Centrotriindane-and triptindane-based polymers of intrinsic microporosity

Article abstract of DOI:10.1016/j.polymer.2013.07.035

In this paper we describe the synthesis and physical characterisation of polymers derived from hydroxyl containing bowl-shaped " centrotriindane" and propellane-type "triptindane" monomers using the dibenzodioxin-forming polymerisation reaction with 2,3,5

Full text of DOI:10.1016/j.polymer.2013.07.035

Polymer 55 (2014) 326e329
Contents lists available at ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Centrotriindane- and triptindane-based polymers of intrinsic
microporosity
*
James Vile, Mariolino Carta, C. Grazia Bezzu, Benson M. Kariuki, Neil B. McKeown
School of Chemistry, Cardiff University, Cardiff, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
In this paper we describe the synthesis and physical characterisation of polymers derived from hydroxyl
containing bowl-shaped “centrotriindane” and propellane-type “triptindane” monomers using the
dibenzodioxin-forming polymerisation reaction with 2,3,5,6-tetrafluoroterephthalonitrile. Evaluation of
the microporosity of the resulting polymers via nitrogen adsorption measurements revealed apparent
Brunauer-Emmett-Teller (BET) surface areas in the range of 555e1039 m2 gꢀ1 which can be related to the
distinct shape of the monomeric units. An evaluation of the shape of the monomers, using X-ray crys-
tallography, helps to explain the degree of microporosity of the polymers from which they are made.
Ó 2013 Elsevier Ltd. All rights reserved.
Received 13 June 2013
Accepted 12 July 2013
Available online 2 August 2013
Keywords:
Intrinsic microporosity
Triptycene
Network polymers
1. Introduction
hydroxylated tribenzotriquinacenes (TBTQ) [22] and propeller-
shaped triptycenes [23] are suitable structural units for the
Recently, considerable efforts have been made in the synthesis
of novel microporous materials which possess high internal sur-
face areas and therefore can have applications in technologies
such as hydrogen storage [1,2], CO2 capture [3,4] and gas separa-
tions [5e7]. Typical examples include metal-organic frameworks
(MOFs) [8,9], zeolites [10,11] and activated carbons [12,13].
Increasingly, polymeric microporous materials built from purely
organic components such as hypercrosslinked polymer networks,
covalent-organic-frameworks (COFs) [14] and polymers of
intrinsic microporosity (PIMs) have attracted much interest [15].
Generally, polymers lack microporosity because they have enough
conformational and rotational freedom to pack space efficiently.
However, PIMs are composed of highly rigid and contorted mac-
romolecules which cannot pack space efficiently, leaving molec-
ular sized interconnected voids [16]. The rigidity is caused by the
polymers being composed of fused ring units and the contorted
structures arises from the incorporation of non-planar sites of
contortion such as spiro-centres [17] or triptycene units [18]. PIMs
have attracted interest for applications in hydrogen storage [19]
membrane-based gas separations [6,7], sensors [20] and hetero-
geneous catalysis [21]. An attractive feature is that their structure
and properties can be tailored by the appropriate choice of
monomer precursors. Previously we have shown that concave
shaped monomers such as cyclotricatechylene (CTC) [19] or
preparation of PIMs by reaction with 2,3,5,6-tetrafluoroterep
hthalonitrile (Fig. 1).
In order to investigate how further changes in the monomer
structure can affect the properties of PIMs, our attention was drawn
to the recent work of Kuck and co-workers [24] in which they
describe the synthesis of the C2-symmetrical 2,3,6,7,10,11-
hexamethoxy-4b,8b,13,14-tetrahydrodiindeno[1,2-a:20,10-b]indene
the hydrocarbon core of which was termed “centrotriindane” and
the
C3v-symmetrical 2,3,6,7,13,14-hexamethoxytriptindane the
core of which was called “triptindane”. These rigid structural units
appeared attractive for the synthesis of novel PIMs. Here we
describe the synthesis of propellane-type “triptindane” monomers
M1, M2 and “centrotriindane” monomer M3 all of which contain
the required peripheral hydroxyl substituents, and their subse-
quent polymerisation with 2,3,5,6-tetrafluoroterephthalonitrile to
give the network polymer Triptindane-PIM A and the non-
network polymers Triptindane-PIM B and Centrotriindane-PIM
(Fig. 2).
2. Experimental
Full experimental details for the multi-step synthesis of mono-
mers and polymers are given in the Supplemental information
along with 13C solid-state NMR spectra. Crystallographic data for
M1 and M3 are also summarised and full data for these compounds
are available on request from the Cambridge Crystallographic Data
centre (CCDC 868321 and 868322).
* Corresponding author. Tel.: þ44 2920 875850.
E-mail address: mckeownnb@cardiff.ac.uk (N.B. McKeown).
0032-3861/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.polymer.2013.07.035

J. Vile et al. / Polymer 55 (2014) 326e329
327
Fig. 1. The structures of network PIMs (a) TBTQ [22] (b) CTC [19] and (c) Triptycene [23].
3. Results and discussion
Cyclization reactions of the resulting diols 7 and 8 were performed
readily in Eaton’s reagent [25], to provide a good yield of centro-
triindanes 9 and 10. Demethylation of 10 with BBr3 afforded the
target novel monomer M3, whereas, demethylation of 9 produced
an unstable compound prone to oxidation, so unfortunately, the
isolation of the desired monomer hexahydroxy-centrotriindane
and its use as a PIM precursor proved impossible.
3.1. Monomer synthesis and structural characterisation
The previous syntheses of 2,3,6,7,10,11-hexamethoxy-
4b,8b,13,14-tetrahydrodiindeno[1,2-a:20,10-b]indene and 2,3,6,
7,13,14-hexamethoxytriptindane reported by Kuck et al. [24]
were modified to give hexahydroxy monomer M1 suitable for
forming network polymer Triptindane-PIM A and tetrahydroxy
monomers M2 and M3 precursors for the non-network polymers
To fully characterise the structure of these monomers, crystals
suitable for XRD analysis of monomer M1 and M3 were grown by
slow diffusion of hexane into THF solutions. Their solid-state mo-
lecular structures are shown in Fig. 3(a and b). Both crystals proved
to be clathrates, with THF found as the included solvent (in Fig. 3
the included THF is omitted for clarity). Clathrate formation is a
typical feature of catechol-containing monomers that prove suit-
able for PIM synthesis, presumably due to their awkward shapes
that cannot form crystals with efficient molecular packing without
the incorporation of solvent molecules [26]. The structures of these
new monomers are significantly different from those of TBTQ [22]
and CTC [19] structures (Fig. 3). For the “triptindane” M1 (Fig. 3a),
the aromatic planes of the catechol units are parallel to the three-
fold molecular axis, as they are in triptycene (Fig. 3d) [27],
whereas in the bowl-shaped TBTQ monomer they are at an angle of
roughly 120ꢁ to the three-fold axis (Fig. 3c). For the “centro-
triindane” M3, the planes of the catechol units are approximately
perpendicular to each other as they are on opposites side of the
spirocentre.
Triptindane-PIM
B and Centrotriindane-PIM, respectively
(Scheme 1). The key intermediates used for preparing monomers
M1eM3 are the bis-ketones 1 and 2, the synthesis of which are
described in detail in the Supplemental information (see Scheme
SI.1) Briefly, the bis-ketones 1 and 2 were prepared directly,
albeit in low yield, by a two-fold benzylation of 5,6-dimethoxy-
1H-indene-1,3(2H)-dione or the commercial available 1H-
indene-1,3(2H)-dione, respectively, using 3,4-dimethoxybenzyl.
Alternatively, better overall yields of bisketones 1 and 2 were
obtained using a multi-step synthetic procedure starting with
the Aldol condensation between veratraldehyde and the appro-
priate diones followed by reduction of the resulting benzylidenes
and then a single benzylation using 3,4-dimethoxybenzyl chlo-
ride. Direct bicyclization of the resulting bis-ketones 1 and 2 with
orthophosphoric acid in refluxing toluene gave the hexa- and
tetramethoxy-substituted triptindanones 3 and 4. Reduction of
triptindanones 3 and 4 was achieved by ionic hydrogenation
using sodium borohydride in trifluoroacetic acid to form the
desired hexa- and tetramethoxy “triptindanes” 5 and 6. Deme-
thylation of 5 and 6 with BBr3 was successful and yielded the
desired novel monomers M1 and M2.
3.2. Polymer synthesis and characterisation
Polymerisation of monomers M1, M2 and M3 with the
commercially available 2,3,5,6-tetrafluoroterephthalonitrile, by a
nucleophilic aromatic substitution reaction commonly used for PIM
synthesis [19,23], gave Triptindane-PIM A and the non-network
polymers Triptindane-PIM B and Centrotriindane-PIM (Fig. 2)
To gain access to the desired centrotriindane monomers, bis-
ketones 1 and 2 were reduced with diisobutylaluminium hydride
(DIBAL-H) to give 7 and 8 in quantitative yields (Scheme 1).
Fig. 2. The structure of propellane-type “triptindane” and “centrotriindane” polymers.

328
J. Vile et al. / Polymer 55 (2014) 326e329
Scheme 1. Reagents and conditions. i. Orthophosphoric acid, toluene, 125 ꢁC, 16 h. ii. NaBH4, DCM, trifluoroacetic acid, 48 h. iii. BBr3, DCM, 0 ꢁC 12e16 h. iv. K2CO3, DMF, 80 ꢁC, 16 h.
v. K2CO3, DMF, 65 ꢁC, 96 h. vi. DIBAL-H, DCM, 4e16 h, 0 ꢁC. vii. Eaton’s reagent, 2e3 h, RT.
in good yields. Despite the anticipated lack of a network structure
for Triptindane-PIM B and Centrotriindane-PIM, all three poly-
mers proved to be insoluble in all solvents tested, which precluded
solution-based characterisation techniques such as Gel Permeation
Chromatography (GPC) and solution based NMR. However, char-
acterisation was achieved by use of solid state 13C NMR which gave
spectra consistent with the ideal structures of the polymer (see SI
for spectra). Thermogravimetric analysis of the three polymers
shows that they all demonstrate excellent thermal stability with no
mass loss below 350 ꢁC.
Nitrogen adsorption measurements at 77 K (Fig. 4) of powdered
samples of the polymer derived from M1eM3 confirmed their
Fig. 3. Single crystal XRD data of monomer M1 (a) and M3 (b), TBTQ [22] (c) and Triptycene (d) [27].

J. Vile et al. / Polymer 55 (2014) 326e329
329
polymer (Triptindane-PIM A) has a large apparent BET surface area
of 1039 m2 ꢀ1, which suggests that the rigid and contorted “trip-
g
tindane” structural unit frustrates packing in the solid state and
leads to enhanced microporosity similar to that of the structurally
related triptycene unit. By appropriate modifications of the estab-
lished synthetic route, tetrahydroxylated monomers of “centro-
triindane” and “triptindane” were prepared that proved successful
for the synthesis of non-network PIMs (Triptindane-PIM B and
Centrotriindane-PIM) but, unfortunately, these proved to be
insoluble in all solvents. The microporosity and insolubility of the
non-network polymer Centrotriindane-PIM is comparable to that
of the related TBTQ polymers. A network polymer based on “cen-
troindane” could not be prepared due to the extreme instability of
its hexahydroxy precursor.
Acknowledgements
We acknowledge funding from EPSRC grants EP/G01244X (MC)
and EP/H024034/1 (CGB) and thank the EPSRC National Solid State
NMR Research Service (Durham University).
Fig. 4. Nitrogen adsorption isotherms of Triptindane-PIM A (A), Triptindane-PIM B
(-) and Centrotriindane-PIM (:) measured at 77 K. Desorption isotherms are shown
in gray.
Appendix A. Supplementary information
Table 1
Summary of polymer properties.
Supplementary information related to this article can be found
at http://dx.doi.org/10.1016/j.polymer.2013.07.035.
Monomer
Polymer
BET surface
Pore volume
a
area (m2 gꢀ1)
(ml gꢀ1
)
M1
M2
M3
n/a
n/a
n/a
n/a
Triptindane-PIM A
Triptindane-PIM B
Centrotriindane-PIM
CTC-PIM [19]
TBTQ-PIM-1 [22]
TBTQ-PIM-3 [22]
Tript-PIM [23]
1039
653
555
830
565
511
1318
0.79
0.46
0.43
0.39
0.36
0.37
0.89
References
[1] Murray LJ, Dinca M, Long JR. Chem Soc Rev 2009;38(5):1294e314.
[2] Zhang LL, Lin T, Pan X, Wang W, Liu T-X. J Mater Chem 2012;22(19):9861e9.
[3] Ren S, Bojdys MJ, Dawson R, Laybourn A, Khimyak YZ, Adams DJ, et al. Adv
Mater 2012;24(17):2357e61.
[4] Chen Q, Luo M, Hammershoej P, Zhou D, Han Y, Laursen BW, et al. J Am Chem
Soc 2012;134(14):6084e7.
a
Calculated from amount of nitrogen adsorbed at 77 K and relative pressure P/
[5] Carta M, Malpass-Evans R, Croad M, Rogan Y, Jansen JC, Bernardo P, et al.
Science 2013;339(6117):303e7.
P0 ¼ 0.9814.
[6] Fritsch D, Bengtson G, Carta M, McKeown NB. Macromol Chem Phys
2011;212(11):1137e46.
[7] Bezzu CG, Carta M, Tonkins A, Jansen JC, Bernardo P, Bazzarelli F, et al. Adv
Mater 2012;24(44):5930e3.
[8] Laessig D, Lincke J, Moellmer J, Reichenbach C, Moeller A, Glaeser R, et al.
Angew Chem Int Ed 2011;50(44):10344e8.
[9] Nguyen LTL, Le KKA, Truong HX, Phan NTS. Catal Sci Technol 2012;2(3):
521e8.
microporosity by demonstrating significant adsorption at low
relative pressures. The apparent BET surface areas and pore vol-
umes of the polymers calculated from these isotherms are shown in
Table 1 along with those of the related PIM networks based on TBTQ
[22], CTC [19] and triptycene units [23].
The network polymer Triptindane-PIM A, demonstrates greater
microporosity, both in terms of apparent BET surface area and pore
volume, than the non-network polymer Triptindane-PIM B. This
finding is consistent with previous observations that polymer
networks are more efficient at generating intrinsic microporosity as
compared to non-network polymers assembled using the same
structural units [16,22]. In addition, Triptindane-PIM A demon-
strates greater microporosity than the related network PIMs
derived from tri-catechol components such as CTC [19] and TBTQ
[22] but less than that from triptycene [23] (Fig. 1). It is possible to
predict from the molecular structure of monomer M1 (Fig. 3a), as
characterised by single crystal X-ray diffraction analysis, that the
shape of the developing polymer network would more closely
resemble that of Tript-PIM than CTC-PIM. In contrast, consideration
of the structure of monomer M3 (Fig. 2) suggests that the Cen-
trotriindane-PIM would form almost linear chains with the
microporosity presumably resulting from the rigid aromatic ring
protruding from the chains blocking efficient packing.
[10] Corma A, Diaz-Cabanas MJ, Martinez-Triguero J, Rey F, Rius J. Nature
2002;418(6897):514e7.
[11] Alvarez-Ayuso E, Garcia-Sanchez A, Querol X. Water Res 2003;37(20):
4855e62.
[12] Jorda-Beneyto M, Suarez-Garcia F, Lozano-Castello D, Cazorla-Amoros D,
Linares-Solano A. Carbon 2007;45(2):293e303.
[13] Xing W, Liu C, Zhou Z, Zhang L, Zhou J, Zhuo S, et al. Energy Environ Sci
2012;5(6):7323e7.
[14] Ding S-Y, Wang W. Chem Soc Rev 2013;42(2):548e68.
[15] Preis E, Widling C, Scherf U, Patil S, Brunklaus G, Schmidt J, et al. Polym Chem
2011;2(10):2186e9.
[16] McKeown NB, Budd PM. Macromolecules 2010;43(12):5163e76.
[17] Carta M, Msayib KJ, McKeown NB. Tetrahedron Lett 2009;50(43):5954e7.
[18] Hashem M, Bezzu CG, Kariuki BM, McKeown NB. Polym Chem 2011;2(10):
2190e2.
[19] McKeown NB, Gahnem B, Msayib KJ, Budd PM, Tattershall CE, Mahmood K,
et al. Angew Chem 2006;118(11):1836e9.
[20] Wang Y, McKeown NB, Msayib KJ, Turnbull GA, Samuel IDW. Sensors
2011;11(3):2478e87.
[21] Mackintosh HJ, Budd PM, McKeown NB. J Mater Chem 2008;18(5):573e8.
[22] Vile J, Carta M, Bezzu CG, McKeown NB. Polym Chem 2011;2(10):2257e60.
[23] Ghanem BS, Hashem M, Harris KDM, Msayib KJ, Xu M, Budd PM, et al. Mac-
romolecules 2010;43(12):5287e94.
[24] Harig M, Kuck D. Eur J Org Chem 2006;2006(7):1647e55.
[25] Eaton PE, Carlson GR, Lee JT. J Org Chem 1973;38(23):4071e3.
4. Conclusions
[26] Bjork JA, Brostrom ML, Whitcomb DR.
J Chem Crystallogr 1997;27(4):
223e30.
Microporous polymers were prepared from novel “centro-
triindane” and “triptindane” monomers. The network “triptindane”
[27] Anzenhofer K, De Boer JJZ. Kristallogr Kristallgeometrie Kristallphys Kris-
tallchem 1970;131(1e2):103e13.