Polymer 55 (2014) 335e339
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Stable benzimidazole-incorporated porous polymer network for
carbon capture with high efficiency and low cost
Muwei Zhang 1, Zachary Perry 1, Jinhee Park, Hong-Cai Zhou*
Department of Chemistry, Texas A&M University, College Station, TX 77842, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 July 2013
Received in revised form
14 September 2013
Accepted 16 September 2013
Available online 25 September 2013
Porous Polymer Networks (PPNs) are an emerging category of advanced porous materials that are of
interest for carbon dioxide capture due to their great stabilities and convenient functionalization processes. In this work, an intrinsically-functionalized porous network, PPN-101, was prepared from
commercially accessible materials via an easy two-step synthesis. It has a BET surface area of 1095 m2/g.
Due to the presence of the benzimidazole units in the framework, its CO2 uptake at 273 K reaches
115 cm3/g and its calculated CO2/N2 selectivity is 199, which indicates its potential for CO2/N2 separation.
The great stability, large CO2/N2 selectivity and low production cost make PPN-101 a promising material
for industrial separation of CO2 from flue gas. Its H2 and CH4 uptake properties were also investigated.
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords:
Porous polymer networks
Polybenzimidazole
Carbon dioxide capture
Elimination of CO2 from mixed component gas streams has
gained a tremendous amount of attention due to a growing concern
of the environmental and climatic impact of greenhouse gas
emissions [1]. Many environmental problems, such as global
warming and ocean acidification, have been primarily attributed to
the escalating level of atmospheric CO2. In order to reduce
anthropogenic CO2 emissions, various Carbon Capture and
Sequestration (CCS) techniques have been investigated as means to
selectively remove CO2 from the flue gas of fossil-fuel-powered
plants and then store it underground [2]. Aqueous alkanolamines,
such as monoethanolamine (MEA) solutions, have been utilized
due to their large CO2 capacity and selectivity [3]. Nevertheless, this
process suffers from a series of complications that have substantially limited their industrial applications, such as the high regeneration cost arising from the large heat capacity of aqueous MEA
solutions, the toxicity, the unpleasant smell and the corrosive nature of amine compounds [4].
As an alternative solution, many solid adsorbents have been
shown to be promising candidates to overcome the downsides of
aqueous alkanolamine solutions. For the past few decades,
advanced porous materials [5] have been extensively investigated
in scientific and technological research due to their capability to
adsorb and interact with atoms, ions and molecules [6]. The
* Corresponding author. Tel.: þ1 979 845 4034; fax: þ1 979 845 1595.
E-mail addresses: zhou@mail.chem.tamu.edu, zhou@chem.tamu.edu (H.-C. Zhou).
1
These authors contributed equally to this work.
0032-3861/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.polymer.2013.09.029
functionalities and prospective applications of the porous materials
are largely dependent on their pore size distribution and pore
surface properties. Under the motivation of achieving larger surface
area and better framework properties, metal-organic frameworks
(MOFs) [7,8] and porous polymer networks (PPNs) [9] emerged as
novel categories of porous materials and are widely applied in areas
such as such as gas storage [9,10], gas separation [11,12], catalysis
[13,14], sensors [15], and other applications. Even though MOFs
possesses many advantages such as enormous surface area, tunable
structures and convenient post-synthetic modifications, most
MOFs suffer from a limited stability, which restrained their practical applications [16]. PPNs, appearing as hyper-crosslinked
organic polymers, have provided an alternative solution to the
construction of ultra-porous materials with high thermal and
chemical stability [17]. Unlike crystalline Covalent Organic Frameworks (COFs) [18], PPNs are amorphous materials with Brunauere
EmmetteTeller (BET) surface areas as large as 6461 m2/g (in PPN4), which is the largest PPN BET surface area to date [9]. Regardless of the enormous surface area and improved stability, those
PPNs suffer from several shortcomings that generally hamper them
from being practically implemented. First, their synthesis involves
the coupling of tetrahedral monomers via the Yamamoto homocoupling reaction [17], which requires an extremely expensive
and air-sensitive reagent, Bis(cyclooctadiene)nickel (Ni(cod)2). This
has significantly increased the production cost of PPNs and prevented their large-scale synthesis for industrial use. Second, many
applications of porous materials require the functionalization of the
inner pore surfaces. Being constructed primarily from phenyl rings
336
M. Zhang et al. / Polymer 55 (2014) 335e339
Scheme 1. The basicity of the basic units that have been incorporated into PPNs.
Fig. 1. The conceptual illustration of an ideal network that incorporates tetrahedral
nodes and linear nodes. Ideally, this will give rise to a network with diamondoid
topology.
or alkyne bridges, the lack of internal functionalization of those
PPNs has significantly limited their applications, especially for the
purposes of carbon dioxide sequestration. Post-synthetic modifications (PSM) of those PPNs appear as an efficient way to improve
their functionality [4,19]; however, the PSM process has not only
drastically impaired their porosity but also further increased their
production cost. It is highly desirable to synthesize functionalized
PPNs with considerably large porosities and reasonable production
costs for industrial CCS processes.
Incorporation of Lewis bases into MOFs [20] and PPNs [4] has
been demonstrated as an effective way to improve CO2 uptake and
CO2/N2 selectivity. These moieties can be incorporated into the
struts of the PPN framework, or can be introduced by postsynthetic modification. Pyrazine [21], triazine [22,23] and polyamines [4] have been successfully incorporated into PPNs. However, none of these materials can remove CO2 from flue gas with
both high efficiency and low cost. Despite the interesting electronic
properties resulting from the conjugate system; the CO2/N2 selectivity of the pyrazine-incorporated PPN, Aza-CMP, was not
investigated, probably due to the low porosity (BET surface area of
24 m2/g when activated at 300 C) and complicated activation
procedure of this material [21]. The incorporation of triazine into
PPN frameworks, as shown in TFM-1 [22] and CTF-PX series [23],
was demonstrated as a moderately efficient way to improve the
CO2/N2 selectivity; however, the selectivity was restrained by the
limited basicity of the triazine units. It is conceivable that incorporation of a more basic unit could further improve the CO2/N2
selectivity and CO2 adsorption enthalpy due to the acidic nature of
CO2. The incorporation of polyamines into PPNs [4] was demonstrated as an efficient way to improve the CO2/N2 selectivity;
however, Ni(cod)2 was required in the synthesis, and the PSM
process further increased the cost of this PPN. In spite of the high
efficiency of CO2 removal in polyamine PPNs, the overall production
cost remains largely problematic. Herein, we introduce a stable
benzimidazole-incorporated porous polymer network, PPN-101,
which has a BET surface area of 1096 m2/g, CO2 uptake of
226.2 mg/g at 273 K, and CO2/N2 selectivity of around 200. The
significantly reduced cost and highly efficient separation of CO2
from N2 make it a promising material for industrial separation of
CO2 from flue gas with both high efficiency and low cost.
Scheme 1 illustrates a few basic units that have been successfully incorporated into PPNs. Benzimidazole incorporation
into the frameworks can be used to improve the CO2/N2 selectivity. Polybenzimidazoles (PBIs) have been widely used in other
areas, such as in proton-exchange membranes in fuel cells [24].
However, few of the PBIs turn out to be suitable materials for
industrial separation of CO2/N2 due to their very limited porosity.
El-Kaderi and coworkers [25] published the first benzimidazoleincorporated PPN named as BILP-1 (Benzimidazole Linked Polymer) by condensation of tetrakis(4-formylphenyl)methane and
2,3,6,7,10,11-hexaaminotriphenylene, with the BET surface area of
1172 m2/g and CO2 uptake of around 180 mg/g at 273 K, which
makes it the first benzimidazole-incorporated PPN with considerable porosity for selective CO2 removal. However, the laborious
syntheses of both the aldehyde [26] and amine [27] monomers
significantly increased the production cost of BILP-1 and hindered its synthesis in large quantities. After the discovery of BILP1, they have provided a series of benzimidazole-incorporated
PPNs where expensive monomers were utilized [28e30]. In this
work, PPN-101 was prepared from a synthetically-accessible silicon-centered aldehyde monomer [31] and a commercially
available amine monomer, which has significantly brought down
Scheme 2. The syntheses of aldehyde monomer (tetrakis(4-formylphenyl)silane) and PPN-101. This PPN can be synthesized within two steps from commercially available compounds, and no extremely expensive reagent is involved.
M. Zhang et al. / Polymer 55 (2014) 335e339
337
Fig. 2. (a) the N2 isotherm at 77 K, 1 bar and (b) the pore size distribution calculated by DFT method.
the production cost of this PPN, while the presence of the
benzimidazole units produces a high CO2/N2 selectivity.
The fact that many PPNs built from tetrahedral monomers
possess both good stability and large surface area has motivated us
to keep exploring monomers with a tetrahedral geometry [9,16,17].
The combination of tetrahedral unit and linear unit will result in a
framework with a diamondoid topology (dia topology). (Fig. 1)
Conceptually, this connectivity will not only retain the framework
stability of the diamond-like structures, but also allow adequate
exposure of the faces and edges of phenyl rings and benzimidazole
rings to the adsorbents, which will increase the surface area and gas
uptake capacities [17]. Additionally, the hypothetical diamondoid
framework offers large cavities and interconnected pores to efficiently eliminate the “dead space” that has no contribution to the
framework porosity [9]. The introduction of silicon centered
monomers, instead of working with the carbon-centered aldehyde
monomer that suffers from a complicated synthesis as in BILP-1,
has significantly simplified the preparation of the tetrahedral
building unit [32] (Scheme 2).
PPN-101 was synthesized by a similar route as BILP-1 with some
modifications [25]. The tetrahedral monomer was dissolved in
anhydrous DMF, while 2 equivalents of the amine monomer dissolved in anhydrous DMF were added dropwise at cryogenic temperature under N2 atmosphere. Orange precipitate was formed very
slowly in the solution. The reaction was kept in cryogenic conditions until no more precipitate was formed and then allowed to
warm to r. t. overnight. Then the reaction mixture was moved to O2
atmosphere and heated under O2 at 130 C for 2 days. (Caution:
This step should be handled in an extremely cautious way to prevent any possible combustions or explosions. Protective shields
should be used during the reaction) Fluffy yellow polymer was
produced in 59% yield.
Similar to other PPNs, PPN-101 is insoluble in commonly used
solvents, which simplifies its purification and activation processes.
Since the framework is constructed exclusively by robust covalent
bonds, PPN-101 exhibits high chemical and thermal stability. The
stability, porosity, and internal functionalization of PPN-101 have
inspired us to study its gas uptake properties. PPN-101 is porous
upon activation (See Electronic Supplementary Materials) and exhibits a type-I isotherm of N2 sorption at 77 K and 1 bar (Fig. 2(a)),
implying the existence of micropores in PPN-101. Its Langmuir
surface area, BrunauerEmmettTeller (BET) surface area and pore
volume are 1798 m2/g, 1095 m2/g and 0.66 cm3/g, respectively. Its
pore size distribution was calculated by the Density Functional
Theory (DFT) method (Fig. 2(b)). Its pore diameters are widely
distributed from 10 A to more than 50 A and no pattern could be
identified from the pore size distribution plot. This implies the
amorphous nature of this PPN, which can further be confirmed by
the Powder XRD (See Electronic Supplementary Materials). The
most prominent pores have diameters of 11, 16 and 19 A, respectively, which is consistent with the microporous nature of PPN-101,
determined from the N2 isotherms. Notably, it has a larger average
pore size than that of BILP-1, in which the pore sizes were found to
be centered at around 6.8 Å [25]. This is probably due to the diamondoid nature of PPN-101, which will inherently provide larger
cavities [9]. Its N2 isotherm at 77 K and pore size distribution plot
Fig. 3. (a) CO2 isotherms at 273 K and 298 K and the N2 isotherm at 273 K. (b) CO2 isotherms at 195 K.
338
M. Zhang et al. / Polymer 55 (2014) 335e339
Table 1
Summary of the BET surface area, CO2/N2 selectivity and production cost of all the aforementioned PPNs with basic units incorporated.
Incorporated Units
Basicity (pKa)
BET Surface Area (m2/g)
CO2 Uptake (273 K, mg/g)
Selectivity of CO2/N2
PSM requiredc
Ref
a
b
c
TFM-1
CTF-P6
Aza-CMP
BILP-1
PPN-101
PPN-6-CH2DETA
Triazine
Barely basic
738
76.1
29 2
No
[22]
Triazine
Barely basic
1152
148.1
16.1
No
[23]
Pyrazine
0.65
24a/1227b
N/A
N/A
No
[21]
Benzimidazole
5.532
1172
188
w70
No
[25]
Benzimidazole
5.532
1096
226.2
199
No
This Work
Polyamine
pKa,1 > 10
555
189
w400
Yes
[4]
Activated at 300 C.
Activated at 500 C.
PSM will reduce the porosity and increase the production cost.
Fig. 4. (a) the H2 isotherm at 77 K and 87 K, 1 bar and (b) CH4 isotherm at 273 K, 1 bar.
can be found in Fig. 2. It should be noted that an isostructural PPN
named as BILP-4 [30] was published by El-Kaderi and coworkers
where the expensive tetrahedral monomer was used.
The benzimidazole-rich PPN-101 was designed to possess an
excellent CO2/N2 selectivity. Fig. 3 (a) shows the gas uptake isotherms of CO2 at 273 K and 296 K and the N2 isotherm at 273 K. The
isotherm for CO2 uptake is fully reversible and the maximum uptake can reach 226.2 mg/g at 273 K, which is higher than the BILP-1
with 188 mg/g at this temperature [25]. This probably resulted from
the fact that PPN-101 has large pores and more Lewis base moieties
than BILP-1. Additionally, its CO2 uptake capacities at cryogenic
conditions were also investigated. At 195 K and 0.95 bar, its CO2
uptake is as high as 498 cm3/g. Notably, a deep rise of adsorption
occurs between 0 and 0.1 bar, which indicates an excellent CO2
affinity for the benzimidazole-incorporated PPN-101. A significant
hysteresis was observed in cryogenic CO2 isotherm, which probably
resulted from the large heat of adsorption of CO2 for PPN-101. (See
Electronic Supplementary Materials).
The selectivity of CO2/N2 of PPN-101 in flue gas conditions was
evaluated by the ratio of the adsorbed gas quantity where the
partial pressure for CO2 is 0.15 bar and N2 is 0.85 bar. Since
minuscule N2 adsorption occurs at 273 K, the calculated singlecomponent CO2/N2 selectivity [3] of PPN-101 in flue gas at 273 K
is as large as 199 (See Electronic Supplementary Materials), which
is significantly larger than other base-incorporated PPNs such as
BILP-1 (Selectivity ¼ 70) [25], TFM-1 (Selectivity ¼ 48.2) [22] and
CTF-PX (Selectivity ¼ 16e24) [23] (Table 1). The selectivity of PPN101 is smaller than polyamine-tethered PPN-6 series (Selectivity of
PPN-6-CH2DETA is around 400, which is the best selectivity among
PPN-6-series) [4], while the BET surface area of PPN-101 is
considerably higher than that of PPN-6-CH2DETA (555 m2/g),
where the PSM process has significantly reduced the surface area of
PPN-6 (4023 m2/g) [19]. The calculated IAST selectivity [33,34] of
CO2/N2 at 273 K for PPN-101 is 45094. (See Electronic
Supplementary Materials) Table 1 summarizes all the aforementioned base-incorporated PPNs. The low production expense,
considerable surface area, and large CO2/N2 selectivity make PPN101 a promising material for industrial CO2 sequestration.
Along with its large CO2 uptake, this PPN has a large H2 and CH4
uptake, which makes it very promising for gas storage applications
due to its reduced cost. Its H2 uptake at 77 K and 1.21 bar
(906 mmHg) is 214.18 cm3/g (1.91 wt %, Fig. 4 (a)), even without the
presence of the unsaturated metal centers, which is similar to that
of BILP-1 (19 mg/g, 1.86 wt %) [25] but larger than many other
MOFs/PPNs that have similar BET surface areas [35]. Its CH4 uptake
at 273 K and 1.08 bar are 23.97 cm3/g (Fig. 4 (b)). The presence of
benzimidazole in the framework appears to be an efficient way to
improve H2 and CH4 uptake as well.
Conclusion
In conclusion, we have successfully synthesized a novel
benzimidazole-incorporated porous polymer network, PPN-101,
which is a very promising material for carbon capture with both
high efficiency and low production cost. Due to the incorporation of
the benzimidazole units, the CO2 uptake of PPN-101 at 273 K reaches 115 cm3/g and its calculated CO2/N2 selectivity reaches 199.
Additionally, it also shows some promise for H2 storage (1.91 wt% at
77 K, 1.21 bar) and methane storage (23.97 cm3/g at 273 K, 1.08 bar).
Acknowledgments
This work was supported as a part of the Center for Gas Separations Relevant to Clean Energy Technologies, an Energy Frontier
Research Center funded by the U.S. Department of Energy (DOE),
Office of Science, Office of Basic Energy Sciences under Award
Number DE-SC0001015. The authors also acknowledge DOE
grants DE-FC36-07GO17033 and DE-AR0000249. We are grateful to
M. Zhang et al. / Polymer 55 (2014) 335e339
Mr. Mathieu Bosch for helpful discussion, Ms. Ying-Pin Chen and
Prof. Dr. Daqiang Yuan (Fujian Institute of Research on the Structure
of Matter, Chinese Academy of Sciences) for simulation work of the
PPN structure, and Mr. Wolfgang M. Verdegaal (SunFire, Germany)
for calculating the IAST selectivity of CO2/N2 of PPN-101.
Appendix A. Supplementary data
Supplementary data related to this article can be found online at
http://dx.doi.org/10.1016/j.polymer.2013.09.029.
References
[1] Pearson PN, Palmer MR. Nature 2000;406(6797):695e9.
[2] Wriedt M, Sculley JP, Yakovenko AA, Ma Y, Halder GJ, Balbuena PB, et al.
Angew Chem Int Ed 2012;51(39):9804e8.
[3] Sumida K, Rogow DL, Mason JA, McDonald TM, Bloch ED, Herm ZR, et al. Chem
Rev 2011;112(2):724e81.
[4] Lu W, Sculley JP, Yuan D, Krishna R, Wei Z, Zhou H-C. Angew Chem Int Ed
2012;51(30):7480e4.
[5] Makal TA, Li J-R, Lu W, Zhou H-C. Chem Soc Rev 2012;41(23):7761e79.
[6] Davis ME. Nature 2002;417(6891):813e21.
[7] Zhou H-C, Long JR, Yaghi OM. Chem Rev 2012;112(2):673e4.
[8] Zhang M, Chen Y-P, Bosch M, Gentle III T, Wang ZU, Zhou H-C. Angew Chem
Int Ed 2013. http://dx.doi.org/10.1002/anie.201307340.
[9] Yuan D, Lu W, Zhao D, Zhou H-C. Adv Mater 2011;23(32):3723e5.
[10] Yuan D, Zhao D, Sun D, Zhou H-C. Angew Chem Int Ed 2010;49(31):5357e61.
[11] Bloch ED, Queen WL, Krishna R, Zadrozny JM, Brown CM, Long JR. Science
2012;335(6076):1606e10.
[12] Ma S, Sun D, Wang X-S, Zhou H-C. Angew Chem Int Ed 2007;46(14):2458e62.
[13] Wu C-D, Hu A, Zhang L, Lin WJ. Am Chem Soc 2005;127(25):8940e1.
[14] Tanabe KK, Siladke NA, Broderick EM, Kobayashi T, Goldston JF, Weston MH,
et al. Chem Sci 2013;4(6):2483e9.
339
[15] Lan A, Li K, Wu H, Olson DH, Emge TJ, Ki W, et al. Angew Chem Int Ed
2009;48(13):2334e8.
[16] Lu W, Yuan D, Zhao D, Schilling CI, Plietzsch O, Muller T, et al. Chem Mater
2010;22(21):5964e72.
[17] Ben T, Ren H, Ma S, Cao D, Lan J, Jing X, et al. Angew Chem Int Ed 2009;48(50):
9457e60.
[18] Côté AP, Benin AI, Ockwig NW, O’Keeffe M, Matzger AJ, Yaghi OM. Science
2005;310(5751):1166e70.
[19] Lu W, Yuan D, Sculley J, Zhao D, Krishna R, Zhou H-C. J Am Chem Soc
2011;133(45):18126e9.
[20] McDonald TM, Lee WR, Mason JA, Wiers BM, Hong CS, Long JRJ. Am Chem Soc
2012;134(16):7056e65.
[21] Kou Y, Xu Y, Guo Z, Jiang D. Angew Chem Int Ed 2011;50(37):8753e7.
[22] Zhu X, Tian C, Mahurin SM, Chai S-H, Wang C, Brown S, et al. Am Chem Soc
2012;134(25):10478e84.
[23] Ren S, Bojdys MJ, Dawson R, Laybourn A, Khimyak YZ, Adams DJ, et al. Adv
Mater 2012;24(17):2357e61.
[24] Asensio JA, Sanchez EM, Gomez-Romero P. Chem Soc Rev 2010;39(8):
3210e39.
[25] Rabbani MG, El-Kaderi HM. Chem Mater 2011;23(7):1650e3.
[26] Fournier J-H, Wang X, Wuest JD. Can J Chem 2003;81(5):376e80.
[27] Chen L, Kim J, Ishizuka T, Honsho Y, Saeki A, Seki S, et al. Am Chem Soc
2009;131(21):7287e92.
[28] Rabbani MG, Reich TE, Kassab RM, Jackson KT, El-Kaderi HM. Chem Commun
2012;48(8):1141e3.
[29] Rabbani MG, Sekizkardes AK, El-Kadri OM, Kaafarani BR, El-Kaderi HM.
J Mater Chem 2012;22(48):25409e17.
[30] Rabbani MG, El-Kaderi HM. Chem Mater 2012;24(8):1511e7.
[31] Pariya C, Marcos YS, Zhang Y, Fronczek FR, Maverick AW. Organometallics
2008;27(17):4318e24.
[32] Zhang M, Chen Y-P, Zhou H-C. CrystEngComm 2013. http://dx.doi.org/
10.1039/C3CE41105B.
[33] Sculley JP, Verdegaal WM, Lu W, Wriedt M, Zhou H-C. Adv Mater 2013;25(29):
3957e61.
[34] Simmons JM, Wu H, Zhou W, Yildirim T. Energy Environ Sci 2011;4(6):
2177e85.
[35] Suh MP, Park HJ, Prasad TK, Lim D-W. Chem Rev 2011;112(2):782e835.