ChemComm
Communication
The solubility is expected to be a consequence of the fluorenyl units selectivities could arise as a consequence of a ‘‘molecular sieving
within the polymer backbone, i.e. the cardo principle.17
effect’’ due to very small pore sizes as observed previously in modified
The thermal transformation of the salts into PIs was monitored by zeolites.21 To see whether the as-made microporous polyimides are
HT-XRD using synchrotron irradiation. The recorded patterns (see indeed capable of separating CO2 and N2, column breakthrough tests
ESI†) reflect a transition from highly crystalline salts to amorphous were conducted. A gas mixture composed of CO2/N2 was allowed to
products. The overall amorphization occurs in both cases between flow through a column filled with 4a particles. The gas composition
250 1C and 275 1C, and has three main reasons. First, from the initial was analyzed by mass-spectrometry straight after the gas left the
arrangement of the two monomers in the salt’s crystal structure column (see ESI†). We found evidence for the successful separation of
(cf. Scheme 1 and Fig. 2b and c) a considerable molecular movement CO2/N2 gas mixtures, i.e. pure N2 left the column before the elution of
is necessary to react to the cyclic imide. Second, two equivalents of CO2. Desorption of N2 and CO2 by He purge could be achieved
water are liberated per imide ring formation. As water molecules move proving the physisorptive nature of the process.
considerably at these temperatures and also tend to phase separate
In summary, the successful solid-state transformations of monomer
from the formed PI chains, we believe that they play a major role in salts to aromatic microporous PIs at ambient pressure have been
the global amorphization. Third, the kinked structure of BAPF 1 evidenced. The crystal structure of 3a has been solved for the very first
should allow for a variety of energetically comparable conformations, time. The polymerization yields structural amorphization. The necessary
additionally disturbing chain-to-chain packing. While the internal internal structural rearrangements do not affect the external crystal
structure amorphizes, the macroscopic crystal habit persists, empha- habit and identical copies of the monomer salt crystallites can be
sizing the solid-state nature of the transformation. The amorphization obtained. The obtained microporous PIs are able to separate CO2 from
goes along with the development of a special type of microporosity, as N2, indicating that these amorphous polymeric materials are an inter-
evidenced by CO2 adsorption experiments conducted at 273.15 K and esting addition to the field of classic crystalline microporous materials. It
298.15 K. The porosity of the PIs synthesized via SSP shows almost can be expected that the presented synthesis routine will have an impact
equal gas uptake (B1.2 mmol gꢀ1 at 273.15 K; see ESI†) as chemically on the sustainable synthesis of a variety of functional HPPs opening
equivalent PIs that have been synthesized using solution-based thereby new possibilities in the area of science and technology.
routines previously, proving the intrinsic nature of the micropores.11,18
This work was supported by the ERC Senior Excellence Grant
Note that the PI particles are apparently non-porous in nitrogen ‘‘HYDRAChem’’, project no. 227639, and the Max Planck Society.
adsorption experiments at 77.4 K, i.e. the developing intrinsic porosity The authors are indebted to Dr Antje Wilke, Martin Blu¨mke and
is very specifically suited for CO2. Consequently, we explored the Dr Anthony Bell (B2 beamline, DESY Hamburg) for experimental
versatility of the materials for gas separation applications. We calcu- support at the synchrotron. Jessica Brandt, Marlies Graewert and
¨
lated the separation selectivity a for the gas pair CO2/N2 based on CO2 Antje Volkel are acknowledged for synthetical support, SEC measure-
and N2 single gas adsorption isotherms obtained at 273.15 K and ment and AUC measurement and data treatment, respectively.
298.15 K, respectively, using the ideal adsorbed solution theory (IAST)
methodology (see ESI†).19 The PIs showed a high a(CO2/N2) of >1000
at 273.15 K, and a a(CO2/N2) of >300 at 298.15 K (gas composition of
Notes and references
1 W. Volksen, R. D. Miller and G. Dubois, Chem. Rev., 2010, 110, 56.
15/85 vol%; Fig. 3).
2 P. M. Hergenrother, High Perform. Polym., 2003, 15, 3.
The capacity and selectivity drop might be explained by the less
pronounced condensation tendency of confined CO2 at higher T,
which is itself connected to the overall uptake. However, the predicted
capacity for CO2 adsorption from the mixed gas is in a comparable
regime to the pure gas adsorption, and thus is highly interesting for
practical applications, where a significant drop in the capacity is often
reported. The scientific literature regarding selectivity and capacity of a
gas is still rather inhomogeneous, but it can be stated that outstanding
gas uptake does not necessarily suggest high selectivities.20 Such high
3 D. Wilson, H. D. Stenzenberger and P. M. Hergenrother, Polyimides,
Blackie & Son Ltd., Glasgow, London, 1990.
4 Polyimides – Thermally Stable Polymers, ed. M. I. Bessonov, M. M. Koton,
V. V. Kudryavtsev and L. A. Laius, Consultants Bureau, N.Y., New York,
1987, vol. 4.
5 W. H. Carothers, J. Chem. Soc., Faraday Trans., 1936, 32, 39.
6 M. Antonietti, Nachr. Chem., 2011, 59, 1027.
7 R. Noyori, Chem. Commun., 2005, 1807.
8 M. M. Unterlass, D. Kopetzki, M. Antonietti and J. Weber, Polym.
Chem., 2011, 2, 1744.
9 C. D. Papaspyrides, Polymer, 1988, 29, 114.
10 Y. Imai, Progress in Polyimide Chemistry, Springer-Verlag, Berlin-Heidelberg,
1999, vol. 140, p. 1.
11 N. Ritter, I. Senkovska, S. Kaskel and J. Weber, Macromolecules,
2011, 44, 2025.
12 J. Dirkx, Analytical Density Gradient Ultracentrifugation, Beckman
Instruments GmbH, Munich, 1964.
¨
¨
13 W. Machtle and L. Borger, Analytical Ultracentrifugation of Polymers
and Nanoparticles, Springer Verlag, Berlin-Heidelberg, 2006.
14 S. K. Bhattacharjee and H. L. Ammon, Acta Crystallogr., Sect. B, 1981,
37, 2082.
15 V. Favre-Nicolin and R. Cerny, J. Appl. Crystallogr., 2002, 35, 734.
16 D. A. Haynes and L. K. Pietersen, CrystEngComm, 2008, 10, 518.
17 V. V. Korshak, et al., J. Macromol. Sci., Rev. Macromol. Chem., 1974, C11, 45.
18 A. Thomas, Angew. Chem., Int. Ed., 2010, 49, 8328.
19 A. L. Myers and J. M. Prausnitz, AIChE J., 1965, 11, 121.
20 R. Dawson, A. I. Cooper and D. J. Adams, Polym. Int., 2013, 62, 345.
21 F. Akhtar, Q. Liu, N. Hedin and L. Bergstrom, Energy Environ. Sci.,
2012, 5, 7664.
Fig. 3 (a) IAST selectivity of 4a for a gas mixture (0.15/0.85 of CO2/N2) at
298.15 K; (b) single gas adsorption isotherms of CO2 and N2 on 4a together with
predicted uptakes from a 0.15/0.85 gas mixture of CO2/N2 (open symbols).
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun.