initial mass gain
at low temperature around ∼600 K followed by a sharp
onset and substantial mass gain initiated between 850 K
and 900 K. The low-temperature regime is ascribed to re-
actions near the surface, whereas the high sorption of
(Figure 4). Li WO maintains a steady capacity of ∼20%
4
5
out to 25 cycles studied here, and the carbonation kinet-
ics improve as the material is cycled. Whereas in the first
cycle after 5 minutes of exposure to CO only ∼6% of the
2
ACS Paragon Plus Environment

Chemistry of Materials
Page 6 of 11
6
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
TABLE III. Theoretically derived carbonation reaction parameters for the preliminary candidates selected for this study compared
to the values obtained from TGA experiments under pCO2 = 1 × 10 Pa.
4
Compound
T
carbonation (K)
CO capacity (gCO2 /gsorbent)
2
Screening Experimental Screening Experimental
Li SbO5
781
846
817
503
1005
890
840
810
500
880
0.186
0.822
0.772
0.151
0.137
0.151
0.352
0.638
0.031
0.058
5
Li MnO4
6
Li ZnO4
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Li WO5
4
Li WO6
6
FIG. 3. Selected carbonation profiles for Li SbO , Li WO and Li WO from extended carbonation cycling experiments. Both
5
5
4
5
6
6
Li SbO and Li WO show a decrease in capacity after 25 cycles, with Li SbO in particular showing a much flatter absorption
5
5
6
6
5
5
profile. Li WO retains both its profile and capacity for the duration of the cycling, albeit with the lowest capacity of the three
4
5
materials.
theoretical sorption capacity is reached, after 25 cycles
this is doubled; Li WO absorbs 13% of the theoretical
capacity after the same exposure time (5 min).
regeneration is surprising considering this behaviour has
never been seen for other similar materials, but may play
a role in the stable cycling capacity observed for the ma-
4
5
terial. In addition to the formation of Li SbO , there is
7
6
also significant formation of Li SbO , which does not re-
3
4
act upon regeneration under inert atmosphere at higher
temperature. TGA experiments on single-phase Li SbO
D. Structural and morphological evolution of Li SbO upon
cycling
5
5
3
4
prepared separately indicate this phase is inert with re-
spect to carbonation (see Supporting Information), so
this represents a permanent loss in capacity, as seen in
the cycling TGA experiments discussed earlier.
Li SbO was found to be the material with the most
5
5
promising thermodynamics of carbonation and capac-
ity retention, so it was subjected to further in-depth
structural and morphological characterisation to exam-
ine changes as a function of cycle number. Samples
were extracted at different points in the cycling experi-
ment and analysed using XRD and scanning electron mi-
croscopy (SEM). The resulting diffractograms were anal-
ysed and the phase fractions in the regenerated mate-
rial were determined through Rietveld refinement (Fig-
ure 5). Diffractograms and fits are available in the Sup-
porting Information.
The phase composition of Li SbO5 eventually does
5
evolve with repeated cycling, and after 26 carbonation
cycles the fraction of Li SbO formed upon regeneration
5
5
decreases to 55 wt.%. However, Li SbO is formed in
7
6
a much greater proportion (21 wt.%) than in the early
cycles, and is present in roughly the same proportion
as Li SbO (24 wt.%). The higher lithium content in
3
4
Li SbO decreases the stability of the phase with respect
7
6
to CO , so unlike Li SbO , we expect Li SbO to be much
2
3
4
7
6
The phase composition of Li SbO after cycling re-
mains relatively stable up to at least three cycles, and is a
more reactive to CO . Consequently, although the frac-
5
5
2
tion of the original phase being regenerated decreases
with increased cycling, the capacity of ”Li SbO ” as a
mixture of ∼83–86 wt.% Li SbO , ∼12–15 wt.% Li SbO
5
5
3
4
5
5
and ∼2–3 wt.% Li SbO (Figure 5). The formation of
material remains relatively stable due to the formation
of Li SbO .
7
6
a secondary phase with a higher lithium content upon
7
6
ACS Paragon Plus Environment

Page 7 of 11
Chemistry of Materials
7
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
FIG. 4. The percentage absorption capacity measured for each
of Li SbO , Li WO and Li WO over the 25 carbonation and
FIG. 5. Phase fraction evolution of regenerated Li SbO as
5
5
5
5
6
6
4
5
a function of cycle number, as determined by Rietveld re-
finement. Upon repeated carbonation cycles, less Li SbO is
regeneration looping cycles. Li SbO displays sustained capac-
5
5
5
5
ity retention after the first 3 cycles; Li WO5 actually shows
4
formed upon regeneration, but overall CO sorption capacity
2
an increase in capacity overall. The Li WO6 displays less
6
is unchanged due to the formation of unreactive Li SbO and
3
4
favourable looping behaviour, with a steady loss in capacity
from cycle 2 onwards. The results for CaO were calculated
using data courtesy of Liu et al.20 and have been included for
comparison.
reactive Li SbO .
7
6
ter understand the key factors in carbonation reactions.
We examine the role of metal ion and crystal struc-
ture by comparing alkali transition metal oxides adopting
the antifluorite-type structure. These phases have similar
crystal structures and compositions, and differ primarily
in the identity of the transition metal in the material. The
materials include Li MnO , Li ZnO , and Li CoO (stud-
The SEM micrographs, shown in Figure 6, also give
an insight into the microstructural evolution of Li SbO5
5
upon cycling. After the first carbonation step, the for-
mation of Li CO can be seen in the apparent coarsen-
2
3
ing of the Li SbO particles, in line with the observa-
5
5
6
4
6
4
6
4
15
12
tions in previous SEM experiments of carbonated mate-
ied in this work), as well as Li FeO4 and Li AlO .
5 5 4
1
5
rials such as Na SbO . This surface area is recreated
These materials all show remarkably similar carbonation
profiles, with the onset of a small initial mass gain at
low temperature around ∼600 K, followed by a sharp on-
set and substantial mass gain at higher temperatures ini-
tiated between 850 K and 900 K. This is sensible given
the carbonation reactions of these materials calculated
3
4
upon regeneration, and the image at higher magnifica-
tion shows that the finer structural features on the sur-
face of the particles are also regenerated. Crucially, the
images taken at the same point in the carbonation cycle
after 26 cycles show a similar surface area and size of
the secondary particles. The primary particles are larger
and less fused together following multiple cycles, which
supports the stable cycling capacity seen in the material.
DFT
in this work have very similar ∆E , with values rang-
ing from −2.16 eV to −2.32 eV. This family of materials
display similar carbonation capacities, with experimen-
tal values ranging from 82% of theoretical capacity for
1
2
Li CoO studied in this work, to 96% for Li AlO .
6
4
5
4
E. Discussion
Looking across these lithium-rich antifluorite-type ma-
terials, the choice of metal in these materials is not
critical, and given the reasonable carbonation proper-
ties across the family, targeting specific structure types
may be an efficient and fruitful direction for future ra-
tional design. Here, the thermodynamic driving force
and high capacities are enabled by the high lithium con-
tent. Meanwhile, the good kinetics and remarkably fast
The benefit of a large scale screening is that it allows
broader insight into trends and underlying causes of ob-
served physical properties. This study is the latest in a
growing body of work experimentally characterising the
CO absorption properties of promising ternary oxides,
2
and the results can be compared to previous work to bet-
ACS Paragon Plus Environment

Chemistry of Materials
Page 8 of 11
8
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
FIG. 6. Scanning electon micrographs of Li SbO after (a) the 1st carbonation, (b) the 1st regeneration, and (c) the 26th re-
generation. Both regenerated specimens display similar morphology and no sign of coarsening, suggesting the microstructure is
conserved upon repeated cycling.
5
5
CO absorption seen at higher temperatures are likely
bonation cycling conditions might be missed.
2
enabled by the many partially-occupied Li sites in the
crystal structure, which lead to high rates of Li trans-
port through the material once defect-mediated trans-
port mechanisms are activated.
14
As seen in previous studies on Ba Sb O , conserva-
4
2
9
tion of capacity and morphology are strongly linked, a
phenomenon that is the subject of ongoing analysis and
effort to also include in future screening methodologies.
The SEM of Li SbO further supports these earlier stud-
ies, even if the underlying causes for microstructure re-
tention are not known. There is nevertheless a degra-
Despite the promising CO2 sorption capacities and
rates, this study found Li MnO and Li ZnO could not
5
5
6
4
6
4
be regenerated and hence cycled, in a similar fashion
15
dation in the absorption rate of Li SbO as the mate-
to the related materials Li FeO and Li CoO . Other
5
5
5
4
6
4
rial is cycled, which is an opportunity for future studies
aimed at achieving capacity retention rates comparable
to those obtained following steam injection in the CaO
system.2
methods will thus be required to make these materials
appropriate for looping applications. One strategy that
has been shown to be successful is the inclusion of water
vapour in the reactant gas stream which led to a signif-
2–24
icant decrease in the temperature required for regener-
The absolute capacity of Li WO is the lowest amongst
4
5
21
ation of Li AlO . A similar strategy that shifts the car-
5
4
the materials in this study, and would certainly require
larger amounts of material to be used in CCS applica-
tions, but this is not necessarily an insurmountable bar-
bonation equilibrium temperature may very well work
with the entire class of antifluorite materials, and future
work is planned to investigate this and the overall role
rier to its implementation. Li WO5 is predominantly
4
water plays in CO absorption and regeneration cycles.
2
composed of Li–O, with a low loading of W. Conse-
quently, the total mass needed to achieve a certain gravi-
The surprising cycling behaviour of Li SbO and the
5
5
metric CO absorption value would be comparable to
formation of increasing amounts of the over-lithiated
phase Li SbO opens a new route towards finding new
2
heavier materials with better capacities, such as Li SbO
5
5
7
6
or Ba Sb O . Furthermore, certain CCS applications do
ternary systems that maintain their cycling capacity. In-
stead of relying upon complete regeneration of the orig-
inal material, if there are other phases with similar com-
position and reactivity that can be accessed during the
reaction, their creation can counteract the loss of ca-
pacity caused by coarsening and the permanent forma-
tion of inert phases. Furthermore the SEM images in-
dicate that Li SbO may indeed form with similar mor-
4
2
9
not require particularly high capacities. In particular,
CCS applications on a smaller scale than power gen-
eration, such as improving hydrogen yields in chemi-
25
cal looping processes or chemical looping partial oxi-
dation of methane,26 rely on efficient and complete re-
generation over many cycles, properties displayed by
Li WO . This finding highlights the advantage of large
4
5
7
6
scale screening methods in that materials can be found
for different target applications in the same experiment,
and further selected on various criteria.
phology as the original Li SbO phase, leading to little
5
5
decline in reactivity with cycling. The fact these phases
might be metastable (Li SbO lies 0.008 eV above the
7
6
hull in the Li–Sb–O phase diagram calculated by the Ma-
terials Project) means future screenings should include
metastable materials close to the hull, otherwise poten-
tial reaction pathways that can be accessed under car-
A final interesting observation arising from the screen-
ing comes when considering the materials that were
predicted to carbonate but failed to do so in experi-
ments. This group of materials had low predicted tem-
ACS Paragon Plus Environment

Page 9 of 11
1
Chemistry of Materials
9
peratures of carbonation, where under atmosphere of
ture that is preserved through repeated cycling; scan-
ning electron micrographs of ”Li SbO ” after regenera-
tion do not reveal any significant evolution of morphol-
ogy after 25 cycles. Further structural analysis at dif-
ferent stages of cycling reveals the formation of a sec-
ondary metastable phase upon reaction, Li SbO , which
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5
pCO < 1 × 10 Pa, T
. 600 K: CaWO (323 K),
carbonation
4
5
5
2
Li WO (503 K), Ca Nb O (606 K), Li NbO (503 K),
2
4
4
2
9
3
4
Li TaO (511 K) and Li SbO (532 K). Considering the
earlier studies investigating the connection between car-
bonation and ionic mobility,
3
4
3
4
27–29
these results suggest a
7
6
material may not carbonate at low temperatures due to
insufficient ionic mobility, even if a material is predicted
to carbonate based on thermodynamics. As the tempera-
ture is increased into a region where ionic mobility might
be sufficient, the equilibrium may favour regeneration
(i.e., decarbonation) rather than carbonation, leading to
further inactivity.
The need to balance two competing factors for the de-
velopment of successful CCS materials is important for
future studies. The carbonation temperature must be
low enough to avoid particle coarsening, which leads to
may prevent coarsening and deactivation of the sorbent.
This phase space, where there are many easily accessible
stoichiometries upon reaction that are close in energy,
could prove fertile ground for the design of new robust
CCS materials.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Further analysis of the candidate materials studied
here in concert with the body of previous research car-
ried out by authors in this work hint at other important
design principles for future research, the most crucial
being the need to use a moderate temperature to sat-
isfy both energetic and diffusion considerations in CO2
absorption materials. The design of new materials, as
well as the chemical and structural modification of ex-
isiting materials, need to consider these two factors to
find a suitable reaction temperature that has favourable
thermodyanamic driving force as well as sufficiently fast
ionic transport. The performance of materials in CCS
applications is the result of a delicate balance between
thermodynamics, ion diffusion, and morphology, and re-
quires a similarly balanced mix of theoretical, synthetic
and analytical approaches to fully understand how to
harness these competing forces.
loss of capacity as in the pure CaO-CaCO system. How-
3
ever, the carbonation temperature cannot be so low as to
hinder ionic diffusion in the material and cause slow car-
bonation kinetics. This also explains the anomalous ma-
terials that carbonate at low temperature: owing to the
+
smaller ionic radius and charge of Li when compared to
2+
Ca , one would expect faster ionic diffusion in Li WO5
4
at lower temperatures than Ca analogues (e.g., CaWO4
and Ca WO ), and hence we observe some carbonation
3
6
of Li WO even at the low temperatures where the Ca-
4
5
based materials do not react. The difficulty in achieving
sufficient Ca ionic diffusion means that even with a rea-
sonably high carbonation temperature the reaction does
not occur. Thus, thermodynamic screening needs to be
carried out in concert with screening for ionic diffusivity,
a methodology that is the focus of ongoing work. Focus-
ing only on decreasing the energy penalty of the chosen
material needs to be balanced with other parameters that
influence actual performance.
ACKNOWLEDGMENTS
The authors thank Zlatko Sarakevic for carrying out
the BET experiments, and Wen Liu, Stuart Scott and John
Dennis for fruitful discussions regarding the work and
assistance with the TGA experiments. The authors also
gratefully acknowledge Anubhav Jain (Computational
Research Division, Lawrence Berkeley National Labora-
tory), Shyue Ping Ong (Department of NanoEngineering,
University of California San Diego) and Kristin Persson
(Department of Materials Science and Engineering, Uni-
versity of California Berkeley) for their advice and assis-
tance in accessing the Materials Project, which itself is
supported through the U.S. Department of Energy, Office
of Basic Energy Sciences, Materials Project Center Grant
No. EDCBEE. MWG is grateful for support from the Euro-
pean Union’s Horizon 2020 research and innovation pro-
gramme under the Marie Skłodowska–Curie grant agree-
ment No. 659764. MTD and CPG acknowledge funding
from EPSRC Grant No. EP/K030132/1. MTD is grateful
for support from Clare College, Cambridge through the
award of a Junior Research Fellowship.
IV. CONCLUSIONS
In this study we used the theoretical screening of struc-
tural databases as a starting point for exploring novel
materials for use in high-temperature carbon capture and
storage (CCS) applications. Diverse candidate materials
were selected, and either displayed desirable theoretical
properties or allowed direct compositional and/or struc-
tural comparisons to previously studied materials. These
candidates were subjected to a suite of structural and
thermogravimetric experiments to determine their per-
formance as CO looping materials.
2
In the first instance, this initial testing reveals that the
theoretical screening accurately predicts the carbonation
temperature of the candidate materials studied here, but
fails to identify materials whose carbonation is hindered
by poor kinetics. Li SbO , Li WO and Li WO all ex-
V. DATA
5
5
6
6
4
5
hibit reversible CO sorption, with Li SbO showing a
2
5
5
particularly high and stable capacity retention over 25
carbonation cycles. Li SbO has a stable microstruc-
All supporting data for this work can be found on
https://www.repository.cam.ac.uk.
5
5
ACS Paragon Plus Environment

Chemistry of Materials
Page 10 of 11
1
0
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
VI. SUPPORTING INFORMATION
17A. A. Coelho, “TOPAS and TOPAS-Academic: An optimization pro-
gram integrating computer algebra and crystallographic objects writ-
ten in C++,” J. Appl. Crystallogr. 51, 210–218 (2018).
XRD diffractograms and Rietveld refinements for the
candidate materials, surface area measurements, XRD
diffractograms of Li SbO at different stages of car-
18
S. Brunauer, P. H. Emmett, and E. Teller, “Adsorption of gases in
multimolecular layers,” J. Am. Chem. Soc. 60, 309–319 (1938).
J. Hauck, “Zur kristallstruktur des Li6WO6,” Z. Naturforsch. B 24,
19
5
5
2
51 (1969).
bonation cycle, and TGA traces for candidate materi-
als. This material is free of charge via the Internet at
http://pubs.acs.org.
20
21
W. Liu, B. Gonz a´ lez, M. T. Dunstan, D. S. Sultan, A. Pavan, C. D.
Ling, C. P. Grey, and J. Dennis, “Structural evolution in synthetic,
Ca-based sorbents for carbon capture,” Chem. Eng. Sci. 139, 15–26
(
2016).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
´
T. L. Avalos-Rend o´ n and H. Pfeiffer, “High CO2 chemisorption in α-
◦
Li5AlO4 at low temperatures (30–80 C): Effect of the water vapor
REFERENCES
addition,” Energy Fuels 26, 3110–3114 (2012).
22
V. Nikulshina, M. E. G a´ lvez, and A. Steinfeld, “Kinetic analysis of
the carbonation reactions for the capture of CO2 from air via the
Ca(OH)2–CaCO3–CaO solar thermochemical cycle,” Chem. Eng. J.
129, 75–83 (2007).
1A. D. Ebner and J. A. Ritter, “State-of-the-art adsorption and mem-
brane separation processes for carbon dioxide production from car-
bon dioxide emitting industries,” Separ. Sci. Technol. 44, 1273–1421
23
24
25
V. Manovic and E. J. Anthony, “Carbonation of CaO-based sorbents
enhanced by steam addition,” Ind. Eng. Chem. Res 49, 9105–9110
(
2009).
2
D. P. Hanak, E. J. Anthony, and V. Manovic, “A review of develop-
ments in pilot-plant testing and modelling of calcium looping process
for CO2 capture from power generation systems,” Energy Environ.
Sci. 8, 2199–2249 (2015).
A. M. Kierzkowska, R. Pacciani, and C. R. M u¨ ller, “CaO-based CO2
sorbents: From fundamentals to the development of new, highly ef-
fective materials,” ChemSusChem 6, 1130–1148 (2013).
E. Bouquet, G. Leyssens, C. Sch o¨ nnenbeck, and P. Gilot, “The
decrease of carbonation efficiency of CaO along calcination–
carbonation cycles: Experiments and modelling,” Chem. Eng. Sci.
64, 2136–2146 (2009).
(
2010).
I. Lind ´e n, P. Backman, A. Brink, and M. Hupa, “Influence of water
vapor on carbonation of CaO in the temperature range 400–550 C,”
◦
Ind. Eng. Chem. Res 50, 14115–14120 (2011).
3
4
M. S. C. Chan, W. Liu, M. Ismail, Y. Yang, S. A. Scott, and J. S.
Dennis, “Improving hydrogen yields, and hydrogen:steam ratio in the
chemical looping production of hydrogen using Ca2Fe2O5,” Chem.
Eng. J. 296, 406–411 (2016).
S. Bhavsar and G. Veser, “Chemical looping beyond combustion: Pro-
duction of synthesis gas via chemical looping partial oxidation of
methane,” RSC Adv. 4, 47254–47267 (2014).
J.-i. Ida and Y. S. Lin, “Mechanism of high-temperature CO2 sorption
on lithium zirconate,” Environ. Sci. Technol. 37, 1999–2004 (2003).
M. T. Dunstan, J. M. Griffin, F. Blanc, M. Leskes, and C. P. Grey, “Ion
dynamics in Li2CO3 studied by solid-state NMR and first-principles
calculations,” J. Phys. Chem. C 119, 24255–24264 (2015).
M. T. Dunstan, H. Laeverenz Schlogelhofer, J. M. Griffin, M. S. Dyer,
M. W. Gaultois, C. Y. Lau, S. A. Scott, and C. P. Grey, “Ion dynam-
ics and CO2 absorption properties of Nb-, Ta-, and Y-doped Li2ZrO3
studied by solid-state NMR, thermogravimetry, and first-principles
calculations,” J. Phys. Chem. C 121, 21877–21886 (2017).
26
5
6
7
8
9
0
1
K. Nakagawa and T. Ohashi, “A novel method of CO2 capture
from high temperature gases,” J. Electrochem. Soc. 145, 1344–1346
27
28
(
1998).
K. Nakagawa and T. Ohashi, “A reversible change between lithium
zirconate and zirconia in molten carbonate,” Electrochem. 67, 618–
6
21 (1999).
A. L o´ pez-Ortiz, N. G. P. Rivera, A. R. Rojas, and D. L. Gutierrez,
Novel carbon dioxide solid acceptors using sodium containing ox-
29
“
ides,” Sep. Sci. Technol. 39, 3559–3572 (2004).
T. Zhao, E. Ochoa-Fern a´ ndez, M. Rønning, and D. Chen, “Prepara-
tion and high-temperature CO2 capture properties of nanocrystalline
Na2ZrO3,” Chem. Mater. 19, 3294–3301 (2007).
T. Yamaguchi, T. Niitsuma, B. N. Nair, and K. Nakagawa, “Lithium
silicate based membranes for high temperature CO2 separation,” J.
Membr. Sci. 294, 16–21 (2007).
1
1
M. J. Venegas, E. Fregoso-Israel, R. Escamilla, and H. Pfeiffer, “Ki-
netic and reaction mechanism of CO2 sorption on Li4SiO4: Study of
the particle size effect,” Ind. Eng. Chem. Res 46, 2407–2412 (2007).
´
T. Avalos-Rend o´ n, J. Casa-Madrid, and H. Pfeiffer, “Thermochem-
ical capture of carbon dioxide on lithium aluminates (LiAlO2 and
Li5AlO4): A new option for the CO2 absorption,” J. Phys. Chem A
1
13, 6919–6923 (2009).
12
´
T. Avalos-Rend o´ n, V. H. Lara, and H. Pfeiffer, “CO2 chemisorption
and cyclability analyses of lithium aluminate polymorphs (α- and β-
Li5AlO4),” Ind. Eng. Chem. Res 51, 2622–2630 (2012).
H. A. Lara-Garc ´ı a, P. Sanchez-Camacho, Y. Duan, J. Ortiz-Landeros,
and H. Pfeiffer, “Analysis of the CO2 chemisorption in Li5FeO4, a
new high temperature CO2 captor material. effect of the CO2 and O2
partial pressures,” J. Phys. Chem. C 121, 2455–3462 (2017).
M. T. Dunstan, W. Liu, A. F. Pavan, J. A. Kimpton, C. D. Ling, S. A.
Scott, J. S. Dennis, and C. P. Grey, “Reversible CO2 absorption by the
1
3
1
4
5
6
H perovskite Ba4Sb2O9,” Chem. Mater. 25, 4881–4891 (2013).
1
M. T. Dunstan, A. Jain, W. Liu, S. P. Ong, T. Liu, J. Lee, K. A. Pers-
son, S. A. Scott, J. S. Dennis, and C. P. Grey, “Large scale compu-
tational screening and experimental discovery of novel materials for
high temperature CO2 capture,” Energy Environ. Sci. 9, 1346–1360
(
2016).
1
6
D. Cruz, S. Bulbulian, E. Lima, and H. Pfeiffer, “Kinetic analysis of
the thermal stability of lithium silicates (Li4SiO4 and Li2SiO3),” J.
Solid State Chem. 179, 909–916 (2006).
ACS Paragon Plus Environment

Page 11 of 11
Chemistry of Materials
1
1
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
TOC graphic
ACS Paragon Plus Environment