Microporous Organic Polyimides for CO2 and H2O Capture and Separation from CH4 and N2 Mixtures: Interplay between Porosity and Chemical Function

Article abstract of DOI:10.1021/acs.chemmater.6b01949

Porous polyimides have been considered to be a promising material class for gas capture and sequestration, leading to the synthesis of a substantial number of individual networks with noteworthy sorption properties. In spite of these efforts, the vision of a chemical control of adsorption and desorption of small molecules, in particular, for the competing uptake of technical relevant gas mixtures, is still hardly investigated. Here, we present a systematic study of five new polyimide networks based on a set of linkers with chemical functionalities covering the full range from hydrophobic to hydrophilic interactions. The corresponding microporous organic polyimides (MOPI-I to -V) were synthesized successfully based on a condensation reaction between amino and anhydride linker molecules in m-cresol at high temperatures, resulting in cross-linking degrees beyond 95% in all cases. Argon and carbon dioxide isotherms reveal surface areas up to 940 m²/g with ultramicroporosity, about 50% microporosity and high thermal stabilities under air with decomposition temperatures up to 480 °C. Sorption screening for variable temperatures revealed remarkable uptakes for carbon dioxide up to 3.8 mmol/g and water vapor up to 19.5 mmol/g combined with a smooth gate opening around 0.25 p/p₀ for MOPI-IV. In contrast, for MOPI-V the water vapor uptake decreases down to 7 mmol/g. Interestingly, the trend of the selectivities calculated by IAST and Henry does not correlate with the uptake behavior. For instance, MOPI-I and MOPI-III exhibit with 78 and 13 the highest CO₂ over N₂ and CH₄ Henry selectivities, although their CO₂ uptake is around 3.0 mmol/g. In total, we attribute the sorption properties for this class of materials mainly to the void size and shape within the ultramicroporous region. The chemical environment of the surfaces seems to have little influence on the uptake and a stronger effect on the separation behavior.

Full text of DOI:10.1021/acs.chemmater.6b01949

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES
Article
2
2
Microporous Organic Polyimides for CO and HO Capture and Separation
4
2
from CH and N Mixtures - Interplay between Porosity and Chemical Function
Christoph Klumpen, Marion Breunig, Thomas Homburg, Norbert Stock, and Jürgen Senker
Chem. Mater., Just Accepted Manuscript • DOI: 10.1021/acs.chemmater.6b01949 • Publication Date (Web): 11 Jul 2016
Downloaded from http://pubs.acs.org on July 17, 2016
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 free 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 accessible to all
readers and 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.
Chemistry of Materials 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 12
Chemistry of Materials
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
Microporous Organic Polyimides for CO
2
and H O Capture and Sepa-
2
ration from CH
4
and N
2
Mixtures – Interplay between Porosity and
Chemical Function
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
†
†
††
††
†*
Christoph Klumpen , Marion Breunig , Thomas Homburg , Norbert Stock , Juergen Senker
†
University of Bayreuth, Inorganic Chemistry III, Universitaetsstr. 30, 95440 Bayreuth, Germany
†
†
Christian-Albrechts-University Kiel, Institute for Inorganic Chemistry, Max-Eyth-Str. 2, 24118 Kiel, Germany
Porous polyimides have been considered to be a promising material class for gas capture and sequestration, leading to the
synthesis of a substantial number of individual networks with noteworthy sorption properties. In spite of these efforts, the
vision of a chemical control of adsorption and desorption of small molecules, in particular, for the competing uptake of
technical relevant gas mixtures, is still hardly investigated. Here we present a systematic study of five new polyimide net-
works based on a set of linkers with chemical functionalities covering the full range from hydrophobic to hydrophilic inter-
actions. The corresponding microporous organic polyimides (MOPI-I to -V) were synthesized successfully based on a con-
densation reaction between amino and anhydride linker molecules in m-cresol at high temperatures, resulting in crosslink-
ing degrees beyond 95 % in all cases. Argon and carbon dioxide isotherms reveal surface areas up to 940 m /g with a ultra-
microporosity, about 50 % microporosity and high thermal stabilities under air with decomposition temperatures up to 480
°C. Sorption screening for variable temperatures revealed remarkable uptakes for carbon dioxide up to 3.8 mmol/g and
water vapor up to 19.5 mmol/g combined with a smooth gate opening around 0.25 p/p for MOPI-IV. In contrast, for MOPI-
0
V the water vapor uptake decreases down to 7 mmol/g. Interestingly, the trend of the selectivities calculated by IAST and
Henry does not correlate with the uptake behavior. For instance, MOPI-I and MOPI-III exhibit with 78 and 13 the highest
2 2 4 2
CO over N and CH Henry selectivities, although their CO uptake is around 3.0 mmol/g. In total, we attribute the sorption
properties for this class of materials mainly to the void size and shape within the ultramicroporous region. The chemical
environment of the surfaces seems to have little influence on the uptake and a stronger effect on the separation behavior.
2
Introduction
natural gas sweetening and the purification of gas mixtures
2
To satisfy the energy requirements of modern society, the
combustion of fossil fuels is still one of the most important
from gasification or water/gas-shift reactor processes. The
first one deals predominantly with CO
2
impurities (> 40%),
1
energy sources nowadays. This implicates the emission of
whereas in the second CO /H separation is of major inter-
2
2
large amounts of carbon dioxide rich gas mixtures, increas-
ing the atmospheric carbon dioxide concentration and in
turn intensifying global warming. According to the World
Meteorological Organization (WMO), CO
portant anthropogenic greenhouse gas, contributing 65 %
to radiative forcing. However, the atmospheric CO level
still rises now reaching 145 % of the pre industrial status.
Until environmentally more friendly techniques are devel-
est. In both cases, elevated temperatures above 40 °C and
high pressures around 30 bar have to be taken into ac-
2
2,4
count. Natural gas and precombustional gas for coal fire
2
is the most im-
plants are usually contaminated with 5-20 % and 40-60 %
2 2
of CO , respectively. CO as main contaminant is not only
2
an initiator for global warming, but has also corroding ef-
fects on the transporting pipeline systems due to its acid-
3
ity. Additionally, captured CO
for the synthesis of various chemical compounds. This
makes the CO capture and sequestration important for
both environmental and industrial interests. Considering
the gigantic amount of annual CO release, any solution
2
could be used as reactant
6
2
oped, the capture and sequestration (CCS) of CO is im-
portant to decrease the global carbon dioxide emission. In
general, CCS techniques address two main fields, precom-
bustion and postcombustion processes, which provide dif-
ferent demands on potential materials.
The typical conditions for a postcombustional gas stream
are nearly ambient pressures (~1 bar) and temperatures
2
2
4
has to be based on reversible processes.
Since 1930 CCS systems were predominantly based on wet
amine solutions (e.g. MEA), which use chemisorptive in-
teractions with carbon dioxide as filtration mechanism.
4
7
around 40-80 °C for typical coal-fired power plants. The
high volume streams consist of 15-16 % CO
2
, 70-75 % N
2
, 5-
and
Precombustion is usually further subdivided into
Due to severe drawbacks, like high cost regeneration, in-
7
% water vapor and other impurities like CO, N
x
O
x
stabilities towards impurities and equipment corrosion,
2
,5
4,8–10
SO
x
.
this method is regarded to be highly inefficient.
Some
1
ACS Paragon Plus Environment

Chemistry of Materials
Page 2 of 12
of the main drawbacks are a substantial parasitic load of
CO capture about 25-30 % and a high energy demand for
the regeneration step which mounts to 25 – 40 % of the
chosen to favor a more or less linear expansion of the net-
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
2
work. Their individual electronic properties, additionally,
allow to vary the chemical nature of the void surfaces in a
systematic fashion among other things from hydrophilic to
hydrophobic. After a thorough characterization of the pol-
2,4,9,11
power plants energy output.
A more elegant solution
could be provided by solid-state materials with high sur-
face areas, which interact with the adsorptives via phy-
sisorption and, in succession, the advantage to facilitate
their regeneration by changing physical parameters to un-
load incorporated molecules like in pressure swing adsorp-
13
15
ymers by IR-, C and N solid-state NMR spectroscopy,
CHN and TG analysis as well as X-ray diffraction, the influ-
ence of surface areas, pore structures and different side
groups on their gas sorption behavior will be discussed. In
4
tion (PSA) or vacuum swing adsorption (VSA) techniques.
addition, the gas uptake and selectivity for N
2 2
, CO and
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
Thus the possibility to modify both adsorbent properties
and process designs strongly influences the effectiveness of
CH , as well as the water vapor uptake into the networks is
4
considered. All these properties are discussed in the con-
text of the chemical nature of the inner surfaces.
1
0
a separation method. Additionally, new materials should
provide further properties like low cost synthesis, recycla-
bility and high selectivity values.
Experimental Section
General Information
1
1–13
In this respect microporous organic polymers
(MOPs)
exhibit outstanding physicochemical properties combined
with high surface areas as well as chemically modifiable
structures and are thus in line with other porous solid ma-
Argon and nitrogen sorption measurements were carried
out on a Quantachrome Autosorb-1 pore analyzer at 87 K
and 298 K, respectively. The data were analyzed using the
ASIQ v 3.0 software package. Pore size distributions were
calculated with methods of quasi stationary density func-
tional theory (QSDFT). The isosteric heat of adsorption
1
4–17
18–21
terials like carbons , zeolites
frameworks
and metal organic
. In fact, many publications on MOPs have
proven their high potential and established strategies for
enhancing the CO uptake like surface area modification,
pore size engineering or isosteric heat of adsorption tun-
2
2–24
2
was calculated using the CO
298 and 313 K. CO and CH
2
adsorption isotherms at 273,
4
adsorption isotherms were
2
4
,12,13,25–27
ing.
Simulations have shown, that CO
2
acting both
measured on a Quantachrome Nova surface analyzer at
273, 298 and 313 K, respectively. Selectivities and adsorp-
tion parameter were calculated using IAST and Henry
as Lewis acid and Lewis base can bind to phenyl rings and
1
3
polar functionalities inside of a porous network. Thus the
incorporation of such moieties is a promising start to tune
3
8,39
methods according to the literature.
specific surface area, pore volumes and pore size distribu-
tions measured with CO at 273 K were carried out using
Calculations of the
2
the CO sorption behavior.
Applying these strategies led to the synthesis of promising
materials like BILP-4, PPN-6-deta and ALP-1 with the re-
2
the nonlocal density functional theory (NLDFT) slit-pore
model for carbon materials. For the Ar based isotherms,
the choice of the QSDFT adsorption branch kernel for cy-
lindrical pores in carbon based materials depended on the
calculated fitting error. Water sorption isotherms were
measured using a BEL JAPAN INC. Belsorpmax instru-
ment. To avoid extremely long measuring times the de-
sorption branch was aborted when the pressure equilibra-
tion exceeded 24 h. Thus for two networks (MOPI-I and
MOPI-IV) the water isotherms are not fully closed. All pol-
markable CO
2
uptake values of 5.3, 4.3 and 5.4 mmol/g at
273 K and 1 bar ,respectively, which are among the best po-
9
,28,29
rous polymer networks up to date.
croporous polimides like PI-NO -1 uptakes up to 4.0
mmol/g at 273 K and 1 bar were reached. Furthermore,
Focusing on mi-
2
30
31
excellent selectivities of CO
2
over N
, e.g. PI-NO -2 (21) or NPI-
2
, like in TPI-2@IC
32
30
(151) or STPI-2 (107), and CH
4
2
33
1 (13), have been reported.
Similar or even better sorption properties were obtained
for microporous, crystalline materials like metal-organic-
-
2
ymers have been degassed under vacuum (10 kPa) at 110
°C for 12 to 16 h before starting the adsorption experiments.
frameworks (MOFs). E.g. mmen-Mg
2
(dobpdc) adsorbs up
1
13
to 6.4 mmol/g at 298 K and 1 bar in combination with a
H and C liquid-NMR spectra of the linker molecules were
carried out on a Bruker 500 MHz spectrometer using either
CDCl or DMSO as solvents. For infrared spectra (IR) a
3
4,35
CO
late framework UTSA-49a shows a CO
4.8 mmol/g at 273 K and 1 bar in combination with a sub-
2
over N
2
selectivity over 96 %.
The zeolitic imidazo-
2
uptake capacity of
3
JASCO FT/IR-6100 fourier transform infrared spectrometer
with attenuated total reflectance (ATR) unit was used.
CHN analysis was carried out at a vario EL-III with
acetanilide as standard (Meas.: C [71.09], H [6.71], N [10.36];
Calc.: C [71.09], H [6.71], N [10.36]). Mass spectrometry
36
2 4
stantial CO over CH selectivity around 35. Nevertheless,
MOFs often suffer from a limited chemical and thermal
stability underlining the competitiveness of MOPs.
Herein we report the synthesis of five new microporous
polyimides (MOPIs) via a polycondensation polymeriza-
tion process of amine and anhydride linker molecules.
While the synthesis procedure is based on our previous
(MS) was performed at
a
Finnigan MAT-8500-
spectrometer with an ionization energy of 70 eV. Thermo-
gravimetric analysis (TGA) was carried out on a Mettler To-
37
e
work on TPIs, here we focus on 3D networks with a cen-
tral tetrahedral core based on tetrakis(4-aminophenyl)me-
thane for all five MOPIs. The corresponding anhydrides are
ledo TGA/SDTA851 . Powder X-ray diffraction (PXRD)
measurements were performed on a PANalytical X`Pert
Pro diffractometer. Here, a region from 2-30° 2ϴ was meas-
ured with a ¼ anti-scatter slit and Cu K
α
-radiation (nickel
2
ACS Paragon Plus Environment

Page 3 of 12
Chemistry of Materials
filtered). All solid-state NMR spectra were acquired on a
Bruker Avance-III HD spectrometer operating at a B field
14 ml of fuming nitric acid in a flask equipped with a mag-
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
netic stirrer, a thermometer and a gas funnel were cooled
down to -10 °C, followed by the portioned addition of 2.5 g
13
15
of 9.4 T. C (δ = 100.6 MHz) and N (δ = 40.6 MHz) MAS
spectra were obtained with ramped cross-polarization (CP)
experiments where the nutation frequency νnut on the pro-
ton channel was varied linearly from 70 – 100 %. The sam-
-3
tetraphenylmethane (7.8·10 mol), 9 ml acetic acid and 4
ml acetic anhydride. The mixture was stirred at -10 °C for 1
h. The resulting yellow solid was filtered, washed with ace-
tic acid and water and dried in an oven at 100 °C. Yield: 2.5
13
15
ples were spun at 12.5 kHz ( C) and 10.0 kHz ( N) in a 4
-
3
1
mm MAS double resonance probe (Bruker). The corre-
g (4.9·10 mol, 63 %). H-NMR (500 MHz, CDCl
3
): δ [ppm]
= 8.2 (d, 2H, Ar−H), 7.3 (d, 2H, Ar−H). C-NMR (500 MHz,
CDCl ): δ [ppm] = 150.8 (C-2), 141.0 (C-5), 136.3 (C-3), 117.9
13
13
sponding νnut on the C channel and the contact time were
1
5
N
adjusted to 70 kHz and 3.0 ms, respectively. On the
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
channel the corresponding νnut and the contact time were
adjusted to 35 kHz and 5.0 ms, respectively. Proton broad-
band decoupling with spinal-64 and nnut = 12.5 kHz was ap-
(C-4), 66.3 (C-1). EA [%]: C [59.67], H [3.35], N [10.67]; Calc.:
C [60.00], H [3.22], N [11.20].MS [M/z] = 500. 278, 332, 302,
-
1
285, 239, 213, 163. IR (ATR): ν [cm ] = 3095, 1590, 1510, 1407,
1339, 1222, 1182, 1106, 1011, 962, 917, 836, 746, 706.
4
0 13
plied during acquisition.
C spectra are referenced with
respect to TMS (tetramethylsilane) using the secondary
1
5
43
standard adamantane. N spectra are referenced with re-
Tetrakis(4-aminophenyl)methane
-
3
spect to CH
3
NO
2
using the secondary standard glycine.
0.75 g of tetrakis(4-nitrophenyl)methane (1.5·10 mol, 1 eq.)
-
3
2 4 2
were suspended in 50 ml THF. 1 g of N H ·H O (20·10 mol,
Synthesis
13 eq.) and ~5 g of Raney nickel were added and the mixture
was refluxed for 3 h. After hot filtration and washing with
THF the solvent was evaporated and the residue washed
All chemicals were purchased at Sigma-Aldrich Chemistry
GmbH, VWR Chemicals, TCI or Grüssing GmbH and were,
if not mentioned otherwise, used without further purifica-
-
4
with DCM to give a white solid. Yield: 0.358 g (9·10 mol,
1
tion (Table S1). m-Cresol was freshly distilled over CaH
2
63 %). H-NMR (500 MHz, CDCl
H), 6.5 (d, 1H, Ar-H), 3.5 (s, 2H, NH
CDCl ): δ [ppm] = 146.1 (C-5), 136.3 (C-2), 131.6 (C-3), 113.1
C-4), 61.5 (C-1). EA [%]: C [77.31], H [6.70], N [14.32]; Calc.:
3
): δ [ppm] =6.9 (d, 1H, Ar-
1
3
and stored under Ar over a molecular sieve M4 prior to use.
All polymerizations were carried out under Ar atmosphere.
2
). C-NMR (500 MHz,
3
(
Synthesis of the linker
C [78.92], H [6.36], N [14.73]. MS [M/z] = 380, 288, 271, 195,
4
1
- 1
Tetraphenylmethane
5 g of trityl chloride (0.054 mol, 1 eq.) and 14.05 ml aniline
0.154 mol, 2.9 eq.) were heated up to 180 °C in a round
180. IR (ATR): ν [cm ] = 3391, 3162, 1609, 1501, 1425, 1268,
1
(
1177, 1191, 1101, 1011, 955, 859, 809.
flask with magnetic stirrer and condenser, until the reac-
tion mixture turned into a violet solid. The heating process
was extended for 10 more minutes. The solid was cooled
down, crushed and suspended in 75 ml MeOH and 75 ml 2
M HCl. The suspension was refluxed for 30 min., filtered
and washed with water. After suspending in ethanol the
reaction mixture was cooled down to -30 °C and 15.75 ml
sulfuric acid and 9.44 g of isopentylnitrite (0.081 mol, 1.5
eq.) were added under vigorous stirring. After stirring for 1
h at -10 °C, 26.9 ml of phosphinic acid (0.609 mol, 11 eq.)
were added slowly and the reaction mixture was refluxed
for 1.5 h. After cooling down, the solid was filtered, washed
General polymerization procedure
Based on a former publication in the field of porous polyi-
mides, all polymerization reactions were carried out under
argon atmosphere in dry three-necked flasks equipped
37
with a magnetic stirrer and a thermometer. For all poly-
mers, m-cresol was used as solvent. The molar ratio of
tetrakis(4-aminophenyl)methane to the corresponding an-
hydride was set to 1:2 to provide equimolar amounts of
amine and anhydride groups. The following, more detailed
synthesis of MOPI-I is given as reference protocol for all
polymerization reactions.
with DMF, H
2
O and Ethanol and subsequently dried in
Synthesis of MOPI-I
-
3
vacuo to get a light brown powder. Yield: 16.1 g (0.05 mol,
68.9 mg tetrakis(4-aminophenyl)methane (0.181·10 mol, 1
eq.) and 130 mg 3,3‘,4,4‘-diphenylsulfonetetracarboxyldi-
1
93 %). H-NMR (500 MHz, CDCl
3
): δ [ppm] = 7.07-7.20 (m,
1
3
-3
5H, Ar-H). C-NMR (500 MHz, CDCl ): δ [ppm] = 146.8 (C-
3
anhydride (0.362·10 mol, 2 eq.) were added to 20 ml of m-
2), 131.2 (C-3), 127.5 (C-4), 125.9 (C-5), 65.0 (C-1). EA [%]: C
[91.73], H [6.91], N [0.24]; Calc.: C [93.71], H [6.29], N [0.00].
cresol under Ar atmosphere. The mixture was stirred for 24
h at 0 °C. Afterwards, a catalytic amount (≈ 1 mg) of iso-
quinoline was added and the mixture was again stirred for
12 h at room temperature. The polymerization reaction was
carried out at 80 °C for 4 h, 120 °C for 4 h, 160 °C for 6 h,
190 °C for 18 h. After cooling down to room temperature,
-1
MS [M/z] = 320, 243, 165. IR (ATR): ν [cm ] = 3050, 1589,
1496, 1434, 1319, 1268, 1185, 1080, 1029, 987, 935, 899, 842,
749, 697, 635.
4
2
Tetrakis(4-nitrophenyl)methane
3
ACS Paragon Plus Environment

Chemistry of Materials
Page 4 of 12
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
Scheme 1: General polymerization reaction and simplified structures of the presented MOPI networks.
the resulting solid was filtered and washed with toluene,
dichloromethane and methanol. Additional purification
was achieved in a soxhlet apparatus with tetrahydrofuran
over night. After drying in an oven at 100 °C, MOPI-I was
obtained as brown powder. Yield: 56 mg (0.055 mmol, 28
= 1780, 1710, 1615, 1497, 1411, 1365, 1294, 1255, 1208, 1161, 1082,
1019, 972, 918, 854, 815, 714.
Synthesis of MOPI-IV
118 mg of tetrakis(4-aminophenyl)methane (0.311·10 mol,
1 eq.) and 0.154 mg of bicyclo[2,2,2]oct-7-en-2,3,5,6-tetra-
-
3
1
3
%
(
[
). C-NMR (CP-MAS, 12.5 kHz): δ [ppm] = 166 (C-6), 147
-
3
C-2, C-9), 132 (C-3, C-7), 124 (C-4), 115 (C-8), 64 (C-1). EA
%]: C [61.12], H [3.66], N [5.11]; Calc.: C [67.18], H [2.72], N
carboxylicdianhydride (0.621·10 mol, 2 eq.) were used as
reactants. MOPI-IV was obtained as brown solid. Yield: 132
-
1
-3
13
[5.40]. IR (ATR): ν [cm ] = 1784, 1721, 1603, 1507, 1420, 1367,
1317, 1223, 1184, 1150, 1094, 1057, 1023, 916, 820, 740.
mg (0.164·10 mol, 53 %). C-NMR (CP-MAS, 12.5 kHz): δ
[ppm] = 177 (C-6), 147 (C-2, C-9), 131 (C-3, C-5), 127 (C-4),
65 (C-1), 42 (C-7), 35 (C-8). EA [%]: C [69.61], H [4.20], N
-
1
Synthesis of MOPI-II
81.9 mg tetrakis(4-aminophenyl)methane (0.215·10 mol, 1
eq.) and 134 mg 4,4‘-oxydiphthalicdianhydride (0.43·10
mol, 2 eq.) were used as reactants. MOPI-II was obtained
as light brown powder. Yield: 170 mg (0.183·10 mol, 85 %).
[6.26]; Calc.: C [73.52], H [3.95], N [6.86]. IR(ATR): ν [cm ]
= 1782, 1714, 1663, 1679, 1495, 1444, 1368, 1334, 1241, 1182, 1123,
1022, 979, 819, 751.
-3
-3
-3
Synthesis of MOPI-V
1
3
-3
C-NMR (CP-MAS, 12.5 kHz): δ [ppm] = 166 (C-6), 159 (C-
80 mg of tetrakis(4-aminophenyl)methane (0.209·10 mol,
9), 146 (C-2), 134 (C-5, C-7), 130 (C-3), 126 (C-4), 117 (C-8`),
108 (C-8), 64 (C-1). EA [%]: C [70.45], H [3.43], N [5.72];
Calc.: C [74.04], H [3.00], N [5.95]. IR(ATR): ν [cm ] = 1780,
1717, 1603, 1507, 1474, 1430, 1360, 1273, 1234, 1080, 1020, 956,
806, 744.
1 eq.) and 186 mg 4,4‘-(hexafluoroisopropyliden)diphthalic
-
3
anhydride (0.418·10 mol, 2 eq.). MOPI-V was obtained as
-
1
-3
13
brown solid. Yield: 167 mg (0.140·10 mol, 67 %). C-NMR
(CP-MAS, 12.5 kHz): δ [ppm] = 166 (C-6), 146 (C-2), 139 (C-
9), 132 (C-7, C-5), 124 (C-3, C-8), 117 (C-11), 64 (C-1), 61 (C-
10). EA [%]: C [62.02], H [2.93], N [4.65]; Calc.: C [63.59], H
-
1
Synthesis of MOPI-III
[2.33], N [4.63]. IR(ATR): ν [cm ] = 1785, 1729, 1617, 1503,
1435, 1366, 1298, 1252, 1206, 1145, 1092, 1023, 986, 916, 886,
856, 818, 741, 711.
-
3
100 mg of tetrakis(4-aminophenyl)methane (0.262·10 mol,
1 eq.) and 169 mg of 3,3‘,4,4‘-benzophenontetracarboxylic-
-
3
dianhydride (0.525·10 mol, 2 eq.) were used as reactants.
MOPI-III was obtained as ocher colored solid. Yield: 168
Results and discussion
Synthesis
-3
13
mg (0.176·10 mol, 67 %). C-NMR (CP-MAS, 12.5 kHz): δ
[ppm] = 193 (C-10), 166 (C-6), 144 (C-2, C-9), 134 (C-5, C-3),
131 (C-4), 124 (C-8), 64 (C-1).EA [%]: C [69.59], H [3.48], N
Tetrakis(4-aminophenyl)methane was synthesized by a
well established procedure, starting with the coupling of
-
1
41
[5.29]; Calc.: C [74.59], H [2.92], N [5.81]. IR(ATR): ν [cm ]
triphenylchloromethane and aniline. After nitration of
4
ACS Paragon Plus Environment

Page 5 of 12
Chemistry of Materials
the resulting tetraphenylmethane the nitro group was re-
duced to tetrakis(4-aminophenyl)methane using Raney
CO stretching vibration for the aryl ether was found at 1273
cm (Figure 2, Figure S9). In the case of MOPI-III the α-
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
42,43
nickel and hydrazine in an overall yield of 37 %.
For the
keto-carbon signal (10) appears at 194 ppm and the aryl ke-
-1
polymerization reaction, tetrakis(4-nitrophenyl)methane
and the correspondent anhydride were used in a molar ra-
tone band was observed at 1211 cm (Figure S11). For MOPI-
IV the aliphatic ß-carbon atom (7) next to the imide group
is assigned to 42 ppm (Figure 1). The β-carbon atoms of the
bicyclooctene group (8) resonate at 35 ppm. In the IR spec-
tra the C=C stretching vibration of the bicyclooctene unit
37
tio of 1:2, following a procedure established by Liebl et al.
The choice of m-cresol as solvent is essential due to its high
boiling point (202 °C), which turned out to be optimal for
37
-1
this kind of polymerization reactions. After dissolving the
is prominent at 1663 cm (Figure S13). The structure of
reactants (Scheme 1) isoquinoline was added as catalyst
MOPI-V is furthermore validated by a signal at 61 ppm, as-
5
,44
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
due to its basic and hygroscopic properties. The reaction
was carried out using a five step temperature program,
starting at room temperature, followed by several heating
periods including 80 °C for 4 h, 120 °C for 4 h, 160 °C for 6 h
and 190 °C for 18 h. This induces a slow polymerization pro-
cess and simultaneously circumvent an early precipitation
signed to the quaternary carbon atoms (10) next to the CF
moieties (11) expected around 113 ppm (Figure 1). Due to
3
1
3
19
strong C F dipole interactions the signal is broadened.
The corresponding -CF stretching vibration was observed
at 1255 and 1191 cm (Figure S15).
3
-
1
37,45
of oligomers.
At the final heating step the precipitation
of the polymer was observed. All polymers were obtained
with good yields between 50 and 85 %. Only for MOPI-I
the yield was considerably lower (28 %).
Characterization
The resulting polymers were characterized by combining
solid-state NMR and IR spectroscopy as well as elemental
analysis. Further information about physicochemical prop-
erties were derived from PXRD (Figure S16) and TG exper-
iments (Figure S17) carried out in air and All five MOPIs
share a tetrakis(phenyl)methane core and imide linkages
as a result of the crosslinking. The most characteristic
1
3
15
peaks in the C and N MAS spectra (Figure 1, Table S2) for
both units are 64 ppm assigned to the tetrahedral carbon
atoms (1) as well as 165 ppm for the imide carbon atoms
(6). The corresponding imide nitrogen atoms resonate at -
1
5
209 ppm in the N NMR spectra. Furthermore, the absence
1
5
of the N amine resonance around -315 ppm (Figure 1) as
well as of the two characteristic anhydride bands around
-
1
-1
2
1830 cm and 1750 cm and of the NH stretching vibration
-
1
around 3500 cm (Figure S4) ensure an almost complete
crosslinking with an exception being MOPI-V, where a
1
8
small amount of unreacted amine end groups remains.
This is in agreement with the results of the CHN analysis
which show only small deviations (< 1 % for H, N and < 5
% for C) from the calculated data (Table 1). In the case of
MOPI-IV the resonances for both the imide carbons (6)
and nitrogens are low field shifted to 177 ppm and -188 ppm
due to the aliphatic character of the corresponding core of
the linker (Figure 1, Table S2). The imide functional groups
were also observed in the infrared spectra of the polymers
-
1
at 1778 cm where the typical imide carbonyl band of cyclic
-1
five membered rings resonates at 740 cm , which is due to
N-H wagging (Figure S4).
1
3
Further distinct but individual C NMR and IR spectro-
scopic features are observed in the case of MOPI-I at 145
ppm, typical for the α-carbon of a sulfone group (9) (Figure
1). The IR-bands of the sulfone group appear at 1177, 1320
1
3
15
Figure 1: Solid-state C (a) and N (b) MAS NMR spectra of
MOPI I-V measured with cross polarization and spinning
speeds of 12.5 kHz and 16 kHz, respectively. Assignment cor-
relates with Scheme 1. The asterisks mark spinning side bands.
-
1
and 1058 cm (Figure S7). For MOPI-II the α-carbon of the
ether group (9) resonates at 159 ppm. The corresponding
5
ACS Paragon Plus Environment

Chemistry of Materials
Page 6 of 12
Table 1: CHN-analysis (wt%) of MOPIs (upper values) and decomposition temperatures measured under air
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
(
lower values) (SInfo. 3.2, Figure S16).
CHN-Analysis
MOPI-I [%]
61.12 (66.12)
3.66 (2.75)
5.11 (6.26)
[°C]
MOPI-II [%]
70.45 (73.71)
3.43 (3.94)
5.72 (6.03)
[°C]
MOPI-III [%]
69.59 (74.37)
3.48 (2.97)
5.29 (5.88)
[°C]
MOPI-IV [%]
69.61 (73.14)
4.20 (4.01)
6.26 (6.96)
[°C]
MOPI-V [%]
62.02 (63.22)
2.92 (2.36)
4.65 (5.62)
[°C]
a
C (Exp./Calc.)
H (Exp./Calc.)
N (Exp./Calc.)
a
a
b
TGA-Analysis
Decomposition
421
440
390
350
475
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
a
b
Repeating units used to derive calculated weight percentages are given in SInfo 3.2. Full decomposition curves are shown in Figure S17 .
TG measurements carried out in air reveal first indications
for structural decomposition between 350 °C for MOPI-IV
and 475 °C for MOPI-V reflecting the thermal persistence
of the core units (Table 1, Figure S16). For MOPI-I addi-
tionally, a weight loss of about 10% was observed at 160 °C
due to desorption of water. PXRD measurements as well as
SEM micrographs confirm the amorphous nature of all pol-
ymers (Figure S15/S35).
at 273 K were measured (Figure 3, SInfo. 4.2) which allows
to detect pores between 0.3 and 1.5 nm.
Surface area and porosity
The porosity of all MOPIs was determined from Ar adsorp-
tion isotherms measured at 87 K (Figure 2, SInfo 4.1). Ar, as
2
monoatomic gas, is in comparison to the often used N en-
gaged in fewer interactions with surfaces, especially of po-
lar compounds and due to its higher boiling point, equilib-
rium processes are significantly shortened. The isotherms
show high Ar uptakes at low relative pressures, typical for
microporous materials. The shapes of the isotherms resem-
ble IUPAC Type I(b) isotherms, indicating broader pore
size distributions including wider micropores and narrow
46
mesopores. For all polymers a hysteresis was observed,
which is generally associated with capillary condensation,
4
6
a phenomenon linked with mesoporosity. The extension
of the hysteresis into the low pressure region, nevertheless,
favors swelling of the networks.
MOPI-I shows, the lowest BET and DFT surface areas
2
around 220 m /g, which is probably due to the high flexi-
bility of the sulfone bridge leading to a more compact
packing of the network. The highest surface area was ob-
2
served for MOPI-V with roughly 930 m /g. For the other
polyimides specific surface areas in between (480 and 730
2
m /g) are found.
For MOPI-I and –III the ratio Vmic/Vtot is around 30 %, in-
dicating a large percentage of mesopores, which correlates
with their lower surface areas in comparison with the other
polyimides (Table 2). This in line with the pore size distri-
butions given in Figure 2 revealing a broad distribution of
mesopores up to 6 nm. MOPI-II and MOPI-V developed
higher values of microporosity around 50 % with only
small mesopores around 2 nm (Figure 2). The highest mi-
croporosity is observed for MOPI-IV (68 %) (Table 2). All
five polyimide networks exhibit a narrow pore size distri-
bution in the microporous region with a center of gravity
between 1.0 nm for MOPI-II and 1.4 nm for MOPI-I (Fig-
Figure 2: Ar isotherms measured at 87 K (a) and normalized
pore size distributions (b) calculated by QSDFT adsorption
branch kernel for cylindrical pores in carbon based materials.
Full symbols characterize adsorption isotherms, hollow sym-
bols the corresponding desorption curve.
2
ure 2). To probe potential ultramicropores CO isotherms
6
ACS Paragon Plus Environment

Page 7 of 12
Chemistry of Materials
Table 2: Surface areas determined from Ar and CO
volumes calculated by DFT methods from Ar and CO
2
isotherms and calculated by BET and DFT methods. Pore
isotherms.
1
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
a
b
c
Polymer
Ar/87.3 K
Ar/87.3 K
[m /g]
CO
[m /g]
2
/273 K
V
tot.,Ar,DFT
V
mic.,Ar,DFT
V
mic./Vtot.
V
tot.,CO2,DFT
[cm /g]
2
2
2
3
3
3
[
m /g]
206
644
443
[cm /g]
[cm /g]
0.05
0.17
Ar, DFT
MOPI-I
MOPI-II
MOPI-III
MOPI-IV
MOPI-V
231
739
468
735
777
0.16
0.31
0.24
676
0.32
0.53
0.19
712
0.27
0.09
0.19
0.33
0.22
660
896
0.28
0.68
0.27
921
939
690
0.44
0.23
0.52
0.23
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
a
b
Calculated based on the BET equation. Calculated using QSDFT for cylindrical pores in carbon based materials for the adsorption branch
only. Calculated based on NLDFT with a slit-pore model for carbon materials.
c
The pore size distributions for all polymers (Figure S18),
calculated from the CO isotherms show a broad distribu-
measurements at 273, 298 and 313 K, respectively, were per-
formed to compare the uptake behavior of the polymers at
1 bar.
2
tion of pore sizes between 0.3 to 1 nm, which indicates an
undirected pore formation due to the amorphous charac-
ter of the networks. For MOPI-I, -III,-V the surface areas
2
derived from the CO isotherms are markedly higher com-
2
pared to the Ar surfaces, in particular for MOPI-I (777 m /g
2
vs. 231 m /g), demonstrating a high degree of ultrami-
cropores (Table 2). In contrast, the surface areas of MOPI-
2
2
II und MOPI-V decrease towards 676 m /g and 690 m /g
2
in comparison to Ar surface areas of 739 m /g and 939
2
m /g, respectively. These values indicate a high amount of
pore sizes beyond 1.5 nm for those networks (Table 2). In
conclusion, MOPI-I, MOPI-III and MOPI-IV possess a
higher ratio of ultramicropores than their related poly-
mers, which is discussed in more detail together with the
sorption behavior towards CO
2 4
and CH .
CO , CH and H O adsorption
2
4
2
The high surface areas and microporous structures render
all MOPIs as suitable candidates for applications such as
gas storage and separation. As mentioned before, the in-
teraction of those materials with industrial relevant gas
mixtures is of vital relevance. In particular, industrial flue
gas mixtures contain high amounts of water (up to 7 %),
which is a potential competitor for binding sites in filtering
materials and would decrease the potential of an appropri-
ate material. In return super hydrophilic materials could
be favorable for alternative applications like water ab-
sorber or heat transformation. In this context, the uptake
of water vapor at low pressures in materials like covalent
organic frameworks and porous carbons for the application
in atmospheric water capture (AWC), heat pumps or elec-
tric dehumidifiers has gained some interest in the litera-
47–49
ture.
Recently Lotsch et al. synthesized the AB-COF for
AWC application with a remarkable water uptake of 15
mmol/g at low relative pressure (p/p < 0.3), which shows
0
the high potential of porous polymer networks in this field
49
of application.
With this in mind the behavior of the synthesized poly-
mers towards CO
tures as well as their selectivity towards specific gas mix-
tures has been analyzed (Tables. 3 and 4). CO adsorption
2
, CH
4
, N
2
and H
2
O at various tempera-
Figure 3: CO
tion for CO
2
uptake at 273 K (a) and isosteric heats of adsorp-
(b) calculated from adsorption isotherms at 273,
2
298 and 313 K. Full symbols characterize adsorption isotherms,
hollow symbols the corresponding desorption curve.
2
7
ACS Paragon Plus Environment

Chemistry of Materials
Page 8 of 12
Table 3: Uptakes taken from individual isotherms. The values for CO and CH were determined at p = 1 bar and
2 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
the ones for N
2
and H
CO
2
O were extracted at p/p
CO
CO
0
= 0.9.
N
298 K
2
CH
273 K
CH
298 K
CH
313 K
2
H O
2
2
2
4
4
4
Polymer
273 K
mmol/g]
3.3
298 K
[mmol/g]
2.2
313K
298 K
[mmol/g]
13.5
[
[mmol/g] [mmol/g] [mmol/g] [mmol/g] [mmol/g]
MOPI-I
MOPI-II
MOPI-III
MOPI-IV
MOPI-V
1.7
1.4
1.5
1.6
1.3
0.11
0.09
0.07
0.10
0.04
0.61
0.61
0.72
0.83
0.49
0.34
0.35
0.19
0.4
0.25
0.24
0.2
2.9
3.0
3.8
2.9
1.9
1.8
2.3
1.6
12.3
9.0
19.5
7.0
0.3
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
0.5
0.17
At 273 K MOPI-I, -II, -III and -V show CO
2
uptakes around
Water adsorption measurements at 298 K provide infor-
mation about the behavior of the polymers in humid envi-
ronments (Table 3). MOPI-IV shows the highest water up-
3
mmol/g which are already extraordinary high compared
5
,30,32,33,50–56
to literature values.
2
The best CO uptake, how-
ever, was obtained for MOPI-IV with 3.8 mmol/g at 273 K,
which is actually the second best value for microporous
take with 6.3 mmol/g at 0.3 p/p
tal amount of 19.5 mmol/g at 0.9 p/p
published ultrahydrophilic covalent organic frameworks
0
, which rises towards a to-
0
(Figure 4). Recently
polyimides next to the previous mentioned PI-NO
2
-1 and
4
9
48
exceeds all other polyimide networks where the uptake is
like AB-COF (22.9 mmol/g), TpPA-1 (24.5 mmol/g) or
3
0–32,52
47
not higher than 3.5 mmol/g.
erature, we attribute the CO
In accordance to the lit-
uptake to the ultrami-
porous carbons like PCC-1 (~22 mmol/g) give comparable
2
results. This is surprising since for MOPI-IV the only polar
groups are the imide linkages and it is thus considered to
be less polar than e.g. MOPI-I with its sulfone bridges,
4
2
croporosity of the networks, for which we use the CO sur-
face areas as indicator. MOPI-IV has by far the largest ul-
tramicroporous surface area, followed by MOPI-I, -III, -V
reaching only 13.5 mmol/g at 0.9 p/p
Thus the high water vapor uptake of MOPI-IV is more
likely to correlate with its high CO surface area which we
take as a measure for the amount of ultramicropores (Table
2). For MOPI-I, -II, -III and –V the CO surface areas are
0
.
and –II. This is in line with the CO
2
uptake given in Table
3
and explains e.g. why MOPI-V and -II exhibits lower CO
2
2
uptakes in spite of the comparably large microporous sur-
faces (determined with Ar). As expected, a decrease of the
2
CO
2
uptake is observed with increasing temperature. Sim-
similar. Here the uptake values of 13.5, 12.3, 9.0 and 7
mmol/g correlate with the polarity of the bridging groups.
Additionally, unlike MOPI-IV the other networks do not
ilar uptakes of around 1.5 mmol/g are found for all MOPIs
at 313K (Table 3).
The isosteric heats of adsorption calculated from the CO
2
0
show an enhanced water uptake below 0.3 p/p .
isotherms lead to values around 30 kJ/mol, which are typi-
cal for strong physisorptive interactions between guest
molecules and the network (Figure 3). Here we observe the
lowest Qst near zero coverage (13 kJ/mol) for MOPI-V,
which is expected to have the weakest interactions with
CO
value (39 kJ/mol) is observed for MOPI-I, which is caused
by the stronger interactions of the CO molecule with the
2 3
due to the presence of CF groups. The highest Qst
2
highly polar sulfone groups. The values for MOPI-II, -III
and -IV were found to be similar around 32 kJ/mol (Table
4
). It is worth noting, that the trend for Qst near zero cov-
erage does not reflect the one for the CO uptake. This sug-
2
gests that the uptake at higher pressures is rather domi-
nated by unselective interactions and not by preferred
binding sites within the ultramicropores. The CH uptake,
4
determined from methane adsorption isotherms measured
at 273, 298 and 313 K (SInfo. 4.3), is remarkably low for all
polymers (around 0.5 to 0.7 mmol/g at 273 K) and de-
creases strongly with increasing temperature, caused by
only weak interactions between the unpolar methane mol-
ecules and the polar network structures.
Figure 4: Water vapor adsorption isotherm measured at 298
K. Full symbols characterize adsorption isotherms, hollow
symbols the corresponding desorption curve.
8
ACS Paragon Plus Environment

Page 9 of 12
Chemistry of Materials
1
Table 4: CO
2
/N
2
, CO
2
/CH
4
and CO
2
/H
2
O selectivities of all MOPIs calculated from the correspondent isotherms
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
and CO isosteric heats of adsorption.
2
a,(b)
c/d,(e)
4
c/d,(e)
c/d,(e)
e
f
Sample
CO
2
/N
2
CO
2
/CH
273 K
CO
2
/CH
298 K
4
CO
2
/CH
313 K
4
H
2
O/CO
298 K
92
2
Q
st
298 K
[kJ/mol]
39
MOPI-I
65.3 (78)
36.4 (51)
50.2 (50)
43.4 (52)
26.4 (33)
16/10 (11)
18/9 (8)
18/12 (13)
17/11 (10)
9/7 (7)
14/11 (9)
9/9 (9)
10/9 (7)
10/12 (8)
12/11 (10)
11/9 (10)
4/4 (5)
6
6
3
3
32
32
31
MOPI-II
MOPI-III
MOPI-IV
MOPI-V
14/11 (10)
15/12 (9)
10/9 (9)
1
05
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
36
13
a
b
c
Calculated by IAST for a mixture of 15 % CO
by IAST from a mixture of 5 % CO
2
and 85 % N
and 95 % CH
2
. Calculated by initial slopes of pure-component sorption isotherms. Calculated
d
e
2
4
. Calculated by IAST from a mixture of 15 % CO
2
and 85 % CH
4
. Calculated by initial
f
2
slopes of pure-component sorption isotherms. Calculated from CO isotherms at 273, 298 and 313 K.
The introduction of CF
duce the water uptake roughly by a factor of 2 and 3 com-
pared to MOPI-I and MOPI-IV, while the CO uptake is
reduced only by 15 to 30 %, respectively. This trend was al-
3
units in MOPI-V allowed to re-
CO
ties were obtained for MOPI-II and MOPI-V leading to a
further decrease of the CO /N selectivities. For this pair
the CF moieties within MOPI-V provide a much lower at-
traction for CO (Table 4) compared to the ether units of
MOPI-II, again accounting for the lower CO /N selectiv-
ity of 26 in the case of MOPI-V. As consequence, in con-
trast to the CO uptakes, the selectivities seem to be influ-
2
adsorption. Even lower but similar ultramicroporosi-
2
2
2
3
5
7
ready reported for FPOP-2 by Han et al. which exhibits
uptakes as low as 1.5 mmol/g. In the case of MOPI-V the
hydrophobicity of the CF groups is probably counterbal-
3
anced by the imide linkages leading to significant higher
values.
2
2
2
2
enced by both the ultramicroporosity and the functionality
of the networks in equal measures. Thus at least to a cer-
tain degree preferred binding sites are important to influ-
ence the separation properties.
2 4 2 2
Selectivities towards CO , CH , N and H O
Since flue gas is a complex mixture of diverse gases, poten-
tial filter materials should possess selective uptake procliv-
ity towards specific molecules to ensure their effectiveness
and efficiency. To estimate selectivities for the here pre-
sented networks, we applied ideal adsorbed solution the-
ory (IAST) for the following industrial relevant gas mix-
2 4
The IAST selectivities for the corresponding CO /CH mix-
tures, reveal similar trends, indicating that the same influ-
encing effects discussed above are relevant. For MOPI-III,
-IV, -I, -II and –V values of 12, 11, 10, 9 and 7 were found in
order to their ultramicroporosity and decreasing affinity of
their functional groups for CO
ering the portion of CO towards a ratio of 5:95 results in
increasing selectivities for all networks with top values of
18 for MOPI-II and -III as well as towards 17 and 16 for
MOPI-IV and -I, respectively. For MOPI-V only a minor
tures CO
2
/N
2
(15/85), CO
2
/CH
4
(5/95 and 15/85) in compar-
2
(Table. 4, SInfo. 5.2). Low-
ison to initial slope calculations in the Henrys law region
2
38,39
(< 0.1 bar) (SInfo. 5.1).
2
The highest CO over N IAST selectivities calculated at
2
298 K, are 65.3 and 50.2 obtained for MOPI-I and MOPI-
III, respectively (Table 4). Especially, MOPI-I competes
CO
4). The increase of the temperature leads towards signifi-
cantly lowered CO affinities towards 12, 11, 10, 10 and 4 for
MOPI-III, -IV, -I, -II and -V, respectively, for a 5:95
CO /CH mixture. At higher CO ratios (15:85) the temper-
2
affinity improvement from 7 to 9 was observed (Table
with materials in the top area of CO
2 2
/N selectivities, only
4
surpassed by networks like modified PPN-6 with values
2
5
8
from 150 to 420 or IRMOF-1-4Li with a selectivity of 395.
We attribute these high values to their pronounced ultra-
microporosity and the presence of keto/sulfone and imide
groups, which provide Lewis acidic and basic preferred ad-
2
4
2
ature seems to have only a minor effect on the selectivities.
Interestingly, in the case of MOPI-II even an increase from
9 to 12 was found hinting at a cooperative effect between
2
sorption sites for CO . For MOPI-IV, MOPI-II and MOPI-
V a further decrease of the selectivity to 43.4, 36.4 and 26.4
is observed, respectively.
CO
2
and CH
4
within the network (Table 4).
O/CO selectivities by initial slopes of
The calculation of H
2
2
This trend does not correlate solely to the CO
thus the ultramicroporosity (CO surface areas). E.g.
MOPI-IV exhibits the by far largest CO surface area lead-
ing to a significant carbon dioxide uptake. Nevertheless, its
CO /N selectivity is with 43 markedly lower compared to
2
uptake and
pure-component sorption isotherms, obtained high values
of 105 and 92 for MOPI-IV and –I, respectively, indicating
a major effect of the ultramicroporosity with respect to the
chemical nature of the polymers. For MOPI-II and -III a
lower selectivity of 63 was observed. Due to its hydropho-
2
2
2
2
MOPI-I (65) and -III (50). In contrast, the diene function-
ality of the corresponding linker within MOPI-IV is less
polar compared to the sulfone and ketone groups for
MOPI-I and -III and thus does not increase the affinity for
pic -CF
36 (Table 4).
3
moieties the selectivity for MOPI-V decreases to
9
ACS Paragon Plus Environment

Chemistry of Materials
Page 10 of 12
Conclusion
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
The condensation polymerization of amino- and anhy-
dride-linker molecules in m-cresol, lead to five new mi-
croporous organic polyimides MOPI-I to -V. Compared to
the recent literature, the here presented compounds are
promising materials for gas capture and sequestration, as
References
(1) Haszeldine, R. S. Carbon Capture and Storage: How
Green Can Black Be? Science 2009, 325, 1647–1652.
(
2)
D’Alessandro, D. M.; Smit, B.; Long, J. R. Carbon
Dioxide Capture: Prospects for New Materials. Angew.
Chemie Int. Ed. 2010, 49, 6058–6082.
9,28,29-32,55-57
well as potential water vapor adsorber.
All polymers show high crosslinking degrees, confirmed by
13
15
(3)
World Meteorological Organization. The State of
Greenhouse Gases in the Atmosphere Based on Global
Observations through 2013. 2014, No. 10, September
C-, N MAS NMR spectroscopy and elemental analysis.
The polyimides exhibit moderate to high specific surface
2
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
areas between 231 m /g (MOPI-I) and 957 m /g (MOPI-V)
and a mainly micro- to ultramicroporous character. Their
porosity in combination with different functional groups
2
014, 1–8.
(
(
4)
5)
Dawson, R.; Cooper, A. I.; Adams, D. J. Chemical
Functionalization Strategies for Carbon Dioxide
Capture in Microporous Organic Polymers. Polym.
Int. 2013, 62, 345–352.
Li, G.; Wang, Z. Microporous Polyimides with
Uniform Pores for Adsorption and Separation of CO 2
Gas and Organic Vapors. Macromolecules 2013, 46,
3058–3066.
leads to remarkable CO
73 K and 1 bar in the case of MOPI-IV, which is among
the highest values obtained for polyimides only surpassed
2
uptake values up to 3.8 mmol/g at
2
3
0
by PI-NO
able IAST selectivity values towards N
I) and CH (18 for MOPI-II, -III) and can compete with
most polyimides (CO
2
-1 (4.03 mmol/g). The polymers exhibit respect-
2
(up to 65 for MOPI-
4
3
0,31,33,52
2
/N
2
< 150 and CO
2
/CH
4
< 20 ).
(6)
Grande, C. A.; Ribeiro, R. P. L.; Oliveira, E. L. G.;
Rodrigues, A. E. Electric Swing Adsorption as
Emerging CO2 Capture Technique. Energy Procedia
2009, 1, 1219–1225.
In addition, water sorption isotherms showed a surpris-
ingly hydrophilic character for MOPI-IV, with uptakes of
6
0 0
.3 mmol/g at 0.3 p/p and 19.5 mmol/g at 0.9 p/p . For
(7)
(8)
Rochelle, G. T. Amine Scrubbing for CO2 Capture.
Science 2009, 325, 1652–1654.
Byun, J.; Je, S.-H.; Patel, H. A.; Coskun, A.; Yavuz, C.
T. Nanoporous Covalent Organic Polymers
Incorporating Tröger’s Base Functionalities for
Enhanced CO 2 Capture. J. Mater. Chem. A 2014, 2,
MOPI-I, -II, -III and -V a decrease of the water uptake in
correlation with the polarity of their functional group
down to 7 mmol/g at 0.9 p/p
observed.
Summarizing, the CO
are dominated by the ultramicroporous character of the
networks. Here, the chemical functionalities introduced by
the individual linkers play only a secondary role indicating
that size discrimination within the frameworks might be
an essential part of the adsorption process. In contrast, for
the selectivities both the porosity as well as the chemical
functionality seem to be equally important. This in turn
0
and 298 K for MOPI-V was
2
and H
2
O uptakes of all polyimides
1
2507.
(
(
9)
Lu, W.; Sculley, J. P.; Yuan, D.; Krishna, R.; Wei, Z.;
Zhou, H.-C. Polyamine-Tethered Porous Polymer
Networks for Carbon Dioxide Capture from Flue Gas.
Angew. Chemie Int. Ed. 2012, 51, 7480–7484.
Bhown, A. S. Status and Analysis of Next Generation
Post-Combustion CO2 Capture Technologies. Energy
Procedia 2014, 63, 542–549.
10)
hints at the existence of preferred adsorption sites for CO
and H O.
2
2
(11)
Dawson, R.; Stöckel, E.; Holst, J. R.; Adams, D. J.;
Cooper, A. I. Microporous Organic Polymers for
Carbon Dioxide Capture. Energy Environ. Sci. 2011, 4,
4239.
Associated Content
(12)
Lu, A.-H.; Hao, G.-P. Porous Materials for Carbon
Dioxide Capture. Annu. Reports Sect. “A” Inorganic
Chem. 2013, 109, 484.
Supporting information
Further spectra, isotherms and selectivity plots are pro-
vided in the supporting information.
(
(
(
13)
14)
15)
Chang, Z.; Zhang, D.-S.; Chen, Q.; Bu, X.-H.
Microporous Organic Polymers for Gas Storage and
Separation Applications. Phys. Chem. Chem. Phys.
Author Information
2
013, 15 (15), 5430–5442.
Corresponding Author
Li, D.; Ma, T.; Zhang, R.; Tian, Y.; Qiao, Y. Preparation
of Porous Carbons with High Low-Pressure CO2
Uptake by KOH Activation of Rice Husk Char. Fuel
*
E-mail: juergen.senker@uni-bayreuth.de
Acknowledgements
2
015, 139, 68–70.
The authors thank Dr. Renée Siegel and Beate Bojer for
performing solid-state NMR measurements. Lena Geiling
for gas sorption measurements, Carina Bojer for taking
SEM images, Dr. Ulrike Lacher for performing mass spec-
trometry experiments and Nadine Popp for consultative
support. We gratefully thank Prof. Josef Breu and Prof.
Rhett Kempe for access to the respective instruments.
Seema, H.; Kemp, K. C.; Le, N. H.; Park, S.-W.;
Chandra, V.; Lee, J. W.; Kim, K. S. Highly Selective
CO2 Capture by S-Doped Microporous Carbon
Materials. Carbon N. Y. 2014, 66, 320–326.
(16)
(17)
Ma, X.; Cao, M.; Hu, C. Bifunctional HNO3 Catalytic
Synthesis of N-Doped Porous Carbons for CO 2
Capture. J. Mater. Chem. A 2013, 1, 913–918.
Sevilla, M.; Valle-Vigón, P.; Fuertes, A. B. N-Doped
10
ACS Paragon Plus Environment

Page 11 of 12
Chemistry of Materials
Polypyrrole-Based Porous Carbons for CO 2 Capture.
Adv. Funct. Mater. 2011, 21, 2781–2787.
Porphyrin Polymer for Selective CO 2 Capture. J.
Chem. 2015, 2015, 1–7.
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
(
18)
Socrates, G. Infrared and Raman Characteristic Group
Frequencies: Tables and Charts, 3. ed., re.; Wiley:
Chichester, 2010.
Fisher, J. C.; Siriwardane, R. V; Stevens, R. W. Zeolite-
Based Process for CO 2 Capture from High-Pressure,
Moderate-Temperature Gas Streams. Ind. {&} Eng.
Chem. Res. 2011, 50, 13962–13968.
(34)
McDonald, T. M.; Lee, W. R.; Mason, J. A.; Wiers, B.
M.; Hong, C. S.; Long, J. R. Capture of Carbon Dioxide
from Air and Flue Gas in the Alkylamine-Appended
Metal-Organic Framework Mmen-Mg2(dobpdc). J.
Am. Chem. Soc. 2012, 134, 7056–7065.
Zhang, Z.; Yao, Z.-Z.; Xiang, S.; Chen, B. Perspective of
Microporous Metal–organic Frameworks for CO 2
Capture and Separation. Energy Environ. Sci. 2014, 7,
2868.
Xiong, S.; Gong, Y.; Wang, H.; Wang, H.; Liu, Q.; Gu,
M.; Wang, X.; Chen, B.; Wang, Z. A New Tetrazolate
Zeolite-like Framework for Highly Selective
CO2/CH4 and CO2/N2 Separation. Chem. Commun.
(Camb). 2014, 50, 12101–12104.
Liebl, M. R.; Senker, J. Microporous Functionalized
Triazine-Based Polyimides with High CO 2 Capture
Capacity. Chem. Mater. 2013, 25, 970–980.
Landa, H. O. R.; Flockerzi, D.; Seidel-Morgenstern, A.
A Method for Efficiently Solving the IAST Equations
with an Application to Adsorber Dynamics. AIChE J.
2013, 59, 1263–1277.
Dawson, R.; Stevens, L. A.; Drage, T. C.; Snape, C. E.;
Smith, M. W.; Adams, D. J.; Cooper, A. I. Impact of
Water Coadsorption for Carbon Dioxide Capture in
Microporous Polymer Sorbents. J. Am. Chem. Soc.
2012, 134, 10741–10744.
Fung, B. M.; Khitrin, A. K.; Ermolaev, K. An Improved
Broadband Decoupling Sequence for Liquid Crystals
and Solids. J. Magn. Reson. 2000, 142, 97–101.
Lu, W.; Yuan, D.; Zhao, D.; Schilling, C. I.; Plietzsch,
O.; Muller, T.; Bräse, S.; Guenther, J.; Blümel, J.;
Krishna, R.; Li, Z.; Zhou, H.-C. Porous Polymer
Networks: Synthesis, Porosity, and Applications in
Gas Storage/Separation. Chem. Mater. 2010, 22 , 5964–
5972.
Lu, J.; Zhang, J. Facile Synthesis of Azo-Linked Porous
Organic Frameworks via Reductive Homocoupling
for Selective CO 2 Capture. J. Mater. Chem. A 2014, 2,
13831.
Ganesan, P.; Yang, X.; Loos, J.; Savenije, T. J.; Abellon,
R. D.; Zuilhof, H.; Sudhölter, E. J. R. Tetrahedral N-
Type Materials: Efficient Quenching of the Excitation
of P-Type Polymers in Amorphous Films. J. Am.
Chem. Soc. 2005, 127, 14530–14531.
Liu, X.; Yang, J.; Guo, L.; Gu, Y. Effects of Catalyst
Isoquinoline in Polymerization Process on the
Aggregation Structures of Polyimide in Solution.
Macromol. Rapid Commun. 2005, 26, 1682–1686.
Popp, N.; Homburg, T.; Stock, N.; Senker, J. Porous
Imine-Based Networks with Protonated Imine
Linkages for Carbon Dioxide Separation from
Mixtures with Nitrogen and Methane. J. Mater. Chem.
A 2015, 3, 18492–18504.
Thommes, M.; Kaneko, K.; Neimark, A. V.; Olivier, J.
P.; Rodriguez-Reinoso, F.; Rouquerol, J.; Sing, K. S. W.
Physisorption of Gases, with Special Reference to the
Evaluation of Surface Area and Pore Size Distribution
(IUPAC Technical Report). Pure Appl. Chem. 2015, 87,
1051–1069.
(19)
(35)
(36)
(
(
(
20)
21)
Yang, S.-T.; Kim, J.-Y.; Kim, J.; Ahn, W.-S. CO2
Capture over Amine-Functionalized MCM-22, MCM-
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
3
6 and ITQ-2. Fuel 2012, 97, 435–442.
Cheung, O.; Hedin, N. Zeolites and Related Sorbents
with Narrow Pores for CO2 Separation from Flue Gas.
RSC Adv. 2014, 4, 14480.
Li, W.; Zhang, Y.; Li, Q.; Zhang, G. Metal−organic
Framework Composite Membranes: Synthesis and
Separation Applications. Chem. Eng. Sci. 2015, 135,
22)
(37)
(38)
2
32–257.
(
23)
Cabello, C. P.; Berlier, G.; Magnacca, G.; Rumori, P.;
Palomino, G. T. Enhanced CO 2 Adsorption Capacity
of Amine-Functionalized MIL-100(Cr) Metal–organic
Frameworks. CrystEngComm 2015, 17 (2), 430–437.
Sabouni, R.; Kazemian, H.; Rohani, S. Carbon Dioxide
Capturing Technologies: A Review Focusing on Metal
Organic Framework Materials (MOFs). Environ. Sci.
Pollut. Res. Int. 2014, 21, 5427–5449.
Zhu, G.; Ren, H. Porous Organic Frameworks;
Springer Berlin Heidelberg: Berlin, Heidelberg, 2015.
Sakaushi, K.; Antonietti, M. Carbon- and Nitrogen-
Based Organic Frameworks. Acc. Chem. Res. 2015, 48,
(24)
(39)
(
(
25)
26)
(40)
(41)
1
591–1600.
(
(
27)
28)
Xu, S.; Luo, Y.; Tan, B. Recent Development of
Hypercrosslinked Microporous Organic Polymers.
Macromol. Rapid Commun. 2013, 34, 471–484.
Arab, P.; Rabbani, M. G.; Sekizkardes, A. K.;
İslamoğlu, T.; El-Kaderi, H. M. Copper(I)-Catalyzed
Synthesis of Nanoporous Azo-Linked Polymers:
Impact of Textural Properties on Gas Storage and
Selective Carbon Dioxide Capture. Chem. Mater. 2014,
26, 1385–1392.
Rabbani, M. G.; El-Kaderi, H. M. Synthesis and
Characterization of Porous Benzimidazole-Linked
Polymers and Their Performance in Small Gas Storage
and Selective Uptake. Chem. Mater. 2012, 24, 1511–1517.
Shen, C.; Wang, Z. Tetraphenyladamantane-Based
(42)
(43)
(29)
(
(
30)
31)
Microporous
Polyimide
and
Its
Nitro-
(44)
(45)
Functionalization for Highly Efficient CO 2 Capture.
J. Phys. Chem. C 2014, 118, 17585–17593.
Wu, S.; Gu, S.; Zhang, A.; Yu, G.; Wang, Z.; Jian, J.; Pan,
C. A Rational Construction of Microporous Imide-
Bridged Covalent–organic Polytriazines for High-
Enthalpy Small Gas Absorption. J. Mater. Chem. A
2
015, 3, 878–885.
(32)
Zhang, C.; Zhai, T.-L.; Wang, J.-J.; Wang, Z.; Liu, J.-M.;
Tan, B.; Yang, X.-L.; Xu, H.-B. Triptycene-Based
Microporous Polyimides: Synthesis and Their High
Selectivity for CO2 Capture. Polymer (Guildf). 2014,
55, 3642–3647.
(46)
(
33)
Neti, Venkata S. Pavan K.; Wang, J.; Deng, S.;
Echegoyen, L. Synthesis of a Polyimide Porous
11
ACS Paragon Plus Environment

Chemistry of Materials
Page 12 of 12
(
47)
Hao, G.-P.; Mondin, G.; Zheng, Z.; Biemelt, T.; Klosz,
TOC-Graphic
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
S.; Schubel, R.; Eychmüller, A.; Kaskel, S. Unusual
Ultra-Hydrophilic, Porous Carbon Cuboids for
Atmospheric-Water Capture. Angew. Chem. Int. Ed.
Engl. 2015, 54, 1941–1945.
Biswal, B. P.; Kandambeth, S.; Chandra, S.; Shinde, D.
B.; Bera, S.; Karak, S.; Garai, B.; Kharul, U. K.;
Banerjee, R. Pore Surface Engineering in Porous,
Chemically Stable Covalent Organic Frameworks for
Water Adsorption. J. Mater. Chem. A 2015, 3, 23664–
(48)
2
3669.
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
(
(
49)
50)
Stegbauer, L.; Hahn, M. W.; Jentys, A.; Savasci, G.;
Ochsenfeld, C.; Lercher, J. A.; Lotsch, B. V. Tunable
Water and CO Sorption Properties in Isostructural
2
Azine-Based Covalent Organic Frameworks through
Polarity Engineering. Chem. Mater. 2015, 27,
7
874.7881
Yang, Y.; Zhang, Q.; Zhang, Z.; Zhang, S. Functional
Microporous Polyimides Based on Sulfonated
Binaphthalene Dianhydride for Uptake and
Separation of Carbon Dioxide and Vapors. J. Mater.
Chem. A 2013, 1, 10368.
Rao, K. V.; Haldar, R.; Maji, T. K.; George, S. J. Porous
Polyimides from Polycyclic Aromatic Linkers:
Selective CO2 Capture and Hydrogen Storage.
Polymer (Guildf). 2014, 55, 1452–1458.
Liao, Y.; Weber, J.; Faul, C. F. J. Fluorescent
Microporous Polyimides Based on Perylene and
Triazine for Highly CO 2 -Selective Carbon Materials.
Macromolecules 2015, 48, 2064–2073.
Farha, O. K.; Bae, Y.-S.; Hauser, B. G.; Spokoyny, A.
M.; Snurr, R. Q.; Mirkin, C. A.; Hupp, J. T. Chemical
Reduction of a Diimide Based Porous Polymer for
Selective Uptake of Carbon Dioxide versus Methane.
Chem. Commun. (Camb). 2010, 46, 1056–1058.
Farha, O. K.; Spokoyny, A. M.; Hauser, B. G.; Bae, Y.-
S.; Brown, S. E.; Snurr, R. Q.; Mirkin, C. A.; Hupp, J. T.
Synthesis, Properties, and Gas Separation Studies of a
(51)
(
(
52)
53)
(
54)
Robust
Diimide-Based
Microporous
Organic
Polymer. Chem. Mater. 2009, 21, 3033–3035.
(55)
Li, G.; Wang, Z. Naphthalene-Based Microporous
Polyimides: Adsorption Behavior of CO 2 and Toxic
Organic Vapors and Their Separation from Other
Gases. J. Phys. Chem. C 2013, 117, 24428–24437.
Rao, K. V.; Haldar, R.; Kulkarni, C.; Maji, T. K.; George,
S. J. Perylene Based Porous Polyimides: Tunable, High
Surface Area with Tetrahedral and Pyramidal
Monomers. Chem. Mater. 2012, 24, 969–971.
Liu, D.-P.; Chen, Q.; Zhao, Y.-C.; Zhang, L.-M.; Qi, A.-
D.; Han, B.-H. Fluorinated Porous Organic Polymers
via Direct C–H Arylation Polycondensation. ACS
Macro Lett. 2013, 2, 522–526.
(
(
(
56)
57)
58)
Liu, J.; Thallapally, P. K.; McGrail, B. P.; Brown, D. R.;
Liu, J. Progress in Adsorption-Based CO2 Capture by
Metal-Organic Frameworks. Chem. Soc. Rev. 2012, 41,
2308–2322.
12
ACS Paragon Plus Environment