Selective CO2 adsorption by a new metal-organic framework: synergy between open metal sites and a charged imidazolinium backbone

Article abstract of DOI:10.1039/c8dt01247d

Metal-organic frameworks (MOFs) are porous, tunable crystalline materials that are attracting widespread scientific attention for their potential use in post-combustion CO₂ capture. In this work, we report the synthesis of a new ligand, 1,3-bis(4-carboxyphenyl)-4,5-dihydro-1H-imidazol-3-ium tetrafluoroborate, H₂Sp5-BF₄, that is subsequently used for the construction of a novel MOF, Cu-Sp5-EtOH. This highly crystalline material has a charged framework that is expected to give rise to high CO₂/N₂ selectivity. However, the pores of the parent structure could not be accessed due to the presence of strongly coordinated ethanol molecules. After solvent exchange with methanol and subsequently heating Cu-Sp5-MeOH under vacuum, we are able to liberate the solvent providing other small molecules like CO₂ access to the inside of the now porous structure, Cu-Sp5. The combination of open metal sites and framework charge leads to an exceptionally high CO₂/N₂ selectivity, as determined by Ideal Adsorbed Solution Theory (IAST) calculations performed on single-component adsorption isotherms. The CO₂/N₂ selectivity of Cu-Sp5 reaches a value of over 200 at pressures typically found in post-combustion flue gas (0.15 bar CO₂/0.85 bar N₂), a value that is among the highest reported to date.

Full text of DOI:10.1039/c8dt01247d

View Article Online
View Journal
Dalton
Transactions
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: I. Kochetygov, S.
Bulut, M. Asgari and W. L. Queen, Dalton Trans., 2018, DOI: 10.1039/C8DT01247D.
This is an Accepted Manuscript, which has been through the
Volume 45 Number
1 7 January 2016 Pages 1–398
Royal Society of Chemistry peer review process and has been
accepted for publication.
Dalton
Transactions
Accepted Manuscripts are published online shortly after
An international journal of inorganic chemistry
www.rsc.org/dalton
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-9226
PAPER
Joseph T. Hupp, Omar K. Farha et al.
Efficient extraction of sulfate from water using
framework
a Zr-metal–organic
rsc.li/dalton

Please do not adjust margins
Dalton Transactions
Page 1 of 10
View Article Online
DOI: 10.1039/C8DT01247D
Dalton Transactions
ARTICLE
Selective CO2 adsorption by a new metal-organic framework:
synergy between open metal sites and a charged imidazolinium
backbone†
Received 00th January 20xx,
Accepted 00th January 20xx
Ilia Kochetygov,‡ Safak Bulut,‡ Mehrdad Asgari and Wendy L. Queen*
DOI: 10.1039/x0xx00000x
www.rsc.org/
Metal-organic frameworks (MOFs) are porous, tunable crystalline materials that are attracting widespread scientific
attention for their potential use in post-combustion CO2 capture. In this work, we report the synthesis of a new ligand, 1,3-
bis(4-carboxyphenyl)-4,5-dihydro-1H-imidazol-3-ium tetrafluoroborate, H2Sp5-BF4, that is subsequently used for the
construction of a novel MOF, Cu-Sp5-EtOH. This highly crystalline material has a charged framework that is expected to
give rise to high CO2/N2 selectivity. However, the pores of the parent structure could not be accessed due to the presence
of non-volatile ethanol molecules. After solvent exchange with methanol and subseuquently heating Cu-Sp5-MeOH under
vacuum, we are able to liberate the solvent providing other small molecules like CO2 access to the inside of the now
porous structure, Cu-Sp5. The combination of open-metal sites and framework charge leads to an exceptionally high
CO2/N2 selectivity, as determined by Ideal Adsorbed Solution Theory (IAST) calculations performed on single-component
adsorption isotherms. The CO2/N2 selectivity of Cu-Sp5 reaches a value of over 200 at pressures typically found in post-
combustion flue gas (0.15 bar CO2
/ 0.85 bar N2), a value that is among the highest reported to date.
adequate gas separation technology. This feat is not easy, as
the differences in the molecules of interest, such as CO2 and
N2, the main components in a post-combustion flue gas, are
minimal. As such, these separations require tailor-made
adsorbent materials with molecule-specific chemical
interactions on their internal surface.
Introduction
Anthropogenic climate change is
a well-recognized
problem.1 According to the U.S. National oceanic and
atmospheric administration, the concentration of CO2, the
main greenhouse pollutant, already exceeds 400 ppm and
continues to rise.2 The two largest CO2 sources include power
plants and internal combustion engines. Due to the expectedly
slow rate at which the world will transition to clean, renewable
energy sources, we anticipate the continued use of fossil fuels
for many years to come.3 Given this, an immediate solution to
reducing CO2 emissions is necessary, which will include capture
of CO2 and its further storage and/or conversion to value-
added chemicals or fuels.
Among a variety of materials suitable for CO2 separation
purposes, metal-organic frameworks (MOFs) have recently
emerged as promising candidates.5 These materials are
constructed by a combination of metal ions or clusters,
referred to as secondary building units (SBUs), which are
interlinked by organic ligands. MOFs offer high accessible
surface areas and an unmatched degree of chemical tunability.
By varying linker geometry and chemical functionality, as well
as metal cluster composition and coordination, one gains
access to thousands of new materials with the properties of
interest for a host of environmental applications. In addition to
facile chemical tunability, MOFs feature high crystallinity,
which allows one to gain knowledge of their structure-derived
function, insight necessary to tune existing materials or
designed new ones for a variety of applications like carbon
capture.
From an industrial point of view, the easiest capture
process to implement is post-combustion capture of CO2 from
flue gas.4 Existing power plants can be retrofitted and new
plants can be designed to effectively cut CO2 emissions,
rendering them carbon-neutral. However, the implementation
of carbon capture is currently constrained by the lack of an
Based on the knowledge of CO2 separation process in
MOFs, several structural features reportedly render materials
with high CO2/N2 selectivity. These methods, which are meant
to enhance the interactions between CO2 and the internal pore
surface, include decorating the MOF with open metal sites,
Lewis base sites (NH2, OH and derivatives), or other strongly
polarizing functional groups (e.g. -F, -CN, -NO2 functionalities,
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 2 of 10
View Article Online
DOI: 10.1039/C8DT01247D
ARTICLE
Journal Name
ion pairs etc.).5-7 Thus, we decided to combine two central atmosphere. NMR was used to watch the reaction proceed,
features for selective CO2 capture: presence of (i) open metal and more triethyl orthoformate was added every 18 h (0.5 eq.
sites and (ii) charged moieties inside the framework. For this each time). After complete conversion to the amidinium salt,
purpose, we designed a new ligand bearing a charged as observed by NMR, the mixture was cooled to room
saturated imidazolinium ring, 1,3-bis(4-carboxyphenyl)-4,5- temperature and acetonitrile was removed using a rotary
dihydro-1H-imidazol-3-ium tetrafluoroborate (H2Sp5-BF4) and evaporator. The remaining solid was suspended in diethyl
subsequently made a novel MOF, [Cu(Sp5)(C2H5OH)]NO3 (Cu- ether (approx. 15 mL solvent per 1 mmol diamine), sonicated
Sp5-EtOH),
and
its
solvent-exchanged
derivative
,
for 0.5 h and filtered over a G4 frit. The cake was further
[Cu(Sp5)(CH3OH)]NO3 (Cu-Sp5-MeOH). The new MOF features washed with 200 mL diethyl ether and finally with 200 mL n-
a combination of copper paddlewheels interlinked by cationic hexane. The product was obtained as a rose-colored powder
ligands, which creates a charged framework that is lined with and dried at room temperature in vacuum (0.1 Pa) for 1 day.
-
ligand-NO3 ion pairs. After the activation of Cu-Sp5-MeOH, in
Experimental data for H2Sp5-BF4. Diamine (8 g, 26.64
situ FTIR using CO as a probe, is employed to show the mmol), triethyl orthoformate (4.145 g, 27.97 mmol), glacial
presence of open-metal coordination sites in Cu-Sp5, which acetic acid (0.16 g, 2.664 mmol), ammonium tetrafluoroborate
are known to provide strong electrostatic interactions for (2.793 g, 26.64 mmol); isolated yield (rose-colored powder):
incoming CO2 molecules.8, 9 With combination of open metal 10.478 g (98.8 %). 1H NMR (400.13 MHz, DMSO-d6, 298 K): δ =
sites and the charge engineered into the framework wall by 13.18 (s, 2H), 10.20 (s, 1H), 8.12 (d, J = 8.4 Hz, 4H), 7.78 (d, J =
the imidazolinium ligand, the material is found to offer one of 8.4 Hz, 4H), 4.82 – 4.48 (m, 4H). 13C NMR (100.61 MHz, DMSO-
the highest CO2/N2 selectivities reported to date, as d6, 298 K): δ = 166.4, 152.9, 139.4, 130.8, 129.1, 118.4, 48.4.
determined by Ideal Adsorbed Solution Theory (IAST) 19F NMR (376.50 MHz, DMSO-d6, 298 K): δ = –148.3. 11B NMR
calculations.
(128.38 MHz, DMSO-d6, 298 K): δ = –1.29. IR (diamond ATR): ꢀȬ
= 415 (w), 454 (s), 475 (m), 512 (s), 522 (s), 546 (s), 634 (m),
689 (s), 771 (vs), 798 (s), 846 (vs), 892 (s), 1002 (vs), 1060 (vs),
1126 (s), 1189 (vs), 1251 (vs), 1275 (vs), 1343 (m), 1427 (vs),
1518 (m), 1581 (vs), 1633 (s), 1687 (vs), 2539 (m), 2826 (m). Td
(TGA) = 213.6/401.1 °C. HRMS (ESI) m/z: [M]+ calcd for
Experimental
Materials and general procedures
The reactions were carried out in air, unless stated
otherwise. Acetonitrile, methanol, ethanol, dichloromethane
and n-hexane were obtained in analytical grade from
commercially available sources and were used without any
further purification or drying. The reactions under water free
conditions were carried out using standard vacuum and
Schlenk techniques. For these reactions, acetonitrile was
distilled after refluxing with CaH2 overnight and stored in a
glove box under a nitrogen atmosphere. 4-aminobenzoic acid,
ammonium tetrafluoroborate and deuterated chloroform
(CDCl3) were obtained from Acros, triethyl orthoformate,
potassium carbonate (anhydrous), Cu(NO3)2·3H2O (99.5 %) and
deuterated dimethyl sulfoxide (dmso-d6) were obtained from
ABCR, sodium hydroxide, hydrochloric acid (37 %) and 1,2-
dibromoethane were obtained from Sigma-Aldrich, sulfuric
acid (95-97 %) and acetic acid (glacial) were obtained from
Merck and used as received without further purification.
+
C17H15N2O4 , 311.1026; found, 311.1025.
Cu-Sp5-EtOH and Cu-Sp5-MeOH synthesis
[Cu(Sp5)(C2H5OH)]NO3 (Cu-Sp5-EtOH). In
a
16 mL
scintillation vial, 24.2 mg of Cu(NO3)2·3H2O (0.1 mmol) and
39.8 mg of H2Sp5-BF4 (0.1 mmol) were mixed with 10 mL of
ethanol. After 10 minutes of sonification, the vial was placed in
the oven and heated to 80 °C at 10 °C/h rate. The mixture was
kept at 80 °C for 48 h and then cooled back to room
temperature at 10 °C/h rate. Large single crystals of Cu-Sp5-
EtOH were washed with copious amount of ethanol and dried
in air. Yield 47% (22.6 mg). By changing the heating and
cooling rate the crystal size can be tuned. Elemental analysis
(%) calc. for C19H20CuN3O8: C, 47.35; H, 4.18; N, 8.72. Found: C,
47.78; H, 4.07; N, 8.05.
[Cu(Sp5)(CH3OH)]NO3 (Cu-Sp5-MeOH)
.
In
a
typical
experiment, 100 mg of Cu-Sp5-EtOH crystals were placed in 20
mL of methanol and soaked for 2 days either at room
temperature or 50 ˚C. Solvent was exchanged to a fresh one
each 3-6 h. Elemental analysis (%) calc. for C18H18CuN3O8: C,
46.21; H, 3.88; N, 8.98. Found: C, 46.16; H, 4.03; N, 8.13.
Synthesis of 1,3-bis(4-carboxyphenyl)-4,5-dihydro-1H-imidazol-3-
ium tetrafluoroborate (H2Sp5-BF4)
The synthesis of the formamidine and diamine compound
on Scheme 1 was carried out using a recently reported
procedure by our group.10 For the synthesis of closed ring
ligand with a five-membered saturated ring (H2Sp5-BF4), the
diamine (1 eq.), triethyl orthoformate (1.05 eq.), glacial acetic
acid (0.1 eq.) and ammonium tetrafluoroborate (1 eq.) were
weighed in a round bottom Schlenk flask that was then
connected to a Schlenk line and evacuated at room
temperature in vacuum (0.1 Pa) for 1 h. Dry acetonitrile
(approx. 15 mL solvent per 1 mmol diamine) was added and
the mixture was refluxed at 120 °C for 3 days under nitrogen
Cu-Sp5-MeOH activation
Cu-Sp5-MeOH was heated in vacuum (< 50 μbar) at 120 °C
for 15 h. The activated sample, Cu-Sp5, was used to perform
the gas adsorption measurements. Elemental analysis (%) calc.
for C17H14CuN3O7: C, 46.85; H, 3.24; N, 9.64. Found: C, 46.40;
H, 3.04; N, 8.16.
X-Ray crystallography
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 3 of 10
View Article Online
DOI: 10.1039/C8DT01247D
Journal Name
ARTICLE
Laboratory single crystal X-ray diffraction was performed at
Thermogravimetric analysis (TGA) and decomposition
100 K using a Bruker D8 Venture diffractometer equipped with temperatures were determined from TGA experiments on a
graphite monochromator and microfocused Mo Kα radiation TGA Q500 device from TA Instruments in 100 μl platinum pans
source (λ = 0.71073 Å). The data was integrated with SAINT11 with an empty pan of the same type as reference. To infer the
and corrected for adsorption with SADABS software.12 decomposition temperatures, the onset points were taken into
Synchrotron single crystal X-ray diffraction data was collected account.
using a multipurpose PILATUS@SNBL diffractometer with Gas adsorption measurements
PILATUS2M detector at the BM01 at European Synchrotron
The CO2 sorption isotherms were obtained using BELSORP-
Radiation Facility (λ = 0.67522 Å). Single crystals were
max system equipped with dry ice-ethanol bath (195 K) or
mounted on a Kapton loop and measured at 293 K. The
temperature-controlled circulating water bath (278 K, 288 K,
structure solution and refinement were performed using
298 K). The nitrogen sorption isotherms were obtained using
ShelXT13 and ShelXL-201314 operated through the Olex2
Micromeritics 3FLEX instrument equipped with liquid nitrogen
interface.15 First, non-hydrogen atoms were located and
Dewar (77 K) or temperature-controlled water bath (278 K,
refined anisotropically, then H atoms were introduced with
288 K, 298 K).
corresponding HFIX commands and refined isotropically.
Synchrotron powder diffraction was performed at the BM02
beamline at ESRF (Grenoble, France) using monochromatic X-
Results and discussion
ray radiation (λ = 0.618375 Å) in transmission mode.
Ligand synthesis
Laboratory powder X-ray diffraction was performed using
To obtain the diamine compound to be converted into the
target ligand, shown in Scheme 1, we performed synthesis
using a recently reported procedure by our group.10 In the first
Bruker D8 Discover instrument equipped with Cu Kα radiation
source operated in Bragg-Brentano geometry. The scan step
was 0.02° and the time per step was 1 s.
step,
a formamidine was prepared using methyl 4-
aminobenzoate and triethyl orthoformate in the presence of a
small amount of glacial acetic acid (0.1 eq.). During the
reaction, the formed ethanol was removed by distillation
pushing the equilibrium forward in favor of the product
formation. The formamidine was isolated as pure solid (up to
20-25 g scales) and employed in step two without any further
purification. In this step, potassium carbonate (K2CO3), a mild
base, was used to deprotonate/activate the formamidine for
the subsequent substitution of the bromines in 1,2-
dibromoethane, and hence promote the C–N bond formation.
Subsequently, the resulting intermediate, formamide was
easily converted into the diamine-carboxylic acid building
block using a standard hydrolysis procedure with sodium
hydroxide. In the final step, the diamine was allowed to react
with triethyl orthoformate and ammonium tetrafluoroborate
in the presence of a small
FTIR measurements
Ex situ FTIR spectrum was recorded at room temperature
on a PerkinElmer FTIR/FIR spectrometer using a Diamond ATR
cell with the PerkinElmer spectrum software package (a
resolution of 4 cm–1 was used). In situ FTIR data were collected
using the same spectrometer. The sample for in situ FTIR
analysis was mixed with KBr and placed into a custom-built cell
with diffuse reflectance Praying Mantis™ accessory. The
sample was first activated by treating at 120 °C in vacuum for
15 h. Then the sample was cooled to -123 °C (150 K) and dosed
with selected CO pressures using a custom-built gas manifold.
The spectrum was collected at 2 cm-1 resolution. All spectra
were background-subtracted.
NMR measurements
Solution NMR spectra were recorded in deuterated
solvents on a BRUKER AVIII HD 400 spectrometer; data are
given in ppm relative to 1 % TMS solution in CDCl3 using the
solvent signals as a secondary reference (1H, 13C). NMR spectra
were recorded at room temperature, if not mentioned
otherwise. The Bruker Topspin software package (version 3.2)
was used for measuring and Mestrenova NMR software
(version 11.0.1) was used for processing of the spectra.
HRMS analysis
High resolution mass spectrometry (HRMS) analysis was
performed using Q Exactive HF Hybrid Quadrupole-Orbitrap
mass spectrometer (Thermo Scientific, Germany) in ESI
ionization mode with ionization source TriVersa NanoMate
(Advion, USA) and quadrupole mass analyzer.
TGA measurements
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 4 of 10
View Article Online
DOI: 10.1039/C8DT01247D
ARTICLE
Journal Name
may lead to double-NHC metal complexes or other interesting
adducts, respectively. Next to each imidazolinium ring, there
are nitrate counter anions with shortest distance of 3.21(1) Å
between the C2 carbon and the O atom of the nitrate. The
anions are well ordered in the structure and hence are readily
observed via structure solution, even at room temperature
(see ESI, Table S1, Figure S2).
Figure 1. Ball and stick model showing the structure of the copper paddlewheel
building unit in Cu-Sp5-EtOH. Atom colors are Cu: turquoise, O: red, N: blue, C:
gray, H: white.
Scheme 1 Synthetic route for the preparation of the closed ring amidinium ligand
with a five-membered saturated ring (H2Sp5-BF4).
amount of glacial acetic acid (0.1 eq.), giving the closed ring
ligand used in this work. All reactions, shown in Scheme 1,
provide products in excellent yields, 97 % for formamidine, 94
% for diamine and 99 % for closed ring ligand with a five-
membered saturated ring (H2Sp5-BF4). It should be noted that
this is the first saturated imidazolinium ligand with carboxylic
acid functionality reported to date in the literature.
Cu-Sp5-EtOH
Synthesis and crystal structure. The synthesis of Cu-Sp5-
EtOH proceeds smoothly under solvothermal conditions with
formation of large single crystals (> 200 μm, Figure S1). Single
crystal X-ray diffraction unveiled the structure of Cu-Sp5-EtOH
,
which is constructed by copper paddlewheel clusters where
carboxylate groups of the Sp5 ligand (Figure 1) are found in the
equatorial position of the cluster and ethanol molecules in
apical positions. The material crystallizes in an orthorhombic
space group, Pbca (№ 61), with unit cell axes of
a =17.705(2)
Å, = 9.252(1) Å, =24.566(2) Å and an overall unit cell
b
c
volume of 4023.9(6) Å3. The SBU and Sp5 ligand interlink in
such a way to form zigzag 2D layers with aperture sizes of 17.3
Å as indicated in Figure 2 (a), (b) and (d). In order to form the
final 3D structure of Cu-Sp5-EtOH, two sets of the 2D layers
are interpenetrated, as shown in Figure 2 (c), (e).
The final structure possesses triangular channels with
several noteworthy features that are highlighted in Figure 2 (f).
First, the C2 atoms of Sp5 ligands point directly towards each
other inside the channel with an interatomic C-C distance of
only 5.490(9) Å. This feature is of particular interest as
imidazolinium ligands readily undergo H+ extraction for the
formation of N-heterocyclic carbenes (NHC).16 With this close
distance, further metalation or modification of the NHC center
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 5 of 10
View Article Online
DOI: 10.1039/C8DT01247D
Journal Name
ARTICLE
0.08
0.06
0.04
0.02
0.00
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
19 20 21 22 23 24 25 26 27 28 29 30
Cu-Sp5-H2O calculated
Cu-Sp5-MeOH experimental
Cu-Sp5-EtOH experimental
Cu-Sp5-EtOH calculated
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30
2θ, ° at λ = 0.618375 Å
Figure 3. Calculated and experimental synchrotron powder patterns for Cu-Sp5
derivatives.
crystal data, as indicated in Table 1.
Thermal analysis and gas adsorption. The gas sorption
isotherms were measured for the sample treated in vacuum at
120 ˚C for 15 h. For nitrogen at 77 K and 278 K, as well as for
CO2 at 195 K and 278 K, there was no sorption detected. This is
no surprise as the space-filling model of Cu-Sp5-EtOH (Figure
S3) clearly indicated limited accessibility of the channels. It is
thought that the pores of the Cu-Sp5-EtOH are not available to
additional guest molecules due to the presence of ethanol
molecules that are coordinated to the Cu2+ ions. This is further
supported by thermogravimetric analysis (TGA) which
indicates no mass loss until 230 ˚C in air (Figure S4), while
afterwards the mass decreases continuously until the
completion of the sample decomposition. In a separate
experiment, we heated the Cu-Sp5-EtOH in an inert
atmosphere (nitrogen) to 230 ˚C and held it at that
temperature for 3 h. Again, there was a continuous mass loss
of more than 40 % without showing a step for the removal of
the ethanol (9.6 % expected as calculated from the molecular
formula of Cu-Sp5-EtOH), and the PXRD indicated the presence
of only metallic copper as residue (Figure S5, S6). Under air,
the final decomposition product was copper (II) oxide, as
determined by PXRD (Figure S7). Its mass (16.3 %) is in a
perfect agreement with the calculated CuO content in Cu-Sp5-
EtOH (16.5 wt. %).
Figure 2. Structural features of Cu-Sp5-EtOH. (a) Side view of a 2D layer formed by
copper paddlewheels and Sp5, (b) Top view of the 2D layer, (c) View along the b axis of
the Cu-Sp5-EtOH structure, where separate sets of layers are shown in red in blue, (d)
Schematic representation of the top view of the 2D layer, (e) Schematic representation
of view along b axis in Cu-Sp5-EtOH indicating interpenetration of 2 layer sets, (f) Close-
up view of a single channel in Cu-Sp5-EtOH, view along b axis. Hydrogen atoms are
omitted for clarity. Atom colors for (a), (b), (f) are Cu: turquoise, O: red, N: blue, C: gray.
In addition to C2 atoms, the terminal paddlewheel sites,
located below in Figure 2 (f), also point inside the channel.
They are occupied by ethanol molecules, whose C-C bonds are
Cu-Sp5-MeOH
Synthesis and crystal structure. Given our inability to
activate the sample containing ethanol, we tried to exchange
ethanol with a more volatile solvent. Upon soaking the
material in methanol at room temperature, the crystal size
decreased significantly (Figure S8), yielding a crystalline
powder with particle sizes that range from 10-20 μm (Figure
S9). However, soaking the starting material in methanol at 50
˚C produced crystals with a relatively larger size distribution,
which were more suitable for SCXRD.
aligned nearly parallel to the
b axis. The distance between two
closest Cu atoms from two different 2D layers is 8.519(1) Å.
The synchrotron powder diffraction pattern of a powder
sample of Cu-Sp5-EtOH matches perfectly with the calculated
pattern obtained from the single crystal structure (Figure 3).
This indicates high phase purity of the material. Moreover, the
Le Bail fit of the powder pattern obtained at 100 K shows good
agreement of lattice parameters and cell volume with the
single
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 6 of 10
View Article Online
DOI: 10.1039/C8DT01247D
ARTICLE
Journal Name
The new material crystallizes in an orthorhombic space displaces methanol. Further, Le Bail refinement carried out on
group, Pbca (№ 61), with unit cell axes of
a
=16.610(4) Å,
the methanol soaked Cu-Sp5-MeOH powder sample measured
at 100 K (Table 1, Figure 3) shows a
Table 1. Unit cell parameters determined from Le Bail fits (powder samples) and single
crystal XRD. All the datasets were obtained at 100 K
Sample
Cu-Sp5-
EtOH
Cu-Sp5-
Cu-Sp5-
Cu-Sp5-
,
EtOH, single H2O, single MeOH,
powder
Pbca
crystal
Pbca
crystal
Pbca
powder
Space
group
a, Å
Pbca
17.775(2) 17.705(2)
9.218(1) 9.2515(9)
16.610(4)
8.957(2)
25.681(6)
3821(2)
16.676(6)
8.953(3)
25.80(1)
3852(3)
b, Å
c, Å
24.621(3) 24.566 (2)
4034.0(8) 4023.9(6)
V, Å
b
= 8.957(2) Å, c =25.681(6) Å and an overall unit cell volume
of 3821(2) Å3. While the structural motif of this material is the
same as that of the parent structure, Cu-Sp5-EtOH, the
interpenetrated 2D layers have shifted so that the channel
shape, the unit cell dimensions, and unit cell volume have
changed significantly, as illustrated in Figure 4. After methanol
Figure 4. View of a channel in Cu-Sp5-H2O structure
slightly larger unit cell volume than the single crystal of Cu-
Sp5-H2O (see Figure 3 for powder pattern comparison), also
implying the presence of methanol rather than water in the
structure of Cu-Sp5-MeOH powder.
exchange, the lengths of
a and b axes decreased while the c
axis elongated. This leads to an overall decrease in unit cell
volume by more than 200 Å3, a change of approximately 5 %.
Importantly, the chemical environment in the channel
Thermal analysis, activation, X-ray diffraction and in situ
FTIR. The key point of methanol exchange with formation of
Cu-Sp5-MeOH was to obtain a material which can be thermally
activated to gain access to a structure that would be porous to
small molecules like CO2 and to impart the formation of open
metal coordination sites. The TGA of Cu-Sp5-MeOH (Figure
S10) clearly indicates a mass loss step of 7.8 % starting at 120
°C with further decomposition beginning at 230 °C. This value
is in agreement with the expected mass loss of the material if
one methanol is liberated from each copper atom, 6.9 %. Thus,
we used the TGA results to develop an activation protocol. Cu-
Sp5-MeOH was heated at 120 °C in vacuum for 15 h to prepare
the Cu-Sp5 for further gas sorption measurement.
changed as well. Since the channel shrinks laterally along the
a
axis, the interatomic distance between two adjacent C2 atoms
decreases to 5.30(3) Å, and the Cu-Cu distance between two
neighboring paddlewheel units decreases to 7.940(3) Å.
Compared to parent Cu-Sp5-EtOH, which has well-resolved
ethanol molecules at the Cu-coordination site, structural
analysis of the solvent exchanged material reveals a single Q
peak in close proximity to the Cu with a distance of 2.16(1) Å.
As such, this excess electron density was refined as oxygen.
Additionally, a second Q peak is found inside the pore, at the
distance of 2.64(2) Å from a Cu-bound oxygen atom. It was
also refined as oxygen with an occupancy of 0.84(4). This
distance is similar to that of the O-O distances observed for
hydrogen bonded water molecules. Due to this and the
absence of additional Q peaks that would indicate the methyl
groups of methanol, we attribute this excess electron density
to surface bound water. For the single crystal measurements,
the operation time needed to mount crystals and carry out
measurements, gives rise to a long air exposure. Since
methanol is highly volatile, it is thought that water displaces it,
leading thus to Cu-Sp5-H2O single crystal instead of Cu-Sp5-
MeOH. The high volatility of the methanol indicated that
activation was now a possibility. It should be noted that while
the single crystals show water, elemental analysis obtained
from a powder sample of Cu-Sp5-MeOH that was stored in the
sealed container under methanol shows a carbon content
which is in a perfect agreement with the Cu-Sp5-MeOH
composition (46.16 %, calc. 46.23 %). For Cu-Sp5-H2O on the
other hand, the calculated carbon content is significantly
lower, 43.53 % only. It indicates that methanol is likely
adsorbed inside the crystal structure upon solvent exchange
and it is only after prolonged air exposure when water
To prove the crystallinity of the material after activation,
we measured synchrotron powder diffraction data of the
activated sample under a nitrogen atmosphere. The data show
an interesting phenomenon of a structural change upon
activation. As can be seen in Figure 5, the powder pattern of
the activated sample features different peak positions and
lower number of peaks in general. The very first peak appears
at higher angle, 3.743° (d = 9.467 Å) vs 2.798° (d = 12.66 Å) for
solvated Cu-Sp5-MeOH
. This clearly indicates a phase
transition presumably due to a shift in the 2D layers following
solvent removal. Importantly, this process does not
significantly alter the crystal quality of Cu-Sp5, as high-angle
diffraction data indicates presence of Bragg reflections as high
as at 2θ = 28.8°. Moreover, upon resolvation of the activated
sample in methanol for 24 h, the MOF gains the initial
structure back (Figure 5), again without any apparent lose in
the crystallinity. While we are currently working on the
structure determination of the activated sample, we can
clearly deduce that the Cu-Sp5-MeOH is stable upon activation
and its structure is reversibly changed during that process.
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 7 of 10
View Article Online
DOI: 10.1039/C8DT01247D
Journal Name
ARTICLE
In order to further evidence the solvent removal from the
pore and from the copper paddlewheel terminal site, we
performed in situ FTIR measurements of the activated material
after dosing it with the CO gas (Figure 6). This technique is a
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.06
2.0
Resolvated in MeOH
1.8
Activated
0.04
1.6
Cu-Sp5-MeOH
1.4
19
20
21
22
23
24
25
2θ, °
26
27
28
29
30
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Resolvated in MeOH
Activated
2200
2175
2150
2125
2100
2075
Cu-Sp5-MeOH
Wavenumber, cm-1
Figure 6. In situ FTIR data in the C-O bond region for activated Cu-Sp5 dosed with
0 (black), 2 (red), 5 (blue), 10 (pink) and 20 (green) Torr CO. Measured at 150 K.
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30
2θ, ° at λ = 0.618375 Å
Figure 5. Synchrotron powder patterns for Cu-Sp5-MeOH, activated Cu-Sp5 and
second located in a channel intersection in ZSM-5 structure,
the associated Cu-bound CO frequencies were blue- and red-
shifted with values of 2157 and 2138 cm-1, respectively.24
resolvated Cu-Sp5-MeOH
common approach for sensing accessible metal sites, not only
in MOFs,17, 18 but also in a variety of other materials.19 During
measurement performed at 150 K, the signal of Cu-bound CO
appeared at 2135 cm-1. Compared to the vibration frequency
of the free CO (2143 cm-1), the signal is red-shifted upon CO
coordination to copper, a phenomenon that is due to classical,
Gas adsorption measurement
.
Upon the standard
measurement of nitrogen sorption at 77 K, the activated
sample Cu-Sp5 did not exhibit significant gas uptake (Figure 7
(a), bottom). However, when CO2 was used as a probe
molecule with the measurement performed at 195 K, rapid
uptake in the low-pressure region with further slow increase
up to a value of 2.38 mmol/g (9.5 wt.%) at 95 kPa (Figure 7 (a),
top) was observed. This indicates that the pores were indeed
accessible for gas adsorption upon solvent removal. The
adsorbed amount is in a good agreement with the calculated
CO2 loading if each copper site adsorbs one CO2 molecule (2.29
mmol/g, 9.2 wt.%).
π
-backbonding interactions. Interestingly, the reported
stretching frequencies of the Cu-bound CO for two structurally
related MOFs, Cu-BTC (HKUST-1)20-22 and Cu-TDPAT23, are blue
shifted with values ranging from 2178-2174 cm-1 and
unperturbed, respectively. Given that these two structures
have a similar Cu-paddlewheel SBU with open Cu2+ sites, this
indicates that the ligand field environment and local pore
structure strongly influence the interaction of Cu2+ with CO.
Similar phenomena has also been illustrated in Cu-ZSM-5
zeolite, which shows a large change in the CO stretching
frequency depending on local environment of the Cu2+ ion. In
particular, it was found that for two copper sites, the first
being inside an oxygen-lined five-membered ring window and
the
The surface area of the material, determined by Langmuir
model of CO2 adsorption at 195 K, is 204 m2/g. However, use
of CO2 as a probe molecule for surface area measurement
could be associated with underestimating the surface area of
the material because of strong CO2-CO2 interactions in the
pore.4, 25, 26. Therefore, the obtained surface area value should
be treated as a lower estimate.
With a high preference of CO2 over N2 shown by the
material at low temperatures, we were eager to investigate
the gas sorption properties of Cu-Sp5 in the ambient
conditions, which are close to relevant conditions for industrial
flue gas. The data were obtained at 3 temperatures, namely
298 K, 288 K and 278 K to examine thermal dependency of the
isotherms and calculate isosteric heats of adsorption of both
gases. As shown in Figure 7 (b), the nitrogen uptake in all cases
is much lower than those of CO2, proving that the material is
selective for CO2 in a range of temperatures close to ambient.
The CO2 uptake at 298 K reaches 1.30 mmol/g (5.41 wt.%),
whereas at 278 K it increases up to 1.65 mmol/g (6.77 wt.%) at
1 bar. The zero-coverage isosteric heat of CO2 adsorption, -Qst,
for Cu-Sp5, calculated via the Clausius–Clapeyron equation
(see the ESI), is 43.1 kJ/mol. Interestingly, it stays nearly
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 8 of 10
View Article Online
DOI: 10.1039/C8DT01247D
ARTICLE
Journal Name
constant during the adsorption as indicated in Figure 8. This selectivity of the MOF in a CO2/N2 mixture, we performed Ideal
indicates the unceasing strong framework interaction with CO2 Adsorbed Solution Theory (IAST) calculations using the single
during the adsorption process. The isosteric heat of nitrogen component isotherms. Based on the calculated data obtained
adsorption at zero coverage, as calculated via similar via pyIAST software27 (see ESI), the selectivity factor for CO2
procedures, is 20.2 kJ/mol (Figure S11). This vast difference over nitrogen in a 15 % CO2/85 % N2 mixture at 1 bar is 207.1
between the isosteric heats of CO2 and N2 clearly shows the at 298 K and 252.5 at 278 K. This value is superior to the
thermodynamic preference of CO2 interaction with the MOF selectivity values of many other reported MOFs,5 exceeding
interior. The trends, in turn, indicate a clear and stable ones bearing open metal sites only, even the top performers
thermodynamic CO2/N2 selectivity for Cu-Sp5 in a wide range such as Mg-MOF-74 (IAST selectivity = 175 at 313 K), which
of loadings and pressures. The observed isosteric heat of CO2 features high densities of open Mg2+ sites (Table S2).28
adsorption corresponds to upper limit of those values
According to this, the striking selectivity of Cu-Sp5 towards
previously reported for MOFs featuring open metal sites as a CO2 could not be explained by presence of open metal sites
main feature for CO2 capture. The highest reported to date is only. As a matter of fact, the other species present in Cu-Sp5
47 kJ/mol observed for Mg-MOF-74 (Table S2). If compared to pores are ion pairs of charged imidazolinium rings of Sp5 and
two other structurally related MOFs that have Cu- nitrate anions. It turned out that limited nitrogen uptake both
paddlewheels as SBUs including Cu-BTC and Cu-TDPAT the at ambient temperature and 77 K was reported for several
observed values are25.9 kJ/mol and 42.2, respectively. MOFs bearing charged moieties.29-32 Since both polarizability
Further, in order to quantitatively investigate the adsorption and quadrupole moment for nitrogen molecule are lower than
(a)
(b)
2.6
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
CO2, 195K
CO2
278K
288K
298K
N2, 77K
N2
0
10
20
30
40
50
60
70
80
90
100
0
10
20
30
40
50
60
70
80
90
100
Absolute pressure, kPa
Absolute pressure, kPa
Figure 7. Gas adsorption isotherms for activated Cu-Sp5. Filled symbols represent adsorption, open symbols represent desorption. (a) Low temperature isotherms. (b)
ambient temperature isotherms
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 9 of 10
View Article Online
DOI: 10.1039/C8DT01247D
Journal Name
ARTICLE
Sp5 becomes decorated with both open Cu2+ sites and ion
pairs, giving rise to an exceptional CO2/N2 IAST selectivity that
reaches 252.5 at 278 K at 0.15 bar CO2 and 0.85 bar N2. This
study indicates that engineering charge into frameworks
combined with other interesting functionalities such as open
metal sites can significantly enhance the performance of
materials towards important environmental challenges like
post-combustion CO2 capture.
45
40
35
30
25
20
Conflicts of interest
There are no conflicts to declare.
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
Loading, mmol/g
Acknowledgements
Part of this work was supported by the Swiss National Science
Foundation under grant number PYAPP2_160581. M.A. was
supported by the Swiss Commission for Technology and
Innovation (CTI). We acknowledge the Swiss-Norwegian Beam
Line BM01 and D2am French CRG Beamline BM02 at European
Synchrotron Radiation Facility for beamtime allocation, Dr.
Dmitry Chernyshov and Dr. Iurii Dovgaliuk for the assistance on
beamline BM01, Dr. Nils Blanc and Dr. Nathalie Boudet for the
assistance on beamline BM02 and Dr. Pascal Schouwink for aid
in single crystal X-ray diffraction at the EPFL.
Figure 8. Isosteric heat of CO2 adsorption for Cu-Sp5
.
for CO2 (Table S3), such conditions lead to very small N2
sorption values, which in combination with reasonable CO2
uptake gives rise to good selectivities.
Thus, Cu-Sp5 combines presence of both open metal sites
and ion pairs inside the channel which together participate in
rendering high CO2 selectivity to the material. Compared to
MOFs bearing open metal sites only, the selectivity of the Cu-
Sp5 is much higher, whereas MOFs with only ion pairs in the
pore present lower isosteric heat values (Table S2) and lower
selectivities. Only one reported MOF with the charged
moieties features a very high zero coverage isosteric heat of
adsorption (58.1 kJ/mol), which is also caused by presence of
open metal sites, uncoordinated nitrogen atoms of tetrazolate
and fluorine atoms.30 However, while the low-pressure CO2
uptake and isosteric heat for this material were very high, the
latter quickly dropped below 40 kJ/mol after 0.75 mmol/g, and
even below 30 kJ/mol after 1.0 mmol/g. In total, this gave rise
to a limited IAST selectivity for 10 % CO2/90 % N2 gas mixture
at 298 K which was only 15 at 1 bar, while Cu-Sp5 presents
almost constant high isosteric heat (43.1 kJ/mol) in the whole
loading range and IAST selectivity 15 % CO2/85 % N2 mixture at
1 bar is 207.1.
Notes and references
1.
IPCC Special Report on Carbon Dioxide Capture and
Storage, Cambridge University Press, Cambridge, United
Kingdom and NewYork, NY, USA, 2005.
2.
Trends in Atmospheric Carbon Dioxide. ESRL Global
Monitoring Division - Global Greenhouse Gas Reference
Network, https://www.esrl.noaa.gov/gmd/ccgg/trends/,
(accessed 15.03.2018).
3.
4.
BP Statistical Review of World Energy, June 2017,
https://www.bp.com/en/global/corporate/energy-
economics/statistical-review-of-world-energy.html,
(accessed 15.03.2018).
B. Smit, J. A. Reimer, C. M. Oldenburg and I. C. Bourg,
Introduction to Carbon Capture and Sequestration,
Imperial College Press, 2014.
This shows that there is a clear synergy of ion pairs and
open metal sites inside the channel, which drastically
enhances the CO2 selectivity towards nitrogen in relevant
conditions. For other materials, showing only one of those
features, either isosteric heats or selectivity values are low.
Therefore, the Cu-Sp5 material uniquely features the
combination of both properties giving rise to combination
outstanding IAST CO2/N2 selectivity and high isosteric heat of
adsorption, introduced by ligand design.
5.
6.
7.
Z. Zhang, Z.-Z. Yao, S. Xiang and B. Chen, Energy &
Environmental Science, 2014, 7.
J. Liu, P. K. Thallapally, B. P. McGrail, D. R. Brown and J.
Liu, Chem Soc Rev, 2012, 41, 2308-2322.
C. A. Trickett, A. Helal, B. A. Al-Maythalony, Z. H. Yamani,
K. E. Cordova and O. M. Yaghi, Nature Reviews Materials,
2017, 2.
8.
9.
W. L. Queen, C. M. Brown, D. K. Britt, P. Zajdel, M. R.
Hudson and O. M. Yaghi, The Journal of Physical Chemistry
C, 2011, 115, 24915-24919.
W. L. Queen, M. R. Hudson, E. D. Bloch, J. A. Mason, M. I.
Gonzalez, J. S. Lee, D. Gygi, J. D. Howe, K. Lee, T. A.
Darwish, M. James, V. K. Peterson, S. J. Teat, B. Smit, J. B.
Conclusions
A new charged imidazolinium ligand, H2Sp5-BF4, gives rise
to a novel metal-organic framework, Cu-Sp5. The activated
framework features triangular channels with imidazolinium
rings interlinked by Cu-paddlewheel SBUs. Upon solvent
exchange and thermal activation, the channel interior of Cu-
Neaton, J. R. Long and C. M. Brown, Chem. Sci., 2014, 5,
4569-4581.
10.
S. Bulut and W. L. Queen, J Org Chem, 2018, DOI:
10.1021/acs.joc.8b00151.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Please do not adjust margins
Dalton Transactions
Page 10 of 10
View Article Online
DOI: 10.1039/C8DT01247D
ARTICLE
11.
Journal Name
Bruker (2012). SAINT. Bruker AXS Inc., Madison,
Wisconsin, USA.
12.
Bruker (2001). SADABS. Bruker AXS Inc., Madison,
Wisconsin, USA.
13.
G. M. Sheldrick, Acta Crystallogr A Found Adv, 2015, 71, 3-
8.
14.
G. M. Sheldrick, Acta Crystallogr C Struct Chem, 2015, 71,
3-8.
15.
O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard
and H. Puschmann, Journal of Applied Crystallography,
2009, 42, 339-341.
16.
17.
M. Fevre, J. Pinaud, A. Leteneur, Y. Gnanou, J. Vignolle, D.
Taton, K. Miqueu and J. M. Sotiropoulos, J Am Chem Soc,
2012, 134, 6776-6784.
E. D. Bloch, M. R. Hudson, J. A. Mason, S. Chavan, V.
Crocellà, J. D. Howe, K. Lee, A. L. Dzubak, W. L. Queen, J.
M. Zadrozny, S. J. Geier, L. C. Lin, L. Gagliardi, B. Smit, J. B.
Neaton, S. Bordiga, C. M. Brown and J. R. Long, Journal of
the American Chemical Society, 2014, 136, 10752-10761.
L. Peng, M. Asgari, P. Mieville, P. Schouwink, S. Bulut, D. T.
Sun, Z. Zhou, P. Pattison, W. van Beek and W. L. Queen,
18.
ACS Appl Mater Interfaces, 2017, 9, 23957-23966.
19.
20.
M. J. Kale and P. Christopher, ACS Catalysis, 2016, 6,
5599-5609.
C. Prestipino, L. Regli, J. G. Vitillo, F. Bonino, A. Damin, C.
Lamberti, A. Zecchina, P. L. Solari, K. O. Kongshaug and S.
Bordiga, Chemistry of Materials, 2006, 18, 1337-1346.
J. Szanyi, M. Daturi, G. Clet, D. R. Baer and C. H. Peden,
Phys Chem Chem Phys, 2012, 14, 4383-4390.
N. Drenchev, E. Ivanova, M. Mihaylov and K. Hadjiivanov,
Phys Chem Chem Phys, 2010, 12, 6423-6427.
C. Y. Wang, P. Ray, Q. Gong, Y. Zhao, J. Li and A. D.
Lueking, Phys Chem Chem Phys, 2015, 17, 26766-26776.
21.
22.
23.
24.
25.
26.
27.
28.
29.
I. C. Hwang, D. H. Kim and S. I. Woo, Catal Lett, 1996, 42
,
177-184.
H.-J. Lee, H. Kwon, J. Sim, D. Song, Y. Kim, J. Kim, K. Kim
and E. Lee, CrystEngComm, 2017, 19, 1528-1534.
K. C. Kim, T.-U. Yoon and Y.-S. Bae, Microporous and
Mesoporous Materials, 2016, 224, 294-301.
C. M. Simon, B. Smit and M. Haranczyk, Computer Physics
Communications, 2016, 200, 364-380.
J. A. Mason, K. Sumida, Z. R. Herm, R. Krishna and J. R.
Long, Energy & Environmental Science, 2011,
4.
J. Y. Lee, J. M. Roberts, O. K. Farha, A. A. Sarjeant, K. A.
Scheidt and J. T. Hupp, Inorganic Chemistry, 2009, 48
,
9971-9973.
30.
D. X. Xue, A. J. Cairns, Y. Belmabkhout, L. Wojtas, Y. Liu,
M. H. Alkordi and M. Eddaoudi, J Am Chem Soc, 2013,
135, 7660-7667.
31.
32.
L. Kong, R. Zou, W. Bi, R. Zhong, W. Mu, J. Liu, R. P. S. Han
and R. Zou, J. Mater. Chem. A, 2014, 2, 17771-17778.
S. Sen, S. Neogi, A. Aijaz, Q. Xu and P. K. Bharadwaj,
Inorganic Chemistry, 2014, 53, 7591-7598.
10 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins