A rational construction of microporous imide-bridged covalent-organic polytriazines for high-enthalpy small gas absorption

Article abstract of DOI:10.1039/c4ta04734f

A series of microporous imide functionalized 1,3,5-triazine frameworks (named TPIs@IC) were designed by an easy-construction technology other than the known imidization method for the construction of porous triazine-based polyimide networks (TPIs) with th

Full text of DOI:10.1039/c4ta04734f

View Article Online
View Journal
Journal of
Materials Chemistry A
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: S. Wu, S. Gu, A.
Zhang, G. Yu, Z. Wang, C. Pan and X. Jian, J. Mater. Chem. A, 2014, DOI: 10.1039/C4TA04734F.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
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
Information for Authors.
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 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.
www.rsc.org/materialsA

View Article Online
DynamicDAOIr: t10ic.1l0e39L/Ci4nTkA0s47►34F
Journal of Materials Chemistry A
Page 1 ofJ8ournal Name
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
ARTICLE TYPE
Rational Construction of Microporous Imide-Bridged Covalent–Organic
Polytriazines for High-enthalpy Small Gas Absorption†
Shaofei Wu,¥a Shuai Gu,¥a Aiqing Zhang,b Guipeng Yu,*a Zhonggang Wang,*c Jigao Jian,c Chunyue
Pan*a
5
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX
DOI: 10.1039/b000000x
A series of microporous imide functionalized 1,3,5-triazine frameworks (named TPIs@IC) were designed
by a easy-construction technology other than the known imidazation method for the construction of
porous triazine-based polyimide networks (TPIs) with same chemical compositions. In contrast to TPIs,
10 TPIs@IC exhibit much higher Brunauer-Emmett-Teller (BET) surface areas (up to 1053 m2 g-1) and
carbon dioxide uptake (up to 3.2 mmol g-1/14.2 wt% at 273K/1bar). The presence of abundant
ultramicropores at 5.4~6.8 Å, mainly ascribed to a high-level cyano cross-linking, allows high heat
absorption and high selective capture of CO2. The Qst (CO2 esoteric enthalpies) from their CO2
adsorptions isotherms at 273 and 298 K are calculated to be in the range 46.1-49.3 kJ mol-1 at low CO2
15 loading, and ideal CO2/N2 separation factors are up to 151, exceeding those of most reported porous
organic polymers to date. High storage capacities of TPIs@IC for other small gases like CH4 (5.01 wt%
at 298 K/22 bar) and H2 (1.47 wt% at 77K/1bar) were also observed, making them promising adsorbents
for gas adsorption and separation.
superior chemical and thermal stability compared to other porous
organic macromolecules. Recently, Han et al reported
50 maximum CO2-adsorption capacity of 5.53 mmol g-1 at 273 K
and bar for perfluorinated covalent triazine-based
20 1. Introduction
a
A variety of solid sorbents including porous coordination
polymers,1 metal-organic frameworks (MOFs)2 and porous
organic polymers (POPs)3 have been intensively investigated as
potential materials for small gases (carbon dioxide, hydrogen and
25 methane) capture for clean energy applications. The former two
frameworks featuring huge surface areas and low cost exhibit
attractive gas uptake capacity, but nevertheless always suffer
from poor physical and chemical stability.4 The main drawback
of such sorbents is their low tolerance to water vapor that is
30 usually present in the gas stream, which undoubtedly damages
their practical popularity. POPs linked by strong covalent bonds
instead of the coordination bonds exceed such limits and have
been demonstrated as competent and promising candidates for
small gas capture and storage relevant to clean energy.
35 POPs constructed solely from light elements like carbon and
hydrogen often possess weak adsorption affinity towards guest
molecules and hence low small gas uptake capacity. One
intensively employed strategy to improve the small gases uptake
is the pore surface engineering by introducing heteroatom
40 functionalities to enhance the binding affinity between host and
guest molecule.5 1,3,5-triazine and imide groups with high
nitrogen content are deemed as ideal building blocks to develop
high performance POPs.6 Such porous polymeric materials,
which can be separately synthesized via the cyclization
45 polymerization of aromatic nitriles7 or the condensation of amine
and aromatic anhydrides,8 are attractive for energy storage
applications because of their high mechanical strength, as well as
1
a
framework.7b More importantly, its uptake of CO2 reaches 1.76
mmol g-1 at 273 K and 0.1 bar with an initial Qst value (35.0 kJ
mol-1). Our group recently applied the concept of polar
55 electronic-rich groups such as triphenylamine, carbazole,
dibenzofuran, and dibenzothiophene to covalent triazine-based
frameworks and observed significantly improved adsorption
performance for small gases (eg. CO2 and H2).6b,9 On the other
hand, remarkable progress has been achieved for the design and
60 pore surface engineering of microporous aromatic polyimides.10
We disclosed the preparation of microporous polyimide networks
with BET surface areas up to 1407 m2 g-1 derived from a rigid
tetrahedral monomer tetra(4-aminophenyl)methane through
imidazation technology in 2010.10a Such hypercrosslinked
65 networks can absorb 3.30 wt% of H2 at 77 K/35 bar. The highest
reported surface area for a porous polyimide is 2213 m2 g-1
derived from tetrahedrally organized perylene, which also
showed 31 wt % of CO2 at 195 K and 1 bar.11 Impressively,
Liebl and coworkers synthesized series of porous triazine-based
70 polyimide
networks (TPIs) combined with the attracting
properties of two heterocyclic polymers.12 TPIs show moderate
BET surface areas ranging from 40 to 809 m2 g-1 as well as broad
pore size distribution and among of them the highest CO2 uptakes
are recorded to 2.45 mmol g-1 (10.8 wt%) at 273 K and 1 bar.
75 Nevertheless, the inconvenience is that TPIs were obtained with a
time-consuming and multi-step synthesis starting from high-
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 1

View Article Online
DOI: 10.1039/C4TA04734F
Journal of Materials Chemistry A
Page 2 of 8
priced halo-substituted nitriles using noble metals as catalysts at
a remarkably low overall yield («35%). Additionally, the low
porosity of TPIs together with their relatively low absorption
capacities, which is strongly enslaved to the solubility of the
respective dianhydride building block in toxic solvent mediums
(like m-cresol), leaves a significant improvement of the carbon
dioxide uptakes to be desired. Moreover, their hydrogen and
methane storage performance still remain unexplored.
5
As part of an ongoing project dealing with porous polymers with
10 tailor-made structures and properties, herein, we present a simple
two-step synthetic strategy to alternatively build imide-based
porous 1,3,5-triazine frameworks (TPIs@IC, namely TPIs
synthesized under ionothermal condition) which have the same
chemical composition as the disclosed TPIs at enhanced overall
15 yields. Comparing with TPIs, considerably improved BET
surface areas and carbon dioxide uptakes were demonstrated by
TPIs@IC. TPIs@IC with abundant ultramicroporisity exhibit
high carbon dioxide absorption heats (up to 49.3 kJ mol-1), which
are among the best reported results for porous organic polymers.
20 Additionally, hydrogen and methane adsorption measurements
for TPIs@IC show excellent uptake capacities despite their
moderate BET surface areas.
Fig. 1 Synthesis routes for TPIs@IC (TPIs).
2. Experimental
General procedure for the preparation of TPIs@IC
measurements. On the contrary, for TPIs the time-consuming
synthesis of 4,4’,4’’-(1,3,5-triamine-2,4,6-triyl)trianiline block
60 was accomplished in the catalysis of corrosive super acid
(CF3SO3H) and volatile organometalliic reagents (LiHMDS and
Pd(dba)2), indicating their harsh reaction conditions and costly
synthetic procedures, which undoubtedly limits their possibility
for scale up. Additionally, TPIs were constructed via the known
65 imidazation technology of the aromatic dianhydride and aromatic
triamines in low boiling point organic solvent mediums (i.e. m-
cresol). Although such polymerization is easier to handle, no high
level of cross-linking was achieved for TPIs. This could be easily
explained, considering the restricted reaction temperatures by
70 organic mediums and the crystallization or precipitation of the
intermediates prior to the formation of the highly crosslinked
networks. We surmised that such higher surface area, porous
organic materials could be achieved by increasing the degree of
cross-linking of the function groups. Indeed, previous reports on
75 microporous hyper-crosslinked organic polymers have suggested
that higher level of cyano cross-linking and hence much higher
network rigidity can lead to higher surface areas for the resulting
materials.15,22
25 The polymeric networks were synthesized by heating a mixture
of 1 g monomer and ZnCl2 (molar ratio, ZnCl2: monomer=10:1)
in a quartz tube (3×10 cm). The tube was evacuated to a high
vacuum, and then sealed rapidly. TPI-1@IC was synthesized by a
o
selected temperature program (Table S1, ESI†) (200 C/5 h, 300
30 oC/5 h, 400 oC/20 h), while TPI-2@IC was obtained by a
o
o
o
temperature program (200 C/5 h, 300 C/5 h, 400 C/10 h, 450
oC/10 h, 500 oC/20 h), and the temperature program for TPI-
o
o
o
o
3@IC (200 C/5 h, 300 C/5 h, 400 C/10 h, 450 C/10 h, 500
oC/10 h, 600 C/20 h) was utilized afterwards. The quartz tube
o
35 was then cooled to room temperature and opened. The reaction
mixture was subsequently ground, and then washed thoroughly
with water to remove most of ZnCl2. Further stirring in diluted
HCl for 15 h was carried out to remove the residual salt. After
this purification step, the resulting black powder was filtered, and
40 washed successively with water and methyl alcohol, following by
Soxhlet extraction using acetone, methyl alcohol and hexane as
eluting solvents sequentially, and dried in vacuum at 150 °C.
Typical isolated yields >90%.
Although our dinitrile compounds display poor solubility in
80 common solvent such as hexane, chloroform, acetone, DMSO
and DMF, they are highly miscible in molten ZnCl2 and strong
acids. Unfortunately, our efforts to accomplish their
polymerization using trifluoromethanesulfonic acid as catalyst
under room temperature failed to obtain objective highly
85 crosslinked networks. This fact may be reasonable considering
that the magic acid has caused breakage within C-N and C-O
linkages of the heterocyclic-containing nitriles. We describe
herein a simple approach for producing TPIs@IC which has been
proved to be very valuable for building highly porous organic
90 polytriazines from thermal/chemical unstable blocks.9 The hyper-
crosslinked polytriazines were synthesized from aromatic nitriles
3. Results and discussion
45 The synthesis routes for microporous imide-functionalized 1,3,5-
triazine frameworks (TPIs@IC) and the reported porous triazine-
based polyimide networks (TPIs) are illustrated in Figure 1. All
TPIs@IC were readily prepared by a two-step reaction starting
from 4-aminobenzonitrile and respective aromatic dianhydrides
50 with overall yields ranging from 56% to 82%, whereas those of
TPIs were apparently much lower (31-35%). The rigid nitrile-
functioned precursors (m1, m2, m3) were synthesized by one-step
solution condensations of 4-aminobenzonitrile (A) with
corresponding rigid aromatic anhydrides (Fig. 1). The progress of
1
55 the reactions was confirmed by H NMR and MALDI-TOF/MS
2
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C4TA04734F
Page 3 of 8
Journal of Materials Chemistry A
in sealed quartz ampoules via multiply-step heating programs
TPI-2@IC exhibits a compact surface, while a strip sponge-like
(Table S1, ESI†). Comparing with the m-cresol applied for TPIs, 60 morphology was observed for TPI-3@IC. The microstructure is
molten ZnCl2 is a much better solvent for monomers and triazine-
based oligomers owing to strong Lewis acid-base interactions.
Furthermore, molten ZnCl2 has been shown to be an excellent
studied by high-resolution transmission electron microscopy
(HR-TEM) and X-ray powder diffraction (PXRD). Typically,
alternately dark and bright area can be clearly observed for
TPIs@IC, implying porous structure (Fig. S6-S8, ESI†). The
5
solvent for the syntheses of polytriazines, since allowing
o
extremely high temperatures (eg., 600 C, etc) which is required 65 PXRD spectra of all networks give an amorphous feature (Fig.
to maintain high-degree cross-linking through the trimerization of
cyano groups. The heating programs for the polymerization of
10 nitrile-functioned precursors were optimized to obtain hyper-
S9-S11, ESI†). Thermo-gravimetric measurements (Fig. S12,
ESI†) show that all three networks deliver excellent
thermodynamic stability, with 5% weight loss exceeding 450 C.
o
crosslinked networks with good porosity (Table S1, ESI†). In
To note out, TPIs@IC still show char yields of over 75% when
cases of naphthalene-centred nitrile and perylene-centred nitrile, 70 heated to 800 oC under a nitrogen atmosphere.
the polymerizations under the conditions similar to those for TPI-
1@IC, only gave soluble products without permanent porosity.
15 When a higher reaction temperature is used (500 or 600 °C, Table
S1, ESI†, entries 4, 5 and 6), it was possible to isolate insoluble
networks from the reaction system.
Three TPIs@IC were coincidently obtained as black monolithic
materials in almost quantitative yields. They are insoluble in
20 water and common organic solvents such as hexane, methanol,
acetone, chloroform, N,N-dimethylacetamide (DMAc), N-
To characterize the porosity of TPIs@IC, cryogenic nitrogen
adsorption isotherms were collected at 77 K (Fig. 2). TPI-1@IC
performs a Type I isotherm in the range of P/P0= 0.05-0.15, and
then possess a very flat adsorption plateau, suggesting the
75 microporous nature. Notably, TPI-1@IC displays some Type II
character with a rise in the nitrogen sorption at high relative
pressures (P/P0>0.9). As a sharp contrast, TPI-2@IC and TPI-
3@IC exhibit typical Type II gas sorption isotherms as well as an
obvious hysteresis phenomenon, which might be attributed to the
methylpyrroli-done (NMP), and tetrahydrofuran (THF) along 80 existence of meso-pore structure and interparticulate voids of the
with diluted HCl solution (10 wt%), implying their good
chemical stability. The imide functionality of these networks
25 were further analyzed by the spectral measurements. For instance,
in the FT-IR spectrum of TPI-3@IC (Fig. 1S) the structure of
samples. The isotherms for all TPIs@IC exhibit a sharp N2
adsorption at low relative pressures (P/P0<0.01), implying the
existence of abundant micropores and ultramicropores. Applying
the Brunauer-Emmett-Teller (BET) model to these isotherms
imide functionality is confirmed by the characteristic bands at 85 gives apparent surface areas of 1053 m2 g-1, 814 m2 g-1 and 963
1658 cm-1 corresponding to the vibrations of the C=O group in
five-membered polyimide rings, which is in good accordance
30 with the solid-state CP/MAS NMR measurement. The broad band
at about 1310-1380 cm-1 correlates to overlap of the C-N-C
m2 g-1 for TPI-1@IC, TPI-2@IC and TPI-3@IC, respectively
(Table 1). The slightly decreased surface area for TPI-2@IC and
TPI-3@IC relative to TPI-1@IC could be due to the following
two reasons: the volume of naphthalene in TPI-2@IC and
stretching vibration of the imide ring and the in-plane stretching 90 perylene in TPI-3@IC is obviously larger than the benzene strut;
vibration of the triazine ring. For all polymers, the disappearance
of the absorption around 2235 cm-1 in FT-IR spectra ascribing to
35 the characteristic carbonitrile stretching suggests almost
completely conversion of cyano groups under the investigated
second, the pi-stacking of the naphthalene and perylene groups is
also unfavorable for the pore formation because of the strong p-
electron-delocalization effect. The BET surface areas of
TPIs@IC are sufficiently higher than the values reported for TPI-
conditions, while the appearance of strong absorption bands at 95 1 (809 m2 g-1), TPI-2 (796 m2 g-1) and TPI-3 (40 m2 g-1) of the
1360 and 1508 cm-1 is indicative of a successful trimerization
reaction. The FTIR spectra and BET surface areas of the samples
40 after refluxing in HCl solution (10 wt%) for 48 h are all kept
comparing with the original networks, further confirmed the
same chemical composition. A plausible explanation for this is
that the high degree of crosslinking and hence high network
rigidity effectively prevent the tight packing or entanglement of
chains, thus resulting in high BET surface areas as discussed
retentivity of the structure. According to other relevant 100 above.
literatures,7d,7e,9 some materials of this type failed to obtain a
meaningful solid-state CP/MAS NMR spectrum. For example,
45 the spectrum of TPI-2@IC (Fig. 2S) only gives characteristic
peaks of aromatic carbons (at 125 ppm) and carbonyl carbons in
polyimide rings (at 162 ppm), while no fine-structure resolution
was found. Nitrogen and oxygen elimination during the
polymerization reaction were observed from elemental analysis
50 (Table S2, ESI†). These deviations may be reasonable
considering the incomplete combustion and trapped adsorbates
including gases and water vapour.7 Inductively Coupled Plasma
(ICP) analysis shows 6.7 wt%, 6.1 wt% and 5.9 wt% zinc
contents for TPI-1@IC, TPI-2@IC and TPI-3@IC, respectively
55 (Table S2, ESI†), which are reasonable and slightly higher than
those of reported NPTNs (4.7~5.9 wt%).9 Morphologies
monitored by scanning electron microscopy (SEM, Fig. S3-S5,
ESI†) demonstrate that TPI-1@IC displays a floppy surface and
Fig.2 N2 adsorption (solid symbols)/desorption (open symbols)
isotherms at 77K
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 3

View Article Online
DOI: 10.1039/C4TA04734F
Journal of Materials Chemistry A
Page 4 of 8
temperatures (eg. 298 K), the carbon dioxide uptake capacities of
45 TPI-1@IC (2.11 mmol g-1) and TPI-3@IC (1.44 mmol g-1) are
slightly decreased under 1 bar, and are also sufficiently higher
than those reported for TPI-1 (1.25 mmol g-1) and TPI-3 (0.43
mmol g-1). This may indicate that narrow ultramicropores (<1 nm)
make a superior contribution to CO2 adsorption than wide
50 micropores and mesopores of TPIs, considering their same pore
surface nature and thermodynamic size of CO2.7 It is worth
mentioning that the uptake capacity of TPI-1@IC at 1 bar and
273 K surpasses those of other heteroatom-containing materials,
including 1,3,5-triazine based conjugated microporous polymers
55 TCMPs (1.22-2.62 mmol g-1),15 azo-linked covalent organic
polymers azo-COPs (1.93-2.55 mmol g-1),16 NOPs(1.31-2.80
mmol g-1),6a,6b NPTNs(2.23-3.20 mmol g-1),9 and triptycene-based
microporous poly(benzimidazole) networks TBIs (2.68-3.17
mmol g-1).17 We attribute this excellent carbon dioxide sorption
60 capability of TPI-1@IC to the combination of appropriate pore
volume and abundant narrow ultramicropores, which afford
strong adsorption potential. In comparison with TPI-1@IC, TPI-
3@IC possesses lower carbon dioxide uptake due to the existence
of mesopores peaked at 2.7 nm, which might be too large to
65 increase the number of double or multiple interactions between
the adsorbed CO2 and the pore walls.18
Fig.3 NLDFT pore size distribution of TPIs@IC from (A) N2 adsorption
(at 77 K) and (B) CO2 adsorption (at 273 K).
Pore size distribution (PSD) of N2 adsorption at 77 K (Fig. 3A)
was estimated from the nonlocal density functional theory
(NLDFT),13 while pore volume was calculated from single point
measurements (P/P0=0.97) and found to be 0.53, 0.76, and 0.74
cm3 g-1 for TPI-1@IC, TPI-2@IC and TPI-3@IC, respectively.
TPIs@IC show higher pore volume than that of TPIs suggesting
10 improved gas capacity, which was further confirmed by their
carbon dioxide adsorption isotherms at 298 and 273 K (Fig. 3). It
is interesting to observe that all TPIs@IC possess abundant
micropores and supermicropores. TPI-1@IC exhibits a narrow
5
pore-size
distribution
and abundant
micropore
and
15 supermicropore with pore diameters centered at around 1.48 nm
and 0.54 nm, respectively. Interestingly, TPI-2@IC and TPI-
3@IC show similar PSD curves with dominant pore width
located at 0.59 nm and 1.27 nm in addition to spot mesopores
peaked at 2.7 nm. One possible application for the small pore size
20 is that the subdividing of the cavity by the segments or rigid strut
moieties because of the network interpenetration. The formation
of an additional mesopore system for TPI-3@IC, which leads to a
drastic increase of the surface area, may be due to another mode
of self-organization of the system at higher temperatures.14 In
25 consideration of the advantages of the CO2 sorption-based model
for pore size distribution (PSD) below 1 nm,7d CO2 micropore
analysis data at 273 K was calculated with NLDFT approach and
exhibits dominant pore width of TPIs@IC range from 0.4 nm to
0.7 nm (Fig. 3B), further demonstrating that ionothermal method
30 facilitates the formation of ultra-micropores. Comparing with
PSD profiles of TPI-2@IC and TPI-3@IC, the curve of TPI-
1@IC without intense peaks at 0.7~1 nm displays more single-
minded pore size distribution at supermicropore range (<0.7 nm),
implying a strong CO2 adsorption potential for TPI-1@IC.
With the lowest BET surface area of TPIs@IC, TPI-2@IC
displays the bottommost carbon dioxide loading (2.06 mmol g-1 at
273 K/1 bar). This could be attributed to the low degree of
70 polymerization correlated to the limited solubility of m2
monomer. With higher temperature, such monomers show a
better solubility in ZnCl2 ionic melts owing to strong Lewis acid-
base interaction, which also contributes to the dissolution of
o
oligomers. At a temperature of 400 C, for instance, the expected
75 polymers were not obtained with the precursors m2 and m3 due
to their poor solubility, while the temperature was adjusted to 500
oC, we gained TPI-2@IC with the lowest BET surface area of
814 m2 g-1. Increasing the reaction temperature to 600 oC resulted
in a successful construction of TPI-3@IC with a BET surface
80 area of 963 m2 g-1. This would be due to the high melting point of
perylene-centred nitrile as well as good dissociation of the metal-
organic complexes prior to consecutive reaction by high
temperature cyclization.14 Moreover, comparing to the
preparation of TPIs in m-cresol at 190 oC,12 the synthesis of
85 TPIs@IC in ionic melt is more effective. Although heated to 190
oC (near the boiling temperature 202.3 oC), m-cresol remains a
deficient solvent for the generation of highly crosslinked network.
35 Table 1. Pore parameters and CO2 uptake of TPIs@IC and TPIs
a
c
Polymer
SBET
SbLAN VTotal
Pore Sized
nm
0.54, 1.48
0.58, 1.27, 2.7
0.58, 1.27, 2.7
0.75-0.85
0.75-0.85
CO2 uptakee
273 K 298 K
m2/g
1053
814
963
809
796
40
m2/g cm3/g
1295
0.53
0.76
0.74
0.45
0.40
3.22
2.06
2.17
2.45
2.45
0.68
2.11
1.43
1.44
1.25
1.23
0.43
TPI-1@IC
TPI-2@IC
TPI-3@IC
TPI-1f
1059
1316
-
-
-
TPI-2f
TPI-3f
aBrunauer-Emmett-Teller surface area. bLangmuir surface area. cPore
volume determined from the N2 isotherm at P/P0=0.99. dPore size derived
from N2 isotherm with the NLDFT approach. eIn mmol/g at 1 bar. fResults
from Ref 12.
40 As we estimated, TPI-1@IC and TPI-3@IC exhibit carbon
dioxide adsorption of 3.22 and 2.17 mmol g-1 under 1 bar and at
273 K (Fig. 4), respectively, which are almost two times of TPI-1
(2.45 mmol g-1) and TPI-3 (0.68 mmol g-1) (Table 1). At elevated
Fig. 4 CO2 adsorption isotherms of TPIs@IC.
4
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C4TA04734F
Page 5 of 8
Journal of Materials Chemistry A
45
Fig. 6 A: Adsorption isotherms of CO2 and N2 gases at 273 K. B: IAST
selectivities of CO2 over N2 for binary gas mixtures 15/85 molar
composition in TPIs@IC at 273k.
Fig. 5 Qst from CO2 adsorption at 273 and 298K.
TPI-3 with sufficient rigidity exhibits a carbon dioxide uptake of
0.68 mmol g-1 at 1 bar and 273 K (Table 1), apparently lower
than our TPI-3@IC which has the same chemical composition.
Therefore, the imidazation technology limits the number of
suitable building blocks for this synthesis pathway. This
demonstrates that our method for doping heteroatom into
frameworks may be more impactful than that used for
5
10 constructing high performance microporous organic polymers.
To determine the binding affinity of TPIs@IC for CO2, we
calculated the isosteric heats of adsorption for CO2 from the data
collected at 273 K and 298 K using the Clausius-Clapeyron
equation (Fig. 5).19 The isosteric heats determined at low loading
15 were 49.3 kJ mol-1 for polymer TPI-1@IC, 46.1 kJ mol-1 for
polymer TPI-2@IC and 49.1 kJ mol-1 for polymer TPI-3@IC,
which were significantly improved compared to TPIs (29.2-34.4
kJ mol-1),12 HCPs (20.0-24.0 kJ mol-1),20 MCTFs (20.5-26.3 kJ
mol-1),7a BILPs (26.5-28.8 kJ mol-1),21 BLPs (20.2-28.3 kJ mol-
20 1),22 CPOP-1 (27.0 kJ mol-1),5a networks (23.3-28.2 kJ mol-1)23
and STPIs (28-36 kJ mol-1)8b and SMPs (23.0-32.0 kJ mol-1).24
The exceptionally high Qst values can be ascribed to the electron-
rich imide rings and high charge density at nitrogen and oxygen
sites that facilitates framework-gas interactions through hydrogen
25 bonding and/or dipole-quadrupole interactions. With the CO2
adsorption increasing from 0 mmol g-1 to 0.3 mmol g-1, a rapid
decrease was observed for these adsorption heat curves,
indicating that the framework-gas interaction is more powerful
than gas-gas interaction. Those profiles keep smooth at high
30 adsorbed amount because of pure gas-gas interactions.
Interestingly, the initial adsorption heats of TPIs@IC are
approximately equal due to the similar ultramicropore and
micropore size in addition to the similar chemical composition.
TPI-2@IC and TPI-3@IC exhibit considerably higher isosteric
35 heats than TPI-1@IC over a wide range of gas loadings,
suggesting that the incorporation of strong p-electron-
delocalization effect into the framework indeed enhances the host
affinity towards CO2.
50
Fig. 7 Adsorption isotherms of CO2 and CH4 gases at 298 K
CO2/N2 gas mixture are 80, 151 and 77 for TPI-1@IC, TPI-2@IC
and TPI-3@IC, respectively (Fig. 6B). The values are comparable
to NOPs (53-81),6b pyrrole-HCP (117)23 and FCTF-1-600 (152),7b
superior to many other porous polymers like NPTNs (22-45),9 PI-
55 NDAT (25),8c PI-NO2 (18-31)8c and networks (26-119).23 For the
sake of comparing the performance of our TPIs@IC and the
known TPIs on selectively separation of small gases, we also
calculated the CO2 uptake and selectivity over N2 involves the
use of initial slope ratios estimated from Henry’s law constants
60 for single-component adsorption isotherms at 273 K (Fig. 6A).
Based on this, we calculated the adsorption selectivity for CO2
over N2 in the pressure range below 0.15 bar (Fig. S19-S21, ESI†)
and the results are presented in Table 2. TPI-1@IC, TPI-2@IC
and TPI-3@IC show CO2/N2 selectivity ratios of 33, 70 and 55,
65 respectively, which are higher than those of TPI-1 (31), TPI-2 (34)
and TPI-3 (35). In comparison with the slightly large pore size of
TPIs, the ample micropores and supermicropores of TPIs@IC,
which match the dimension of CO2 better for gas sorption,
contribute a lot to the high CO2/N2 selectivity.
70 To further evaluate the potential use of TPIs@IC in gas
separation applications, the selectivity of TPIs@IC toward CO2
over CH4 was investigated by collecting isotherms at 298 K (Fig.
7). At 298 K and 1 bar, TPI-1@IC, TPI-2@IC and TPI-3@IC
exhibit CO2 uptakes of 2.11, 1.43 and 1.44 mmol g-1, respectively,
75 whereas CH4 adsorptions of TPI-1@IC (0.020 mmol g-1), TPI-
2@IC (0.021 mmol g-1) and TPI-3@IC (0.034 mmol g-1) are
almost negligible. On the basis of initial slope calculations in the
pressure range of 0 to 0.2 bar (Fig. S22-S24, ESI†), the estimated
adsorption selectivity for CO2 over CH4 are 11, 8 and 2 (Table 2).
80 The values of TPI-1@IC and TPI-2@IC exceed those reported
for activated carbon and ZIFs. The high selectivity of CO2 over
N2 and over CH4 indicates that TPIs@IC are promising
To evaluate the adsorption selectivity of TPIs@IC for
40 postcombustion flue gas, the prevalent ideal adsorbed solution
theory (IAST) model was adopted to imitate the selectivity of
CO2 over N2 at an equilibrium partial pressure of 0.85 bar (N2)
and 0.15 bar (CO2) in the bulk phase (Fig. S13–S18, ESI†). The
data in Table 2 show that ideal selectivities at 273 K/1 bar for
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 5

View Article Online
DOI: 10.1039/C4TA04734F
Journal of Materials Chemistry A
Page 6 of 8
Table 2. CO2, H2 and CH4 sorption of TPIs@IC
adsorption and separation studies. TPIs@IC can also store up to
7.35 mmol g-1 H2 at 77 K/1 bar and 3.13 mmol g-1 CH4 at 298
K/22 bar, comparable with many other porous organic polymers.
45 The synthetic availability, structural precision and highly
microporous nature open a new opportunity for the large-scale
industrial preparation of high performance carbonaceous
absorbents and their application in small gas absorption and
separation.
a
b
c
Polymer SCO2/N2 SCO2/N2 SCO2/CH4
Qst H2 uptaked CH4 uptakee
kJ/mol mmol/g 1 bar 22 bar
80
32.2
69.6
55.1
30.9
33.5
35.1
11
8
2
ND
ND
ND
49.3
46.1
49.1
34.4
31.4
ND
6.15
6.80
7.35
ND
ND
ND
0.020 2.04
0.021 1.61
0.034 3.13
TPI-1@IC
TPI-2@IC
TPI-3@IC
TPI-1f
151
77
NDg
ND
ND
ND
ND
ND
ND
ND
ND
TPI-2f
TPI-3f
aSCO2-N2 is calculated by the IAST model from 85% N2 and 15% CO2, at 1
50 Acknowledge
b
bar/ 273 K. SCO2-N2 is calculated from initial slope calculations at 273K.
d
cSCO2-CH4 is calculated from initial slope calculations at 298K. at 77 K/1
We acknowledge the financially support from the National Science
Foundation of China (Nos. 21204103 and 21376272), China Postdoctoral
Science Foundation (2012M521535), China Postdoctoral Science
Foundation Specialized Funded project (2014T70787), State Key
55 Laboratory of Fine Chemicals (KF1206), Key Laboratory of Catalysis
and Materials Science of the State Ethnic Affairs Commission & Ministry
of Eduction (CHCL12006), and State Key Laboratory of Advanced
Technolgy for Materials and Processing (2015-KF-8).
5 bar. ein mmol g-1. fResults from Ref 12. gNo detection.
Notes and references
60 aCollege of Chemistry and Chemical Engineering , Central South
University, Changsha 410083, China. E-mail:gilbertyu@csu.edu.cn;
panchunyue@sina.com.
Fig. 8 Adsorption isotherms of CH4 at 298K (A) and H2 at 77K (B)
bKey Laboratory of Catalysis and Materials Science of the State Ethnic
Affairs Commission & Ministry of Eduction, Hubei Province, South-
65 Central University for Nationalities, Wuhan 430074, P. R. China
cDepartment of Polymer Science & Materials, State Key Laboratory of
Fine Chemicals, Dalian University of Technology, Dalian 116012, China.
¥These authors contribute to this work equally.
candidates for the separation and purification of CO2 from
various CO2/CH4 or CO2/N2 mixtures.
10 We have also considered TPIs@IC in hydrogen absorption and
high pressure methane storage studies because both gases are
attractive candidates for renewable and clean energy. The
† Electronic Supplementary Information (ESI) available: [Synthesis and
methane uptakes for TPIs@IC (1.61-3.13 mmol g-1) at 298 K/22 70 Characterization data and properties]. See DOI: 10.1039/b000000x/
bar (Fig. 8A) are comparable to COF-1 (2.5 mmol g-1, 298 K/85
1
2
(a) P. Li, Y.-B. He, J Guang, L.-H Weng, J. C.-G. Zhao, S. C. Xiang
and B. L. Chen, J. Am. Chem. Soc., 2014, 136, 547–549. (b) M.
Mastalerz and I. M. Oppel, Angew. Chem., Int. Ed., 2012, 51, 5252-
5255.
(a) A. M. Fracaroli, H. Furukawa, M. Suzuki, M. Dodd, S. Okajima,
F. Gandara, J. A. Reimer and O. M. Yaghi, J. Am. Chem. Soc., 2014,
136, 8863-8866. (b) F. Gandara, H. Furukawa, S. Lee and O. M.
Yaghi, J. Am. Chem. Soc., 2014, 136, 5271-5274. (c) F. Jeremias, V.
Lozan, S. K. Henninger and C. Janiak, Dalton Trans., 2013, 42,
15967-15973. (d) K. Schutte, H. Meyer, C. Gemel, J. Barthel, R. A.
Fischer and C. Janiak, Nanoscale, 2014, 6, 3116-3126.
(a) Y. Yuan, F.-X. Sun, L. Li, P. Cui and G.-S. Zhu, Nat. Commun.,
2014, 5, 4260-4267. (b) X. Gao, X.-Q. Zou, H.-P. Ma, S. Meng and
G.-S. Zhu, Adv. Mater., 2014, 26, 3644–364. (c) S.-Y. Ding and W.
Wang, Chem. Soc. Rev., 2013, 42, 548-568. (d) X.-H. Liu, C.-Z.
Guan, S.-Y. Ding, W. Wang, H.-J. Yan, D. Wang and L.-J. Wan, J.
Am. Chem. Soc., 2013, 135, 10470-10474.
15 bar),25 zeolites (1.94-4.56 mmol g-1, 298 K/85 bar)26 and
mesoporous silicas (2.5 mmol g-1, 298 K/85 bar).26 Hydrogen
storage capacity for TPIs@IC are 1.23 wt%, 1.36 wt% and 1.47
wt%, respectively, measured at 77 K and 1.0 bar (Fig. 8B).
Similar to the CO2 and CH4 isotherms, the H2 uptake has not
20 reached saturation in the investigated pressure range, implying
that high storage can be expected at higher pressures. The
calculated hydrogen uptake ability is obviously higher than those
reported for COF-11À (1.22 wt%),27 EOFs (0.94-1.21 wt%),28
polyanilines (0.38-0.85 wt%),29 Porph-PIM (1.20 wt%),30 PIM-1
25 (0.95 wt%),30a,31 HCP-BDM (1.11 wt%)32 and HCP-BA (0.97
wt%).32 The high adsorption capacity of TPIs@IC can refer to
their narrow ultra-micropores as well as polar surface due to the
presence of triazine and imide rings.
75
80
85
90
3
4
5
H.Furukawa, K. E. Cordova, M. O. Keeffe and O. M. Yaghi, Science,
2013, 341, 1230444-1-1230444-12.
(a) Q. Chen, M. Luo, P. Hammershøj, D. Zhou, Y. Han, B. W.
Laursen, C.-G. Yan and B.-H. Han, J. Am. Chem. Soc., 2012, 134,
6084–6087. (b) Q. Chen, D.-P. Liu, M. Luo, L.-J. Feng, Y.-C. Zhao
and B.-H. Han, Small, 2014, 10, 308-315.
4. Conclusions
30 We have demonstrated a facile method for the rapid and cost-
effective synthesis of highly microporous imide functionalized
1,3,5-triazine frameworks (TPIs@IC) other than the deficient
imidazation method. The obtained networks, which have the
same chemical compositions as the known TPIs, exhibit higher
35 BET surface areas (up to 1053 m2·g -1) and better carbon dioxide
uptake ability (up to 3.2 mmol·g-1/14.2 wt%). The presence of
abundant ultramicropores affords TPIs@IC with remarkably high
absorption enthalpies (up to 49.3 kJ mol-1) as well as high
selectivities: CO2/N2 (151 at 273 K) based on IAST calculations
40 and CO2/CH4 (11 at 298 K) derived from the ratios of initial
slopes of isotherms, making them very attractive in CO2
95 6 (a) S.-H. Xiong, X. Fu, L. Xiang, G.-P. Yu, J. G. Guan, Z.-G Wang,
Y. Du, X. Xiong and C.-Y Pan, Polym. Chem., 2014, 5, 3424–3431.
(b) Y. Liu, S.-F. Wu, G. Wang, G.-P. Yu, J. G. Guan, C.-Y Pan and
Z.-G Wang, J. Mater. Chem. A., 2014, 2, 7795-7801. (c) L. Stegbauer,
K. Schwinghammer and B. V. Lotsch, Chem. Sci., 2014, 5, 2789-
100
2793. (d)G.-Y. Li and Z.-G Wang, Macromolecules, 2013, 46, 3058-
3066.
7
(a) X.-M. Liu, H. Li, Y.-W. Zhang, B. Xu, Sigen A, H. Xia and Y.
Mu, Polym. Chem., 2013, 4, 2445–2448. (b) Y.-F. Zhao, K.-X. Yao,
B.-Y. Teng, T. Zhang and Y. Han, Energy Environ. Sci., 2013, 6,
3684–3692. (c) Sevilla, M.; Valle-Vigꢀn, P.; Fuertes, A. B. Adv.
Funct. Mater. 2011, 21, 2781−2787. (d) A. Bhunia, I. Boldog, A.
105
6
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C4TA04734F
Page 7 of 8
Journal of Materials Chemistry A
Moller and C. Janiak, J. Mater. Chem. A., 2013, 1, 14990-14999. (e)
A. Bhunia, V. Vasylyeva and C. Janiak, Chem. Commun., 2013, 49,
3961-3963.
8
(a) C. J. Shen, Y.-J. Bao and Z.-G Wang, Chem. Commun., 2013, 49,
3321-3323. (b) C. Zhang, T.-L. Zhai, J.-J. Wang, Z. Wang, J.-M. Liu,
B.-E. Tan, X.-L. Yang and H.-B. Xu, Polymer, 2014, 55, 3642-3647.
(c) C.-J. Shen and Z.-G. Wang, J. Phys. Chem. C, 2014, 118, 17585-
17593.
5
9
S.-F. Wu, Y. Liu, G.-P. Yu, J.-G. Guan, C.-Y. Pan, Y. Du, X. Xiong
and Z.-G. Wang, Macromolecules, 2014, 47, 2875-2882.
10
10 (a) Z.-G. Wang, B.-F. Zhang, H. Yu, L.-X. Sun, C.-L. Jiao and W.-S.
Liu, Chem. Commun., 2010, 46, 7730-7732. (b) J. Weber, Q. Su, M.
Antonietti and A. Thomas, Macromol. Rapid Commun., 2007, 28,
1871-1876. (c) B. S. Ghanem, N. B. McKeown, P. M. Budd, N. M.
Al-Harbi, D. Fritsch, K. Heinrich, L. Starannikova, A. Tokarev and Y.
Yampolskii, Macromolecules, 2009, 42, 7881-7888. (d) S. A. Sydlik,
Z.-H. Chen and T. M. Swager, Macromolecules, 2011, 44, 976-980.
11 K. V. Rao, R. Haldar, C. Kulkarni, T. K. Maji and S. J. George,
Chem. Mater., 2012, 24, 969-971.
15
20 12 M. R. Liebl and J. Senker, Chem. Mater., 2013, 25, 970-980.
13 J. Weber, J. Schmidt, A. Thoma and W. Bohlmann, Langmuir, 2010,
26, 15650-15656.
14 P. Kuhn, A. Thomas and M. Antonietti, Macromolecules, 2009, 42,
319-326.
25 15 S.-J. Ren, R. Dawson, A. Laybourn, J.-X. Jiang, Y. Khimyak, D. J.
Adams and A. I. Cooper, Polym. Chem., 2012, 3, 928-934.
16 H. A. Patel, S. H. Je, J. Park, D. P. Chen, Y. S. Jung, C. T. Yavuz and
A. Coskun, Nat. Commun., 2013, 4, 1357-1364.
17 Y.-C. Zhao, Q.-Y. Cheng, D. Zhou, T. Wang and B.-H. Han, J. Mater.
30
Chem., 2012, 22, 11509-11514.
18 (a) J.-X. Jiang, F. Su, A. Trewin, C. D. Wood, H.-J. Niu, J. T. A.
Jones, Y. Z. Khimyak and A. I. Cooper, J. Am. Chem. Soc., 2008,
130, 7710-7720. (b) R. Dawson, A. I Cooper and D. J Adams, Polym.
Int., 2013, 62, 345-352.
35 19 V. Krungleviciute, L. Heroux, A. D. Migone, C. T. Kingston and B.
Simard, J. Phys. Chem. B, 2005, 109, 9317-9320.
20 C. F. Martín, E Stöckel, R. Clowes, D. J. Adams, A. I. Cooper, J. J.
Pis, F Rubiera and C Pevida, J. Mater. Chem., 2011, 21, 5475-5483.
21 (a) M. G. Rabbani and H. M. El-Kaderi, Chem. Mater., 2011, 23,
40
45
1650–1653. (b) M. G. Rabbani, T. E. Reich, R. M. Kassab, K, T.
Jackson and H. M. El-Kaderi, Chem. Commun., 2012, 48, 1141-1143.
(c) M. G. Rabbani and H. M. El-Kaderi, Chem. Mater., 2012, 24,
1511-1517.
22 (a) K. T. Jackson, M. G. Rabbani, T. E. Reich and H. M. El-Kaderi,
Polym. Chem., 2011, 2, 2775-2777. (b) T. E. Reich, S. Behera, K. T.
Jackson, P. Jena and H. M. El-Kaderi, J. Mater. Chem., 2012, 22,
13524-13528.
23 S. W. Yao, X. Yang, M. Yu, Y. H. Zhang and J.-X. Jiang, J. Mater.
Chem. A, 2014, 2, 8054-8059.
50 24 B.-Y. Li, Z.-H. Guan, X.-J. Yang, W. D. Wang, W. Wang, I. Hussain,
K.-P. Song, B.-E. Tan and T. Li, J. Mater. Chem. A, 2014, 2, 11930–
11939.
25 H. Furukawa and O. M. Yaghi, J. Am. Chem. Soc., 2009, 131, 8875-
8883.
55 26 V. C. Menon, S. Komarneni, J. Porous. Mater., 1998, 5, 43-58.
27 R. W. Tilford, S. J. Mugavero III, P. J. Pellechia and J. J. Lavigne,
Adv. Mater., 2008, 20, 2741-2746.
28 M. Rose, W. Böhlmann, M. Sabo and S. Kaskel, Chem. Commun.,
2008, 2462-2464.
60 29 J. Germain, J. M. J. Fréchet and F. Svec, J. Mater. Chem., 2007, 17,
4989-4997.
30 (a) N. B. McKeown, P. M. Budd and D. Book, Macromol. Rapid
Commun., 2007, 28, 995-1002. (b) B. S. Ghanem, M. Hashem, K. D.
M. Harris, K. J. Msayib, M.-C. Xu, P. M. Budd, N. Chaukura, D.
65
70
Book, S. Tedds, A.Walton and N. B. McKeown, Macromolecules,
2010, 43, 5287-5294.
31 P. M. Budd, B. S. Ghanem, S. Makhseed, N. B. McKeown, K. J.
Msayi and C. E. Tattershall, Chem. Commun., 2004, 230-231.
32 Y.-L. Luo, S.-C. Zhang, Y.-X. Ma, W. Wang and B.-E. Tan, Polym.
Chem., 2013, 4, 1126-1131.
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 7

Journal of Materials Chemistry A
Page 8 of 8
View Article Online
DOI: 10.1039/C4TA04734F
Graphic abstract
This paper presents a facile and efficient method for the construction of microporous imide
functionalized 1,3,5-triazine frameworks (ITFs), which exhibit high sorption capacities of CO2,
CH4 and H2, showing potentials in small gas separation and recovery.