New PIM-1 copolymers containing 2,3,6,7-anthracenetetrayl moiety and their use as gas separation membranes

Article abstract of DOI:10.1016/j.mencom.2020.11.015

New PIM-1 copolymers with 2,3,6,7-anthracenetetrayl moiety have been synthesized as fine powders from 2,3,6,7-tetrahydroxy-9,10-di(p-fluorophenyl)anthracene monomer in DMSO, with crystal structure of the monomer as well as its precursors and the product o

Full text of DOI:10.1016/j.mencom.2020.11.015

Mendeleev
Communications
Mendeleev Commun., 2020, 30, 734–737
New PIM-1 copolymers containing 2,3,6,7-anthracenetetrayl moiety
and their use as gas separation membranes
Igor I. Ponomarev,a Kirill M. Skupov,a Konstantin A. Lyssenko,b Inesa V. Blagodatskikh,a Alexander V. Muranov,a
Yulia A. Volkova,a Dmitry Yu. Razorenov,*a Ivan I. Ponomarev,a Liudmila E. Starannikova,c
Denis A. Bezgin,c Alexander Yu. Alentievc and Yuri P. Yampolskiic
a A. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, 119991 Moscow,
Russian Federation. E-mail: razar@ineos.ac.ru, gagapon@ineos.ac.ru
b Department of Chemistry, M. V. Lomonosov Moscow State University, 119991 Moscow, Russian Federation
c A. V. Topchiev Institute of Petrochemical Synthesis, Russian Academy of Sciences, 119991 Moscow,
Russian Federation
DOI: 10.1016/j.mencom.2020.11.015
New PIM-1 copolymers with 2,3,6,7-anthracenetetrayl
moiety have been synthesized as fine powders from
2,3,6,7-tetrahydroxy-9,10-di(p-fluorophenyl)anthracene
monomer in DMSO, with crystal structure of the monomer
as well as its precursors and the product of its model side
crosslinking reaction being determined using X-ray
diffraction. The microporosity and surface characteristics of
the copolymers powders have been measured. As-cast
membrane from the copolymer with 5 mol% of the monomer
synthesized revealed promising gas separation properties as
compared with original PIM-1.
Keywords: gas separation membrane, ladder polymer, polymer synthesis, copolymer, micropores, PIM-1, sonochemistry, BET.
Polymers of intrinsic microporosity (PIMs) represent non-cross-
linked soluble organic polymers with micropores originated
from distorted packing of macromolecular chains. PIMs were
synthesized for the first time by Budd and McKeown and have
attracted considerable attention since 2004 due to their
remarkable gas separation properties.1–5 The most investigated
PIM representative is a partially ladder spiroindanobenzodioxin
polymer known as PIM-1, which is formed from commercially
available 2,3,5,6-tetrafluoroterephthalonitrile (TFTPN) and
5,5',6,6'-tetrahydroxy-3,3,3',3'-tetramethylspiro-1,1'-bisindane
(TTSBI). In the last 15 years many efforts have been applied to
improve the PIM-1 gas separation performance as well as to
address its physical aging and plasticization. One of possible
solutions consists in selection of co-monomers for poly-
condensation, which affects polymer chain conformation and
thus membrane microporosity and morphology.6 A lot of
F
O
CN
CN
O
O
O
O
O
O
O
CN
CN
n
PIM-1
F
n
F
F
1
O
O
O
CN
CN
CN
CN
CN
CN
CN
CN
O
O
O
O
O
O
O
O
O
O
O
O
O
m
k
n
n
F
2
F
3a–e
(see Table 1)
© 2020 Mendeleev Communications. Published by ELSEVIER B.V.
on behalf of the N. D. Zelinsky Institute of Organic Chemistry of the
Russian Academy of Sciences.
– 734 –

Mendeleev Commun., 2020, 30, 734–737
ꢁaꢃ
ꢀꢁꢂꢃ
HO
O
MeO
MeO
OMe
OMe
OMe
i
+
CH
OMe
Oꢁꢂꢃ
Oꢁ2ꢃ
Oꢁꢄꢃ
Oꢁꢅꢃ
F
F
F
F
ꢀꢁ2ꢃ
ꢀꢁꢂꢃ
ꢁbꢃ
MeO
OMe
OMe
MeO
MeO
OMe
OMe
ii
+
Cꢁꢂꢅꢃ
MeO
Oꢁ2ꢃ
Oꢁꢂꢃ
Cꢁꢂ2ꢃ
Cꢁꢂꢂꢃ
F
F
30:70
F
5
4
iii
Figure 1 Molecular structures of compounds (a) 4 and (b) 5∙2CHCl3.
Displacement ellipsoids are drawn at 50% probability level.
Compound 4 spontaneously transformed in chloroform
solution in air into fully aromatic derivative 5, which was
crystallized as a corresponding solvate. The X-ray data obtained
were indispensable to a conformation analysis of the PIM-1
copolymers based on tetrahydroxyanthracene moiety because,
HO
HO
OH
OH
for example, dihydro derivative
4 was not planar. By
demethylation of tetramethyl ether 5, 2,3,6,7-tetrahydroxy-9,10-
di(4-fluorophenyl)anthracene 6 was obtained and its structure
was also confirmed by X-ray data (Figure 2). Polycondensation
F
6
Scheme 1 Reagents and conditions: i, H3PW12O40, MW,
5
min;
ii, p-fluorobenzaldehyde, Ac2O, H3PW12O40, 90 °C, 2 h; iii, aq. HBr, AcOH,
reflux, 6 h.
(5597 independent reflections, Rint = 0.1591) and used in the refinement,
which converged to wR2 = 0.1734, GOF = 0.915 for all independent
reflections [R1 = 0.0633 for 2472 reflections with I > 2σ(I)]. 2qmax was
56° (completeness 100%). Residual electron density 0.342/–0.327 e Å–3.
Crystal data –for 5∙2CHCl3. C32H26Cl6F2O4, M = 725.27, triclinic,
space group P1, T = 120 K, a = 5.7424(4), b = 11.0710(8) and
different tetrahydroxy-aromatics have been introduced into the
PIM-1 synthesis instead of some structural parts of TTSBI.1,2
However, only in few cases the polymers obtained demonstrated
improved selectivity for several gas pairs compared with
PIM-1.7
In this work, PIM-1 analogues 1–3 were synthesized starting
from 2,3,6,7-tetrahydroxy-9,10-di(4-fluorophenyl)anthracene as
a new monomer.
The monomer was obtained from veratrole and
p-fluorobenzaldehyde using the known method8 (Scheme 1).
Reaction of veratrole with substituted benzaldehyde was
carried out in two steps. First, after 3–5 min under 120 W
microwave irradiation in a household oven in the presence of
phosphorus–tungsten heteropolyacid (HPA), the corresponding
triphenylmethane derivative was generated (for details, see
Online Supplementary Materials). This intermediate reacted
with another portion of aldehyde, HPA and acetic anhydride
resulting in relatively stable 2,3,6,7-teramethoxy-9,10-di(4-
fluorophenyl)-9,10-dihydroanthracene 4 as by-product and the
desired 2,3,6,7-tetramethoxy-9,10-di(4-fluorophenyl)anthracene
5 with a molar ratio of 30:70 in 70% total yield. Structures of
compounds 4 and 5 were confirmed using NMR and X-ray
diffraction (Figure 1).†
c
= 13.0250(10) Å, a = 96.4190(10), b = 93.0650(10) and
g = 97.0720(10)°, V = 814.70(10) Å3, Z = 1, Z' = 0.5, dcalc = 1.478 g cm–3,
μ(MoKa) = 5.75 mm–1, F(000) = 370. 9827 reflections were collected
(4318 independent reflections, Rint = 0.0173) and used in the refinement,
which converged to wR2 = 0.0879, GOF = 1.018 for all independent
reflections [R1 = 0.0323 for 3865 reflections with I > 2σ(I)]. 2qmax was
58° (completeness 100%). Residual electron density 0.613/–0.643 e Å–3.
Crystal data for 6∙4Me2CO. C38H40F2O8, M = 662.70, monoclinic,
space group P21/n, T = 120 K, a = 14.2580(4), b = 7.9447(2) and
c = 15.1400(4) Å, b = 93.2002(11)°, V = 1712.32(8) Å3, Z = 2, Z' = 0.5,
dcalc
= 1.285 g = = 700.
cm–3, μ(MoKa) 0.96 mm–1, F(000)
178050 reflections were collected (11030 independent reflections,
Rint = 0.0690) and used in the refinement, which converged to
wR2 = 0.1713, GOF = 1.032 for all independent reflections [R1 = 0.0522
for 7880 reflections with I > 2σ(I)]. 2qmax was 81° (completeness 100%).
Residual electron density 0.625/–0.409 e Å–3.
–
Crystal data for 7. C54H12F30O4, M =1294.64, triclinic, space group P1,
T = 120 K, a = 5.8796(9), b = 11.363(2) and c = 18.586(3) Å,
a = 99.456(4), b = 93.980(3) and g = 99.719(3)°, V = 1201.2(3) Å3, Z = 1,
Z' = 0.5, dcalc = 1.790 g cm–3, μ(MoKa) = 1.92 mm–1, F(000) = 638.
13056 reflections were collected (9827 independent reflections,
Rint = 0.0205) and used in the refinement, which converged to
wR2 = 0.0958, GOF = 1.030 for all independent reflections [R1 = 0.0359
was calculated for 3865 reflections with I > 2σ(I)]. 2qmax was 55°
(completeness 100%). Residual electron density 0.309/–0.344 e Å–3.
CCDC 1991888–1991891 contain the supplementary crystallographic
data for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via http://www.ccdc.cam.ac.uk.
†
Crystal data for 4. C30H26F2O4, M = 488.53, monoclinic, space group
P21/c, T = 120 K, a = 13.349(3), b = 5.4211(10) and c = 32.210(6) Å,
b = 95.504(5)°, V = 2320.2(8) Å3, Z = 4, Z' = 1, dcalc = 1.398 g cm–3,
μ(MoKa) = 1.02 mm–1, F(000) = 1024. 21285 reflections were collected
– 735 –

Mendeleev Commun., 2020, 30, 734–737
ꢀꢁ20Cꢃ
ꢀꢁꢂꢃ
ꢀꢁꢂꢃ
ꢀꢁ20Bꢃ
ꢀꢁꢂꢉꢃ
ꢀꢁ20Aꢃ
ꢀꢁꢂꢆꢃ
OꢁꢂSꢃ
Oꢁꢂꢃ
ꢀꢁꢂꢈꢃ
Oꢁꢂꢃ
ꢀꢁ2ꢆꢃ
Oꢁ2ꢃ
ꢀꢁꢂꢇꢃ
ꢀꢁ2ꢄCꢃ
Oꢁ2Sꢃ
ꢀꢁ2ꢄBꢃ
ꢀꢁ2ꢄAꢃ
ꢀꢁ2ꢇꢃ
Oꢁ2ꢃ
ꢀꢁ2ꢅꢃ
ꢀꢁ22ꢃ
Figure 2 Molecular structure of compound 6∙4Me2CO. Displacement
ellipsoids are drawn at 50% probability level.
Figure 3 Molecular structure of compound 7. Displacement ellipsoids are
drawn at 50% probability level.
of monomer 6 with TFTPN in DMA or DMSO in the presence of
K2CO3 was performed by the known methods9,10 and resulted in
formation of ladder polymer 1, which was not soluble even in
strong inorganic acids like methanesulfonic or sulfuric ones. An
attempted approach to copolymer of compound 6, TFTPN and
TTSBI at the TTSBI to monomer 6 ratio (m/k) of 1:1 led again
to fully insoluble product 2. The insolubility of polymers 1 and 2
precluded a membrane casting and hence the intended application
of these compounds.
In general, the hydroxyl groups in monomer 6 can react either
coherently, resulting in dibenzodioxin cycles, or independently,
producing branched and/or crosslinked structures. The latter way
was confirmed by reaction of compound 6 with fourfold excess
of perfluorotoluene as a highly active compound suitable for
aromatic nucleophilic substitution. Complete reaction at all four
PIM-1 and its copolymers represent microporous materials,
therefore special attention has been paid to investigation of their
microporosity by surface analysis. Adsorption and desorption
isotherms for polymers PIM-1 and 3a–d are shown in Figures
S8–S12 (see Online Supplementary Materials). These isotherms
do not coincide for adsorption and desorption processes, which
is typical of a microporous material. The values of total specific
surface area Ssp calculated according to BET are similar
(849–866 m2 g–1) for polymers PIM-1 and 3a–c with 0–15% of
monomer 6 and then drops when the incorporation of compound
6 reaches 20%, being 750 m2 g–1 (see Table 1). The micropore
specific surface area values Smicro were obtained according to the
t-method. Copolymer 3a with 5% of monomer 6 has higher
Smicro value than pure PIM-1, namely 411 vs. 394 m2 g–1, which
can be explained by the PIM-1 microstructure disruption by low
amount of monomer 6. Further increase in the monomer 6
incorporation, namely to 10, 15 and 20%, lowers the Smicro values
up to 364 m2 g–1 due to different packing pattern of the polymer
chains. Polymers 3e and 2 with 30 and 50% of monomer 6
incorporation, respectively, demonstrate a further decrease in the
Smicro values. External surface area Sext obtained by the t-method
and related to meso- and macropores, is similar to the BJH
surface area for pores with diameter d > 2 nm, which confirms
reliability of the methods applied. BJH pore size distribution
(Figures S8–S12) for a pore diameter range of 2–10 nm in the
copolymers differs from that attributed to PIM-1, which allows
one to assume that gas diffusion properties for the copolymers
with anthracene moieties will also be different compared with
those for PIM-1. To confirm this assumption, as-cast films of
polymer 3a (5% of monomer 6) with the highest micropore
specific surface area value were obtained from CHCl3 and
explored by gas chromatography,‡ resulting in the following
preliminary data on the gas permeability (in Barrer): 1133 (He),
2985 (H2), 1493 (O2), 633 (N2), 10338 (CO2) and 1387 (CH4).
These values can be compared with the ones for as-cast PIM-1
hydroxyl groups of monomer
6 resulted in 9,10-bis(4-
fluorophenyl)-2,3,6,7-tetrakis[2,3,5,6-tetrafluoro-4-(trifluoro-
methyl)phenoxy]anthracene 7, whose structure was confirmed
by NMR and X-ray analysis (Figure 3).
Thus, the mixed reactivity of 2,3,6,7-tetrahydroxyanthracenes
in aromatic nucleophilic substitution with TFTPN and in their
triple copolycondensation with TTSBI was verified. Therefore,
we changed the copolymer synthesis strategy and the monomer
loading scheme towards soluble high-molecular weight products
with elevated content of PIM-1 units. The best outcome was
obtained when the mixture of TTSBI and monomer 6 was
pretreated with K2CO3 at room temperature for 0.6–1 h, which
led to their activated deprotonated forms. Further addition of
TFTPNandheatingto80°Cfor5hinDMSOwithultrasonication
(37 kHz) resulted in completely soluble high-molecular weight
copolymers of TFTPN and compound 6, namely products
3a–d, with TTSBI–monomer 6 ratios (m/k) of 95:5, 90:10,
85:15 and 80:20, which had the reduced viscosity hred values in
1,1,2,2-tetrachloroethane (TCE) of 0.70–0.81 dl g–1 and other
physical properties collected in Table 1.
Table 1 Physical properties of the synthesized copolymers.
Ratio of TTSBI:6
Polymer
hreda/dl g–1 MW/kDa (GPC) PDI (GPC) Sspb/m2 g–1 (BET) Smicro /m2 g–1 Sext /m2 g–1 S/m2 g–1 (BJH) Vsp/cm3 g–1
units (m/k)
PIM-1
3a
3b
3c
3d
3e
2
100:0
95:5
1.17
120
86.6
80.6
47.6
38.2
–
3.8
3.6
3.3
4.3
4.0
–
856
866
849
861
750
829
605
394
411
394
370
394
350
351
463
455
455
491
386
478
254
424
439
423
441
346
453
240
0.575
0.631
0.598
0.667
0.618
0.604
0.374
0.81
90:10
85:15
80:20
70:30
50:50
0.76
0.70
0.70
insoluble
insoluble
–
–
a hred is reduced viscosity measured at 25 °C for 0.5 g dl–1 solutions in TCE for PIM-1 and 3a,b or 0.25 g dl–1 solutions in TCE–CF3COOH (97:3) for 3c,d.
b Ssp, Smicro, Sext and S are total specific surface area according to BET, micropore specific surface area, external surface area and surface area according to
BJH, respectively, measured for polymer powders after drying at 150 °C and residual pressure of 10–5 bar.
– 736 –

Mendeleev Commun., 2020, 30, 734–737
film, namely 760 (He), 1630 (H2), 580 (O2), 180 (N2), 4390
References
(CO2) and 310 (CH4), obtained using the same equipment under
1 Z.-X. Low, P. M. Budd, N. B. McKeown and D. A. Patterson, Chem.
Rev., 2018, 118, 5871.
the same experimental conditions.11
2 N. B. McKeown, Sci. China: Chem., 2017, 60, 1023.
3 P. M. Budd, E. S. Elabas, B. S. Ghanem, S. Makhseed, N. B. McKeown,
K. J. Msayib, C. E. Tattershall and D. Wang, Adv. Mater., 2004, 16, 456.
4 P. M. Budd, B. S. Ghanem, S. Makhseed, N. B. McKeown, K. J. Msayib
and C. E. Tattershall, Chem. Commun., 2004, 230.
5 Y. Wang, X. Ma, B. S. Ghanem, F. Alghunaimi, I. Pinnau and Y. Han,
Mater. Today Nano, 2018, 3, 69.
6 I. I. Ponomarev, K. A. Lyssenko, D. Yu. Razorenov, Yu. A. Volkova,
Iv. I. Ponomarev, K. M. Skupov, Z. S. Klemenkova, L. E. Starannikova,
A. Yu. Alentiev and Yu. P. Yampolskii, Mendeleev Commun., 2019, 29,
663.
In summary, this work demonstrates a new approach to the
synthesis of PIMs. The increase in gas permeability originating
from the presence of 5 mol% monomer 6 can be attributed to the
2,3,6,7-anthracenetetrayl moieties that engage the statistically
distributed extended ladder parts in the polymer structure.
Another reason is the 4-fluorophenyl substituents, which are
known to increase the permeability.12 Further investigation of
gas permeation and selectivity for this type of copolymers is
planned.
7 D. Fritsch, G. Bengtson, M. Carta and N. B. McKeown, Macromol.
Chem. Phys., 2011, 212, 1137.
8 I. Mohammadpoor-Baltork, M. Moghadam, S. Tangestaninejad,
V. Mirkhani, K. Mohammadiannejad-Abbasabadi and H. R. Khavasi,
Eur. J. Org. Chem., 2011, 1357.
9 I. I. Ponomarev, I. V. Blagodatskikh, A. V. Muranov, Yu. A. Volkova,
D. Yu. Razorenov, Iv. I. Ponomarev and K. M. Skupov, Mendeleev
Commun., 2016, 26, 362.
This work was supported by the Russian Foundation for Basic
Research (project no. 18-08-00472 A). Elemental analysis was
supported by the Ministry of Science and Higher Education of
the Russian Federation and was carried out using the equipment
of ‘Center for Molecular Composition Studies’ of the
A. N. Nesmeyanov Institute of Organoelement Compounds,
Russian Academy of Sciences.
10 H. R. Kricheldorf, N. Lomadze, D. Fritsch and G. Schwarz, J. Polym.
Sci., Part A: Polym. Chem., 2006, 44, 5344.
11 P. M. Budd, N. B. McKeown, B. S. Ghanem, K. J. Msayib, D. Fritsch,
L. Starannikova, N. Belov, O. Sanfirova, Y. Yampolskii and
V. Shantarovich, J. Membr. Sci., 2008, 325, 851.
Online Supplementary Materials
Supplementary data associated with this article can be found
in the online version at doi: 10.1016/j.mencom.2020.11.015.
12 Yu. P. Yampolskii, N. A. Belov and A. Yu. Alentiev, Russ. Chem. Rev.,
2019, 88, 387.
‡
Permeability coefficients were determined with respect to He, H2, O2,
N2, CO2 and CH4 for the prepared films using gas chromatographic
method at 20–22 °C. A steady state stream of tested gas at 1 atm contacted
with the inlet surface of a film, while the penetrated stream was diluted
by the carrier gas (typically He or alternatively N2 when permeation of
He or H2 was measured). Partial pressure of the penetrant gases below the
films was negligible compared with atmospheric pressure above the
films. The permeability coefficients were calculated from the response of
the thermal conductivity detector and an estimation for the stream flow.
Received: 28th April 2020; Com. 20/6208
– 737 –