Kinetically controlled porosity in a robust organic cage material

Article abstract of DOI:10.1002/anie.201209922

Won by the hare: A new crystalline microporous solid was prepared that is composed of discrete organic cages made entirely from carbon-carbon bonds. The porosity of this material can be controlled by simple kinetic methods, which afford reproducible access to either a non-porous or porous polymorph. Copyright

Full text of DOI:10.1002/anie.201209922

Angewandte
Chemie
DOI: 10.1002/anie.201209922
Porous Molecular Solids
Kinetically Controlled Porosity in a Robust Organic Cage Material**
Antonio Avellaneda, Peter Valente, Alexandre Burgun, Jack D. Evans, Adrian W. Markwell-
Heys, Damien Rankine, David J. Nielsen, Matthew R. Hill, Christopher J. Sumby,* and
Christian J. Doonan*
Microporous materials are of significant interest owing to
their central role in gas storage, separation processes, and
catalysis.[1–4] Recently, microporous molecular solids com-
posed of discrete, shape-persistent organic cages have
received growing attention[1] because they possess unique
properties that set them apart from conventional, extended
network materials, such as zeolites,[2] metal–organic frame-
works,[3] and covalent organic frameworks.[4] For example,
molecular solids are readily solution-processable,[5] provide
facile access to multicomponent materials by mix-and-match
synthesis,[6] and, by virtue of their noncovalent intermolecular
packing, can exhibit advanced properties, such as adsorbate-
triggered on/off porosity switching.[7]
cage molecule (C1) that is constructed entirely from thermo-
dynamically robust carbon–carbon bonds and has the molec-
ular formula C112H62O2 (Scheme 1). Furthermore, we demon-
strate kinetically controlled access to two crystalline poly-
morphs C1a and C1b that possess dramatically different N2
porosities: polymorph C1a, which is nonporous to N2, and
polymorph C1b, which affords a BET surface area of
1153 m2 gÀ1.
Unlike extended networks, where solvent-accessible voids
are linked through rigid covalent framework solids composed
of discrete organic cages predominantly aggregate by rela-
tively weak dispersion forces. Predicting the crystal structures
of such weakly aggregating materials is a long-standing
challenge in solid-state chemistry,[8] and is, in this field,
inherently coupled to estimating the ultimate porosity of
a molecular solid from its building units, as different
polymorphs can afford solids with dramatically different
surface areas.[9] Accordingly, relatively few examples of
porous organic solids have been reported.[1d] Nevertheless,
recent work from the laboratories of Cooper and Mastalerz
have demonstrated that the porosity of such materials can be
modified through crystal engineering strategies and synthetic
processing.[5a,10] Herein we describe the synthesis and charac-
terization of a novel, permanently porous, shape-persistent
Scheme 1. Procedure for the synthesis of trigonal-prismatic cage C1.
Molecule C1 was synthesized by Eglinton homocoupling
of two rigid, alkyne-terminated building units (Scheme 1; 2).
Such reactions, which are often conducted with a stoichio-
metric excess of copper reagents, have been widely employed
in macrocycle synthesis.[11] The cage precursor, compound 2,
can be elaborated from a tripodal building block, 4-[tris(4-
iodophenyl)methyl]phenol,[12] by sequential phenol methyl-
ation, Sonogashira coupling, and silyl deprotection reactions
in 53% yield over three steps.[13] The ultimate homocoupling
step proceeds under high-dilution conditions with a large
excess of catalyst to maximize the yield of the kinetic product
C1. The yield of C1 (20%) is remarkable given the
irreversible nature of the bonding involved and the fact that
one incorrect bond formation step during cage synthesis will
direct the reaction towards the formation of oligomers. No
other major products are isolated in this reaction that requires
three Eglinton homocoupling reactions. The energy-mini-
mized structure of C1 is best described as a distorted
triangular prism with internal vertical and horizontal diam-
eters of 13.5 ꢀ and 12 ꢀ, respectively.[14]
[*] Dr. A. Avellaneda, Dr. P. Valente, Dr. A. Burgun, J. D. Evans,
A. W. Markwell-Heys, D. Rankine, Dr. C. J. Sumby, Dr. C. J. Doonan
School of Chemistry and Physics, The University of Adelaide
Adelaide, South Australia, 5005 (Australia)
E-mail: christian.doonan@adelaide.edu.au
Dr. D. J. Nielsen
Human Protection and Performance Division
Defence Science and Technology Organisation
506 Lorimer St, Fishermans Bend, Victoria, 3207 (Australia)
Dr. M. R. Hill
CSIRO, Materials Science and Engineering
Private Bag 33, Clayton South MDC, Victoria 3169 (Australia)
[**] C.J.D. and C.J.S. would like to acknowledge the Australian Research
Council for funding (DP 120103909 (C.J.D.) and FT 100100400
(C.J.D.), and FT0991910 (C.J.S.)). D.J.N. and C.J.D. acknowledge
support from the DSTO Fellowship programme. J.D.E. thanks
CSIRO Materials Science and Engineering for a top-up Ph.D.
scholarship.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201209922.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
The most common strategy used to synthesize organic
cages is to employ covalent dynamic imine chemistry to
facilitate isolation of the thermodynamic molecular product
from a one-step reaction. Inspired by analogous chemistry,[15]
we aimed to expand the reaction space of such porous
molecular solids by synthesizing a thermodynamically robust
shape-persistent cage molecule by homocoupling of a single
component. Furthermore, we note that the one-step synthesis
of related multicomponent cages may be possible, under such
bond-forming conditions, by judicious choice of templating
strategies.[16]
chloroform, dichloromethane, and benzene, but insoluble in
alcohols, H2O, and other highly polar solvents.
Large colorless block-shaped crystals of C1 formed in
approximately 24 h from slow evaporation of a dichlorome-
thane/methanol solution of C1. The crystal structure of C1[17]
(Figure 1) closely resembles the energy-minimized structure
identified by computational approaches. The vertical and
horizontal outer dimensions of C1 are circa 3.1 nm by 1.6 nm,
and these enclose an internal cavity of the dimensions noted.
The volume occupied by a cage molecule is about 1300 ꢀ3.
The preorganized tripodal building block adopts the geom-
etry anticipated from initial modeling and structure predic-
tion with angles of 104.1(3)–111.5(3)8 around the tetraphenyl
carbon atom. The dialkynyl struts are close to linear and all
alkyne moieties have the expected bond lengths (in the range
1.171(5)–1.224(5) ꢀ). This single-crystalline polymorph, C1a,
crystallizes in the orthorhombic space group Pbcn with four
molecules in the unit cell. Consideration of the packing
reveals that the cages pack in a herringbone-type arrange-
ment, if the cages are treated as rods along their long
molecular axis. Each individual molecule of C1 packs closely
with four other molecules of C1 in the same orientation and
two sets of four additional cages, with a near-orthogonal
direction of their molecular axis, at the poles of the first cage
(Figure 1b). This has the effect of placing at least two
molecules of C1 into each window of an individual cage.
Owing to the lack of functional groups directing the packing,
the primary intercage forces in the crystal packing are van der
Waals interactions and edge-to-face p-stacking interactions
The formation of C1 was confirmed by 1H NMR and
13C NMR spectroscopy, which showed aromatic resonances in
the range 7.76–6.85 ppm and a resonance for the methoxy
1
group at 3.83 ppm in the H NMR spectrum; these are all
consistent with the cage structure.[13] The alkyne proton of 2 is
notably absent from the 1H NMR spectrum of C1, and
electrospray ionization mass spectrometry (ESI-MS) showed
a peak for the parent ion at m/z 1439, which corresponds to
+
ꢀ
[C1]H . Two weak IR bands for the C C stretches at 2219 and
2207 cmÀ1 were readily apparent. Bulk samples of C1 were
readily desolvated and stable up to about 4008C, as deter-
mined by concomitant thermal gravimetric analysis–differ-
ential scanning calorimetry (TGA-DSC) experiments. It is
noteworthy that the DSC trace showed no evidence of
chemical transformations below 4008C. This is quite remark-
able given the close proximity of three diyne moieties but
points to the overall rigidity and thermal robustness of the
structure. C1 is soluble in common organic solvents, such as
Figure 1. a) Representations of the structure of C1 and b) wire-framed depiction of the packing of C1a down the b axis. c) N2 accessible surface
area of C1a and d) the simulated mean-squared displacement of N2 and H2 through C1a.
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
involving both phenyl and alkyne moieties.[18] The crystals of
C1 contain residual solvent electron density peaks that could
not be definitively identified and the SQUEEZE routine of
Platon[19] was applied to the collected data.[20] Powder X-ray
diffraction (PXRD) was used to confirm that the single crystal
structure was representative of the bulk sample and deter-
mine that the a polymorph was retained subsequent to
evacuation of residual solvent.[13]
freeze drying is very unusual for porous molecular solids,
however, we note that PXRD methods are silent to the
presence of an amorphous component. Scanning electron
microcoscopy (SEM) indicated that polymorph C1b forms
thin plate-like crystallites, in contrast to polymorph C1a that
form block-shaped crystals.[13] This plate-like morphology of
the C1b polymorph accounts for the broadness of the peaks
and weak high-angle diffraction in the PXRD. Our contention
that access to crystalline polymorphs C1a and C1b is a kineti-
cally driven process is supported by PXRD experiments
carried out on samples of C1 generated from supersaturated
solutions on a rotary evaporator (Figure 2). Solvent evapo-
ration from a solution of C1 in dichloromethane gives rise to
crystalline solids with PXRD patterns that are consistent with
a mixture of both polymorphs. On the qualitative time scales
investigated in this work, solvent evaporation (minutes) lies in
the intermediate range between single crystal growth by slow
evaporation (hours to days) and rapid precipitation (seconds).
We acknowledge that exploration of other crystallization
techniques, solvent combinations, and temperatures may
provide access to additional polymorphs. Nevertheless, we
clearly demonstrate predictable and reproducible access to
polymorphs C1a and C1b by simple kinetic control. These
observations suggest that crystallization of C1 follows Ost-
waldꢁs rule, as C1b is kinetically trapped in a metastable
crystalline phase that upon dissolution and slow crystalliza-
tion affords the thermodynamically favored form C1a.[23]
Although the formation of C1a and C1b is kinetically
driven, variable-temperature PXRD experiments showed no
evidence that a thermodynamic phase transition occurs in the
solid state.
We assessed the permanent porosity of polymorph C1b by
first evacuating solvent molecules from its pores (12 hours,
2 mTorr, 298 K) and then measuring a N2 isotherm at 77 K
(Figure 3). The isotherm shape is best described as type 1,
which is consistent with pore diameters of less than 2 nm. The
slight hysteretic behavior suggests the presence of structural
inhomogeneity or poor uniformity in crystal size distribution
and this has been observed in other porous molecular solids
and flexible metal–organic framework materials.[10a] BET
analysis of the isotherm in Figure 3 indicates that C1b has
a surface area of 1153 m2 gÀ1. It is noteworthy that surface
areas in excess of 1000 m2 gÀ1 are rare for molecular cages.[1]
Additionally, polymorph C1b can be dissolved and precipi-
The simulated accessible pore space of N2 displayed in
Figure 1c shows that the structure of C1a contains one-
dimensional channels comprised of adjacent cages connected
by windows of about 4 ꢀ. Given that the kinetic diameter of
N2 is 3.64 ꢀ[21] it was expected that that these windows would
restrict the diffusion of N2 through the material. To support
this hypothesis we employed molecular dynamics to simulate
the diffusion of H2 and N2 within the pore structure of C1a.[22]
This was determined by measuring the mean-square displace-
ment of single N2 and H2 molecules for 3 ns after 1 ns of
equilibration at 77 K. Figure 1d clearly shows that the motion
of N2 is constrained within polymorph C1a while the smaller
H2 is able to diffuse through the structure by the circa 4 ꢀ
windows. In accordance with these structure simulations, 77 K
N2 isotherms indicated that activated samples of C1a were
non-porous to N2 but porous to H2, affording a total uptake of
approximately 40 cm3 gÀ1 at 77 K. However, C1a can be
considered a “soft” porous crystal, and it is plausible that with
greater gas loading pressures and temperature, slight struc-
tural deformations may allow N2 to diffuse through the
framework.
Rapid precipitation of C1 was found to reliably form
a second, kinetically trapped polymorph C1b. Addition of an
antisolvent to solutions of C1 or freeze drying from benzene
both form microcrystalline powders with identical PXRD
patterns (Figure 2). Upon solvent removal and drying,
polymorph C1b retains crystallinity and yields a PXRD
pattern that corresponds to the solvated forms, indicating
structural uniformity subsequent to guest removal. The
propensity of C1 to form a crystalline material following
Figure 2. PXRD patterns of desolvated samples of C1a (red) and C1b
(mauve). The green PXRD pattern shows a mixture of polymorphs
C1a, and C1b. The blue arrow on the right of the Figure is a guide for
the timescale in which each solid sample was crystallized.
*
*
Figure 3. N2 77 K isotherm of C1b. adsorption, desorption.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
[5] a) T. Hasell, S. Y. Chong, K. E. Jelfs, D. J. Adams, A. I. Cooper,
J. Am. Chem. Soc. 2012, 134, 588; b) T. Hasell, A. I. Cooper, Adv.
Mater. 2012, 24, 5732.
[6] a) S. Jiang, J. T. A. Jones, T. Hasell, C. E. Blythe, D. J. Adams, A.
Trewin, A. I. Cooper, Nat. Commun. 2011, 2, 207; b) T. Hasell,
S. Y. Chong, M. Schmidtmann, D. J. Adams, A. I. Cooper,
Angew. Chem. 2012, 124, 7266; Angew. Chem. Int. Ed. 2012,
51, 7154.
[7] J. T. A. Jones, D. Holden, T. Mitra, T. Hasell, D. J. Adams, K. E.
Jelfs, A. Trewin, D. J. Willock, G. M. Day, J. Bacsa, A. Steiner,
A. I. Cooper, Angew. Chem. 2011, 123, 775; Angew. Chem. Int.
Ed. 2011, 50, 749.
tated several times without diminishing the accessible surface
area. These properties highlight the facile processability of
C1. Pore size distributions calculated by nonlocal density
functional theory from the adsorption data shows two main
peaks centered at approximately 6 ꢀ and 11 ꢀ. The larger
pore size corresponds well to the predicted internal pore
diameter of C1 and the presence of a second pore suggests
solvent accessible extrinsic voids. However, in the absence of
structural data, the contribution of intrinsic and extrinsic
porosity to the total surface area cannot be confirmed.
In summary, we have described the synthesis and charac-
terization of a robust organic cage that is constructed entirely
from carbon–carbon bonds. Solids of C1 can be predictably
crystallized by kinetic control into two separate polymorphs
C1a and C1b. Rapid precipitation of C1 leads to the
permanently porous polymorph C1b, which has a notably
high surface area of 1153 m2 gÀ1 for a molecular solid;
however, slow crystallization methods yield C1a, which was
found to be nonporous to N2 gas. Such control of polymorph-
ism is of great interest, as the properties of polymorphic
materials can, as in this present case, show remarkable
variation. Furthermore, fine control of polymorphism can
provide insight into the mechanism of multistage polymorphic
transitions from the beginning of crystallization to the
formation of stable solids. We are currently investigating if
the kinetic trapping methods observed in this work can be
generally applied to derivatives of C1. Additionally, we are
also synthesizing C1 analogues functionalized with moieties
designed to enhance its selective gas adsorption properties.
[8] G. R. Desiraju, Nat. Mater. 2002, 1, 77.
[9] T. Tozawa, J. T. A. Jones, S. I. Swamy, S. Jiang, D. J. Adams, S.
Shakespeare, R. Clowes, D. Bradshaw, T. Hasell, S. Y. Chong, C.
Tang, S. Thompson, J. Parker, A. Trewin, J. Basca, A. M. Z.
Slawin, A. Steiner, A. I. Cooper, Nat. Mater. 2009, 8, 973.
[10] a) M. J. Bojdys, M. E. Briggs, J. T. A. Jones, D. J. Adams, S. Y.
Chong, M. Schmidtmann, A. I. Cooper, J. Am. Chem. Soc. 2011,
133, 16566; b) M. W. Schneider, I. M. Oppel, H. Ott, L. G.
Lechner, H.-J. S. Hauswald, R. Stoll, M. Mastalerz, Chem. Eur. J.
2012, 18, 836.
[11] a) J. J. Li, Name Reactions: A Collection of Detailed Mechanisms
and Synthetic Applications, Springer, Berlin, 2009; b) H. A.
Stefani, A. S. Guarezemini, R. Cella, Tetrahedron 2010, 66, 7871.
[12] M. E. Gallina, B. Baytekin, C. Schalley, P. Ceroni, Chem. Eur. J.
2012, 18, 1528.
[13] See the Supporting Information for full experimental details.
[14] The structure of C1 was optimized to B3LYP/6-31G(d,p) theory
with default convergence criteria. The vertical diameter was
measured between the two sp3 carbon atoms of the cage
framework and the horizontal diameter was defined as twice
the average distance between the cage centroid and the diyne
moiety. Full reference to these methods is in the Supporting
Information.
Received: December 12, 2012
Published online: && &&, &&&&
[15] a) C. Zhang, C.-F. Chen, J. Org. Chem. 2007, 72, 9339; b) C.
Zhang, Q. Wang, H. Long, W. Zhang, J. Am. Chem. Soc. 2011,
133, 20995.
[16] a) R. L. E. Furlan, S. Otto, J. K. M. Sanders, Proc. Natl. Acad.
Sci. USA 2002, 99, 4801; b) C. D. Meyer, C. S. Joiner, J. F.
Stoddart, Chem. Soc. Rev. 2007, 36, 1705.
Keywords: cage compounds · crystallization landscapes ·
.
kinetic control · polymorphism · porosity
[17] Selected data for C1: C112H62O2, Mr 1439.72, T= 150(2) K,
orthorhombic, Pbcn, a = 31.9812(14), b = 18.5193(8), g =
[1] a) J. R. Holst, A. Trewin, A. I. Cooper, Nat. Chem. 2010, 2, 915;
b) A. I. Cooper, Angew. Chem. 2011, 123, 1028; Angew. Chem.
Int. Ed. 2011, 50, 996; c) M. Mastalerz, Angew. Chem. 2010, 122,
5164; Angew. Chem. Int. Ed. 2010, 49, 5042; d) M. Mastalerz,
Chem. Eur. J. 2012, 18, 10082.
20.3265(7) ꢀ, V= 12038.8(8) ꢀ3, Z = 4, 1calc. = 0.794 MgmÀ3
,
m = 0.046 mmÀ1
, F(000) 3000, colorless block, 0.33 ꢄ 0.31 ꢄ
0.06 mm3, 56266 reflections collected, 14052 independent
reflections [Rint = 0.1347], completeness 99.9%, data/restraints/
parameters 14052/0/515, GOF 0.871, R1 = 0.0888, wR2 = 0.2291,
(all data: R1 = 0.2581, wR2 = 0.2839). CCDC 913184 contains the
supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[18] S. Dawn, M. B. Dewal, D. Sobransingh, M. C. Paderes, A. C.
Wibowo, M. D. Smith, J. A. Krause, P. J. Pellechia, L. S. Shimizu,
J. Am. Chem. Soc. 2011, 133, 7025.
[2] M. E. Davis, Nature 2002, 417, 813.
[3] a) M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M.
OꢁKeeffe, O. M. Yaghi, Science 2002, 295, 469; b) J. L. C.
Rowsell, O. M. Yaghi, Microporous Mesoporous Mater. 2004,
73, 3; c) S. Kitagawa, R. Kitaura, S. Noro, Angew. Chem. 2004,
116, 2388; Angew. Chem. Int. Ed. 2004, 43, 2334; d) O. K. Farha,
J. T. Hupp, Acc. Chem. Res. 2010, 43, 1166.
[4] a) A. P. Cꢂtꢃ, A. I. Benin, N. W. Ockwig, A. J. Matzger, M.
OꢁKeeffe, O. M. Yaghi, Science 2005, 310, 1166; b) H. M. El-
Kaderi, J. R. Hunt, J. L. Mendoza-Cortꢃs, A. P. Cꢂtꢃ, R. E.
Taylor, M. OꢁKeeffe, O. M. Yaghi, Science 2007, 316, 268; c) F. J.
Uribe-Romo, J. R. Hunt, H. Furukawa, C. Klock, M. OꢁKeeffe,
O. M. Yaghi, J. Am. Chem. Soc. 2009, 131, 4570; d) J. R. Hunt,
C. J. Doonan, J. D. LeVangie, A. P. Cꢂtꢃ, O. M. Yaghi, J. Am.
Chem. Soc. 2008, 130, 11872; e) F. J. Uribe-Romo, C. J. Doonan,
H. Furukawa, K. Oisaki, O. M. Yaghi, J. Am. Chem. Soc. 2011,
133, 11478.
[19] A. L. Spek, Acta Crystallogr. Sect. D 2009, 65, 148.
[20] The SQUEEZE routine of PLATON was applied to the
collected data, which resulted in significant reductions in R1
and wR2 and an improvement in the GOF. R1, wR2 and GOF
before SQUEEZE routine: 18.6%, 52.4% and 1.22; after
SQUEEZE routine: 8.88%, 28.4% and 0.87.
[21] L. M. Robeson, J. Membr. Sci. 1991, 62, 165 – 185.
[22] T. X. Nguyen, S. K. Bhatia, J. Phys. Chem. C 2007, 111, 2212.
[23] T. Threlfall, Org. Process Res. Dev. 2003, 7, 1017.
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Porous Molecular Solids
A. Avellaneda, P. Valente, A. Burgun,
J. D. Evans, A. W. Markwell-Heys,
D. Rankine, D. J. Nielsen, M. R. Hill,
C. J. Sumby,*
C. J. Doonan*
&&&& — &&&&
Kinetically Controlled Porosity in a Robust
Organic Cage Material
Won by the hare: A new crystalline
microporous solid was prepared that is
composed of discrete organic cages
made entirely from carbon–carbon
bonds. The porosity of this material can
be controlled by simple kinetic methods,
which afford reproducible access to either
a non-porous or porous polymorph.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!