Ammonia capture in porous organic polymers densely functionalized with Bronsted acid groups

Article abstract of DOI:10.1021/ja4105478

The elimination of specific environmental and industrial contaminants, which are hazardous at only part per million to part per billion concentrations, poses a significant technological challenge. Adsorptive materials designed for such processes must be engendered with an exceptionally high enthalpy of adsorption for the analyte of interest. Rather than relying on a single strong interaction, the use of multiple chemical interactions is an emerging strategy for achieving this requisite physical parameter. Herein, we describe an efficient, catalytic synthesis of diamondoid porous organic polymers densely functionalized with carboxylic acids. Physical parameters such as pore size distribution, application of these materials to low-pressure ammonia adsorption, and comparison with analogous materials featuring functional groups of varying acidity are presented. In particular, BPP-5, which features a multiply interpenetrated structure dominated by <6 A pores, is shown to exhibit an uptake of 17.7 mmol/g at 1 bar, the highest capacity yet demonstrated for a readily recyclable material. A complementary framework, BPP-7, features slightly larger pore sizes, and the resulting improvement in uptake kinetics allows for efficient adsorption at low pressure (3.15 mmol/g at 480 ppm). Overall, the data strongly suggest that the spatial arrangement of acidic sites allows for cooperative behavior, which leads to enhanced NH₃ adsorption.

Full text of DOI:10.1021/ja4105478

Article
pubs.acs.org/JACS
Ammonia Capture in Porous Organic Polymers Densely
Functionalized with Brønsted Acid Groups
Jeffrey F. Van Humbeck,† Thomas M. McDonald,† Xiaofei Jing,§ Brian M. Wiers,† Guangshan Zhu,§
and Jeffrey R. Long*,†,‡
†Department of Chemistry, University of California, Berkeley, California 94720, United States
‡Material Science Division, Lawrence Berkeley National Laboratory, Berkeley, California, 94720, United States
§State Key Laboratory of Inorganic Synthesis and Preparative Chemistry, Jilin University, Changchun, China 130012
S
* Supporting Information
ABSTRACT: The elimination of specific environmental and industrial contami-
nants, which are hazardous at only part per million to part per billion concentrations,
poses a significant technological challenge. Adsorptive materials designed for such
processes must be engendered with an exceptionally high enthalpy of adsorption for
the analyte of interest. Rather than relying on a single strong interaction, the use of
multiple chemical interactions is an emerging strategy for achieving this requisite
physical parameter. Herein, we describe an efficient, catalytic synthesis of
diamondoid porous organic polymers densely functionalized with carboxylic acids.
Physical parameters such as pore size distribution, application of these materials to
low-pressure ammonia adsorption, and comparison with analogous materials
featuring functional groups of varying acidity are presented. In particular, BPP-5,
which features a multiply interpenetrated structure dominated by <6 Å pores, is
shown to exhibit an uptake of 17.7 mmol/g at 1 bar, the highest capacity yet
demonstrated for a readily recyclable material. A complementary framework, BPP-7, features slightly larger pore sizes, and the
resulting improvement in uptake kinetics allows for efficient adsorption at low pressure (3.15 mmol/g at 480 ppm). Overall, the
data strongly suggest that the spatial arrangement of acidic sites allows for cooperative behavior, which leads to enhanced NH3
adsorption.
1. INTRODUCTION
The adsorption of contaminants that are present in vanishingly
low concentrationparts per million or belowpresents a
significant technical challenge in both environmental and
industrial settings. To achieve meaningful adsorption capacity,
an extremely high enthalpy of adsorption (ΔHads) is required.
This energetic requirement typically lies well outside the range
of physical adsorption processes and will instead require the
development of materials that interact chemically with the
analyte of interest.1 Noteworthy progress has been achieved in
the selective adsorption of 390−400 ppm carbon dioxide, as a
first step toward its direct capture from air,2−4 and presents an
interesting conceptual approach.5−8 Appropriate adsorption
enthalpies are achieved in these cases through multiple
Figure 1. Strategy for the development of low-pressure ammonia
adsorbents. (a) Multiple interactions for adsorption of low-pressure
chemical interactions: Initial interaction of a basic amine with
the electrophilic carbon of CO2 yields a carbamic acid, which is
carbon dioxide. (b) Proposed approach to ammonia adsorption.
further stabilized through either full proton transfer to yield an
ammonium carbamate ion pair9 or through hydrogen-bonding
interactions (Figure 1a).10 A similar approach can be
concentrations. With regards to human health, ammonia itself
has a recommended CAL-OSHA permissible exposure limit of
only 25 ppm, which can be encountered in numerous industrial
envisioned, whereby multiple acidic sites located in close
proximity result in the strong adsorption of Lewis basic
pollutants (Figure 1b).
Basic gases, such as ammonia, can lead to significant
environmental and industrial concerns at similarly small
Received: October 14, 2013
Published: January 23, 2014
© 2014 American Chemical Society
2432
dx.doi.org/10.1021/ja4105478 | J. Am. Chem. Soc. 2014, 136, 2432−2440

Journal of the American Chemical Society
Article
settings.11 Highly toxic amines have even more stringent safety
limits (e.g., diethanolamine, 0.46 ppm). In perhaps the most
extreme example, volatile organic amines can disrupt photo-
lithography at only tens of parts per billion concentration,
resulting in characteristic ‘T-top’ channel features that render
the resulting silicon wafer useless (Figure 2a).12,13 As integrated
lesser degree. Trikalitis reported a MOF-20527 analog
containing free phenolic −OH groups, which demonstrated
excellent low- and moderate-pressure ammonia capacity.28
However, the basic zinc acetate-type structure was not stable to
ammonia exposure, with framework collapse suggested by
powder X-ray diffraction and gas adsorption experiments, in
line with previous observation made on analogous zinc-based
materials (MOF-5 and MOF-177).29,30 In an effort to generate
ammonia adsorbents that would be stable, and therefore
potentially reusable, Yaghi investigated a zirconium-based UiO-
66 analog31 featuring anilinium cations as the Brønsted acid
source.32 Although only one-third of the available aniline sites
in the material had been protonated, a meaningful increase in
NH3 adsorption was noted, and the framework survived
exposure up to 1 bar of pressure.
Recently, porous organic polymers featuring a diamondoid-
type structure have been introduced as ‘element−organic
frameworks’ (e.g., EOF-1),33 ‘porous aromatic frameworks’
(e.g., PAF-1),34 and ‘poly(aryleneethynylene) networks’ (e.g.,
PAE-E1),35 with initial investigations on ammonia capture
using isolated metal catechol36 and polyimide37 functional
groups reported. Two particular features of these materials, and
PAF-1 in particular, suggested that they would be an excellent
platform for the development of stable low-pressure ammonia
adsorbents utilizing cooperative interactions. In addition to the
high specific surface area of PAF-1, the physicochemical
stability of this material is especially noteworthy. It exhibits
pH stability ranging from anhydrous chlorosulfonic acid38 to
KOH/DMSO ‘superbase’39 (vide infra). Not only does this
suggest that exposure to ammonia will be easily tolerated, it also
allows for a diverse range of postsynthetic chemical trans-
formations. Already chloromethylation,7 nitration,40 and
sulfonation,38 which occur under vigorously acidic conditions,
have been demonstrated. We surmised that common Brønsted
acidic functional groups (e.g., −CO2H) that are difficult to
include in MOFs, due to their metal complexation abilities,
could be included into these networks, in a density that would
be otherwise difficult to achieve.
Figure 2. Target applications for ammonia capture and storage based
on ammonia/amine pressure and adsorbent stability requirements. (a)
Air purification for semiconductor manufacture. (b) Ammonia as a
first model for chemical warfare agents. (c) Use of ammonia as an on-
board reducing agent to mitigate vehicle emissions.
circuits with narrower feature sizes are pursued, air purity
requirements will become even more stringent,14 though these
separation challenges are tempered by the fact that the low
ammonia concentration will present only moderate demands
on material stability. Outside of this specific application, low-
pressure ammonia adsorption can also serve as the first model
for a generic acid−base interaction, taking the place of
important but difficult to handle analytes such as the V-series
of nerve agents (e.g., VX 1, Figure 2b). At higher pressures, safe
and reversible on-board storage for vehicle applications could
be an enabling technology for ammonia-based fuel cells15 or
more likely as a reducing agent for the mitigation of nitrogen
oxide emissions (Figure 2c).16,17 Exposure to high ammonia
pressures and the need for hundreds-to-thousands of
adsorption cycles will present an exceptional demand on
material stability, as will the diversity of potential environments
where chemical weapons could be encountered.
Herein, we give a full account of the conceptual and
experimental development that led from Brønsted acidic MOFs
with single-point ammonia binding sites, to new porous organic
polymers that display excellent low-pressure ammonia
adsorption, with multiple functional groups present in a spatial
arrangement that leads to cooperative reactivity.
2. EXPERIMENTAL SECTION
General experimental information, specific synthetic procedures, and
full characterization for all new small molecules and organic polymers
are available in the Supporting Information.
Over the past 15 years, metal−organic frameworks (MOFs)
have demonstrated their utility in numerous applications,
including gas storage, molecular separations, sensing, and size-
selective catalysis.18−24 Some preliminary investigations into the
use of these materials for ammonia adsorption have also been
conducted. The overwhelming majority of these examples use
Lewis acidic framework sites to increase the strength of
adsorption for NH3. In materials such as MOF-74, exposed
metal cations provide the desired adsorption sites.25 In related
materials known as covalent organic frameworks (COFs), it has
been demonstrated that three-coordinate boron centers can
behave in a similar Lewis acidic fashion, with the framework
generated from hexahydroxytriphenylene and biphenyldibor-
onic acid (COF-10) displaying high uptake at moderate
pressure.26
Generic Procedure for Bis(cyclooctadiene)nickel(0) Poly-
merization. The procedure applied was that reported for the original
synthesis of PAF-1.34 Bis(cyclooctadiene)nickel(0) (5.2 equiv was
charged in an oven-dried Schlenk flask inside an inert atmosphere
glovebox. The flask was sealed, removed, and connected to a standard
manifold for further manipulation. Vacuum-dried 2,2′-bipyridine (5.2
equiv) was added against positive nitrogen pressure, followed by 1,5-
cyclooctadiene (6.6 equiv) and anhydrous DMF (0.05 M concen-
tration, relative to C−Br bonds). The solution was heated to 80 °C for
1 h. [Important note: At this point, the solution should be a vibrant,
dark violet color, without any hint of black or brown. Occasionally,
especially when older sources of DMF were used, a significant dulling
of the purple color of the reagent was noted, and these reactions
universally gave unsatisfactory results. This issue was never
encountered when using freshly obtained anhydrous DMF of high
commercial grade or DMF that had been stored under rigorously
In the context of MOFs, the use of Brønsted acidic centers
for the adsorption of ammonia has been explored to a much
2433
dx.doi.org/10.1021/ja4105478 | J. Am. Chem. Soc. 2014, 136, 2432−2440

Journal of the American Chemical Society
Article
anhydrous conditions, protected from light and heat.] Solid tetrakis-
(arylbromide) (1.0 equiv, 4.0 equiv C−Br functional groups) was
added to the vibrant purple solution against positive nitrogen pressure.
The reaction was left to stir at 80 °C for 72 h. At that point, it was
allowed to cool to room temperature, concentrated hydrochloric acid
was added (one-half the volume of DMF), and the suspension was left
to stir overnight. The resulting solid was collected by filtration and
washed with water (5 washes with HCl volume), absolute ethanol (5
washes with HCl volume), and THF (5 washes with HCl volume). It
was further purified by Soxhlet extraction with THF (24 h) and dried
under vacuum at the specified temperature to yield the desired
polymer.
The ideal structure would deliver a moderate density (3.35
mmol/g) of anilinium sites, along with the weakly acidic
hydroxide protons found in the inorganic cluster (2.23 mmol/
g). For a representative example of a stronger Brønsted acid,
Fe-MIL-101-SO3H41 provides a comparable number of total
acidic sites to UiO-66-NH3Cl, distributed between the protic
sulfonic acids (2.19 mmol/g) and Lewis acidic unsaturated
metal centers (3.28 mmol/g). The room temperature (298 K)
ammonia uptakes for these materials are shown in Figure 3.
Generic Procedure for Palladium-Catalyzed Polymer Syn-
thesis. An oven-dried round-bottomed flask, cooled under N2, was
charged with a bis(pinacolborane)aryl monomer (2.1 equiv), tetrakis-
(4-bromophenyl)methane (1.0 equiv), and chloro(2-dicyclohexylphos-
phino-2′,6′-dimethoxy-1,1′-biphenyl)[2-(2′-amino-1,1′-biphenyl)]-
palladium(II) (0.70−1.1 mol % relative to −Br functional groups) and
was purged for 10 min with flowing nitrogen. Degassed THF (0.042 M
relative to tetrabromide) and degassed aqueous potassium carbonate
(2 M, 10% of THF volume) were added, and the solution was heated
to 60 °C for 48−72 h. During the course of the polymerization, the
reaction became an extremely viscous gel. The gel was cooled to room
temperature and was transferred onto a Buchner funnel with the aid of
̈
additional THF. With constant vacuum applied, the gel eventually
collapsed into a free-flowing powder, which was washed with hot
hydrochloric acid (3 N, 5 washes with triple THF volume), hot water
(5 washes with triple THF volume), hot ethanol (5 washes with triple
THF volume), and hot CHCl3 (5 washes with triple THF volume).
The resulting powder was then further purified by Soxhlet extraction
with THF (24 h). The polymer was activated at the appropriate
temperature under vacuum to deliver the desired material.
Generic Procedure for Side-Chain Cleavage to Yield Free
Terephthalic Acid Polymers. A solid terephthalic ester polymer was
charged, under N2, in an oven-dried round-bottomed flask. Solid
potassium hydroxide was added (135 wt %), followed by anhydrous
DMSO (to 0.4 M KOH). The flask was heated to 150 °C for 24 h.
After cooling to room temperature, the solid was collected by filtration,
washed with methanol (3 washes with DMSO volume), and allowed to
air dry on the Buchner funnel over vacuum for 15 min. The resulting
̈
solid was resubjected to identical basic conditions for another 24 h
using fresh KOH/DMSO. This cycle was repeated a total of two to
three times, depending on the substrate. After filtering and washing
with methanol the final time, the resulting solid was suspended in
hydrochloric acid (1 N, 50% of DMSO volume) at room temperature
for 8 h. The acid was removed with a syringe, with care not to remove
any polymer. Next, the solid was exposed to hydrochloric acid of
higher concentration (3 N, 50% of DMSO volume) at room
temperature for ∼12 h. Again, the acid was removed by syringe, and
water was added (50% of DMSO volume). The vial was left to sit at 80
°C for 1 h, at which point the water was removed by syringe and
replaced with fresh water (50% of DMSO volume). This exchange was
repeated twice more, before the polymer was finally collected by
filtration. Soxhlet extraction with THF (16 h) delivered the final
product, which was activated at the appropriate temperature under
vacuum to deliver the desired porous acidic polymer.
Figure 3. Ammonia adsorption at 298 K in Brønsted acidic MOFs
(Fe-MIL-101-SO3H: red circles, UiO-NH3CI: blue circles) and a
nonacidic structure (UiO-66-NH2: green circles).
The improvement in uptake, especially at low pressure, shown
by UiO-66-NH3Cl as compared to the parent framework (i.e.,
0.93 mmol/g at 488 ppm for UiO-66-NH2 vs 2.64 mmol/g at
663 ppm for UiO-66-NH3Cl) confirms our prediction that
Brønsted acidic adsorption sites could lead to meaningful
ammonia adsorption at low pressure. Gratifyingly, the more
acidic framework Fe-MIL-101-SO3H yielded an exceptional
adsorbent, which displayed improved uptake at low pressure
(3.52 mmol/g at 510 ppm) and, owing to its significantly
higher specific surface area, at higher pressure as well (17.80
mmol/g at 1004 mbar).
Porous Organic Polymers with Isolated Brønsted
Acidic Substituents. Although encouraging performance
was seen for low-pressure ammonia adsorption in UiO-66-
NH3Cl and Fe-MIL-101-SO3H, these materials face certain
practical limitations. Although UiO-66-NH3Cl maintained
crystallinity after controlled HCl addition, ammonia adsorption
in this specific case results in the formation of NH4Cl, which
3. RESULTS AND DISCUSSION
Brønsted Acidic MOFs. As reported previously, inclusion
of hydrochloric acid equivlents in the 2-aminoterephthalic acid
derivative of UiO-66 ([Zr6O4(OH)4][p-(CO2)2C6H3NH2]6)
allows for a significant increase in ammonia uptake, even with
only one-third of the organic linkers protonated.32 In our
hands, exposure of the neutral parent framework UiO-66-NH2
to anhydrous hydrochloric acid in 1,4-dioxane did not result in
any loss of crystallinity, although analysis of powder X- ray
diffraction data could not unequivcally locate the chloride
counterion in the presumed anilinium chloride UiO-66-NH3Cl.
2434
dx.doi.org/10.1021/ja4105478 | J. Am. Chem. Soc. 2014, 136, 2432−2440

Journal of the American Chemical Society
Article
prevents the reactivation and recycling of this material by a
simple temperature and/or vacuum swing. Fe-MIL-101-SO3H
seemed like a more promising material for further optimization,
with one significant caveat: with the appended sulfonic acids
already delivering very high Brønsted acidity, it is not clear how
the performance of this material could be improved further.
With high specific surface area and exceptional chemical
stability, PAF-1 seemed like an ideal platform for the
development of acidic adsorbents, especially those that must
function for extended periods of time in high-pressure
ammonia environments. Already, Zhou has described the
conversion of PAF-1 into the analogous sulfonic acid-
functionalized material (i.e., PPN-6-SO3H: [(C6H4-
C6H3SO3H)2(C)]),38 and while we recognized that frameworks
substituted with anilinium chloride groups (i.e., −NH3Cl)
would represent a suboptimal material due to NH4Cl
formation, it would allow for the effect of the framework
backbone, as opposed to the functional group complement, to
be directly interrogated. Our approach to the synthesis of the
required materials is shown in Figure 4. The parent material
(i.e., PAF-1) was synthesized according to the original
procedure.34 Menke conditions42a far milder alternative to
HNO3introduce nitrogen as an aromatic nitro group, which
was fully reduced to the corresponding aniline by sodium
dithionite to deliver [(C6H4-C6H3NH2)2(C)] (BPP-1: Berkeley
porous polymer-1), as indicated by FTIR analysis (see
Supporting Information). Simple protonation with anhydrous
hydrochloric acid provided the anilinium chloride-function-
alized material [(C6H4-C6H3NH3Cl)2(C)] (BPP-2). Following
the reported procedure,38 exposure to chlorosulfonic acid
affords the arylsulfonic acid polymer PPN-6-SO3H. The
nitrogen adsorption characteristics of these materials are also
presented in Figure 4. Functional group addition to PAF-1
results in a significant reduction in the BET surface area, from
4240 m2/g in the parent framework to 1400 m2/g in BPP-1,
965 m2/g in BPP-2, and 1200 m2/g in PPN-6-SO3H.
As expected, surface area as determined by N2 adsorption is
not predictive of ammonia uptake (Figure 5). Instead, there is a
Figure 5. Room temperature (298 K) ammonia adsorption for PAF-1
(green circles), BPP-1 (red circles), BPP-2 (yellow circles), and PPN-
6-SO3H (blue circles).
clear relationship between functional group acidity and
adsorption, at both our low-pressure value of interest (500
ppm) and at the highest pressure investigated (1 bar). With
relative acidity seemingly dominating adsorption performance,
it is unsurprising that the MOFs investigated show better
uptake for the same functional group substitutions (i.e., 2.64
mmol/g at 663 ppm for UiO-66-NH3Cl vs 0.97 mmol/g at 636
ppm for BPP-2; 3.52 mmol/g at 510 ppm for Fe-MIL-101-
Figure 4. Synthesis and 77 K nitrogen adsorption charateristics of
PAF-1 (green circles) and functionalized derivatives: BPP-1 (red
circles), BPP-2 (yellow circles), and PPN-6-SO3H (blue circles). Open
circles represent desorption data.
2435
dx.doi.org/10.1021/ja4105478 | J. Am. Chem. Soc. 2014, 136, 2432−2440

Journal of the American Chemical Society
Article
SO3H vs 1.54 mmol/g at 919 ppm for PPN-6-SO3H). The
electron-withdrawing ability of the two carboxylate groups in
the terephthalate linkers of the UiO-66 and MIL-101 series
should increase the acidity of the ammonium chloride and
sulfonic acid, respectively.
Porous Organic Polymers with Cooperative Brønsted
Acidic Substituents. Given the ammonia uptake results
obtained for MOFs and porous aromatic frameworks with
isolated Brønsted acidic functional groups, it seemed likely that
we would not be able to increase uptake substantially, especially
at low pressure, through further tuning of the functional group,
as sulfonic acids represented some of the most acidic groups
that are readily accessible synthetically. Traditional adsorbents
such as zeolites and activated carbons, which have been
investigated in the context of ammonia adsorption, are similarly
limited in their chemical tunability.30,43−46 Instead, we were
inspired by recent computational results suggesting that the
cooperative activity of multiple groups could have a strong
positive effect on low-pressure adsorption.47 Although the
materials derived from PAF-1 displayed lower uptake, we
preferred to base further development around this platform for
a simple reason: Binding sites that contained multiple acidic
functional groups (e.g., −CO2H) close enough in space to
interact with a single ammonia molecule would likely also bind
metal atoms strongly, potentially disrupting MOF synthesis.
The monomer required to produce a close analog to PAF-1
(and also to EOF-1), with functional groups poised to interact
cooperatively, was successfully synthesized employing the
method presented in Scheme 1. Selective metalation of
Scheme 1. Sythesis of meta-Substituted Porous Organic
a
Polymer Precursors
Figure 6. Polymerization, nitrogen adsorption characteristics (77 K),
and pore size distribution for BPP-3. Open circles represent
desorption data.
could explain such a low surface area, two important
observations strongly suggest an alternative explanation.
Analysis of residual bromine by energy dispersive X-ray
spectroscopy (EDX) indicated no remaining bromine, to the
limit of detection, suggesting that the low surface area observed
was not due to a low degree of polymerization. Most
importantly, pore size distribution revealed that the source of
the low observed nitrogen uptake was framework inter-
penetration,48,49 with the dominant diameters being much
smaller than the ∼11 Å expected for an open-pore structure. As
the steric parameters of functionalized monomer 5 are not
remarkably different from the monomer used to synthesize
PAF-1 (i.e., tetrakis(4-bromophenyl)methane), it seems likely
that an increased attractive interaction between independent
networks, due to the dipole−dipole interaction between the
aldehyde substituents, drives interpenetration. Although net-
work interpenetration results in lower specific surface area, it
also offers a new design element that could potentially be
leveraged for materials synthesis. If functional group sub-
stitution leads to an associative interaction, this self-assembly
between individual polymer networks could allow for the
creation of strong binding sites that occur between independent
networks. This approach has a number of significant
advantages. In addition to removing the rigorous requirement
a
Conditions: (a) n-Buli, THF, −78 °C, then SiCl4, −78 °C. 72%. (b)
Benzoyl peroxide, N-bromosuccinimide, CCl4, reflux 62%. (c)
Dimethyl sulfoxide/acetic acid/water (90/5/5 v/v), 105 °C, 66%.
bromoiodoarene 2 and 4-fold addition to silicon tetrachloride
produced tetra-meta-methylated arylsilane 3. Selective 8-fold
bromination under radical conditions yields 4. All attempts to
monobrominate each benzylic position selectively resulted in a
mixture of doubly-, singly-, and nonbrominated methyl groups.
Hydrolysis in wet and slightly acidic DMSO provides
tetrakis(3-formyl-4-bromophenyl)silane 5, which has both the
halide atoms requisite for polymerization and a versatile
functional group handle for postsynthetic modification.
Polymerization of 5 proceeded as expected under identical
conditions as used previously to yield [(2,2′-C6H3CHO-
C6H3CHO)2(Si)] (BPP-3), which bears functional groups in
the positions directly adjacent to the biphenyl bond, poised to
interact cooperatively during gas adsorption (Figure 6).
However, we were surprised to find a surface area
approximately one-half of what would be expected, based on
the functionalized PAF-1 derivatives presented above (570 m2/
g BET vs 965−1400 m2/g). While inefficient polymerization
2436
dx.doi.org/10.1021/ja4105478 | J. Am. Chem. Soc. 2014, 136, 2432−2440

Journal of the American Chemical Society
Article
for substitution at both ortho positions across the same
biphenyl bond, the spatial relationship of cooperative functional
groups could be finer tuned by adjusting network packing. As a
practical advantage, this strategy also allows high-performance
materials to be generated without recourse to using expensive
and air-sensitive Ni(cod)2, as in typical Yamamoto polymer-
ization50 conditions (vide infra).51
The syntheses of PAF materials fulfilling these properties are
presented in Scheme 2. Monomer synthesis begins with 2,5-
Scheme 2. Sythesis of Terephthalic Ester Monomers,
Polymerization, And Deprotection to Yield Acidic Porous
a
Networks
Figure 7. Low-temperature nitrogen adsorption characteristics of
terphenyl-linked porous organic polymers (77 K). BPP-4: green
circles, BPP-5: blue circles, BPP-6: red circles, BPP-7: yellow circles.
Open circles represent desorption data.
subsequent reacidification with 3 N HCl, delivered the desired
Brønsted acidic adsorbent [(C6H4-p-C6H2(CO2H)2-
C6H4)2(C)] (BPP-5) with a slightly increased BET surface
area of 700 m2/g. Similar polymerization with 1-nonyl
terephthalate ester 10 produced the nonporous polymer
[(C6H4-p-C6H2(CO2n-C9H19)2-C6H4)2(C)] (BPP-6) However,
after side chain cleavage, the resulting acidic material, [(C6H4-p-
C6H2(CO2H)2-C6H4)2(C)] (BPP-7), displayed porosity effec-
tively identical to the acidic polymer derived from 9 (705 m2/g
BPP-7 vs 700 m2/g BPP-5). Again, two crucial features
suggested that these frameworks were highly interpenetrated.57
Residual bromine was below the limit of detection by EDX
spectroscopy, suggesting an efficient polymerization to form
terphenyl bridges. Most importantly, whereas a noninterpen-
trated structure would be expected to possess a pore size on the
order of 2 nm, the dominant pore size observed in methyl ester-
derived acidic polymer BPP-5 is near the lower limit of N2
detection, at ∼5.4−5.6 Å (Figure 8). Interestingly, this pore size
is entirely absent in the 1-nonyl ester-derived material BPP-7,
likely the result of the larger 1-nonanol side chains preventing
extremely dense network packing. A corresponding increase is
seen in pore sizes ranging from ∼6.0−6.5 Å.
Not only do these polymers display a high density of
functional groups (up to 6.0 mmol/g for an ideal polymer-
ization), they also function to simultaneously interrogate the
hypotheses mentioned in our design plan. If the ammonia
uptake demonstrated by the materials is only dependent upon
gas-phase pKa,58−61 then the uptake at low pressure would be
expected to be vastly inferior to the much more acidic PPN-6-
SO3H. However, if cooperative interactions could significantly
enhance the adsorption enthalpy, the opposite could be
observed. Furthermore, this would strongly suggest that a
cooperative interaction between individual polymer networks is
possible, given the para substitution in these particular
materials. Additionally, if the interaction is between polymer
networks, then the packing effects that are suggested by
differences in pore size distribution should also affect ammonia
uptake. Ammonia adsorption data for these materials are shown
in Figure 9. Significant differences can be observed at both low
a
Conditions: (a) (COCl)2, PhH; ROH, pyridine, CH2Cl2, 79−89%.
(b) Pd(dppf)Cl2, B2pin2, KOAc, 1,4-dioxane, 61−67%. (c) 9 or 10, 11,
cat. 12, K2CO3, THF/H2O, 84−86%. (d) KOH, DMSO, 81−86%.
dibromoterephthalic acid (6), which is commercially available
or easily generated from inexpensive 1,4-dibromo-2,5-dime-
thylbenzene.52 Esterification with methanol (7) or 1-nonanol
(8) and Miyaura borylation53 delivers functional cross-coupling
partners 9 and 10. Suzuki polymerization with the same
monomer used in PAF-1 synthesis (11) is enabled by
Buchwald’s recently disclosed palladacycle precatalyst 12,54
delivering insoluble materials at acceptably low catalyst loadings
(0.7−1.1 mol % relative to new carbon−carbon bonds). Less
reactive catalysts that have been successful in the synthesis of
other PAFs (e.g., Pd(PPh3)4)55,56 were ineffective in this
challenging case. Nitrogen adsorption data obtained for the
acidic polymers, and their ester precursors, are shown in Figure
7. The polymer derived from methyl terephthalic ester 9,
[(C6H4-p-C6H2(CO2CH3)2-C6H4)2(C)] (BPP-4), presented a
BET surface area of 665 m2/g. Saponification under extremely
vigorous conditions (KOH, anhydrous DMSO, 150 °C), and
2437
dx.doi.org/10.1021/ja4105478 | J. Am. Chem. Soc. 2014, 136, 2432−2440

Journal of the American Chemical Society
Article
Figure 8. Effect of terephthalate ester size on pore size distribution
after saponification. Methyl ester derived polymer BPP-5 = blue. 1-
Nonyl ester derived polymer BPP-7 = yellow.
Figure 9. Comparison of room temperature (298 K) ammonia
adsorption for terephthalic acid-based porous organic polymers and
PPN-6-SO3H. Carboxylic acid derived from methyl ester (BPP-5):
blue circles. Carboxylic acid derived from 1-nonyl ester (BPP-7):
yellow circles. Sulfonic acid (PPN-6-SO3H): red circles. Open circles
represent desorption data.
and high pressure. Most importantly, the carboxylic acid-
substituted materials display vastly increased ammonia
adsorption, especially at low pressure, where binding to the
acidic functional groups takes place. Given that these results
cannot be explained by gas-phase acidity (e.g., gas-phase
ionization ΔG for C6H5CO2H = 1393 kJ/mol vs C6H5SO3H <
1318 kJ/mol).58−61 Additional information that can be
gathered from including desorption data indicates that kinetics
effects cannot be ignored in these materials. Even with
extended equilibration times, BPP-5 displays lower uptake at
and below 1 mbar, yet retains more ammonia at these pressures
upon desorption. This strongly suggests that there is a
difference in kinetics resulting from the differing pore size
distributions. Very small porespotentially, even below our
limit of detection with N2 adsorption analysisare difficult to
access, and as such, adsorption does not occur on a reasonable
time scale until higher pressures are applied. At 1000 mbar,
BPP-5 does display a meaningful increase over BPP-7 (17.7
mmol/g for BPP-5 vs 16.1 mmol/g for BPP-7), effectively
matching Fe-MIL-101-SO3H. Upon desorption, the higher
fraction of very small pore binding sites in BPP-5 leads to
greater amounts of residual adsorption (5.3 mmol/g at 203
ppm for BPP-5 vs 4.5 mmol/g at 204 ppm desorption for BPP-
7).
4. CONCLUSIONS
MOFs featuring highly acidic Brønsted acid sites have been
developed in the context of proton conducting materials62−64
and heterogeneous catalysis.65 We have shown herein that these
materials can display excellent ammonia adsorption character-
istics, which are competitive with more common approaches
based on Lewis acidic adsorbents. Porous organic polymers
based on the predictable modular assembly of rigid building
blocks66 have the potential to provide the same designed
binding environments found in MOFs, while affording the
exceptional chemical stability more commonly associated with
traditional adsorbents such as zeolites and activated carbons,
which may be required for challenging applications such as
high-pressure reversible ammonia storage. The foregoing
results detail a general strategy that can provide access to
porous adsorbents displaying a combination of functional group
density and chemical stability that rivals current champion
materials. Suzuki polymerization using readily accessible
precursors, along with modern catalysts showing high activity
and functional group tolerance, will allow for the introduction
of diverse binding sitesincluding those that may be difficult
to incorporate in MOFs. As a specific realization of this
2438
dx.doi.org/10.1021/ja4105478 | J. Am. Chem. Soc. 2014, 136, 2432−2440

Journal of the American Chemical Society
Article
(11) Permissible Exposure Limits for Chemical Contaminants; Cal/
OSHA: Oakland, CA; http://www.dir.ca.gov/title8/5155table_ac1.
html (accessed on October 3, 2013).
(12) MacDonald, S. A.; Hinsberg, W. D.; Wendt, H. R.; Clecak, N. J.;
Willson, C. G.; Snyder, C. D. Chem. Mater. 1993, 5, 348.
(13) Lin, I.-K.; Bai, H.; Wu, B.-J. Aerosol Air Qual. Res. 2010, 10, 245.
(14) Kitajima, H.; Shiramizu, Y. IEEE Trans. Semicond. Manuf. 1997,
10, 267.
concept, the successful invention of a high-capacity ammonia
adsorbent has been demonstrated. The superior stability of this
material under extremely basic conditions suggests that
multiple adsorption/desorption cycles and long-term ammonia
exposure will be tolerated. Fundamental studies toward
expanding the range of pore sizes available in these materials,
further characterization of the proposed binding site, and
additional application-driven investigations are currently under-
way.
(15) Schuth, F.; Palkovits, R.; Schlogl, R.; Su, D. S. Energ. Environ. Sci.
2012, 5, 6278.
(16) Li, J.; Chang, H.; Ma, L.; Hao, J.; Yang, R. T. Catal. Today 2011,
175, 147.
ASSOCIATED CONTENT
■
(17) Brandenberger, S.; Krocher, O.; Tissler, A.; Althoff, R. Catal.
̈
S
* Supporting Information
Rev. 2008, 50, 492.
Full experimental procedures, thermal gravimetric analysis,
infrared, 1H, 13C, and 11B NMR data, powder X-ray diffraction,
EDX spectra, and elemental analyses are provided. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(18) O’Keeffe, M.; Yaghi, O. M. Chem. Rev. 2012, 112, 675.
(19) Getman, R. B.; Bae, Y.-S.; Wilmer, C. E.; Snurr, R. Q. Chem. Rev.
2012, 112, 703.
(20) Sumida, K.; Rogow, D. L.; Mason, J. A.; McDonald, T. M.;
Bloch, E. D.; Herm, Z. R.; Bae, T.-H.; Long, J. R. Chem. Rev. 2012,
112, 724.
(21) Suh, M. P.; Park, H. J.; Prasad, T. K.; Lim, D.-W. Chem. Rev.
2012, 112, 782.
AUTHOR INFORMATION
■
Corresponding Author
jrlong@berkeley.edu
(22) Li, J.-R.; Sculley, J.; Zhou, H.-C. Chem. Rev. 2012, 112, 869.
(23) Wang, C.; Xie, Z.; de Krafft, K. E.; Lin, W. J. Am. Chem. Soc.
2011, 133, 13445.
(24) Yanai, N.; Kitayama, K.; Hijikata, Y.; Sato, H.; Matsuda, R.;
Kubota, Y.; Takata, M.; Mizuno, M.; Uemura, T.; Kitagawa, S. Nat.
Mater. 2011, 10, 787.
(25) Glover, T. G.; Peterson, G. W.; Schindler, B. J.; Britt, D.; Yaghi,
O. Chem. Eng. Sci. 2011, 66, 163.
(26) Doonan, C. J.; Tranchemontagne, D. J.; Glover, T. G.; Hunt, J.
R.; Yaghi, O. M. Nat. Chem. 2010, 2, 235.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was supported through the Center for Gas
Separations Relevant to Clean Energy Technologies, an Energy
Frontier Research Center funded by the U.S. Department of
Energy, Office of Science, Office of Basic Energy Sciences
under award DE-SC0001015. We thank the 11-BM and 17-BM
staff at the Advanced Photon Source at Argonne National
Laboratory for assisting with powder X-ray diffraction experi-
ments. Use of the Advanced Photon Source at Argonne
National Laboratory was supported by the U.S. Department of
Energy, Office of Science, Office of Basic Energy Sciences,
under contract no. DE-AC02-06CH11357. We thank the 11-
BM and 17-BM staff at the Advanced Photon Source at
Argonne National Laboratory, Dr. Wendy L. Queen, and Jarad
A. Mason for assisting with powder X-ray diffraction experi-
ments. Dr. C.-N. Kuo is thanked for helpful discussions.
(27) Furukawa, H.; Ko, N.; Go, Y. B.; Aratani, N.; Choi, S. B.; Choi,
̈
E.; Yazaydin, A. O.; Snurr, R. Q.; O’Keeffe, M.; Kim, J.; Yaghi, O. M.
Science 2010, 329, 424.
(28) Spanopoulos, I.; Xydias, P.; Malliakas, C. D.; Trikalitis, P. N.
Inorg. Chem. 2013, 52, 855.
(29) Saha, D.; Deng, S. J. Colloid Interface Sci. 2010, 348, 615.
(30) Petit, C.; Bandosz, T. J. Adv. Funct. Mater. 2010, 20, 111.
(31) Cavka, J. H.; Jakobsen, S. r.; Olsbye, U.; Guillou, N.; Lamberti,
C.; Bordiga, S.; Lillerud, K. P. J. Am. Chem. Soc. 2008, 130, 13850.
(32) Morris, W.; Doonan, C. J.; Yaghi, O. M. Inorg. Chem. 2011, 50,
6853.
(33) Rose, M.; Bohlmann, W.; Sabo, M.; Kaskel, S. Chem. Commun.
2008, 2462.
(34) Ben, T.; Ren, H.; Ma, S.; Cao, D.; Lan, J.; Jing, X.; Wang, W.;
Xu, J.; Deng, F.; Simmons, J. M.; Qiu, S.; Zhu, G. Angew. Chem., Int.
Ed. 2009, 48, 9457.
(35) Stockel, E.; Wu, X.; Trewin, A.; Wood, C. D.; Clowes, R.;
Campbell, N. L.; Jones, J. T. A.; Khimyak, Y. Z.; Adams, D. J.; Cooper,
A. I. Chem. Commun. 2009, 212.
(36) Weston, M. H.; Peterson, G. W.; Browe, M. A.; Jones, P.; Farha,
O. K.; Hupp, J. T.; Nguyen, S. T. Chem. Commun. 2013, 49, 2995.
(37) Peterson, G.; Farha, O.; Schindler, B.; Jones, P.; Mahle, J.;
Hupp, J. J. Porous Mater. 2012, 19, 261.
REFERENCES
■
(1) Chen, B.; Xiang, S.; Qian, G. Acc. Chem. Res. 2010, 43, 1115.
(2) Goeppert, A.; Czaun, M.; May, R. B.; Prakash, G. K. S.; Olah, G.
A.; Narayanan, S. R. J. Am. Chem. Soc. 2011, 133, 20164.
(3) McDonald, T. M.; Lee, W. R.; Mason, J. A.; Wiers, B. M.; Hong,
C. S.; Long, J. R. J. Am. Chem. Soc. 2012, 134, 7056.
(4) Didas, S. A.; Kulkarni, A. R.; Sholl, D. S.; Jones, C. W.
ChemSusChem 2012, 5, 2058.
(5) Demessence, A.; D’Alessandro, D. M.; Foo, M. L.; Long, J. R. J.
Am. Chem. Soc. 2009, 131, 8784.
(6) McDonald, T. M.; D’Alessandro, D. M.; Krishna, R.; Long, J. R.
Chem. Sci. 2011, 2, 2022.
(7) Lu, W.; Sculley, J. P.; Yuan, D.; Krishna, R.; Wei, Z.; Zhou, H.-C.
Angew. Chem., Int. Ed. 2012, 51, 7480.
(8) Das, A.; Choucair, M.; Southon, P. D.; Mason, J. A.; Zhao, M.;
Kepert, C. J.; Harris, A. T.; D’Alessandro, D. M. Microporous
Mesoporous Mater. 2013, 174, 74.
(9) Danon, A.; Stair, P. C.; Weitz, E. J. Phys. Chem. C 2011, 115,
11540.
(38) Lu, W.; Yuan, D.; Sculley, J.; Zhao, D.; Krishna, R.; Zhou, H.-C.
J. Am. Chem. Soc. 2011, 133, 18126.
(39) Trofimov, B. A. Sulfur Reports 1992, 11, 207.
(40) Merino, E.; Verde-Sesto, E.; Maya, E. M.; Iglesias, M.; San
́
chez,
F.; Corma, A. Chem. Mater. 2013, 25, 981.
(41) Akiyama, G.; Matsuda, R.; Sato, H.; Takata, M.; Kitagawa, S.
Adv. Mater. 2011, 23, 3294.
(42) Menke, J. B. Recl. Trav. Chim. Pays-Bas 1925, 44, 141.
(43) Helminen, J.; Helenius, J.; Paatero, E.; Turunen, I. J. Chem. Eng.
Data 2001, 46, 391.
(44) Petit, C.; Bandosz, T. J. J. Phys. Chem. C 2009, 113, 3800.
(45) Petit, C.; Kante, K.; Bandosz, T. J. Carbon 2010, 48, 654.
(46) Petit, C.; Bandosz, T. J. Microporous Mesoporous Mater. 2008,
114, 137.
(10) Planas, N.; Dzubak, A. L.; Poloni, R.; Lin, L.-C.; McManus, A.;
McDonald, T. M.; Neaton, J. B.; Long, J. R.; Smit, B.; Gagliardi, L. J.
Am. Chem. Soc. 2013, 135, 7402.
2439
dx.doi.org/10.1021/ja4105478 | J. Am. Chem. Soc. 2014, 136, 2432−2440

Journal of the American Chemical Society
Article
(47) Yu, D.; Ghosh, P.; Snurr, R. Q. Dalton Trans. 2012, 41, 3962.
(48) Ermer, O. J. Am. Chem. Soc. 1988, 110, 3747.
́
(49) Zhang, Y.-B.; Su, J.; Furakawa, H.; Yun, Y.; Gandara, F.; Duong,
A.; Zou, X.; Yaghi, O. M. J. Am. Chem. Soc. 2013, 135, 16336.
(50) Kanbara, T.; Saito, N.; Yamamoto, T.; Kubota, K. Macro-
molecules 1991, 24, 5883.
(51) Palladium-catalyzed cross coupling has been used successfully to
produce terphenyl-linked PAFs that do not have functional groups:
Rose, M.; Klein, N.; Bohlmann, W.; Bohringer, B.; Fichtner, S.; Kaskel,
S. Soft Matter 2010, 6, 3918.
(52) Yao, Y.; Tour, J. M. Macromolecules 1999, 32, 2455.
(53) Ishiyama, T.; Murata, M.; Miyaura, N. J. Org. Chem. 1995, 60,
7508.
(54) Kinzel, T.; Zhang, Y.; Buchwald, S. L. J. Am. Chem. Soc. 2010,
132, 14073.
(55) Jing, X.; Sun, F.; Ren, H.; Tian, Y.; Guo, M.; Li, L.; Zhu, G.
Microporous Mesoporous Mater. 2013, 165, 92.
(56) Zhang, K.; Kopetzki, D.; Seeberger, P. H.; Antonietti, M.; Vilela,
F. Angew. Chem., Int. Ed. 2013, 52, 1432.
(57) The illustration in Scheme 2 is meant to illustrate multifold
interpenetration (>2×), rather than precise and uniform 3-fold
interpenetration.
(58) McMahon, T. B.; Kebarle, P. J. Am. Chem. Soc. 1977, 99, 2222.
(59) Koppel, I. A.; Taft, R. W.; Anvia, F.; Zhu, S.-Z.; Hu, L.-Q.; Sung,
K.-S.; DesMarteau, D. D.; Yagupolskii, L. M.; Yagupolskii, Y. L. J. Am.
Chem. Soc. 1994, 116, 3047.
(60) Smith, J. D.; O’Hair, R. A. J.; Williams, T. D. Phosphorus, Sulfur
Silicon Relat. Elem. 1996, 119, 49.
(61) Taft, R. W.; Bordwell, F. G. Acc. Chem. Res. 1988, 21, 463.
(62) Sadakiyo, M.; Yamada, T.; Kitagawa, H. J. Am. Chem. Soc. 2009,
131, 9906.
(63) Hurd, J. A.; Vaidhyanathan, R.; Thangadurai, V.; Ratcliffe, C. I.;
Moudrakovski, I. L.; Shimizu, G. K. H. Nat. Chem. 2009, 1, 705.
(64) Yoon, M.; Suh, K.; Natarajan, S.; Kim, K. Angew. Chem., Int. Ed.
2013, 52, 2668.
(65) Lee, J.; Farha, O. K.; Roberts, J.; Scheidt, K. A.; Nguyen, S. T.;
Hupp, J. T. Chem. Soc. Rev. 2009, 38, 1450.
(66) Zou, X.; Ren, H.; Zhu, G. Chem. Commun. 2013, 49, 3911.
2440
dx.doi.org/10.1021/ja4105478 | J. Am. Chem. Soc. 2014, 136, 2432−2440