Finely Tuning the Free Volume Architecture in Iptycene-Containing Polyimides for Highly Selective and Fast Hydrogen Transport

Article abstract of DOI:10.1021/acs.macromol.6b00485

Iptycene-based polyimides have attracted extensive attention recently in the membrane gas separation field due to their unique structural hierarchy and chemical characteristics that enable construction of well-defined yet tailorable free volume architecture for fast and selective molecular transport. We report here a new series of iptycene-based polyimides that are exquisitely tuned in the monomer structure to afford preferred microcavity architecture for hydrogen transport. In particular, a triptycene-containing dianhydride (TPDAn) was prepared to react with two iptycene-containing diamines (i.e., TPDAm and PPDAm) or 2,2′-bis(3-amino-4-hydroxyphenyl)hexafluoropropane (6FAP) to produce entirely or partially iptycene-based polyimides. The incorporation of iptycene units effectively disrupted chain packing, which resulted in ultrafine microporosity in the membranes with a desired bimodal size distribution with maxima at ～3 and ～7 ?, respectively. Depending on the combination of diamine and dianhydride, the microporosity was feasibly tuned and optimized to meet the needs of challenging H₂ separations, especially for H₂/N₂ and H₂/CH₄ gas pairs. Particularly, a H₂ permeability of 27 barrers and H₂/N₂ and H₂/CH₄ selectivities of 142 and 300, respectively, were obtained for TPDAn-6FAP.

Full text of DOI:10.1021/acs.macromol.6b00485

Article
pubs.acs.org/Macromolecules
Finely Tuning the Free Volume Architecture in Iptycene-Containing
Polyimides for Highly Selective and Fast Hydrogen Transport
†
†
†
‡
‡
Shuangjiang Luo, Jennifer R. Wiegand, Barbara Kazanowska, Cara M. Doherty, Kristina Konstas,
‡
,†
Anita J. Hill, and Ruilan Guo*
†Department of Chemical and Biomolecular Engineering, University of Notre Dame, Notre Dame, Indiana 46556, United States
‡The Commonwealth Scientific and Industrial Research Organization (CSIRO) Manufacturing, Private Bag 10, Clayton South,
Victoria 3169, Australia
*
S Supporting Information
ABSTRACT: Iptycene-based polyimides have attracted ex-
tensive attention recently in the membrane gas separation field
due to their unique structural hierarchy and chemical
characteristics that enable construction of well-defined yet
tailorable free volume architecture for fast and selective
molecular transport. We report here a new series of
iptycene-based polyimides that are exquisitely tuned in the
monomer structure to afford preferred microcavity architecture
for hydrogen transport. In particular, a triptycene-containing
dianhydride (TPDAn) was prepared to react with two
iptycene-containing diamines (i.e., TPDAm and PPDAm) or
2
,2′-bis(3-amino-4-hydroxyphenyl)hexafluoropropane (6FAP) to produce entirely or partially iptycene-based polyimides. The
incorporation of iptycene units effectively disrupted chain packing, which resulted in ultrafine microporosity in the membranes
with a desired bimodal size distribution with maxima at ∼3 and ∼7 Å, respectively. Depending on the combination of diamine
and dianhydride, the microporosity was feasibly tuned and optimized to meet the needs of challenging H separations, especially
2
for H /N and H /CH gas pairs. Particularly, a H permeability of 27 barrers and H /N and H /CH selectivities of 142 and
2
2
2
4
2
2
2
2
4
300, respectively, were obtained for TPDAn-6FAP.
7
−12
1
. INTRODUCTION
KAUST-PIs)
base,
and polyimides containing Tro
̈
ger’s
13,14
were reported, some of which demonstrated high
Precise control over the free volume architecture (size and size
distribution) in polymer materials is central to fast and selective
molecular transport that concerns many critical separation
applications such as membrane gas separations implemented in
hydrogen recovery (H /CO , H /N , H /CH ), air separation
separation performance for several gas pairs exceeding the
Robeson’s upper bounds. However, it has been observed that
the significant improvement in permeabilities of these new
polyimides was not always accompanied by the improvement of
selectivities, and the gas separation properties, especially H2-
related separations, of some of these new polyimides are still
challenged by the permeability−selectivity trade-off.
2
2
2
2
2
4
(
O /N ), natural gas purification (CO /CH ), and carbon
2 2 2 4
1
−4
capture (CO /N ).
However, polymeric membranes are
2
2
frequently challenged by a trade-off between permeability and
selectivity, characterized by the well-known Robeson’s upper
bounds, and the challenge originates largely from the limited
amount of free volume and very broad free volume size
15−17
Our recent research
as well as the studies reported by
showed that polyimide membranes
10,11,18−22
5
,6
other groups
prepared from iptycene-based monomers have great potential
as gas separation membranes. It has been shown that iptycene
structures are instrumental in generating large fractional free
volume as well as constructing a well-defined free volume
architecture that deliver impressive gas separation properties in
the resulting polyimide membranes for various separation
needs. In particular, we demonstrated that the intrinsic
molecular cavities (i.e., internal free volume) defined by the
shape of the iptycene molecules, especially the pentiptycene
structure, provide great opportunity for architectural tailoring
1
−4
distribution.
In this regard, high free-volume macro-
molecules with well-defined yet tailorable free volume
architecture are potential materials for high performance
polymeric membranes for gas separations.
Polyimides (PIs) are among the most widely used materials
for membrane gas separation, owing to their exceptional
thermal and chemical stability, excellent mechanical properties,
a large family of monomer variants, and good gas selectivities.
However, most common polyimides have relatively low
fractional free volume (FFV) and thus low gas permeabilities,
limiting their practical applications in large-scale gas separa-
tions. Recently, high free-volume polyimides, such as
polyimides with intrinsic microporosity (PIM-PIs and
Received: March 7, 2016
Revised: April 6, 2016
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.macromol.6b00485
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 1. Synthesis of Triptycene-Containing Dianhydride (TPDAn)
Scheme 2. Synthesis of Iptycene-Based Polyimides Based on TPDAn
2
3,24
through incorporation of different substituent groups that may
partially fill the molecular cavities depending on their size. As
such, precise control and fine manipulation of molecular sieving
properties of the membranes have been achieved. In these
studies, the aromatic polyimides containing iptycene structures
were prepared via condensation polymerization between
commercial 6FDA dianhydride and custom-synthesized ipty-
cene-based diamines that were systematically varied in the
structure (i.e., triptycene- or pentiptycene-based, CH - or CF -
containing dianhydride (TPDAn)
and our previously
reported iptycene-containing diamines as well as a −C(CF ) −
3
2
bridged diamine (i.e., 6FAP, 2,2′-bis(3-amino-4-hydroxy-
phenyl)hexafluoropropane) that mirrors 6FDA dianhydride
structure in comparable polyimides. This new series of
triptycene−dianhydride based polyimide membranes are
discussed in depth regarding their synthesis (both the
monomer and polymers) and characterization with a focus on
establishing the correlation between the monomer/polymer
structure and free volume architecture in the membranes by
comparing them with our previously reported iptycene-
containing polyimides. Pure-gas transport properties in terms
of permeability and selectivity are tested for H , N , Ar, CH ,
3
3
1
5,16
substituted).
We have found that the detailed monomer
structure has significant impact on the chain rigidity, fractional
free volume, and consequent gas transport properties of the
solution-cast polymeric membranes. A study reported by
Pinnau et al. on a series of triptycene-based PIM−PIs has
also demonstrated the impact of the dianhydride−diamine
2
2
4
and CO at 10 atm and 35 °C, which are assessed via
2
comprehensive analyses of diffusivity and solubility coefficients
to provide fundamental understanding on their transport
properties. As such, combining with the data we reported
11
combination on the properties of polyimide membranes.
In this study, we extend the efforts to develop a new series of
entirely iptycene-based polyimides derived from a triptycene-
15,16
earlier on iptycene−diamine-based polyimides
and 6FDA-
B
DOI: 10.1021/acs.macromol.6b00485
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
2
5
6
FAP polyimide, this study provides a broad set of
NMR (500 MHz, CDCl
J = 8.90 Hz, 2H), 6.84 (s, 2H), 6.87 (s, 2H), 7.01−7.02 (m, 4H),
3
): δ 5.56 (s, 2H), 5.59 (s, 2H), 6.66−6.68 (d,
dianhydride−diamine combinations to elucidate the respective
role of iptycene structure, either as diamine or dianhydride of
the imide structure, in tailoring free volume architecture and
impacting gas transport properties of the membranes.
7
7
7
.02−7.04 (m, 4H), 7.20−7.22 (m, 4H), 7.24−7.25 (m, 2H), 7.25−
.26 (m, 4H), 7.43−7.45 (m, 2H), 7.50−7.51 (d, J = 2.10 Hz, 2H),
.90−7.91 (d, J = 2.10 Hz, 2H), 7.94−7.96 (d, J = 8.30 Hz, 2H). ATR-
−1
FTIR (membrane, ν, cm ): 1782 (imide asym CO str), 1723
(
imide sym CO str), 1374 (imide −C−N), 720 (imide CO
2
. EXPERIMENTAL SECTION
.1. Materials. Triptycene- and pentiptycene-based diamine
monomers (i.e., TPDAm and PPDAm) were prepared according to
our previously reported procedures, which were dried at 100 °C under
vacuum before use.
fluoropropane (6FAP, >98.5%) was purchased from Akron Polymer
Systems and dried at 65 °C under vacuum overnight before use.
Anthracene (Alfa Aesar, 97%), p-benzoquinone (Sigma-Aldrich,
98%), hydrobromic acid (Sigma-Aldrich, 48 wt %), 4-nitro-
phthalonitrile (TCI, >98.0%), anhydrous N,N-dimethylacetamide
DMAc, EMD, 99.8%), acetic anhydride (Sigma-Aldrich, >98.0%),
anhydrous pyridine (EMD, 99.8%), anhydrous 1-methyl-2-pyrrolidi-
none (NMP, Sigma-Aldrich, 99.5%), and 1,2-dichlorobenzene (o-
DCB, Sigma-Aldrich, 99%) were used as received.
bending). Molecular weight measured by size exclusion chromatog-
4
raphy (SEC) (DMF eluent, polystyrene standards): M = 1.92 × 10 g
2
n
−
1
4
−1
mol , M = 3.98 × 10 g mol , PDI = 2.08.
w
TPDAn−PPDAm was synthesized following a similar procedure
1
5,16
with TPDAn−TPDAm, and the product was obtained as light-yellow
2,2′-Bis(3-amino-4-hydroxyphenyl)hexa-
1
fibrous solid (yield 95%). H NMR (500 MHz, CDCl ): δ 5.44 (s,
3
4
H), 5.60 (s, 2H), 6.27−6.29 (d, J = 8.65 Hz) 6.41−6.43 (d, J = 8.95
Hz) (2H), 6.85 (s, 2H), 6.89−6.94 (m, 8H), 6.96−6.97 (m, 4H), 7.04
s, 4H), 7.25−7.27 (m, 8H), 7.33 (s, 2H), 7.34−7.41 (m, 2H), 7.52 (s,
H), 7.95−7.97 (d, J = 8.20 Hz, 2H), 8.00−8.02 (d, J = 10.25 Hz, 2H).
(
2
≥
−1
ATR-FTIR (membrane, ν, cm ): 1780 (imide asym CO str), 1723
imide sym CO str), 1374 (imide −C−N), 723 (imide CO
bending). Molecular weight (SEC, DMF eluent, polystyrene stand-
(
(
4
−1
4
−1
ards): M = 2.10 × 10 g mol , M = 5.15 × 10 g mol , PDI = 2.45.
n
w
TPDAn−6FAP was synthesized via solution thermal imidization
instead of chemical imidization to avoid the reactions between
hydroxyl groups of 6FAP and acetic anhydride: diamine 6FAP (1.1969
g, 3.3 mmol) and anhydrous NMP (13 mL) were added into a flame-
dried, three-neck flask equipped with a mechanical stirrer and nitrogen
purge at room temperature. After complete dissolution of 6FAP,
dianhydride TPDAn (1.8905 g, 3.3 mmol) was added, and the flask
was warmed gradually to 80 °C and held for 30 min to dissolve
TPDAn. The solution was then cooled down to room temperature and
stirred overnight to form viscous poly(amic acid). To conduct solution
thermal imidization, a Dean−Stark trap and a reflux condenser were
connected to the flask and o-dichlorobenzene (6 mL) was added as an
azeotropic reagent. The flask was gradually heated to 190 °C and
refluxed for 12 h before the solution was cooled down and precipitated
in a mixture of methanol and water (v/v = 1/1, 500 mL). Fibrous,
white solid was collected, washed with fresh methanol, and then dried
2
.2. Synthesis of Triptycene-Containing Dianhydride.
Triptycene-based dianhydride monomer (TPDAn) was synthesized
following the previously reported procedures with some modifications,
shown in Scheme 1.
two phenolic hydroxyl groups at 1,4-positions was first constructed via
Diels−Alder cycloaddition of p-benzoquinone to anthracene followed
by reduction, which was then substituted with 4-nitrophthalonitrile via
23,24
In general, the triptycene skeleton containing
nucleophilic aromatic substitution (S Ar) reaction. The obtained
N
tetranitrile compound was hydrolyzed into tetracarboxylic acid, which
was further dehydrated into the final triptycene-based dianhydride.
Modifications to the reported procedures were made during the
hydrolysis step as follows: triptycene-based tetranitrile compound (5.0
g, 9.3 mmol) and 20 g of potassium hydroxide (KOH, 0.356 mol)
were suspended/dissolved in an ethanol/water mixture (50 mL/50
mL), heated, and refluxed for 48 h. The resulting clear solution was
acidized to pH = 1.0 with 6 M HCl solution, and white precipitates of
tetracarboxylic acid were produced, which were filtered and washed
with deionized (DI) water repeatedly. After drying under vacuum at
1
at 180 °C under vacuum overnight (2.96 g, yield 96%). H NMR (500
MHz, DMSO-d ): δ 5.71 (s, 2H), 6.96 (s, 4H), 7.01 (s, 2H), 7.08−
6
1
00 °C overnight, the obtained tetracarboxylic acid compound was
7.10 (d, J = 8.75 Hz, 2H), 7.24 (s, 2H), 7.26 (s, 2H), 7.27 (s, 4H), 7.31
(s, 2H), 7.39 (s, 2H), 7.90−7.92 (d, J = 8.20 Hz, 2H), 10.52 (s, 2H).
refluxed in acetic anhydride (15 mL) for 6 h, followed by filtration and
washing with petroleum ether. The final product of triptycene−
dianhydride (TPDAn, 4.89 g, yield 81%) was obtained as white
powder after drying at 120 °C in a vacuum oven overnight; mp
−1
ATR-FTIR (membrane, ν, cm ): 3394 (broad, −OH), 1782 (imide
asym CO str), 1729 (imide sym CO str), 1378 (imide −C−N),
724 (imide CO bending). Molecular weight (SEC, DMF eluent,
4
−1
5
(
melting point from differential scanning calorimetry at a heating rate
polystyrene standards): M
mol , PDI = 3.12.
n
= 3.97 × 10 g mol , M
w
= 1.24 × 10 g
−1
1
−1
of 5 °C min ): >300 °C. H NMR (500 MHz, DMSO-d ): δ 5.66 (s,
6
2
2
H), 6.98−7.01 (m, 4H), 7.05 (s, 2H), 7.27−7.29 (m, 4H), 7.37 (s,
2.4. Film Casting. Thin films of the iptycene-based polyimides
−1
H), 7.45−7.47 (d, J = 8.35 Hz, 2H), 8.09−8.11 (d, J = 8.40 Hz, 2H).
were prepared by casting polymer NMP solutions (∼7% w/v, g mL )
onto clean, leveled glass plates. Solvent was evaporated slowly under
an infrared lamp (Staco Energy Products Co., 120 V) at ∼60 °C
overnight; the obtained isotropic films were peeled off and soaked in
methanol for 24 h for solvent exchange and finally dried at 180 °C
under vacuum for 24 h. TPDAn-6FAP film was further thermally
treated at 300 °C under nitrogen purge for 2 h to achieve fully
imidized structure. All the films prepared are solvent-free using the
above drying procedures, which was confirmed by thermogravimetric
analysis (TGA). Film thickness (40−80 μm) was measured with a
digital micrometer, and effective area of the films in the gas permeation
tests was determined using a digital scanner (LiDE120, Canon) and
ImageJ software.
2.3. Synthesis of Iptycene-Based Polyimides. Iptycene-based
polyimides were prepared by condensation polymerization between
triptycene-containing dianhydride (TPDAn) and iptycene-containing
diamine (i.e., TPDAm and PPDAm) or 6FAP through either chemical
imidization or solution thermal imidization (Scheme 2). Specifically,
TPDAn−TPDAm was synthesized using the following procedures:
TPDAm (1.1731 g, 2 mmol) and anhydrous DMAc (12 mL) were
added at room temperature into a flame-dried, three-necked flask
equipped with a nitrogen inlet and a mechanical stirrer. After TPDAm
was dissolved by stirring the mixture for several minutes, an equimolar
amount of TPDAn (1.2259 g, 2 mmol) was added, and the flask was
immersed in an oil bath, warmed gradually to 80 °C, and held for 30
min to dissolve TPDAn. The oil bath was then removed, and the
solution was cooled down to room temperature and stirred overnight
under N purge to form a viscous poly(amic acid) solution. Cyclization
of poly(amic acid) into polyimide was done by adding acetic anhydride
1.2 mL) and pyridine (1.2 mL), and the mixture was stirred for
another 24 h to allow complete imidization. The resulting polyimide
solution was precipitated in excess amount of methanol (500 mL) to
give fibrous polyimide product, which was collected, washed with fresh
methanol, and then dried at 160 °C under vacuum overnight giving
2.5. Characterization Methods. 1H NMR spectra of both
monomers and polymers were recorded on a Bruker 500 spectrometer
using either deuterated chloroform or deuterated dimethyl sulfone as a
solvent. The attenuated total reflection mode Fourier transform
infrared (ATR-FTIR) spectra of the films were acquired on a Jasco
2
(
−
1
FT/IR-6300 spectrometer with a resolution of 4 cm and 64 scans.
Molecular weight and molecular weight distribution of the polyimides
were measured by size exclusion chromatography (SEC, Waters GPC
System) using polystyrene as an external standard and DMF as the
eluent. A Waters 515 HPLC pump and three Polymer Standards
1
white solid as TPDAn−TPDAm polyimide (2.26 g, yield 94%). H
C
DOI: 10.1021/acs.macromol.6b00485
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1
Figure 1. H NMR spectra of iptycene-based polyimides (note: CDCl was used as the solvent for TPDAn-TPDAm and TPDAn-PPDAm and
3
DMSO-d for TPDAn-6FAP).
6
4
3
2
Service (PSS) columns (GRAM, 10 , 10 , and 10 Å) were equipped in
in the free volume elements of the films. The size of free volume
elements was calculated using the Tao−Eldrup equation:
−
1
29,30
the DMF SEC at 55 °C with a DMF flow rate of 1 mL min , and a
Waters 2414 refractive index detector was connected using PSS
WinGPC 7.5 software.
Thermogravimetric analysis (TGA) was carried out under a
nitrogen atmosphere using a TGA Q500 instrument (TA Instruments)
at a heating rate of 10 °C min . Differential scanning calorimetry
DSC) analyses were performed under nitrogen purge (50 mL min )
on a DSC Q2000 (TA Instruments) at a heating rate of 10 °C min
⎡
⎛
⎞⎤
⎟⎥
τ = 1 ⎢1 −
R
R0
1
2π
2πR
0
+
sin⎜
2
⎢
⎣
⎝ R ⎠⎥
⎦
(2)
−1
−
1
where τ is the lifetime, R is the radius of the pores, and R
the radius of potential well, where ΔR was determined to be 1.66 Å
due to the thickness of the electron layer within the potential well. The
0
= R + ΔR is
(
−
1
−
1
and cooling rate of 20 °C min , and the thermal transition
temperatures were reported based on the second heating cycle in
the temperature range of 150−400 °C. Glass transition temperature
31
pore size distributions were calculated using PAScual software.
Pure gas permeabilities of the membranes were measured at 35 °C
with five gases (H , N , Ar, CH , and CO ) using the constant-volume
2
2
4
2
(
T ) was determined based on the automatic mode of TA Universal
g
32
variable-pressure method. All membranes were degassed on both
sides overnight before permeation tests. Upstream pressure was
maintained at a predetermined value (i.e., 3.0, 6.4, 9.8, 13.0, and 16.6
atm), and the downstream pressure increase was recorded as a
function of time. Gas permeability was then calculated based on the
steady-state rate of pressure increase in the fixed downstream volume
as follows:
Analysis software. Wide-angle X-ray diffraction (WAXD) patterns were
recorded in the reflection mode at 40 mA and 40 kV on a Bruker D8
Advance Davinci diffractometer with Cu Kα radiation (wavelength λ =
1
.54 Å), and the scan speed and rate were 5 s per step and 0.02° per
step, respectively.
Film densities were determined with a density kit of an analytical
balance (ML204, Mettler Toledo) in deionized (DI) water using the
buoyancy method at room temperature. Fractional free volume (FFV)
of the polymer films was calculated as follows:
⎡
⎣
⎤
⎥
V l
⎛ dp ⎞
⎛ dp ⎞
10
d
P = 10
⎢⎜
⎟
− ⎜
⎟
p TRA⎢⎝ dt ⎠
⎝ dt ⎠
⎥
⎦
up
ss
leak
(3)
V0 − 1.3Vw
FFV =
V0
(1)
where P (barrer, 1 barrer = 10−10 cm 3 (STP) cm/(cm s cmHg)) is
2
the gas permeability, l is the film thickness (cm), V is the calibrated
d
where V is the molar volume of the films derived from the density
0
3
downstream volume (cm ), p is the upstream pressure (cmHg), A is
up
measurement, and Vw is the van der Waals volume estimated from
2
26,27
the effective film area (cm ), (dp/dt) and (dp/dt) are the steady-
ss
leak
Bondi’s group contribution method.
state pressure increment in downstream and the leak rate of the system
cmHg/s), respectively; T is the test temperature (K), and R is the gas
The cavity size and size distribution within polymer membranes
(
were investigated by positron annihilation lifetime spectroscopy
3
3
−5
constant (0.278 cm cmHg/(cm (STP) K)). The ideal selectivity
αA/B) for two different gases A and B was defined as the ratio of pure
(
PALS), which was performed under vacuum (1 × 10 Torr) at
(
room temperature using an EG&G Ortec (Oak Ridge, TN) fast−fast
coincidence spectrometer. Polymer membranes were cut into ∼1 × 1
cm pieces, which were then stacked to a thickness of 2 mm on either
gas permeability of the two gases:
PA
6
22
side of the positron source (1.5 × 10 Bq of NaCl sealed in a Mylar
αA/B ≡
PB
(4)
envelope). At least five files were collected for every sample, and each
6
file contained 4.5 × 10 integrated counts. The obtained data were
2
−1
28
The apparent diffusion coefficient D (cm s ) was obtained from the
analyzed with LT v9 software using a source correction (1.609 ns
and 4.002%) and modeled as the sum of four components. The first
two components were attributed to the para-positronium self-
annihilation (τ , 0.125 ns) and the free annihilation (τ , ∼0.4 ns).
3
2
time-lag method using the equation
2
D = l
6θ
1
2
(5)
The last two components were due to ortho-positronium (o-Ps) decay
D
DOI: 10.1021/acs.macromol.6b00485
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
where l is the film thickness (cm) and θ is the lag time (s). Solubility
coefficient (S, cm (STP)/(cm atm)) was calculated via S = P/D.
groups of poly(amic acid) intermediates is observed in the
spectra, suggesting all polyimides are fully imidized.
3
3
Molecular weights and molecular weight distributions of the
iptycene-based polyimides were measured with SEC, and the
results are listed in Table 1. All the synthesized polyimides
exhibit molecular weights greater than 19 200 g mol− with
polydispersity in the range of 2.0−3.1, suggesting sufficient
reactivity and reasonably good purity of the triptycene-
dianhydride monomer synthesized in this work. It is expected
that higher molecular weight polyimides can be feasibly
obtained by applying more rigorous monomer purification
processes and/or refining polycondensation reaction condi-
tions. Thermal properties of these polymers were evaluated
with DSC and TGA, and the results are tabulated in Table 1.
No melting and crystallization peaks were observed in the DSC
curves (Figure S1 in the Supporting Information), suggesting a
completely amorphous structure of all three polyimides. Glass
transition temperatures of TPDAn-TPDAm, TPDAn-PPDAm,
and TPDAn-6FAP were determined to be 284, 350, and 305
3
. RESULTS AND DISCUSSION
.1. Polyimides Synthesis and Characterization. The
3
1
triptycene-based dianhydride monomer (TPDAn) was pre-
pared via dehydration of the intermediate tetracarboxylic acid
compound with high yield (>80%) and high purity. Iptycene-
based diamine monomers, i.e., TPDAm and PPDAm, were
15,16
synthesized according to our previous reports,
and a non-
iptycene diamine monomer (6FAP) was used for comparison.
Three iptycene-based polyimides were synthesized via two-step
condensation polymerization between TPDAn and an iptycene
diamine or 6FAP, in which poly(amic acid) intermediates were
prepared first followed by either chemical imidization to obtain
TPDAn-TPDAm and TPDAn-PPDAm polyimides or solution
thermal imidization to TPDAn-6FAP polyimide. TPDAn-6FAP
films were further treated at 300 °C under a nitrogen purge in a
furnace to achieve complete imidization. All polyimides have
1
fully imidized structures as confirmed by H NMR and ATR-
°
C, respectively, suggesting very high chain rigidity of these
FTIR spectra. As shown in Figure 1, peak assignments in the
polymers. TPDAn-PPDAm polyimide shows the highest T due
g
1
H NMR spectra clearly verify the designed chemical structures
to the presence of very bulky pentiptycene units that tend to
of the polyimides. Specifically, the characteristic bridgehead
protons of iptycene units (peaks 3 and c) are presented in the
range of δ = 5.3−5.8 ppm, suggesting successful incorporation
of iptycene units in the main chains. The isometric structures of
pentiptycene diamine (PPDAm) arising from restricted
rotation of the ether bonds are also shown in the NMR
restrict chain motion. TPDAn-TPDAm shows a lower T than
g
TPDAn-6FAP, most likely due to the flexible ether bonds of
TPDAm monomer that decrease the chain rigidity. Entirely
iptycene-based polyimides (e.g., TPDAn-TPDAm and TPDAn-
PPDAm) exhibit high thermal stability, as shown in Table 1
and Figure S1. The thermal decomposition temperatures at 5%
and 10% weight loss are in the range of 551−553 and 581−586
16
spectra, which is in consistent with our previous studies.
ATR-FTIR spectra of the polyimides (Figure 2) show the
°
C, respectively, and the char yields at 800 °C in nitrogen are
all higher than 64%. For TPDAn-6FAP polyimide, a two-step
weight loss profile was observed. This is a characteristic feature
of o-hydroxyl-containing polyimides that can be thermally
rearranged at high temperatures (>400 °C) into polybenzox-
25,33−35
azole (PBO) structures.
Studies of thermal rearrange-
ment (TR) of TPDAn−6FAP polyimide will be reported in a
different paper to avoid excessive information in this paper. It
should be noted that no weight loss was observed up to 300 °C
in all TGA curves, suggesting that all films were solvent-free
after vacuum drying and the nonsolvent extraction procedure.
Aromatic polyimides generally have poor solubility in
common organic solvents due to the rigid imide ring structure
36
and the strong interchain interactions, which frequently
results in difficulties in membrane fabrication and processing
for practical applications. However, the iptycene-based
polyimides prepared in this study exhibit excellent solubility
in a wide range of organic solvents such as DMF, chloroform,
NMP, and DMSO, shown in Table 2. The good organic
solubility of these iptycene-containing polyimides can be
attributed primarily to the Y- or H-shaped structures of
Figure 2. ATR-FTIR spectra of the iptycene-based polyimide
membranes.
−1
characteristic imide bands at ∼1784 cm (imide carbonyl,
−1
asymmetric stretching), ∼1726 cm (imide carbonyl, sym-
iptycene units as well as the bulky CF substituent groups, both
3
−1
metric stretching), ∼1374 cm (imide C−N stretching), and
of which tend to reduce the interchain interactions in the
polyimides by disrupting efficient chain packing and increasing
1
∼
720 cm− (imide−carbonyl bending). No residual carbonyl
Table 1. Molecular Weights, Thermal Properties, and Fractional Free Volume (FFV) of Iptycene-Based Polyimides
a
b
c
density (g cm−3
)
polymer
M (g mol−1)
Xn
PDI (M /M )
Tg (°C)
Td,5% (°C)
Td,10% (°C)
char yield (%)
FFV
n
w
2.08
2.45
3.12
n
TPDAn-TPDAm
TPDAn-PPDAm
TPDAn-6FAP
19, 200
21, 000
39, 700
17
16
44
284
350
305
551
553
424
581
586
490
64
67
60
1.291
1.260
1.339
0.175
0.187
0.170
a
b
c
Td,5%: decomposition temperature at 5% weight loss. Td,10%: decomposition temperature at 10% weight loss. Residual weight retention after
heated to 800 °C in a nitrogen atmosphere.
E
DOI: 10.1021/acs.macromol.6b00485
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Table 2. Room Temperature Organic Solvent Solubility of Iptycene-Based Polyimides
a
solvent
b
polymer
DMF (12.1)
DMSO (12.0) NMP (11.2) DMAc (10.8) DCM (9.9) acetone (9.8) THF (9.5) CHCl (9.2) toluene (8.9)
3
TPDAn-TPDAm
TPDAn-PPDAm
TPDAn-6FAP
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
−
−
+
+
+
+
+
+
+
−
+/−
+/−
a
−1
b
Polymer solution concentration, 10 mg mL ; “+”, soluble at room temperature; “−”, insoluble at room temperature; “+/−”, partially soluble. The
numbers in parentheses denote the solubility parameters of testing solvents.
3
7
interchain distance. Particularly, the presence of bulky −C-
(
CF ) − groups in TPDAn−6FAP polyimide makes it soluble
3
2
in almost all the solvents tested at room temperature.
Additionally, the ether linkages introduced in the backbone
structures is also believed to promote organosolubility of these
entirely aromatic polyimides. As a result, these aromatic
polyimides could be easily solution-casted into dense thin
films (Figure S2).
3.2. Polymer Chain Packing and Microporosity. Free
volume architecture (i.e., size and size distribution) is
considered as a key structural parameter that determines gas
transport properties of polymer separation membranes. Multi-
ple characterization techniques were utilized to obtain a clear
picture of how polymer chain packing and consequential free
volume architecture are affected by incorporating iptycene
structures in polymer backbone. Fractional free volume (FFV)
values of these membranes were estimated using group
contribution method coupled with density measurements, and
the results are shown in Table 1. The iptycene-based
polyimides easily fall into the family of high FFV polymers
with FFV in the range of 0.170−0.187, which are higher than
commercial membrane materials such as Matrimid polyimide
Figure 3. WAXD patterns of iptycene-based polyimides (A, B, and C
designate diffraction peak maxima).
Table 3. WAXD Diffraction Peak Positions (2θ) and
Calculated d-Spacing Values of Iptycene-Based Polyimides
and Some Literature Data for Comparison
38
(0.17) and polysulfones (0.13−0.17). Entirely iptycene-based
polyimides, i.e., TPDAn-TPDAm and TPDAn-PPDAm, have
higher fractional free volume than partially iptycene-based
polyimide of TPDAn-6FAP, indicating that the incorporation
of iptycene structure into the polymer backbone is an effective
approach to disrupt chain packing and introduce large amount
of free volume. In particular, TPDAn−PPDAm polyimide
containing both triptycene and pentiptycene units shows the
highest FFV, which is due to the larger internal free volume of
2θ (deg)
d-spacing (Å)
polymer
A
B
C
A
B
C
TPDAn-PPDAm
TPDAn-TPDAm
TPDAn-6FAP
9.7
11.6
13.7
14.4
13.8
17.0
16.7
20.9
18.4
23.3
9.1
7.6
6.5
5.3
4.3
4.8
3.8
1
8
6
FDA-mTMPD
6.15
6.4
16
39
PIM-6FDA-OH
pentiptycene unit as well as bulkier structure of pentiptycene.
4
0
6
FDA-DADBSBF
5.2
Polymer chain packing of the polyimide membranes was
investigated with wide-angle X-ray diffraction (WAXD);
diffraction patterns are shown in Figure 3, and the peak
positions and calculated d-spacing values are tabulated in Table
rings of iptycene moieties (peak C). For the entirely triptycene-
based TPDAn-TPDAm membrane, two broad peaks were
observed, which represent the main interchain distance
disrupted by triptycene units and the average interchain
distance of imide backbone, while one dominant broad
amorphous halo was detected for TPDAn-6FAP, a partially
triptycene-based polyimide. The main interchain distance (peak
A) follows the expected trend of TPDAn-PPDAm > TPDAn-
TPDAm > TPDAn-6FAP, which is consistent with the trend of
FFV results determined by density measurements (Table 1). In
particular, the entirely iptycene-based polyimides have much
higher main interchain distance values than most of reported
3. Broad halos instead of sharp peaks were observed in all
WAXD patterns, suggesting a generally amorphous structure of
the iptycene-based polyimides. This result is consistent with
DSC results, in which no melting or crystallization peaks were
observed. Depending on the monomer structure (i.e.,
triptycene or pentiptycene, entirely or partially iptycene-
based), quite different diffraction patterns were observed for
these polyimide membranes signifying different chain packing
structures induced by iptycene moieties. TPDAn-PPDAm
membrane that contains alternating triptycene and pentipty-
cene units along the polymer backbone shows three diffraction
peaks (i.e., A, B, and C) located at 2θ = 9.7°, 16.7°, and 23.3°,
which correspond to d-spacing values of 9.1, 5.3, and 3.8 Å,
respectively. These d-spacing values are likely associated with
the main interchain distance disrupted by the rigid
pentiptycene units (peak A), the average interchain distance
of the imide backbone (peak B), and π−π stacking of arene
18,39,40
noniptycene-containing polyimides (Table 3),
confirm-
ing the disruption of tight chain packing by iptycene units.
Positron annihilation lifetime spectroscopy (PALS) measure-
ments were conducted on TPDAn-TPDAm and TPDAn-
PPDAm membranes to quantitatively analyze the cavity size
and size distribution. Two o-Ps components, i.e., τ and τ ,
3
4
which have annihilation lifetime of ∼0.9 and ∼3.0 ns,
F
DOI: 10.1021/acs.macromol.6b00485
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Table 4. Cavity Size Characterization of Iptycene-Based Polyimides Measured by PALS Compared with Some High FFV
Polymers Reported Previously
polymer
τ3 (ns)
I3 (%)
cavity diam d (Å)
τ4 (ns)
I4 (%)
cavity diam d (Å)
3
4
TPDAn-TPDAm
TPDAn-PPDAm
0.87 ± 0.10
0.94 ± 0.09
1.7
8.8 ± 1.6
8.2 ± 1.1
7.4
2.87 ± 0.38
3.10 ± 0.33
5.1
2.70 ± 0.03
3.07 ± 0.03
8.8
17.0 ± 0.4
16.9 ± 0.3
34
6.84 ± 0.04
7.36 ± 0.04
12.4
4
1
PTMSP
4
2
aPBO
1.06 ± 0.09
2.06
7.1 ± 0.7
6.15
3.51 ± 0.28
5.8
3.90 ± 0.04
6.28
12.7 ± 0.7
18.6
8.37 ± 0.04
10.6
4
1
PIM-1
respectively, were observed, indicating two populations of
micropores, or bimodal size distribution of free volume
elements, in the membranes. Based on eq 2, the average cavity
size was calculated, and the results are shown in Table 4 and
compared with a few reported high-free-volume polymers. Two
kinds of microcavities can be defined: ultrafine micropores with
an average diameter of d ∼ 3 Å (τ ∼ 0.9 ns) and micropores
resembles the shape of an “hourglass”, whereby the large
openings allow fast adsorption/desorption at the surface while
the narrow necking enables size discrimination, leading to both
10,11,33,41,42
high permeability and high selectivity.
In this study,
the small ∼3 Å cavities are likely associated with the (partially
filled) internal free volume between the “blades” of iptycene
1
5,16
3
3
units,
which might provide interconnections between the
33
with an average diameter of d ∼ 7 Å (τ ∼ 3.0 ns). As
large ∼7 Å cavities, resulting in improved gas permeabilities.
Additionally, these small cavities have an average diameter of
4
4
expected, TPDAn-PPDAm polyimide containing alternating
triptycene and pentiptycene units shows larger pore sizes (7.36
and 3.10 Å) than those of fully triptycene-based TPDAn-
TPDAm membrane (6.84 and 2.87 Å). On the other hand,
∼3 Å, which is slightly larger than the kinetic diameter of H
2
(2.89 Å) but smaller than those of other test gases in this study
(i.e., CO 3.35 Å, Ar 3.40 Å, N 3.64 Å, and CH 3.80 Å). As
2
2
4
almost the same annihilation intensities (I ) were observed for
will be discussed later, this microstructural feature is responsible
for the excellent separation performance of these iptycene
dianhydride-based polyimides in hydrogen-related gas separa-
tion.
4
these two membranes, indicating the same concentration of
micropores. The results are consistent with FFV data
determined from density measurements as well as the
interchain spacing results by WAXD. It is evident that iptycene
structure, especially bulky pentiptycene structure, is very
effective in disrupting chain packing and generating large
3
.3. Gas Permeation Properties. Pure gas permeation
tests of iptycene-based polyimides for five gases (H , N , Ar,
2
2
CH , and CO ) were conducted at 35 °C with a wide range of
4
2
10,11,18−22
micropores enabling fast gas transport.
feed pressure of 3−17 atm. As shown in Figure S3, all gas
permeabilities exhibited very weak dependence on feed
Conventional polyimides generally show only one o-Ps
component, which typically suggests a very broad free volume
pressure. In the case of CO permeation, this observation
2
3
3,42
size distribution.
In this study, two o-Ps components were
indicates that the films are resistant to plasticization up to 17
atm feed pressure. Table 5 summarizes the gas transport
properties measured at 10 atm and 35 °C along with the
calculated diffusivity and solubility coefficients. The data of
several commercial membranes are listed for comparison.
For all three iptycene-based polyimides, gas permeabilities
follow the order of P(H ) > P(CO ) > P(Ar) > P(N ) >
observed in both entirely iptycene-based polyimide membranes,
suggesting a bimodal microcavity size distribution (Figure 4).
2
2
2
P(CH ), which is consistent with the kinetic diameters of the
4
testing gases. The same trend is also observed in diffusivity
coefficient, suggesting the size sieving mechanism in these
separation membranes and the dominant role of gas diffusivity.
It is evident that the variation in monomer structure
significantly influenced the gas transport properties of the
resulting iptycene-based polyimides. Compared with TPDAn-
6
FAP, TPDAn-TPDAm exhibits higher gas permeabilities for
all gases. For example, CO permeability of TPDAn-TPDAm is
2
2
times that of TPDAn−6FAP (Table 5). This is because
TPDAm diamine has very bulky triptycene unit and CF
3
substituent groups that are more effective in disrupting chain
packing than 6FAP diamine, leading to a higher free volume
fraction. Additionally, the hydroxyl groups in 6FAP diamine
induced strong interchain hydrogen bonds, which tend to
decrease the interchain distance (Figure 3) and thus diffusivity
(Table 5). On the other hand, TPDAn-TPDAm and TPDAn-
6FAP polyimides showed very similar solubility coefficients (S),
indicating that the separation mechanism is kinetically
dominant in these membranes. As expected, replacing
triptycene unit with bulkier pentiptycene unit in the diamine
structure further increased the gas permeabilities, and TPDAn-
PPDAm shows the highest gas permeabilities for all gases
among the three polyimides. Although the solubility coefficients
Figure 4. Microcavity size distribution of TPDAn-TPDAm and
TPDAn-PPDAm from positron annihilation lifetime spectroscopy
(
PALS) measurements.
Compared with TPDAn-TPDAm, both peaks of TPDAn−
PPDAm shifted toward larger cavity sizes, which is consistent
with FFV and WAXD interchain spacing data, indicating both
larger ultrafine cavities and larger microcavities induced by the
pentiptycene units. Bimodal free volume size distribution,
which has been observed in high performance gas separation
membranes such as PIM, is considered to be beneficial for gas
separation membranes in that such free volume architecture
G
DOI: 10.1021/acs.macromol.6b00485
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Table 5. Pure Gas Permeability (P), Diffusivity Coefficient (D), Solubility Coefficient (S), and Ideal Selectivity of Iptycene-
Based Polyimides Compared with Commercial Membranes
test gas
Ar
ideal selectivity (α)
polymers
H2
CO2
N2
1.8
CH4
H /N2
H /CO2
H /CH4
2
CO /CH
2
2
2
4
a
TPDAn-PPDAm
P
96
39
3.9
35
1.5
11
53
46
2.5
29
0.09
3.7
59
64
98
26
D
S
1112
39
24
3.4
7.7
31
0.66
7.7
9.7
14
0.85
0.95
12
0.57
0.44
8.6
1.0
1.2
82
0.66
116
304
0.38
300
483
0.65
56
TPDAn-TPDAm
TPDAn-6FAP
P
D
S
36
0.31
2.7
820
95
5.2
6.1
52
0.33
5.3
4.7
7.2
5.0
5.6
4.8
10
0.63
0.38
5.3
0.39
0.19
3.7
0.87
0.09
1.3
0.85
142
170
0.85
56
0.06
5.7
P
D
S
27
628
87
5.5
9.8
22
0.33
0.55
0.39
0.25
0.15
0.32
0.51
0.25
0.15
0.28
0.1
0.07
2.5
4
3
polysulfone
P
P
P
P
14
b
44
CA-2.45
12
80
2.5
80
32
4
5−47
matrimid
18
56
1.8
64
36
4
8
aramid
24.5
245
a
Units: P, 10−10 cm (STP) cm/(cm s cmHg); D, 10 cm /s; S, cm (STP)/(cm atm). CA-2.45: cellulose acetate with degree of acetylation of
3
2
−9
2
3
3
b
2
.45.
Figure 5. Effect of the diamine and dianhydride structure on (a) H permeability and (b) H /N selectivity for iptycene-based polyimides (note: the
2
2
2
15
16
triptycene diamine of 1,4-trip_CF3 and pentiptycene diamine of PPDA(CF3) were renamed from the published ones as TPDAm and PPDAm,
respectively, in this paper to keep consistency).
16
(S) of TPDAn-PPDAm are slightly higher than those of
PPDA(CF3), differ from the ones in this study in that a
noniptycene 6FDA dianhydride was used. Given the same
diamine monomers (i.e., TPDAm or PPDAm), TPDAn-
TPDAm and TPDAn-PPDAm polyimides in this study exhibit
relatively lower gas permeabilities compared with their 6FDA-
based counterparts, despite that TPDAn has a very bulky
triptycene unit. It has been demonstrated that the hexafluoro
substituted carbon (i.e., −C(CF ) −) of 6FDA dianhydride
TPDAn-TPDAm and TPDAn-6FAP membranes, it is the
significant increases in the diffusivity coefficients (D) of
TPDAn-PPDAm membrane that are responsible for the
dramatic increase of gas permeabilities, signifying the dominant
role of diffusivity in these size-sieving membranes. The
permeability results are consistent with FFV analysis, the
interchain packing results from WAXD, and the cavity size and
size distribution analysis from PALS, as discussed previously.
The ideal selectivities generally showed a decreasing trend with
increasing gas permeabilities. However, the ideal selectivities
decreased at a slower rate than what has been typically observed
for glassy membranes, leading to improved overall separation
performance.
3
2
increases the stiffness of the polymer chain and frustrates chain
packing due to the steric hindrance from the CF groups,
3
leading to large fractional free volume in the corresponding
3
,49
polyimides.
On the other hand, the flexible ether bonds in
TPDAn structure likely promote tight interchain packing,
resulting in lower overall fractional free volume despite of the
bulkiness of triptycene structure. However, gas selectivities of
TPDAn-based polyimides are higher than those of 6FDA-based
polyimides, especially for hydrogen-related gas separations
To obtain better understanding of the respective role of
diamine and dianhydride structure in affecting chain packing
and consequent transport behavior in the iptycene-based
membranes, gas transport properties of the new membranes
in this study are compared with our previously reported
iptycene-based polyimides as shown in Figure 5. The previously
(Figure 5b). For example, H /N2 selectivity of TPDAn-
2
PPDAm is almost 2 times that of 6FDA-PPDAm given the
same diamine. A similar trend of improved H -related
2
15
reported polyimides, i.e., 6FDA-1,4-trip_CF3 and 6FDA-
selectivity is observed for other gas pairs such as H /CO
2
2
H
DOI: 10.1021/acs.macromol.6b00485
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 6. Permeability-selectivity upper bound plots of TPDAn-PPDAm (■), 6FDA-PPDAm (□), TPDAn-TPDAm (●), 6FDA-TPDAm15 (○),
1
6
2
5
TPDAn-6FAP (▲), and 6FDA-6FAP (△) for (a) H /N and (b) H /CH gas pairs. Data points of commercial materials are included for
comparison: 1, Matrimid;
2
2
2
4
45−47
43
44
2, polysulfone; 3, CA-2.45.
and H /CH as shown in Figure S4. This observation suggests
with a hydrogen permeability of 27 barrers and H /N and H /
2
4
2
2
2
that iptycene units are particularly useful in constructing
ultrafine micropores for selective hydrogen transport, which is
in agreement with the bimodal pore size distribution analysis by
PALS discussed earlier.
CH selectivities of 142 and 300, respectively.
4
4
. CONCLUSIONS
A general methodology was reported for the synthesis of
iptycene-based polyimides using a triptycene-based dianhydride
(TPDAn) with select iptycene-based diamine (TPDAm or
PPDAm) or noniptycene 6FAP. All polyimides exhibited good
solubility in common organic solvents and excellent thermal
stability. Owing to the incorporation of iptycene units, they all
showed high FFV, amorphous chain packing structures, and
bimodal microcavity size distribution with two maxima at ∼3
and ∼7 Å, respectively. The entirely iptycene-based polyimides,
i.e., TPDAn-TPDAm and TPDAn-PPDAm, exhibited higher
FFV and higher interchain distances than those of a partially
iptycene-based TPDAn-6FAP polyimide, and pentiptycene-
containing TPDAn-PPDAm polyimide showed the highest FFV
and main interchain distance indicating the effectiveness of
bulkier pentiptycene in disrupting tight chain packing. In
comparison with previous reports on 6FDA-based polyimides,
Comparisons between TPDAn-6FAP and our previously
25
reported 6FDA-6FAP further demonstrated that iptycene-
based structure is instrumental to boost both H permeability
2
and selectivities. For instance, TPDAn-6FAP not only exhibits
the highest H /N selectivity of 142 and H /CH selectivity of
2
2
2
4
3
00 but also outperforms 6FDA-6FAP in terms of H2
permeability. All above observations suggest that the bimodal
size distribution of free volume elements created/induced by
iptycene structure in the membranes holds the key to the fast
and ultraselective transport of H molecules. Compared with
2
commercially relevant polymer membranes that are typically
1
43
44
used for hydrogen recovery, such as polysulfone, CA-2.45,
4
8
and aramid, iptycene-based polyimides in this study show
much higher H permeability and, more importantly, they show
2
equal or even higher selectivities than these commercial
membranes. For example, given the same FFV, TPDAn-6FAP
exhibits a 1.5 times H permeability and around 3 times H /N
2
2
2
selectivity when compared to Matrimid (Table 5). These gas
permeation results reveal the synergetic effect of iptycene units
on the improvements of both permeability and selectivity for
hydrogen separation.
TPDAn-based polyimides showed much higher H selectivity,
2
most likely due to the bimodal size distribution of microcavities
that features a population of ultrafine microcavities. In general,
incorporation of iptycene units resulted in both higher
permeabilities and higher selectivities when compared with
commercial membrane materials for hydrogen separation.
Particularly, H permeability of 27 and H /N and H /CH
4
Gas transport data of iptycene-based polyimides were plotted
against the Robeson’s upper bounds for H /N and H /CH
4
2
2
2
5
,6
gas pairs to evaluate their gas separation performance, shown
in Figure 6. Commercial membranes, i.e., Matrimid, poly-
sulfone, and CA-2.45, are included for comparison. Generally,
iptycene-based polyimides membranes exhibit significantly
improved performance in terms of the permeability/selectivity
trade-off when compared with commercial polymers, and some
of the membranes display separation performance beyond the
2
2
2
2
selectivities of 142 and 300 was obtained for TPDAn-6FAP,
suggesting its great potential for hydrogen purification.
ASSOCIATED CONTENT
■
*
S
Supporting Information
1991 upper bound. The permeability/selectivity trade-off of
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.macro-
mol.6b00485.
TPDAn-PPDAm outperforms that of TPDAn-TPDAm,
suggesting preferred cavity architecture was obtained in
TPDAn-PPDAm due to the incorporation of pentiptycene
units. Compared with 6FDA-based polyimides, TPDAn-6FAP
and TPDAn-PPDAm show better hydrogen-related gas
separation performance, and TPDAn-6FAP shows the best
hydrogen separation performance among the three polyimides
DSC and TGA curves for the iptycene-based polyimides,
pictures of thin films of the polyimides, the feed pressure
dependence of gas permeabilities, and effect of diamine
and dianhydride structures on gas selectivities (PDF)
I
DOI: 10.1021/acs.macromol.6b00485
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(
14) Wang, Z.; Wang, D.; Jin, J. Microporous Polyimides with
AUTHOR INFORMATION
■
*
Rationally Designed Chain Structure Achieving High Performance for
Gas Separation. Macromolecules 2014, 47 (21), 7477−7483.
Corresponding Author
Tel +1-574-631-3453; Fax +1-574-631-8366; e-mail rguo@nd.
edu (R.G.).
(15) Wiegand, J. R.; Smith, Z. P.; Liu, Q.; Patterson, C. T.; Freeman,
B. D.; Guo, R. Synthesis and Characterization of Triptycene-based
Polyimides with Tunable High Fractional Free Volume for Gas
Separation Membranes. J. Mater. Chem. A 2014, 2, 13309−13320.
Notes
The authors declare no competing financial interest.
(16) Luo, S.; Liu, Q.; Zhang, B.; Wiegand, J. R.; Freeman, B. D.; Guo,
R. Pentiptycene-based Polyimides with Hierarchically Controlled
Molecular Cavity Architecture for Efficient Membrane Gas Separation.
J. Membr. Sci. 2015, 480, 20−30.
ACKNOWLEDGMENTS
■
R. Guo gratefully acknowledges the financial support of the
Division of Chemical Sciences, Biosciences, and Geosciences,
Office of Basic Energy Sciences of the U.S. Department of
Energy (DOE) under Award DE-SC0010330. S. Luo thanks the
partial financial support from the Center of Sustainable Energy
at Notre Dame via the ND Energy Postdoctoral Fellowship
Program. We thank Prof. Haifeng Gao at University of Notre
Dame for the use of SEC. C. M. Doherty is supported by the
Australian Research Council (DE140101359).
(
17) Luo, S.; Stevens, K. A.; Park, J. S.; Moon, J. D.; Liu, Q.;
Freeman, B. D.; Guo, R. Highly CO -Selective Gas Separation
2
Membranes Based on Segmented Copolymers of Poly(ethylene oxide)
Reinforced with Pentiptycene-containing Polyimide Hard Segments.
ACS Appl. Mater. Interfaces 2016, 8 (3), 2306−2317.
(
18) Cho, Y. J.; Park, H. B. High Performance Polyimide with High
Internal Free Volume Elements. Macromol. Rapid Commun. 2011, 32
7), 579−586.
19) Mao, H.; Zhang, S. Synthesis, Characterization, and Gas
Transport Properties of Novel Iptycene-based Poly bis-
(
(
(
benzimidazobenzisoquinolinones). Polymer 2014, 55 (1), 102−109.
REFERENCES
(20) Carta, M.; Croad, M.; Malpass-Evans, R.; Jansen, J. C.;
■
Bernardo, P.; Clarizia, G.; Friess, K.; Lanc, M.; McKeown, N. B.
Triptycene Induced Enhancement of Membrane Gas Selectivity for
Microporous Troger’s Base Polymers. Adv. Mater. 2014, 26 (21),
(
1) Sanders, D. F.; Smith, Z. P.; Guo, R.; Robeson, L. M.; McGrath, J.
E.; Paul, D. R.; Freeman, B. D. Energy-efficient Polymeric Gas
Separation Membranes for a Sustainable Future: A Review. Polymer
3
(
526−3531.
2
013, 54 (18), 4729−4761.
2) Baker, R. W.; Low, B. T. Gas Separation Membrane Materials: A
Perspective. Macromolecules 2014, 47 (20), 6999−7013.
3) Xiao, Y.; Low, B. T.; Hosseini, S. S.; Chung, T. S.; Paul, D. R. The
21) Ghanem, B. S.; Swaidan, R.; Ma, X.; Litwiller, E.; Pinnau, I.
(
Energy-efficient Hydrogen Separation by AB-type Ladder-polymer
Molecular Sieves. Adv. Mater. 2014, 26 (39), 6696−6700.
(
(22) Swager, T. M. Iptycenes in the Design of High Performance
Strategies of Molecular Architecture and Modification of Polyimide-
based Membranes for CO Removal from Natural Gas: A Review.
Polymers. Acc. Chem. Res. 2008, 41 (9), 1181−1189.
2
(23) Zhang, Q.; Li, S.; Li, W.; Zhang, S. Synthesis and Properties of
Prog. Polym. Sci. 2009, 34 (6), 561−580.
Novel Organosoluble Polyimides Derived from 1,4-bis 4-(3,4-
dicarboxylphenoxy) Triptycene Dianhydride and Various Aromatic
Diamines. Polymer 2007, 48 (21), 6246−6253.
(
4) Yampolskii, Y. Polymeric Gas Separation Membranes. Macro-
molecules 2012, 45 (8), 3298−3311.
5) Robeson, L. M. Correlation of Separation Factor versus
Permeability for Polymeric Membranes. J. Membr. Sci. 1991, 62 (2),
65−185.
6) Robeson, L. M. The Upper Bound Revisited. J. Membr. Sci. 2008,
20 (1−2), 390−400.
7) Ghanem, B. S.; McKeown, N. B.; Budd, P. M.; Selbie, J. D.;
(
(24) Hsiao, S.-H.; Wang, H.-M.; Chou, J.-S.; Guo, W.; Lee, T.-M.;
Leu, C.-M.; Su, C.-W. Triptycene Poly(ether-imide)s with High
Solubility and Optical Transparency. J. Polym. Res. 2012, 19 (1), 9757.
1
(
3
(
(25) Kushwaha, A.; Dose, M. E.; Smith, Z. P.; Luo, S.; Freeman, B.
D.; Guo, R. Preparation and Properties of Polybenzoxazole-based Gas
Separation Membranes: A Comparative Study Between Thermal
Rearrangement (TR) of Poly(hydroxyimide) and Thermal Cyclo-
dehydration of Poly(hydroxyamide). Polymer 2015, 78 (5), 81−93.
Fritsch, D. High-performance Membranes from Polyimides with
Intrinsic Microporosity. Adv. Mater. 2008, 20 (14), 2766−2771.
(
8) Ghanem, B. S.; McKeown, N. B.; Budd, P. M.; Al-Harbi, N. M.;
(
1
(
26) Bondi, A. van der Waals Volumes and Radii. J. Phys. Chem.
964, 68 (3), 441−451.
27) Park, J. Y.; Paul, D. R. Correlation and Prediction of Gas
Fritsch, D.; Heinrich, K.; Starannikova, L.; Tokarev, A.; Yampolskii, Y.
Synthesis, Characterization, and Gas Permeation Properties of a Novel
Group of Polymers with Intrinsic Microporosity: PIM-polyimides.
Macromolecules 2009, 42 (20), 7881−7888.
Permeability in Glassy Polymer Membrane Materials via a Modified
Free Volume Based Group Contribution Method. J. Membr. Sci. 1997,
(
9) Rogan, Y.; Malpass-Evans, R.; Carta, M.; Lee, M.; Jansen, J. C.;
1
(
25 (1), 23−39.
Bernardo, P.; Clarizia, G.; Tocci, E.; Friess, K.; Lanc, M.; McKeown,
N. B. A Highly Permeable Polyimide with Enhanced Selectivity for
Membrane Gas Separations. J. Mater. Chem. A 2014, 2 (14), 4874−
28) Yampolskii, Y.; Shantarovich, V. Materials Science of Membranes
for Gas and Vapor Separation; John Wiley & Sons Ltd.: Chichester,
2006; p 21.
4
877.
(
10) Ghanem, B. S.; Swaidan, R.; Litwiller, E.; Pinnau, I. Ultra-
(29) Tao, S. J. Positronium Annihilation in Molecular Substances. J.
Chem. Phys. 1972, 56 (11), 5499−5510.
(30) Eldrup, M.; Lightbody, D.; Sherwood, J. N. The Temperature-
dependence of Positron Lifetimes in Solid Pivalic Acid. Chem. Phys.
1981, 63 (1−2), 51−58.
(31) Pascual-Izarra, C.; Dong, A. W.; Pas, S. J.; Hill, A. J.; Boyd, B. J.;
Drummond, C. J. Advanced Fitting Algorithms for Analysing Positron
Annihilation Lifetime Spectra. Nucl. Instrum. Methods Phys. Res., Sect. A
2009, 603 (3), 456−466.
(32) Lin, H.; Freeman, B. D. Permeation and Diffusion. In Springer
Handbook of Material Measurement Methods; Czichos, H., Saito, T.,
Smith, L., Eds.; Springer: New York, 2006; pp 371−387.
(33) Park, H. B.; Jung, C. H.; Lee, Y. M.; Hill, A. J.; Pas, S. J.; Mudie,
S. T.; Van Wagner, E.; Freeman, B. D.; Cookson, D. J. Polymers with
Cavities Tuned for Fast Selective Transport of Small Molecules and
Ions. Science 2007, 318 (5848), 254−258.
microporous Triptycene-based Polyimide Membranes for High-
performance Gas Separation. Adv. Mater. 2014, 26 (22), 3688−3692.
(
11) Swaidan, R.; Al-Saeedi, M.; Ghanem, B.; Litwiller, E.; Pinnau, I.
Rational Design of Intrinsically Ultramicroporous Polyimides Con-
taining Bridgehead-substituted Triptycene for Highly Selective and
Permeable Gas Separation Membranes. Macromolecules 2014, 47 (15),
5
(
104−5114.
12) Budd, P. M.; Ghanem, B. S.; Makhseed, S.; McKeown, N. B.;
Msayib, K. J.; Tattershall, C. E. Polymers of Intrinsic Microporosity
PIMs): Robust, Solution-processable, Organic Nanoporous Materials.
Chem. Commun. 2004, 2, 230−231.
13) Zhuang, Y.; Seong, J. G.; Do, Y. S.; Jo, H. J.; Cui, Z.; Lee, J.; Lee,
Y. M.; Guiver, M. D. Intrinsically Microporous Soluble Polyimides
Incorporating Troger’s Base for Membrane Gas Separation. Macro-
molecules 2014, 47 (10), 3254−3262.
(
(
̈
J
DOI: 10.1021/acs.macromol.6b00485
Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(
34) Guo, R.; Sanders, D. F.; Smith, Z. P.; Freeman, B. D.; Paul, D.
R.; McGrath, J. E. Synthesis and Characterization of Thermally
Rearranged (TR) Polymers: Influence of ortho-positioned Functional
Groups of Polyimide Precursors on TR Process and Gas Transport
Properties. J. Mater. Chem. A 2013, 1 (2), 262−272.
(
35) Guo, R.; Sanders, D. F.; Smith, Z. P.; Freeman, B. D.; Paul, D.
R.; McGrath, J. E. Synthesis and Characterization of Thermally
Rearranged (TR) Polymers: Effect of Glass Transition Temperature of
Aromatic Poly(hydroxyimide) Precursors on TR Process and Gas
Permeation Properties. J. Mater. Chem. A 2013, 1 (19), 6063−6072.
(
36) Ghosh, A.; Sen, S. K.; Banerjee, S.; Voit, B. Solubility
Improvements in Aromatic Polyimides by Macromolecular Engineer-
ing. RSC Adv. 2012, 2 (14), 5900−5926.
(
37) Polymer Handbook, 4th ed.; Brandrup, J., Immergut, E. H.,
Grulke, E. A., Eds.; John Wiley: New York, 1999; pp 675−711.
38) Huang, Y.; Wang, X.; Paul, D. R. Physical Aging of Thin Glassy
Polymer Films: Free Volume Interpretation. J. Membr. Sci. 2006, 277
1−2), 219−229.
39) Ma, X.; Swaidan, R.; Belmabkhout, Y.; Zhu, Y.; Litwiller, E.;
(
(
(
Jouiad, M.; Pinnau, I.; Han, Y. Synthesis and Gas Transport Properties
of Hydroxyl-functionalized Polyimides with Intrinsic Microporosity.
Macromolecules 2012, 45 (9), 3841−3849.
(
40) Kim, Y. H.; Kim, H. S.; Kwon, S. K. Synthesis and
Characterization of Highly Soluble and Oxygen Permeable New
Polyimides Based on Twisted Biphenyl Dianhydride and Spirobi-
fluorene Diamine. Macromolecules 2005, 38 (19), 7950−7956.
(
41) Staiger, C. L.; Pas, S. J.; Hill, A. J.; Cornelius, C. J. Gas
Separation, Free Volume Distribution, and Physical Aging of a Highly
Microporous Spirobisindane Polymer. Chem. Mater. 2008, 20 (8),
2
(
606−2608.
42) Han, S. H.; Misdan, N.; Kim, S.; Doherty, C. M.; Hill, A. J.; Lee,
Y. M. Thermally Rearranged (TR) Polybenzoxazole: Effects of Diverse
Imidization Routes on Physical Properties and Gas Transport
Behaviors. Macromolecules 2010, 43 (18), 7657−7667.
(
43) Aitken, C. L.; Koros, W. J.; Paul, D. R. Effect of Structural
Symmetry on Gas-transport Properties of Polysulfones. Macromolecules
992, 25 (13), 3424−3434.
44) Puleo, A. C.; Paul, D. R.; Kelley, S. S. The effect of degree of
1
(
acetylation on gas sorption and transport behavior in cellulose-acetate.
J. Membr. Sci. 1989, 47 (3), 301−332.
(
45) Huang, Y.; Paul, D. R. Effect of Film Thickness on the Gas-
permeation Characteristics of Glassy Polymer Membranes. Ind. Eng.
Chem. Res. 2007, 46 (8), 2342−2347.
(
46) Vu, D. Q.; Koros, W. J.; Miller, S. J. Mixed Matrix Membranes
Using Carbon Molecular Sieves - I. Preparation and Experimental
Results. J. Membr. Sci. 2003, 211 (2), 311−334.
(
47) Zhang, Y.; Musseman, I. H.; Ferraris, J. P.; Balkus, K. J., Jr. Gas
Permeability Properties of Matrimid® Membranes Containing the
Metal-organic Framework Cu-BPY-HFS. J. Membr. Sci. 2008, 313 (1−
2
(
), 170−181.
48) Ekiner, O. M.; Vassilatos, G. Polyaramide Hollow Fibers for
Hydrogen Methane Separation - Spinning and Properties. J. Membr.
Sci. 1990, 53 (3), 259−273.
(
49) Bernardo, P.; Drioli, E.; Golemme, G. Membrane Gas
Separation: A Review/State of the Art. Ind. Eng. Chem. Res. 2009,
4
8 (10), 4638−4663.
K
DOI: 10.1021/acs.macromol.6b00485
Macromolecules XXXX, XXX, XXX−XXX