Directing the crystallization of dehydro[24]annulenes into supramolecular nanotubular scaffolds

Article abstract of DOI:10.1021/jacs.6b01939

The self-assembly of a series of dehydro[24]annulene derivatives into columnar stacks has been examined for its latent ability to form -conjugated carbon-rich nanotubular structures through topochemical polymerizations. We have studied the parameters affe

Full text of DOI:10.1021/jacs.6b01939

Subscriber access provided by TULANE UNIVERSITY
Article
Directing the Crystallization of Dehydro[24]annulenes
into Supramolecular Nanotubular Scaffolds
Mitsuharu Suzuki, Juliet F. Khosrowabadi Kotyk, Saeed Khan, and Yves Rubin
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b01939 • Publication Date (Web): 18 Apr 2016
Downloaded from http://pubs.acs.org on April 23, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 24
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
Mitsuharu Suzuki,† Juliet F. Khosrowabadi Kotyk, Saeed I. Khan, and Yves Rubin*
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Department of Chemistry and Biochemistry, University of California, Los Angeles, 607 Charles E. Young Drive East,
Los Angeles, CA 90095, United States
ABSTRACT: The self-assembly of a series of dehydro[24]annulene derivatives into columnar stacks has been examined for its
latent ability to form -conjugated carbon-rich nanotubular structures through topochemical polymerizations. We have studied the
parameters affecting self-assembly, including the nature of the substituent and crystallization conditions, using ten different dehy-
dro[24]annulene derivatives. In particular, hydrogen-bonding interactions through carbamate groups were found to be especially
useful at directing the formation of nanotubular supramolecular assemblies. We have also evaluated the electronic coupling be-
tween neighboring dehydroannulene molecules within these supramolecular assemblies. Density functional calculations on the
stacked supramolecular nanotube assemblies show that transfer integrals vary considerably between the three columnar assemblies,
ranging from moderate to high (59–98 meV for the highest occupied molecular orbitals, 63–97 meV for the lowest unoccupied mo-
lecular orbitals), depending on the local molecular topology. In addition, the dehydro[24]annulenes derivatives afforded distinct
architectures in the crystal, including nanochannel arrays, sheets with solvent-filled pores, and lamellae. This work is an essential
step toward a controlled formation of covalently-linked carbon-rich nanostructures generated from molecular precursors with a
latent diacetylene reactivity.
Scheme 1 shows the potential of this approach in which dehy-
dro[24]annulene macrocycles are stacked in a crystal to form
nanotubular assemblies. If these assemblies fulfill the required
packing parameters for a fourfold butadiyne topochemical
polymerization, it could occur on all four sides of the molecule.
The chemical and physical properties of organic molecules
can be closely correlated with their packing arrangement in the
crystalline state. In particular, several chemical reactions have
been investigated on the basis of the topochemical principle, in
The ensuing product would be a fully -conjugated carbon-
which the course and outcome of a reaction is primarily de-
fined by the solid-state arrangement of the reactants.1–7 The
rich nanotubular structure having a distinct framework topolo-
gy.
electronic and optoelectronic characteristics of molecular sem-
iconductors are also strongly affected by the solid-state pack-
Scheme 1. Fourfold topochemical polymerization of
a dehydro-
ing and morphology of individual molecules, making these
considerations directly relevant to the performance of organic
devices such as field-effect transistors and photovoltaic devic-
es.8–17 Accordingly, much research has been devoted to under-
standing how intermolecular interactions dictate molecular
packing, and how these interactions can be engineered into
molecular solids for their intended applications.18
[24]annulene.a
We have been working on such a molecular engineering ap-
proach to supramolecular architectures based on the dehydro-
annulenes,19 a class of compounds with a fully carbon-rich -
conjugated macrocycle containing several diacetylene units.
We chose the dehydroannulenes because of the inherently high
kinetic strain contained in their diacetylenic carbons, which are
easily prone to expand their coordination. Along this line, the
pioneering results by the groups of Vollhardt20,21 and Bunz22
has shown that dehydroannulenes can produce carbon nano-
tubes and carbon onions in explosive, high-temperature solid-
state reactions. Although these reactions afforded a messy
mixture of structures, we envisioned that an exquisitely con-
trolled solid-state reaction of dehydroannulenes using the top-
ochemical polymerization principle would give a pristine pol-
ymeric structure resulting from the highly organized state of
molecules within a crystal and the minimal amount of atom
movements in most topochemical polymerizations.7,23–32
aA columnar assembly of dehydro[24]annulene molecules that can poten-
tially undergo the butadiyne-to-polydiacetylene topochemical polymeriza-
tion at each of the four macrocycle edges (left, the reaction pathway is
indicated by red dashed lines), providing a π-conjugated nanotubular scaf-
fold (right, newly formed covalent C=C bonds shown in red). Ideal pack-
ing parameters for the butadiyne topochemical polymerization are: repeat
distance r = 4.9 Å, tilt angle  = 45°, and C1···C4' distance d = 3.5 Å.
Note that small deviations from these values are acceptable for topochemi-
cal polymerization.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 24
Chart 1. Structures of dehydro[24]annulene derivatives 1a–1j.
sity-functional calculations show that the transfer integrals
1
2
3
4
5
6
7
8
range from moderate to high for both the highest occupied
molecular orbitals (HOMOs) and the lowest unoccupied mo-
lecular orbitals (LUMOs), depending on the local arrangement
of molecules, confirming that there is a significant degree of
interaction between the dehydroannulene units in these stacks.
Overall, our experimental findings constitute an important first
step in the exploration of the chemical and physical properties
of rigid molecular solids constructed from dehydroannulenes.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Molecular Design. Among a wide variety of conceivable
dehydroannulene structures for the aim of this project (Scheme
1), dehydro[24]annulene 1 incorporating four butadiyne units
(Chart 1) was chosen as the best test system because of its
folded conformation and relatively high synthetic accessibility.
As reported in our preliminary communication,19 fourfold sub-
stitution with small polar groups such as hydroxymethyl (1a)
considerably improves the solid-state stability of the otherwise
highly reactive dehydro[24]annulene framework.55 Benzo-
annulation was avoided even though this motif is highly effec-
tive at stabilizing dehydroannulenes. The steric encumbrance
of benzene rings adjacent to butadiyne units has been known to
impede topochemical polymerization.56,57 To facilitate self-
association in the crystal, all the dehydro[24]annulene substit-
uents in compounds 1a–1j were endowed with either hydro-
gen- or halogen-bonding functionalities so that highly orga-
nized molecular packing motifs could be gained through direc-
tional noncovalent interactions. The methylene “hinge” be-
tween each functional group and the dehydro[24]annulene
framework was expected to allow a good degree of conforma-
tional flexibility to assist self-association processes, as well as
to accommodate structural changes during the topochemical
polymerization reactions in the solid state.
The topochemical polymerization of butadiynes has been
extensively studied, and thus, the principles of molecular-
design and reaction conditions for forming poly(butadiyne)s
are well established.7,23–33 In fact, the butadiyne topochemical
polymerization has not been limited to simple, linear substrates,
but also applied to macrocyclic systems.25,34–41 However, the
use of dehydroannulenes as monomers in topochemical
polymerization remains a significant challenge, even though
this concept has been invoked in several reports.42–48 Only one
successful example has been reported so far, in which a triben-
zodehydro[14]annulene was polymerized in the single-
crystalline state under high-pressure conditions.49 The scarcity
of controlled polymerization of dehydroannulenes appears to
be due, in part, to the general instability of these compounds in
the solid state. Indeed, dehydroannulenes are largely un-
derrepresented in the field of crystal engineering. The system-
atic investigation of their solid-state packing is limited to a few
rigid and relatively stable systems such as benzo-fused dehy-
dro[12]annulenes.43–47,50–52 For conformationally flexible de-
rivatives not endowed with stabilizing benzoannulation, crys-
tallographic characterization of molecular structures has been
rather elusive.53 Furthermore, the weak intermolecular π–π
interactions between acetylene units do not promote the for-
mation of face-to-face stacks of dehydroannulenes.54 Conse-
quently, tubular assemblies based on flexible, non-
benzoannulated dehydroannulenes are absent in the literature.
Conformational flexibility of the dehydro[24]annulene
framework and pseudopolymorphism. Diffraction parame-
ters and crystal data for all the crystal structures discussed in
this paper are presented in Table 1. Tetraol 1a, the simplest
derivative in the series, provided different packing motifs de-
pending on the solvent system employed for crystallization.
While all the dehydro[24]annulene derivatives 1a–1j are pale-
yellow in solution, red plates were obtained when 1a was crys-
tallized by slow diffusion of a small, linear or oblong-shaped
solvent (MeOH, MeCN, or CH2Cl2) into a THF solution. Sin-
gle-crystal X-ray diffraction analysis revealed that the dehy-
dro[24]annulene framework in these red crystals adopts an
unusual planar conformation, resulting in the formation of an
array of solvent-accessible channels, each of which is walled
by highly polarizable  electrons in the diacetylene units (Fig-
ure 1).
We are reporting here an extensive crystal-engineering study
on a series of the tetrasubstituted dehydro[24]annulenes 1a–1j
(Chart 1). By employing directional noncovalent interactions,
we have achieved nanotubular assemblies that possess near-
ideal packing parameters for the fourfold topochemical
polymerization described in Scheme 1. We have also accessed
a variety of other dehydro[24]annulene-based supramolecular
architectures through modification of the substituents and the
crystallization conditions. In addition, we have evaluated the
electronic coupling between neighboring dehydroannulene
molecules within the obtained supramolecular nanotubes. Den-
As the packing motifs are essentially the same for 1a crys-
tallized from all three solvent systems (THF with MeOH,
MeCN, or CH2Cl2), only the pseudopolymorph containing
CH2Cl2 (referred to as 1a-I) is described here. The planarized
macrocyclic framework in 1a-I has nonequivalent minimum
distances for the two sets of mutually opposed edges, indicat-
ing a strong distortion due to crystal packing effects: i.e.,
7.546(3) Å for C1∙∙∙C2ʹ and 8.228(3) Å for C7∙∙∙C8ʹ (Figure
1a). The Cl–C–Cl planes of the occluded CH2Cl2 molecules
are approximately parallel to the longer edge-to-edge axis. All
the four hydroxy groups of each macrocycle are engaged in
ACS Paragon Plus Environment

Page 3 of 24
Journal of the American Chemical Society
intermolecular hydrogen bonds with a cyclic tetragonal motif,
between aliphatic alcohols.58 The corrugated sheets stack on
top of each other, mediated by short C(sp2)–H∙∙∙O contacts of
2.39(2) Å (H5∙∙∙O2, Figure 1b) as well as C(sp2)∙∙∙C(sp) con-
tacts of 3.389(2) Å (C4∙∙∙C9, see Supporting Information (SI)).
1
2
3
4
which results in the formation of a corrugated sheet structure
comprising the dehydro[24]annulene frameworks (Figure 1b).
The corresponding intermolecular O1∙∙∙O2 distances of
2.814(1) and 2.764(1) Å are typical values for hydrogen bonds
5
6
Table 1. Single-crystal X-ray diffraction parameters and crystal data.
7
8
1a-I
1a-II
1a-III
1b
1c
1d
1f
C28H16O4
·CH2Cl2
C28H16O4
·(C4H10O)2
C28H16O4
·C3H6O
C56H32O12
·(C4H8O)5
C56H32O12
·(CH2Cl2)3
C56H32O16
·(C3H6O)3
9
formula
C56H28I4O8
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
formula weight
crystal system
space group
a (Å)
501.33
564.65
474.49
1257.34
1151.59
1135.05
1336.38
Triclinic
Triclinic
Triclinic
Monoclinic
Triclinic
Triclinic
Triclinic
̅
P1
̅
P1
̅
P1
̅
P1
̅
P1
̅
P1
P21/n
7.127(2)
7.465(2)
11.374(4)
96.247(3)
90.085(4)
100.706(4)
590.9(3)
1
8.6953(14)
8.8490(14)
20.955(3)
83.764(2)
81.190(2)
89.372(2)
1583.9(4)
2
16.538(10)
6.995(4)
22.907(14)
90
9.613(5)
13.191(7)
13.847(7)
70.530(6)
89.281(6)
79.425(6)
1625.2(14)
1
7.8633(11)
12.4758(18)
14.475(2)
87.013(2)
79.310(2)
78.040(2)
1364.9(3)
1
8.0494(17)
12.367(3)
14.371(3)
87.308(2)
76.651(2)
77.200(3)
1357.3(5)
1
9.2751(14)
11.8804(18)
13.115(2)
103.213(2)
104.214(2)
112.564(1)
1206.6(3)
1
b (Å)
c (Å)
(deg)
 (deg)
(deg)
V (ų)
106.113(6)
90
2546(3)
4
Z
T (K)
100(2)
100(2)
100(2)
100(2)
100(2)
100(2)
100(2)
crystal description red plate
red block
yellow block
red prism
red plate
red plate
red plate
crystal size (mm3) 0.25 × 0.15 × 0.10 0.15 × 0.05 × 0.05 0.30 × 0.10 × 0.05 0.20 × 0.10 × 0.05 0.40 × 0.30 × 0.05 0.35 × 0.20 × 0.10 0.20 × 0.10 × 0.04
7.64, 58.56
6909
7.40, 57.21
8025
7.80, 58.24
7115
7.22, 58.04
6995
7.46, 56.66
5867
2min, 2max (deg)
no. of refl. (unique) 3103
2761


7663
4530
4852
4873
5008
5093
5092
no. of refl. (I>2(I))
R1, wR2 (all data) 0.0518, 0.1268
0.1194, 0.2063
0.0673, 0.1776
1.034
0.0741, 0.1172
0.0458, 0.1050
1.028
0.1067, 0.1677
0.0585, 0.1411
1.013
0.1065, 0.2109
0.0770, 0.1910
1.114
0.0914, 0.1938
0.0662, 0.1765
1.086
0.0294, 0.0573
0.0232, 0.0545
1.026
R, wR (I > 2(I))
GOF on F2
, e Å–3
0.0466, 0.1224
1.044
0.818, –0.569
0.367, –0.233
0.248, –0.292
0.589, –0.360
1.009, –0.606
0.537, –0.593
0.780, –0.392
1b·2b*
C56H32O12
·C14H14N4O2
1167.11
Monoclinic
C2/c
1a·2a
C28H16O8
·C14H14N4O2
1g-I*
1g-II*
C60H44N4O8
·(C4H8O)2
1093.20
Orthorhombic
Pca21
1h
1i*
1j
C60H44N4O12
·(C3H6O)4
formula
C60H44N4O8
C56H68N4O8
C56H36N4O8
formula weight
crystal system
space group
a (Å)
686.70
948.99
Monoclinic
C2
925.14
892.89
1245.30
Triclinic
̅
P1
Triclinic
̅
P1
Triclinic
̅
P1
Tetragonal
̅
I4
38.92(2)
4.918(3)
38.53(2)
90
8.896(3)
9.089(3)
10.731(4)
88.821(4)
89.682(5)
85.649(4)
865.0(5)
1
21.686(9)
4.784(2)
27.556(13)
90
9.9571(10)
17.4414(19)
42.952(5)
90
10.0954(13)
16.195(2)
16.930(2)
105.474(2)
95.831(2)
91.576(2)
2649.5(6)
2
25.112(7)
25.112(7)
8.489(3)
90
10.2111(8)
10.4183(8)
16.0100(12)
78.912(1)
83.712(1)
72.751(1)
1593.7(2)
1
b (Å)
c (Å)
(deg)
 (deg)
(deg)
V (ų)
101.655(12)
90
102.913(10)
90
90
90
90
90
7224(7)
4
2786(2)
2
7459.3(14)
4
5353(3)
4
Z
T (K)
100(2)
100(2)
100(2)
100(2)
100(2)
100(2)
100(2)
crystal description yellow needle
red block
yellow needle
yellow block
yellow block
yellow needle
red prism
crystal size (mm3) 0.60 × 0.20 × 0.05 0.25 × 0.20 × 0.15 0.40 × 0.25 × 0.05 0.50 × 0.35 × 0.15 0.40 × 0.30 × 0.05 0.35 × 0.15 × 0.10 0.50 × 0.30 × 0.15
8.14, 57.06
no. of refl. (unique) 9096
5366
7.60, 57.64
4377
7.52, 56.72
3820
7.82, 58.22
19972
7.54, 56.56
12848
7.56, 56.94
3593
7.80, 58.18
8275
2min, 2max (deg)
3863
2705
14515
8008
2184
6862
no. of refl. (I>2(I))
R1, wR2 (all data) 0.1166, 0.2056
0.0432, 0.1019
0.0381, 0.098
1.018
0.0994, 0.2067
0.0754, 0.1932
1.040
0.0935, 0.2147
0.0718, 0.2008
1.037
0.1279, 0.2157
0.0764, 0.1951
1.094
0.1078, 0.1738
0.0637, 0.1556
1.073
0.0499, 0.1058
0.0396, 0.0991
1.030
0.0729, 0.1852
0.962
R, wR (I>2(I))
GOF on F2
, e Å–3
0.291, –0.277
0.453, –0.244
0.254, –0.216
0.490, –0.320
0.344, –0.297
0.258, –0.212
0.377, –0.187
*The disordered solvent molecules could not be properly refined, and the corresponding diffused electron density was treated with the SQUEEZE routine of
the PLATON program.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 24
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Representations of the pseudopolymorph 1a-I. Selected bond angles and interatomic distances are given in degrees and angstroms, respectively.
(a) Thermal-ellipsoid plot of tetraol 1a at 50% probability. (b) Sheet motif observed in 1a-I. Inset: Close-up view highlighting the intrasheet tetragonal
hydroxy hydrogen-bonding motif (green dashed lines) and intersheet C(sp2)–H∙∙∙O interactions (magenta dashed lines). (c, d) Molecular packing in 1a-I
viewed along and perpendicular to the nanochannel axis, respectively. Disordered solvent molecules (CH2Cl2) within the channels are omitted for clarity in
(c), whereas they are shown as yellow ball-and-stick models in (d).
The nanochannel array obtained here is reminiscent of the
organic crystal with wide channels reported by Moore et al.59
The authors showed that a phenolic derivative of a hexagonal-
ly shaped m-phenylene-ethynylene macrocycle formed extend-
ed porous sheets in which the associated hydrogen-bonding
motif could be described as an 푅33(33) graph set according to
Etter’s terminology.60 The sheets stacked with a proper registry
to construct solvent-accessible channels with ca. 9 Å in diame-
ter. The authors attributed the observed inter-sheet alignment
to π–π stacking interactions between the aromatic rings. In the
present case of our system 1a-I, the square-shaped tetraol
forms a much “denser” hydrogen-bonding motif of four hy-
droxy groups that can be described as an 푅44(8) graph set, and
the resulting sheet structure is not flat but corrugated. The cor-
rugation is probably there to accommodate optimal hydrogen-
bonding interactions with minimal angle strain. The – inter-
actions between dehydroannulene frameworks are weak54 and
are likely less responsible for the observed intersheet registry
as compared to other factors such as effective solvent inclusion
and the formation of the C(sp2)–H∙∙∙O contacts.
narized macrocycles and two disordered Et2O molecules. The
two macrocycles are referred to as 1a-II(A) and 1a-II(B) and
differentiated by grey and light blue colors, respectively (Fig-
ure 2). Inspection of the extended crystal structure reveals
channel-like structures based on the macrocyclic frameworks
(Figure 2c). However, in contrast to the case of pseudopoly-
morph 1a-I, the solvent molecules in 1a-II are sandwiched
between two macrocycles rather than aligned along the chan-
nel axis (Figure 2d,e). Thus, the motif in the latter structure
would be better described as a “pseudochannel”. Several in-
termolecular interactions are observed between neighboring
molecules of 1a: The hydroxy–hydroxy interactions afford 1D
extended ladder structures (Figure 2b, green dashed lines),
which can be described as an 푅22(22) graph set. All the re-
maining hydroxy hydrogen atoms interact with the oxygen
atoms of Et2O (Figure 2e, red dashed lines for 1a-II(A); see SI
for 1a-II(B)). Consequently, there are no hydroxy hydrogen
bonds between neighboring ladder structures, and instead,
C(sp2)–H∙∙∙O contacts are present (Figure 2b, magenta dashed
lines). The closest intermacrocyclic C∙∙∙C contacts are found
between C21 and C25 of 1a-II(A) with a distance of 3.386(3)
Å (Figure 2e, light blue dashed lines).
A different pseudopolymorph 1a-II was obtained by slow
diffusion of Et2O into a THF solution of 1a (Figure 2). In this
case, the unit cell consists of two symmetry-independent pla-
ACS Paragon Plus Environment

Page 5 of 24
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Representations of pseudopolymorph 1a-II. Selected bond angles and interatomic distances are given in degrees and angstroms, respectively. (a)
Thermal-ellipsoid plot of molecule 1a-II(A) at 50% probability. (b) Ladder motif observed in 1a-II. Carbon atoms are grey for 1a-II(A), and blue for 1a-
II(B). Hydroxy–hydroxy and C(sp2)–H∙∙∙O interactions are indicated with green and magenta dashed lines, respectively. (c, d) Molecular packing in 1a-II
along and perpendicular to the pseudochannel structures, respectively. Disordered solvent molecules are omitted for clarity in (c), while they are shown as
yellow ball-and-stick models in (d). (e) Close-up view highlighting the sandwiched Et2O molecules and intermacrocyclic short contacts (light blue dashed
lines) in 1a-II(A). The interactions between the hydroxy groups of 1a-II(A) and the oxygen atoms of Et2O are indicated with red dashed lines. Each Et2O
molecule is disordered adopting two different conformations with a 50:50 probability, and only one conformation for each is shown for clarity.
The longitudinal and lateral edge-to-edge distances of each
macrocyclic framework in 1a-II are very similar (7.783(4) and
7.855(4) Å for 1a-II(A), 7.826(4) and 7.842(3) Å for 1a-II(B))
in clear contrast to the case of 1a-I (Figures 1a and 2a). In
connection with this observation, a DFT calculation predicts
that the edge-to-edge distances are essentially the same be-
tween the two orthogonal directions (7.881 and 7.913 Å) for
the planarized macrocyclic framework of 1a in the gas phase
(Figure S51 in SI).61 Additionally, numerous short contacts are
observed between solvent molecules (CH2Cl2) and the inner
wall of macrocyclic frameworks in 1a-I, all of which are along
the longer edge-to-edge axis of the framework. Based on these
observations, it is presumed that the macrocycles in 1a-I are
distorted to accommodate the CH2Cl2 molecules within the
inner cavity.62 In the case of 1a-II, the solvent Et2O is sterical-
ly more demanding and cannot fit within the macrocyclic cavi-
ty; thus, the solvents are sandwiched instead. In other words,
the formation of these two different pseudopolymorphs may be
primarily attributed to the differences in physical dimension of
the occluded solvents.
The third type of pseudopolymorph of tetraol 1a (1a-III)
was obtained using a ‘branched’ solvent. Single crystals grown
from a heterogeneous solid–liquid mixture of compound 1a
and acetone are light yellow and block-shaped. Their appear-
ance is in sharp contrast to those of pseudopolymorphs 1a-I
and 1a-II, which are both obtained as red plates. X-ray diffrac-
tion analysis revealed that pseudopolymorph 1a-III has
1a·(CH3)2CO stoichiometry, and the macrocyclic framework
of tetraol 1a adopts a folded conformation (Figure 3). All the
hydroxy groups are engaged in intermolecular hydroxy–
hydroxy interactions to form a helical motif along the b axis
(Figure 3c). The associated O∙∙∙O distances (2.657(2)–2.752(2)
Å) are slightly shorter than those observed in 1a-I or 1a-II,
probably reflecting the higher degree of polarization-
enhancement effect in the extended chain motif.58,63,64 Each
acetone molecule sits in the middle of a dehydro[24]annulene
framework (Figure 3a,b) without forming any hydroxy hydro-
gen bonds; instead, the acetone oxygen is in contact with three
C–H hydrogen atoms from two methylene groups and one
C(sp2)–H unit of 1a (Figure 3d).
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 24
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Representations of pseudopolymorph 1a-III. Selected bond angles and interatomic distances are given in degrees and angstroms, respectively. (a,
b) Packing diagrams along and perpendicular to the b axis, respectively. (c) Close-up view of the 1D extended hydroxy hydrogen-bonding network (red
dashed lines) (d) Close-up view highlighting the three C(sp2)–H∙∙∙O interactions associated with an acetone oxygen atom (green dashed lines). (e) Thermal-
ellipsoid plot of 1a at 50% probability.
Two of the three pseudopolymorphs of tetraol 1a involve
planar dehydro[24]annulene frameworks, which are conforma-
tionally strained and formally antiaromatic. The observed
near--perfect planarity is unusual in that the planarization is
achieved without any conformational constraints via covalent
bonds; e.g., small-ring annulation65,66 or inter-annular bridg-
ing.67 At the same time, this observation is reasonable if one
considers the fact that the energy required to planarize the
macrocyclic framework from the most stable nonplanar con-
formation is estimated to be less than 4 kcal mol–1 in the gas
phase by DFT calculations (Table S1, SI). The small enthalpy
requirement is easily compensated by stabilizing noncovalent
interactions in the crystalline state, such as hydrogen bonding
and van der Waals interactions. The conformational strain
associated with the planarization is manifested by the differ-
ences in bond angles between the macrocyclic frameworks in
1a-II and 1a-III (Figure 2a and 3e). The smallest difference is
observed for the substituted sp2 carbons (0.7° on average),
while the largest difference is for the sp carbons  to the sub-
stituted sp2 carbons (3.1° on average). The extent of the differ-
ences well represents the high bond-angle flexibility of sp-
hybridized carbon atoms.68 On the other hand, corresponding
bond lengths are essentially the same between the planar and
folded macrocyclic frameworks.
zen” into various degrees of planarity in the crystalline state
via noncovalent interactions. This conformational flexibility
may be beneficial in a multifold topochemical polymerization
reaction in that different manifolds of polymerization events
can be expected depending on the macrocycle conformation
and its solid-state arrangement, as proposed in our preliminary
communication.19 In addition, the experimental observations
described above carry useful implications for subsequent crys-
tal-engineering studies; i.e., the use of small, linear solvents
such as MeOH or MeCN may promote the formation of the
intended tubular assemblies, because these solvent molecules
can effectively fill the inner cavity of the macrocyclic frame-
work, thereby stabilizing the tubular structures. The size of
CH2Cl2 appears to be at the upper limit of best fit, and the use
of any larger or branched solvents will most likely result in
non-tubular structures.
Crystal packing of benzoate derivatives 1b–1f. Encour-
aged by the high solid-state stability and crystallinity of tetraol
1a, we then synthesized the three hydroxy-benzoate deriva-
tives 1b–1d, as well as the two iodobenzoate derivatives 1e
and 1f, to study their crystal packing modes. Functionalized
benzoate groups have been commonly employed as “handles”
in molecules for the construction of topochemically polymer-
izable assemblies of linear oligoynes.7
The pseudopolymorphism observed with 1a clearly demon-
strates that the dehydro[24]annulene framework can be “fro-
ACS Paragon Plus Environment

Page 7 of 24
Journal of the American Chemical Society
Single crystals of 4-hydroxybenzoate 1b were obtained as
1
red plates by vapor diffusion of n-pentane into a THF solution.
X-ray diffraction analysis revealed that 1b crystallized with a
unit-cell formula of 1b·(THF)5. The macrocyclic framework
again has a planar conformation as expected from the red color
of the crystals. All the substituents are nearly coplanar with
the macrocyclic core, and the maximum deviation of a nonhy-
drogen atom in the substituents is only 0.85 Å from the mean
plane of the macrocyclic framework. The four phenolic hy-
droxy groups are all hydrogen-bonded to THF oxygen atoms,
with corresponding O∙∙∙O distances of 2.62(1)–2.77(1) Å. As a
consequence, there is no direct hydrogen bonding between
hydroxy groups (Figure 4a). The one remaining THF molecule
in the unit cell is engaged in an Aryl–H∙∙∙O(THF) contact with
a H∙∙∙O distance of 2.43(2) Å. Neighboring molecules of 1b
are interacting through cyclic double Aryl–H∙∙∙O=C interac-
tions (푅22(10) motif) and Aryl–H∙∙∙O–H contacts, where the
corresponding H∙∙∙O distances are 2.68(3) and 2.50(3) Å, re-
spectively. Overall, these hydrogen-bonding interactions lead
to the formation of a 2D extended sheet structure. The 2D
sheets stack in such a way that the intersheet offset is approx-
imately one macrocyclic-framework distance, with no chan-
nels seen in the resulting packing (Figure 4b). There are no
direct short contacts between macrocyclic frameworks, and the
only intersheet C∙∙∙C short contacts are found between the
alkynyl carbon C12 and the aryl carbon C16 with an intera-
tomic distance of 3.275(3) Å (see SI). The inner cavity of each
macrocyclic framework is occupied with a THF molecule
from an adjacent layer either above or below with 50:50 prob-
ability. This uncertainty originates from the solvent disorder
shown in Figures 4c and 4d. Given Nature’s strong propensity
to minimize void spaces in molecular crystals, it is reasonable
to assume that this solvent arrangement, in which THF mole-
cules effectively fill both the intermolecular spaces as well as
the macrocyclic cavities, is an important factor in forming the
observed interlayer registry.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
Slow evaporation of a MeOH/acetone/H2O solution of 3,5-
dihydroxybenzoate 1d resulted in the formation of red plates.
Single-crystal X-ray diffraction analysis revealed that the mo-
lecular conformation of 1d in the crystalline state is similar to
that of 4-hydroxybenzoate 1b (Figures 4a and 5a); Namely,
the macrocyclic framework and the four substituents in 1d are
coplanar, and the maximum deviation of a substituent nonhy-
drogen atom from the mean plane of the macrocyclic frame-
work is small, being only 0.37 Å. In contrast, the hydrogen-
bonding motifs between these two crystal structures are very
different. The hydroxy groups of 1b are mainly involved in
interactions with the occluded solvent molecules, whereas
those of 1d are mostly involved in direct interactions with
neighboring macrocycles forming cyclic O–H∙∙∙O=C interac-
tions and O–H∙∙∙O–H hydrogen bonds (O∙∙∙O distances are
2.847(2) Å and 2.783(2) Å, respectively). Molecules are ar-
ranged to form a 2D extended sheet motif via these interac-
tions (Figure 5a), which stack so that the two different types of
vacant areas (inside and outside of the macrocyclic frame-
work) are located approximately on top of each other (Figure
5b). The two types of voids are occupied with disordered sol-
vent molecules. There are no direct short contacts between
dehydro[24]annulene frameworks, and the shortest intersheet
C∙∙∙C contacts are found between carbonyl and aryl groups
with an interatomic distance of 3.314(3) Å (see SI).
Figure 4. Representations of the crystal structure of 1b. (a) Pack-
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ing diagram showing the 2D extended sheet structure. Associated
hydrogen bonds are indicated with dashed lines, and the corre-
sponding interatomic distances are shown in angstroms. Hydrogen
atoms of THF molecules are omitted for clarity. Molecules shown
in yellow are disordered. (b) View perpendicular to the plane of
macrocycles to show the registry between three consecutive sheet
structures (top: navy, middle: green, bottom: gray). Hydrogen
atoms and solvent molecules are omitted for clarity. (c,d) Disor-
dered THF molecules viewed perpendicular to and along the plane
of macrocycles, respectively. Two sets of molecular arrangements
related by an inversion center are observed with a 50:50 probabil-
ity, which are differentiated by dark and light colors. The two
“standing” THF molecules in (d) fill the inner cavity of dehydro-
annulene framework in the next upper or lower sheet.
Single crystals of 3-hydroxybenzoate 1c were obtained as
red plates by vapor diffusion of n-hexane into a THF/CH2Cl2
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 24
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Representations of the crystal structure of 1f. Selected
interatomic distances and angles are shown in angstroms and de-
grees, respectively. (a) Packing diagram highlighting the quadru-
ple aryl stack sandwiched between two dehydro[24]annulene
frameworks. Associated I∙∙∙C(sp) short contacts are indicated with
green dashed lines. (b) Halogen- and hydrogen-bonding interac-
tions observed between 3-iodophenyl moieties. The two comple-
mentary sets of three short contacts (I∙∙∙I, I∙∙∙H, and H∙∙∙O) for
each neighboring molecules lead to the formation of a 1D extend-
ed chain motif based on 1f.
tially the same packing motif, pointing to the robustness of the
associated intermolecular interactions.
Similarly to hydrogen bonding, halogen bonding is a highly
directional noncovalent interaction69 that is commonly used in
the rational construction of supramolecular assemblies includ-
ing those intended for topochemical polymerization.7,70,71 In
this study, we synthesized the two regioisomeric iodobenzo-
ates 1e and 1f to examine their crystal-packing behavior.
However, no single crystal of 4-iodobenzoate 1e suitable for
X-ray diffraction analysis was obtained despite multiple trials
under different crystallization conditions. The very low solu-
bility of 1e led to the formation of powdery precipitates in all
cases.
Figure 5. Representations of the crystal structure of 1d. (a) Pack-
ing diagram showing the 2D extended sheet structure. Associated
hydroxy hydrogen bonds are indicated with dashed lines. (b)
View perpendicular to the plane of the macrocycle to show the
registry between three consecutive sheet structures based on 1d
(top: navy, middle: green, bottom: gray). Hydrogen atoms and
solvent molecules are omitted for clarity.
solution. X-ray diffraction analysis showed that each 3-
hydroxyphenyl group is disordered to adopt two different ori-
entations in 1:1 ratio, and the group appears as if it was 3,5-
dihydroxyphenyl on average. The packing of 1c is very similar
to that of 1d, and hence it is not described in detail here; how-
ever, it is worth noting that the very different solvent systems
used between the crystallizations of 1c and 1d provide essen-
The 3-iodobenzoate 1f is considerably more soluble than 1e,
and can be crystallized by slow diffusion of n-hexane into a
CH2Cl2 solution to form red plates. Single-crystal X-ray dif-
fraction analysis revealed that 1f has crystallized without in-
corporating any solvent molecules. The 3-iodophenyl moieties
are slightly tilted against the planarized dehydro[24]annulene
framework by 29° or 34°, effectively filling the macrocyclic
ACS Paragon Plus Environment

Page 9 of 24
Journal of the American Chemical Society
pores and inter-macrocyclic spaces in the crystal. Here, inter- Crystal engineering via a host–guest strategy. N,N'-
1
molecular I∙∙∙C(sp) contacts with a distance of 3.619(2) Å are
observed between I1 and C12 (Figure 6a). The aryl groups
form isolated stacks, each of which is sandwiched between
two dehydroannulene frameworks. The stack consists of four
iodophenyl units from two neighboring molecules, and all the
four aryl rings are parallel to each other having interplane an-
gles of 0° or 5°. Their stacking distances are approximately 3.5
Å, as observed for common -stacking assemblies such as
hexabenzocoronenes.72 Intermolecular iodine–iodine contacts
are also observed (I1∙∙∙I2 = 3.949(1) Å), each of which is ac-
companied with an Aryl–H∙∙∙O=C interaction (H26∙∙∙O2 =
2.57(3) Å) and an Aryl–H∙∙∙I contact (H26∙∙∙I2 = 3.07(3) Å), as
shown in Figure 6b. Two sets of these interactions connect
neighboring molecules of 1f to form a linear chain motif. The
two I∙∙∙I–C angles associated with the I∙∙∙I contact are 90.9(1)°
and 154.4(1)°. This arrangement slightly differs from the ideal
right-angle mode for halogen-halogen bonding, in which the
corresponding angles are 90° and 180°, respectively.69 The
deviation is assumed to be the result of compromise to gain
the optimal total stabilization from the halogen- and hydrogen-
bonding interactions associated with this intermolecular inter-
action motif.
Disubstituted oxalamides are known to form 1D extended
assemblies via hydrogen bonding. The repeat distance between
hydrogen-bonded oxalamide units is ca. 5 Å, which matches
the distance requirement for butadiyne topochemical polymer-
ization shown in Scheme 1. Upon the installation of comple-
mentarily interacting functional groups between the oxalamide
and dehydro[24]annulene components, it should be possible to
form cocrystals in which the 5 Å repeat distance is transferred
from the oxalamide assembly to the stack of dehy-
dro[24]annulene molecules. Fowler, Laugher, Goroff, and
their coworkers have successfully employed this host‒guest
strategy for the topochemical polymerization of various linear
butadiyne derivatives, in which they achieved several remark-
able single-crystal-to-single-crystal polymerizations revealed
by X-ray diffraction.7,28 In this study, we examined the cocrys-
tallization of dehydro[24]annulenes 1a–1f (guests) with N,N'-
bis(pyridylmethyl)oxalamides 2a and 2b (hosts, Chart 2). Ei-
ther hydroxy-to-pyridine (hydrogen-bonding) or iodine-to-
pyridine (halogen-bonding) interactions were expected to op-
erate in these host–guest combinations.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chart 2. Structures of oxalamide derivatives 2a and 2b.
The crystal structures of benzoate derivatives 1b–1d and 1f
share a common feature—the planarized dehydro[24]annulene
framework. This frequent appearance of the planar confor-
mation, including the two pseudopolymorphs 1a-I and 1a-II,
makes an interesting contrast to the previous observations in
which nonplanar conformations are observed. For example,
derivatives with small, non-hydrogen-bonding substituents
such as acetoxymethyl19 adopt nonplanar macrocyclic confor-
mations in the crystalline state. The same is true for the parent
dehydro[24]annulene19 and its fully benzo-fused derivative.73
Although it is still difficult to predict any solid-state confor-
mation of a macrocyclic framework, it is safe to say that a less
stable, planarized conformation can be expected in future ex-
amples.
Cocrystals of 4-hydroxybenzoate 1b and oxalamide 2b were
obtained as yellow needles from a THF/MeOH/MeCN mixed-
solvent system. X-ray diffraction analysis revealed that the
cocrystal has the monoclinic C2/c space group with Z = 4. As
desired, oxalamide 2b forms 1D stacks via hydrogen bonding
with a repeat distance of 4.918(3) Å (Figure 7). This is suc-
cessfully transferred to molecules of 1b through phenol–
pyridine interactions to afford a stacked, tubular array based
on 1b that extends along the crystallographic b axis. The co-
crystal contains the two components in 1:1 ratio, even though
the crystallization was set up with a molar ratio of 1b:2b = 1:2.
Only two of the four phenol groups in 1b participate in phe-
nol–pyridine interactions (O∙∙∙N = 2.646(4) Å), while the other
two form intermolecular phenol–phenol hydrogen bonds
(O∙∙∙O = 2.737(2) Å) (see SI). The macrocyclic framework is
nonplanar in this case, and the inner cavity within each tubular
structure is filled with solvent molecules highly disordered
along the channel axis (axis b).
In contrast to the similarity of macrocyclic conformations in
the examples above, the molecular-packing modes vary con-
siderably between the benzoate derivatives depending on the
substitution pattern of the aryl groups. Neighboring substitu-
ents in meta-hydroxybenzoates 1c and 1d form the self-
complementary cyclic 푅22(14) motifs that involve both hy-
droxy and carbonyl groups thanks to the proper relative ar-
rangement of these two functionalities. meta-Iodobenzoate 1f
also forms a self-complementary cyclic motif based on halo-
gen and hydrogen bonding. In contrast, the hydroxy groups of
para-hydroxybenzoate 1b do not participate in any strong
intermolecular interactions other than the hydrogen bonding
with solvent molecules. Thus, where crystal engineering
through a cocrystallization approach is concerned, it appears
favorable that the “handles” on the compound of interest—
hydroxy or iodo functionalities in the cases of benzoates 1b–
1f—are readily available for noncovalent interactions with
other species. The formation of strong, self-complementary
interactions between handle units (e.g., those observed for
meta-functionalized derivatives 1c, 1d, and 1f) is unfavorable
in this context, perhaps because of the bent geometry of a me-
ta-substituted benzene. Based on these considerations, we
assumed that the para-substituted derivative 1b would most
easily influence the crystal packing mode through noncovalent
interactions, whereas the other benzoate derivatives would not.
This is discussed in the following section.
The two symmetry-nonrelated sets of butadiyne moieties are
essentially equivalent in terms of the packing parameters rele-
vant to the butadiyne topochemical polymerization—the ob-
served values are the repeat distance r = 4.9 Å, tilt angle  =
53°, and C1∙∙∙C4' distance d = 4.0 Å (Figure 7b). Although the
values of  and d differ slightly from the optimal (45° and 3.5
Å, respectively), the obtained packing structure is within the
range in which many butadiyne topochemical polymerizations
can operate according to published examples. For example,
2,4-hexadiyne-1,6-diyl dibenzoate was reported to undergo
pressure-induced topochemical polymerization in the crystal-
line state. Its monomeric units are aligned with packing pa-
rameters of r = 4.3Å,  = 59°, and d = 4.0 Å.74 Another study
involved two different diiodobutadiyne–oxalamide (2a or 2b)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 24
polymerizations.75,76 Thus, the tubular packing obtained in the
present case (r = 4.9 Å) is highly promising for our intended
1
2
polymerization.
3
4
5
6
7
8
9
Contrary to 4-hydroxybenzoate 1b, the other benzoate de-
rivatives 1c–1f did not form cocrystals suitable for single-
crystal X-ray analysis with either oxalamide 2a or 2b. All our
attempts resulted in the formation of crystals of either of the
two separate components, or the precipitation of powdery sol-
ids. As described earlier, 1b is the only derivative whose
“handles” are left available for solvents, in that they are not
participating in any strong, self-complementary interactions.
Thus, it is presumed that the reluctance of 1b to form strong
self-associative interactions facilitates the formation of cocrys-
tals with host molecules. A related observation has been re-
ported by Fowler, Lauher and coworkers, in which the ther-
modynamic stability associated with self-hydrogen-bonded
packing of 4,4'-(oxalyldiimino)dibutyric acid was suspected as
the reason why this compound did not cocrystallize with di-
pyridyldiacetylenes.56 Although the experimental observations
by others and us do not yet provide a general rule, these results
point out the importance of carefully examining crystal struc-
tures of each component in designing host–guest cocrystals.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Another cocrystallization was achieved in the combination of
tetraol 1a and oxalamide 2a. The crystals were grown by va-
por diffusion of n-pentane into a THF/MeOH solution contain-
ing 1a and 2a in 1:2 molar ratio. Single-crystal X-ray diffrac-
tion analysis revealed that the crystal has the triclinic P space
group with Z = 1 and the formula 1a·2a. There are no solvent
molecules occluded in the cocrystal, and the pyridyl groups of
2a effectively fill the cavities of planarized macrocyclic
frameworks of 1a. These pyridyl groups form energetically
favorable antiparallel, rather than parallel,  dimers77 with an
associated plane-to-plane distance of 3.45 Å. Unexpectedly,
the molecules of oxalamide 2a form neither the commonly
observed oxalamide–oxalamide interaction nor any direct hy-
drogen bonds to each other. Instead, each molecule of 2a is
engaged in six hydrogen-bonding interactions with neighbor-
ing molecules of 1a; namely, two N–H∙∙∙O–H (N2∙∙∙O1 =
2.978(2) Å), two C=O∙∙∙H–O (O2∙∙∙O3 =2.754(1) Å), and two
N(pyridyl)∙∙∙H–O interactions (N1∙∙∙O1 = 2.849(2) Å) (Figure
8a, green dashed lines). The sextuple hydrogen-bonding motif
ties up the planarized macrocycles to form a 2D extended
sheet structure. Adjacent molecules of 1a have short
C(sp2)∙∙∙C(sp2) contacts of 3.382(2) Å (Figure 8a, red dashed
lines). The planarized dehydro[24]annulene framework is con-
siderably distorted from the near-square configurations ob-
served in 1a-I and 1a-II (Figure 1a and 2a, respectively) into a
rhombus shape (Figure 8b). The two non-equivalent substitut-
ed sp2 carbons in 1a·2a have considerably different bond an-
gles (121.6(1)° for C5 and 125.5(1)° for C10), and sp carbons
C9 and C9' are bent from linearity by more than 15°. This
significant in-plane distortion of the macrocyclic framework is
probably there to accommodate the multiple hydrogen-
bonding interactions between 1a and 2a. They are most likely
enabled by the high flexibility of acetylene units in the dehy-
dro[24]annulene framework.68
Figure 7. Molecular-packing diagrams for the cocrystal of 1b and
2b. (a) Perspective view along the column axis, or the crystallo-
graphic b axis. (b) Close-up view of three consecutive dehy-
dro[24]annulene frameworks within a tubular stack. Nonequiva-
lent butadiyne units are differentiated by light blue and yellow.
Packing parameters relevant to butadiyne topochemical polymeri-
zation are shown in angstroms or degrees. Substituents are omit-
ted for clarity.
The cocrystals of 1a·2a do not form the expected host–guest
assembly. In a host–guest strategy, the host is expected to dic-
tate the packing of the guest, and thus a distinct hierarchy can
be ascribed between the two components. In cocrystals of
1a·2a, the two components mutually affect each other and
essentially collaborate to form a unique packing motif for this
cocrystals with packing parameters of r = 5.1 Å,  = 51°, d =
3.9 Å, or r = 5.0 Å,  = 65°, d = 4.9 Å.75 Both cocrystals were
also found to undergo topochemical polymerization under
high-pressure conditions. Additionally, it has been suggested
that the repeat distance r is more important than the C1∙∙∙C4'
distance d in determining the success of topochemical
ACS Paragon Plus Environment

Page 11 of 24
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 8. Representations of cocrystal 1a·2a. Selected interatom-
ic distances and bond angles are shown in angstroms and degrees,
respectively. (a) Molecular packing viewed approximately per-
pendicular to the 2D extended sheet structure. Hydrogen bonds
are shown with green dashed lines, and short C∙∙∙C contacts with
red dashed lines. (b) Thermal-ellipsoid plot of 1a in the cocrystal
at 50% probability.
Figure 9. Representations of pseudopolymorph 1g-I. (a) Perspec-
tive view along the crystallographic axis. Each dehy-
b
dro[24]annulene-based tubular structure is associated with four
linear hydrogen-bonding networks of carbamate units along the b
axis. Substituents are partially disordered and only the main con-
formations are shown. (b) Close-up view of three consecutive
dehydro[24]annulene frameworks within the tubular motif. Non-
equivalent butadiyne units are differentiated by light blue and
yellow. Packing parameters relevant to butadiyne topochemical
polymerization are shown in angstroms or degrees. Substituents
are omitted for clarity.
specific combination. An obvious and important factor that
facilitates the formation of what we can term “collaborative
cocrystals” is the structural matching between the involved
components so as to accommodate stabilizing complementary
interactions. In relation to this, it is worth reporting that our
attempts to grow cocrystals of tetraol 1a with oxalamide 2b, a
regioisomer of 2a, resulted in the formation of discrete crystals
of either of the two compounds. Another factor that potentially
operates in the formation of cocrystals of 1a·2a is the insuffi-
cient differentiation in the strengths of hydrogen bonds be-
tween the donor and acceptor. Specifically, the hydrogen-
bond-donating abilities of the hydroxymethyl moieties of 1a
and the amide groups of 2a may be similar enough to perturb
the formation of the desired oxalamide– oxalamide interaction
under the given crystallization conditions. This assumption is
consistent with the earlier example of host–guest assembly
1b·2b, in which all the stronger hydrogen-bond donors (phe-
nolic hydroxy groups) selectively interact with the stronger
hydrogen-bond acceptors (pyridyl nitrogen atoms). Thus, even
though the combination of the hydroxy–pyridine and the oxa-
lamide–oxalamide interactions is known to be an effective
setup for the formation of host–guest assemblies,78 fine-tuning
of the hydrogen-bond donor/acceptor strengths79 may still be
needed to resolve difficult cases such as the present one.
Crystal-packing behavior of carbamate derivatives 1g–
1j. Another strategy to achieve favorable molecular packing
for a diacetylene topochemical polymerization is to covalently
introduce functional groups that provide the appropriate repeat
distances of around 4.9 Å in crystals. Based on this approach,
a number of linear butadiyne derivatives have been successful-
ly organized to form topochemically polymerizable assemblies
in single crystals,30,80–84 liquid crystals,85 or organo-
gels.27,29,31,32,86–88 Although macrocyclic systems are signifi-
cantly underrepresented compared to linear ones, there are
several successful examples currently known.25,35,37,38,40,41
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 24
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 10. Molecular-packing representations of pseudopolymorph 1g-II. The colors of carbon atoms are differentiated based on the
symmetry operations that correlate molecules; grey: identity, yellow: twofold screw axis, light blue: glide plane. Salient distances and an-
gles are shown in angstroms and degrees, respectively. Substituents are partially disordered and only the main conformation is shown. (a)
Perspective view along the crystallographic a axis. The 2D-sheet motif extends along the ab plane. (b) Packing diagram perpendicular to
the a axis, highlighting the hydrogen-bonding networks of carbamates (green dashed lines). Only two substituents per molecule are shown
for clarity. (c) Arrangement of dehydro[24]annulene frameworks within the sheet structure viewed along the c axis. Short C∙∙∙C contacts
associated with one macrocyclic framework (bottom-middle) are indicated by red dotted lines. Substituents are omitted for clarity. Atoms
highlighted with larger balls corresponds to the carbon atoms shown in (d). (d) Close-up view of the butadiyne-unit sequence extending
along axis a. Neighboring units are related by a glide plane.
In the present study, we synthesized tetracarbamate deriva-
tives 1g–1j and examined their packing behavior in the crys-
talline state. The carbamate–carbamate hydrogen bonding
motif is among the most commonly used interactions in di-
acetylene topochemical polymerization.80,81,83,84,88,89 Addition-
ally, previous experimental observations suggest that it is
preferable to have an odd number of methylene units between
a rigid reacting core and a carbamate moiety in order to obtain
an adequate tilt angle (ca. 45°) between the butadiyne units.80
Thankfully, and in accordance with this observation, each
carbamate group in derivatives 1g–1j is linked through a sin-
gle methylene unit to the macrocyclic core.
Single-crystal X-ray diffraction analysis revealed that 1g crys-
tallizes in the chiral space group C2 with Z = 2 under these
conditions. In fact, two pseudopolymorphs of 1g were ob-
tained, and this is the first structure that is referred to as 1g-I.
As desired, the macrocycles form beautiful columnar stacks
via a fourfold hydrogen-bonding network (Figure 9). The nar-
row cavities both inside and outside of the dehy-
dro[24]annulene-based tubular stacks are filled with severely
disordered solvent molecules, most likely MeCN. As expected
from the light-yellow color of the crystals, the macrocyclic
framework adopts a folded conformation in this case. The
butadiyne units of 1g are arranged in a similar manner to that
observed in the cocrystal of 1b·2b. In the case of 1g-I, the
observed packing parameters for the two symmetry-nonrelated
Slow evaporation of a THF/MeCN solution of benzyl car-
bamate 1g resulted in the formation of light-yellow needles.
ACS Paragon Plus Environment

Page 13 of 24
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 11. Representations of the crystal structure of 1h. The molecules differentiated by light blue and yellow correlate with each other
by inversion. One of the four n-hexyl groups per molecule is disordered to adopt two different conformations, and only the main confor-
mation is shown. (a) Perspective representation along the a axis. The hydrogen-bond-mediated 2D-sheet motif extends along the crystallo-
graphic ab plane. (b) Packing diagram perpendicular to the a axis, highlighting two hydrogen-bond networks of carbamate units (green
dashed lines). The corresponding hydrogen-bond distances are shown in angstroms. Substituents are partially omitted for clarity.
sets of butadiyne units are: repeat distance r = 4.8 Å, tilt angle
 = 53°, C1∙∙∙C4' distance d = 4.0 Å, or r = 4.8 Å,  = 55°, d =
4.0 Å (Figure 9b). On the other hand, the arrangement of sub-
stituents in the crystal structures of 1g-I and 1b·2b is consid-
erably different. The allyloxy moieties (-C=C–CH2–O-) are
almost coplanar in the cocrystal 1b·2b (the C–C–C–O dihedral
angles are 2.5(3)° and 2.8(3)°), whereas they adopt near right-
angle dihedrals in the columnar stacks of 1g (115.9(5)° and
115.6(5)°). These observations indicate that the hydroxyme-
thyl substituent conformation does not have a significant effect
on the macrocyclic arrangement within columnar stacks.
cyclohexane and/or THF), some of which could not be refined
properly because of severe disorder.
There are many short C∙∙∙C contacts between dehy-
dro[24]annulene frameworks in pseudopolymorph 1g-II. Ten
of the 24 carbons in each framework are in contact with
neighboring frameworks, with the shortest interatomic dis-
tance of 3.222(3) Å for C4∙∙∙C16 (Figure 10c). In spite of these
many short contacts, the crystals of 1g-II were stable in the
mother liquor at 25 °C for at least a month, and they did not
show any explosive behavior upon mechanical stimuli. This is
in stark contrast to the extremely high reactivity of the parent
dehydro[24]annulene in the solid state, which explodes upon
touching with a needle or lab spatula.19 Six of the macrocyclic
carbons in the parent dehydro[24]annulene are similarly en-
gaged in intermolecular short C∙∙∙C contacts in the single-
crystalline state. The high stability of 1g-II is probably due to
the tight molecular packing mediated by the multiple hydro-
gen-bonding networks, which may provide considerable re-
sistance against crystal-packing deformation associated with
potential solid-state reactions. It is also worth pointing out that
near-parallel sequences of butadiyne units are observed along
axis a, in which neighboring units are not translationally
equivalent, but related by a glide plane (Figures 10c,d). The
associated repeat distance is 5.255(3) Å (calculated as the in-
ter-centroid distance, shown as r in Figure 10d), and the
C1∙∙∙C4' distance is 3.358(3) Å. It has been reported that glide-
plane-related butadiyne units of deca-4,6-diyne-2,9-diyl
bis(phenylcarbamate) undergo topochemical polymerization.81
However, the polymerization seems unlikely in 1g-II because
the reaction would cause unfavorable deformation of molecu-
A completely different packing motif resulted when benzyl-
carbamate 1g was crystallized by slow diffusion of cyclohex-
ane into a THF solution. Single crystals of pseudopolymorph
1g-II are yellow blocks, and X-ray diffraction analysis re-
vealed that the crystal has the orthorhombic Pca21 space group
with Z = 4. In this case, the carbamate hydrogen bonds of
pseudopolymorph 1g-II led to the formation of a 2D extended
sheet structure (Figure 10) in contrast to the 1D extended co-
lumnar motif observed for 1g-I. The difference in crystal-
packing modes appears to be largely related to the availability
of a small, linear solvent such as MeCN, which can stabilize
tubular structures by filling their inner cavity, as discussed
earlier. The dehydro[24]annulene-based sheet structure is
formed via carbamate–carbamate hydrogen-bonding networks
extending along the crystallographic a axis (Figures 10a and
b). Each of the dehydroannulene-based sheet structures is
“sandwiched” between benzyl-group layers (Figure 10a). The
spaces between the macrocyclic framework and the benzyl
groups are filled with cyclic solvent molecules (most likely
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 24
The pseudopolymorphism of benzyl carbamate 1g implies
1
that the two hydrogen-bond mediated packing motifs (i.e., the
column and sheet motifs) based on the tetracarbamate system
are close in energy. Thus, it should be possible to alter crystal-
packing patterns between these two motifs not only by chang-
ing crystallization solvents, but also by changing substituent
structures. This assumption was experimentally confirmed—
when the N-substituents were changed from benzyl to n-hexyl,
a sheet motif was selectively formed even in the presence of a
small, linear solvent. Specifically, slow diffusion of MeCN
into a THF solution of n-hexylcarbamate 1h provided yellow
plates, which possess the molecular packing shown in Figure
11. Similarly to the case of 1g-II, the carbamate hydrogen
bonds of 1h form a 2D extended sheet motif along the ab
plane. Here again, the crystals of 1h are stable at room tem-
perature despite numerous short C∙∙∙C contacts between mac-
rocyclic frameworks. In this case, 12 of the 24 carbon atoms in
each macrocyclic framework are in close-contact with neigh-
boring frameworks, with the shortest contact of 3.253(5) Å
(see SI). The n-hexyl groups are interdigitated to effectively
fill the space between dehydro[24]annulene-based sheet struc-
tures, and there are no solvent molecules incorporated in the
crystal. Because of this interdigitation, the sheet structures
form a slipped stack that fits in the triclinic P space group,
while the layers in 1g-II stack straight to fit in the orthorhom-
bic space group.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Phenylcarbamate 1i, on the other hand, forms again the ex-
pected columnar motif (Figure 12). Slow diffusion of MeOH
into a THF solution of 1i resulted in the precipitation of light
yellow needles. Single-crystal X-ray analysis revealed that 1i
crystallized in the tetragonal I space group, and the macrocy-
clic framework is slightly flattened in comparison to the other
folded conformations described above. The stacking pattern of
1i is unlike the two previously described columnar structures
of 1b (Figure 7) and 1g (Figure 9) in that molecules of 1i stack
alternatively within a column. Thus, each column based on 1i
is racemic and associated with a fourfold-rotoinversion axis at
its center along the crystallographic c axis (Figure 12a). The
carbamate–carbamate hydrogen bonds in this structure form
zigzag chains, rather than the straight chains observed in pseu-
dopolymorph 1g-I. Accordingly, the repeat distance between
the stacked butadiyne units of 1i is only 4.2 or 4.3 Å (center-
to-center distances between neighboring butadiyne units, Fig-
ure 12b), being considerably shorter than the corresponding
distances in the two former cases of 1b and 1g (4.9 and 4.8 Å,
respectively). Furthermore, the associated tilt angles (62–63°)
and the C1∙∙∙C4' distances (ca. 4.2 Å) are considerably larger
than the optimal values for butadiyne topochemical polymeri-
zation (45° and 3.5 Å, respectively, Scheme 1). Although the
reasons for this alternating/non-alternating variation between
1g and 1i are unclear,83,90,91 it is evident that the conformation-
al flexibility of the dehydro[24]annulene framework is an im-
portant factor in providing the large diversity of packing mo-
tifs seen in this work.
Figure 12. Representations of the crystal structure of phenylcar-
bamate 1i. (a) Perspective view along the c axis. Each dehy-
dro[24]annulene-based columnar structure is associated with four
threads of carbamate hydrogen-bonding networks formed along
the c axis. A fourfold rotoinversion axis runs through the center of
each column along the c axis, and neighboring columns are relat-
ed by a twofold screw axis. Substituents are partially disordered
and only the main conformations are shown. (b) Three consecu-
tive dehydro[24]annulene frameworks within a columnar stack.
Non-equivalent butadiyne units are differentiated by light blue
and yellow. Packing parameters relevant to butadiyne topochemi-
cal polymerization are shown in angstroms or degrees. Substitu-
ents are omitted for clarity. Note that the positions of the hydro-
gen atoms switch between neighboring frameworks because of the
alternating-stack mode.
An extreme case of macrocyclic conformation was observed
for p-methoxyphenyl carbamate 1j (Figure 13). Single crystals
of 1j were obtained as red prisms by slow diffusion of acetone
into a THF solution. X-ray diffraction analysis revealed that 1j
crystallized into the triclinic P space group with 1j·(acetone)4
stoichiometry. As expected from the red color, the macrocy-
clic framework adopts a planar conformation. Two of the four
substituents on each macrocycle are coplanar with the macro-
cyclic framework, while the other two are near perpendicular
lar structure and crystal packing as mentioned above. Indeed,
1g-II did not show any sign of topochemical polymerization
upon heating up to 120 °C or by UV irradiation at 254 nm.
ACS Paragon Plus Environment

Page 15 of 24
Journal of the American Chemical Society
to the framework and antiparallel to each other (Figure 13b). which no direct hydrogen bonding between carbamate groups
1
2
3
4
5
6
7
8
Overall, the molecule has a windmill-like conformation asso-
ciated with an inversion center in the middle of the macrocy-
clic framework. Unexpectedly, there is no direct hydrogen
bonding between the carbamate groups of 1j, and instead each
carbamate hydrogen atom interacts with a carbonyl oxygen of
the occluded acetone (N∙∙∙O = 2.982(1) or 2.910(2) Å, green
dashed lines in Figure 13a). The acetone molecules also serve
as hydrogen-bond donors toward the carbamate carbonyls
(C∙∙∙H = 2.40(2) or 2.59(3) Å, only the former is shown in
Figure 13a with blue dashed lines). Neighboring molecules of
1j interact through self-complementary O∙∙∙H contacts involv-
ing two C(vinyl)–H∙∙∙O(ether) interactions (H7∙∙∙O6 = 2.51(2)
Å) and two C(aryl)–H∙∙∙O=C interactions (H26∙∙∙O5 = 2.420(1)
Å) as shown in Figure 13a with red dashed lines. The se-
quence of this hydrogen-bonding motif leads to a zigzag-chain
structure extending along the crystallographic c axis. In addi-
tion, there are a number of other hydrogen bonds and C–H∙∙∙
interactions between the zigzag-chain structures (see SI).
is even observed (derivative 1j). This crystal-packing diversity
is likely the result of the rotational freedom of the methylene
hinges in addition to the flexibility of the dehydro[24]annulene
framework. Indeed, the allyloxy moieties adopt a wide range
of dihedral angles in order to accommodate favorable hydro-
gen-bonding interactions. For example, the substituents in 1g-I
are oriented “exo” to the nonplanar macrocyclic framework
with allyloxy torsion (C–C–C–O) angles of –115.9(5)° and –
115.6(5)°, while those in 1g-II are directed “endo” with the
corresponding torsion angles of ±127.6(2)°. In the case of 1j,
the observed allyloxy torsion angles are ±1.1(2)° and
±132.9(1)°
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Transfer integral analysis. Columnar -stacks allow effi-
cient exciton and charge-carrier migration along the column
axis. For example, Zang, Moore and coworkers reported that
nanofibril self-assemblies of arylene ethynylene macrocycles
could be used for explosive sensing based on a fluorescence
quenching mechanism.92 The high sensitivity of the system
was assumed to originate from the efficient exciton migration
in columnar stacks. In addition, the authors pointed out the
importance of the intrinsic porosity originating from the mac-
rocyclic molecular structure in terms of analyte diffusion with-
in the film. Charge-carrier mobility in 1D columnar -stacks
was also explored in many theoretical and experimental stud-
ies.9,93–103 The high sensitivity of charge-carrier mobility to
local molecular arrangement was highlighted in, among others,
the 2009 report by Müllen, Andrienko and coworkers.94
For the purpose of examining the extent of intermolecular
electronic coupling, we calculated the transfer integrals of the
columnar motifs obtained in this work. Although calculated
transfer integrals cannot always be linked directly to experi-
mental charge-carrier mobilities, they serve as useful indica-
tors of the strength of intrinsic electronic coupling between
neighboring molecules. Here, we employed the PW91 func-
tional with a Slater-type triple- plus polarization (TPZ) basis
set, which has been shown to give reliable results.104,105
Figure 14 shows the calculated transfer integrals for the
highest occupied molecular orbitals (HOMOs) and the lowest
unoccupied molecular orbitals (LUMOs). The highest values
are observed in the case of compound 1g; transfer integrals for
HOMOs (tH) and LUMOs (tL) are 98.3 and 96.8 meV, respec-
tively (Figure 13b). This tH value exceeds those calculated for
high-performance p-type semiconductors such as pentacene
(79.3 meV) or rubrene (90.5 meV) (Figure S52a,b). In addi-
tion, the tL value is comparable to that of 6,13-bis(triisopropyl-
Figure 13. Representations of the crystal structure of 1j. (a) View
parallel to the ab plane. Selected distances are shown in ang-
stroms. The self-complementary quadruple hydrogen-bond inter-
actions between molecules of 1j are highlighted with red dashed
lines. The occluded acetone molecules serve as both hydrogen-
bond acceptors (green dashed lines) and donors (blue dashed
lines). (b) Side view of the same portion of the zigzag chain motif
as shown in (a).
silylethynyl)-5,7,12,14-tetraazapentacene (102.5 meV),
a
high-performance n-type semiconductor (Figure S52c). The
other columnar structures obtained in this work gave consider-
ably different values (Figure 14a,c), reflecting that minor vari-
ations in the relative arrangement of neighboring molecules
can greatly affect those values.17,94,106 Importantly, a shorter
repeat distance between the macrocycles does not necessarily
lead to a larger transfer integral, as can be seen in the cases of
1g (r = 4.78 Å) and 1i (r = 4.23 Å). It is known that intermo-
lecular electronic interactions can be optimized through the
control of relative offsets or rotation angles between neighbors
for planar -systems.17,94,106 For nonplanar -systems, as in the
case of the dehydro[24]annulenes, one also needs to carefully
optimize the degree of framework folding to achieve favorable
orbital–orbital interactions.
The crystal structures of 1g–1j explicitly demonstrate that
the N-substituent of carbamate moieties has strong impact over
the crystal-packing behavior of these dehydro[24]annulene
tetracarbamate systems. The rich diversity in crystal-packing
modes and the significant variation in the associated hydro-
gen-bond motifs are remarkable, given that all these com-
pounds share the same core structure (the dehy-
dro[24]annulene framework) and the same type of hydrogen-
bonding functionality (the carbamate group). In fact, a case in
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 24
In addition to the nanotubular assemblies, a wide variety of
1
dehydro[24]annulene-based molecular architectures were ob-
tained, such as an array of nanochannels (1a-I), stacked sheets
with solvent-filled pores (1b–1d), and lamellar assemblies
(1g-II and 1h). The significant diversity in packing modes
arises from the conformational flexibility of the macrocyclic
core and the rotational freedom of the methylene hinges. In-
deed, the flexibility of the dehydro[24]annulene core is clearly
demonstrated with the pseudopolymorphs of tetraol 1a, in
which the macrocyclic framework adopts either planar (1a-I
and 1a-II) or nonplanar (1a-III) conformations. In addition,
the planarized macrocyclic framework in the cocrystal 1a·2a
shows considerable in-plane deformation from a square to a
rhombus shape that is solely induced through noncovalent
interactions. The pseudopolymorphs of 1a are associated with
different motifs of intermolecular hydroxy hydrogen bonds,
which show an interesting transition from a compact cyclic
motif leading to the formation of sheet structures (1a-I), to a
more extended cyclic motif that generates ladder structures
(1a-II), and to an open-chain motif that forms spiral columnar
structures (1a-III).
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
These results will serve as a valuable guide to the rational
design of molecular assemblies based on flexible dehydroan-
nulenes, which have been significantly underrepresented in the
field of crystal engineering. Here, it would be worth noting
again that a fourfold substitution with small, simple groups
such as hydroxymethyl can effectively increase the solid-state
stability of an otherwise highly unstable system such as the
dehydro[24]annulene macrocycle. It is hoped that this contri-
bution will encourage further exploration into the solid-state
chemistry of a wide variety of dehydroannulenes and related
acetylenic macrocycles, especially of those generally consid-
ered “unstable”. Along these lines, we will further pursue the
topochemical polymerization of dehydroannulenes to achieve
the controlled synthesis of novel carbon-rich nanostrutures.
Figure 14. Transfer integrals between HOMOs (tH) or LU-
MOs (tL) calculated at the PW91/TZP level of theory for the
columnar assemblies based on 1b (a), 1g (b), or 1i (c).
Synthetic procedures, spectroscopic data, computation, X-ray
crystallographic analysis, crystallographic information files
(CIFs). This material is available free of charge via the Internet at
http://pubs.acs.org.
Three different tubular architectures based on the dehy-
dro[24]annulene framework were obtained as part of the pre-
sent crystal-engineering study. These supramolecular struc-
tures are among the rare examples of isolated (i.e., non-
*rubin@chem.ucla.edu
intermeshed)
dehydroannulene-based
columnar
stacks.45,52,107,108 They are also the first examples of columnar
supramolecular architectures based on flexible, nonplanar de-
hydroannulenes with solvent-accessible inner space. Gratify-
ingly, the tubular assemblies obtained with derivatives 1b or
1g possess close-to-ideal packing parameters for the intended
multifold topochemical polymerization. Thus, the nanotubular
materials proposed in Scheme 1 may be within reach. Fur-
thermore, the calculation of transfer integrals indicates that
dehydroannulenes with nonplanar conformations may have
significant electronic coupling in the solid state upon careful
optimization of the molecular arrangement, and thus, should
be suitable for detection studies on gaseous or explosive mole-
cules.
†Graduate School of Materials Science, Nara Institute of Science
and Technology (NAIST), 8916-5 Takayama-cho, Ikoma, Nara
630-0192, Japan
The manuscript was written through contributions of all authors.
All authors have given approval to the final version of the manu-
script.
We thank the National Science Foundation for financial support
through an International Collaboration grant NSF-CHE-1125054,
and for instrumentation grants NSF-CHE-9871332 (X-ray) and
ACS Paragon Plus Environment

Page 17 of 24
Journal of the American Chemical Society
NSF-CHE-9974928 (NMR). MS thanks Dr. Robert D. Kennedy
for valuable discussions.
(27) van den Heuvel, M.; Prenen, A. M.; Gielen, J. C.; Chris-
1
2
3
4
5
6
7
8
tianen, P. C. M.; Broer, D. J.; Löwik, D. W. P. M.; van
Hest, J. C. M. J. Am. Chem. Soc. 2009, 131, 15014–15017.
(28) Li, Z.; Fowler, F. W.; Lauher, J. W. J. Am. Chem. Soc.
2009, 131, 634–643.
(29) Jahnke, E.; Weiss, J.; Neuhaus, S.; Hoheisel, T. N.; Frau-
enrath, H. Chem. – Eur. J. 2009, 15, 388–404.
(30) Xu, R.; Schweizer, W. B.; Frauenrath, H. J. Am. Chem.
Soc. 2008, 130, 11437–11445.
(31) Frauenrath, H.; Jahnke, E. Chem. – Eur. J. 2008, 14,
2942–2955.
(32) Dautel, O. J.; Robitzer, M.; Lère-Porte, J.-P.; Serein-
Spirau, F.; Moreau, J. J. E. J. Am. Chem. Soc. 2006, 128,
16213–16223.
(33) Zhu, L.; Tran, H.; Beyer, F. L.; Walck, S. D.; Li, X.;
Ågren, H.; Killops, K. L.; Campos, L. M. J. Am. Chem.
Soc. 2014, 136, 13381–13387.
(34) Xu, W. L.; Smith, M. D.; Krause, J. A.; Greytak, A. B.;
Ma, S.; Read, C. M.; Shimizu, L. S. Cryst. Growth Des.
2014, 14, 993–1002.
(35) Rondeau-Gagné, S.; Néabo, J. R.; Desroches, M.; La-
rouche, J.; Brisson, J.; Morin, J.-F. J. Am. Chem. Soc.
2013, 135, 110–113.
(36) Rondeau-Gagné, S.; Néabo, J. R.; Desroches, M.;
Levesque, I.; Daigle, M.; Cantin, K.; Morin, J.-F. Chem.
Commun. 2013, 49, 9546–9548.
(37) Nagasawa, J. ’ichi; Yoshida, M.; Tamaoki, N. Eur. J. Org.
Chem. 2011, 2247–2255.
(38) Xu, Y.; Smith, M. D.; Geer, M. F.; Pellechia, P. J.; Brown,
J. C.; Wibowo, A. C.; Shimizu, L. S. J. Am. Chem. Soc.
2010, 132, 5334–5335.
(39) Baughman, R. H.; Biewer, M. C.; Ferraris, J. P.; Lamba, J.
S. Synth. Met. 2004, 141, 87–92.
(40) Banerjie, A.; Lando, J. B.; Yee, K. C.; Baughman, R. H. J.
Polym. Sci. Polym. Phys. Ed. 1979, 17, 655–662.
(41) Baughman, R. H.; Yee, K. C. J. Polym. Sci. Polym. Chem.
Ed. 1974, 12, 2467–2475.
(42) Tahara, K.; Inukai, K.; Hara, N.; Johnson, C. A.; Haley, M.
M.; Tobe, Y. Chem. – Eur. J. 2010, 16, 8319–8328.
(43) Hisaki, I.; Shigemitsu, H.; Sakamoto, Y.; Hasegawa, Y.;
Okajima, Y.; Nakano, K.; Tohnai, N.; Miyata, M. Angew.
Chem. Int. Ed. 2009, 48, 5465–5469.
(44) Hisaki, I.; Sakamoto, Y.; Shigemitsu, H.; Tohnai, N.;
Miyata, M. Cryst. Growth Des. 2009, 9, 414–420.
(45) Nishinaga, T.; Nodera, N.; Miyata, Y.; Komatsu, K. J.
Org. Chem. 2002, 67, 6091–6096.
(46) Bunz, U. H. F.; Enkelmann, V. Chem. – Eur. J. 1999, 5,
263–266.
(47) Haley, M. M.; Bell, M. L.; English, J. J.; Johnson, C. A.;
Weakley, T. J. R. J. Am. Chem. Soc. 1997, 119, 2956–
2957.
(1)
(2)
Kissel, P.; Murray, D. J.; Wulftange, W. J.; Catalano, V.
J.; King, B. T. Nat. Chem. 2014, 6, 774–778.
Kory, M. J.; Wörle, M.; Weber, T.; Payamyar, P.; Poll, S.
W. van de; Dshemuchadse, J.; Trapp, N.; Schlüter, A. D.
Nat. Chem. 2014, 6, 779–784.
(3)
(4)
(5)
(6)
(7)
(8)
(9)
Colson, J. W.; Dichtel, W. R. Nat. Chem. 2013, 5, 453–
465.
Biradha, K.; Santra, R. Chem. Soc. Rev. 2013, 42, 950–
967.
Chernick, E. T.; Tykwinski, R. R. J. Phys. Org. Chem.
2013, 26, 742–749.
Rondeau-Gagné, S.; Morin, J.-F. Chem. Soc. Rev. 2013,
43, 85–98.
Lauher, J. W.; Fowler, F. W.; Goroff, N. S. Acc. Chem.
Res. 2008, 41, 1215–1229.
Ryno, S. M.; Risko, C.; Brédas, J.-L. J. Am. Chem. Soc.
2014, 136, 6421–6427.
Idé, J.; Méreau, R.; Ducasse, L.; Castet, F.; Bock, H.;
Olivier, Y.; Cornil, J.; Beljonne, D.; D’Avino, G.; Roscio-
ni, O. M.; Muccioli, L.; Zannoni, C. J. Am. Chem. Soc.
2014, 136, 2911–2920.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(10) Akkerman, H. B.; Mannsfeld, S. C. B.; Kaushik, A. P.;
Verploegen, E.; Burnier, L.; Zoombelt, A. P.; Saathoff, J.
D.; Hong, S.; Atahan-Evrenk, S.; Liu, X.; Aspuru-Guzik,
A.; Toney, M. F.; Clancy, P.; Bao, Z. J. Am. Chem. Soc.
2013, 135, 11006–11014.
(11) Mei, J.; Diao, Y.; Appleton, A. L.; Fang, L.; Bao, Z. J. Am.
Chem. Soc. 2013, 135, 6724–6746.
(12) Dang, M. T.; Wuest, J. D. Chem. Soc. Rev. 2013, 42,
9105–9126.
(13) O’Neill, M.; Kelly, S. M. Adv. Mater. 2011, 23, 566–584.
(14) Liu, H.; Xu, J.; Li, Y.; Li, Y. Acc. Chem. Res. 2010, 43,
1496–1508.
(15) Pramod, P.; Thomas, K. G.; George, M. V. Chem. – Asian
J. 2009, 4, 806–823.
(16) Zang, L.; Che, Y.; Moore, J. S. Acc. Chem. Res. 2008, 41,
1596–1608.
(17) Coropceanu, V.; Cornil, J.; da Silva Filho, D. A.; Olivier,
Y.; Silbey, R.; Brédas, J.-L. Chem. Rev. 2007, 107, 926–
952.
(18) Desiraju, G. R. J. Am. Chem. Soc. 2013, 135, 9952–9967.
(19) Suzuki, M.; Comito, A.; Khan, S. I.; Rubin, Y. Org. Lett.
2010, 12, 2346–2349.
(20) Boese, R.; Matzger, A. J.; Vollhardt, K. P. C. J. Am. Chem.
Soc. 1997, 119, 2052–2053.
(21) Dosa, P. I.; Erben, C.; Iyer, V. S.; Vollhardt, K. P. C.;
Wasser, I. M. J. Am. Chem. Soc. 1999, 121, 10430–10431.
(22) Laskoski, M.; Steffen, W.; Morton, J. G. M.; Smith, M.
D.; Bunz, U. H. F. J. Am. Chem. Soc. 2002, 124, 13814–
13818.
(23) Jin, H.; Young, C. N.; Halada, G. P.; Phillips, B. L.;
Goroff, N. S. Angew. Chem. Int. Ed. 2015, 54, 14690–
14695.
(24) Diegelmann, S. R.; Tovar, J. D. Macromol. Rapid Com-
mun. 2013, 34, 1343–1350.
(25) Hsu, T.-J.; Fowler, F. W.; Lauher, J. W. J. Am. Chem. Soc.
2012, 134, 142–145.
(26) Deschamps, J.; Balog, M.; Boury, B.; Ben Yahia, M.;
Filhol, J.-S.; van der Lee, A.; Al Choueiry, A.; Barisien,
T.; Legrand, L.; Schott, M.; Dutremez, S. G. Chem. Mater.
2010, 22, 3961–3982.
(48) Zhou, Q.; Carroll, P. J.; Swager, T. M. J. Org. Chem.
1994, 59, 1294–1301.
(49) Baldwin, K. P.; Matzger, A. J.; Scheiman, D. A.; Tessier,
C. A.; Vollhardt, K. P. C.; Youngs, W. J. Synlett 1995,
1215–1218.
(50) Hisaki, I.; Senga, H.; Shigemitsu, H.; Tohnai, N.; Miyata,
M. Chem. – Eur. J. 2011, 17, 14348–14353.
(51) Hisaki, I.; Senga, H.; Sakamoto, Y.; Tsuzuki, S.; Tohnai,
N.; Miyata, M. Chem. – Eur. J. 2009, 15, 13336–13340.
(52) Hisaki, I.; Sakamoto, Y.; Shigemitsu, H.; Tohnai, N.;
Miyata, M.; Seki, S.; Saeki, A.; Tagawa, S. Chem. – Eur.
J. 2008, 14, 4178–4187.
(53) Spitler, E. L.; Johnson, C. A.; Haley, M. M. Chem. Rev.
2006, 106, 5344–5386.
(54) Shetty, A. S.; Zhang, J.; Moore, J. S. J. Am. Chem. Soc.
1996, 118, 1019–1027.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 24
(55) McQuilkin, R. M.; Garratt, P. J.; Sondheimer, F. J. Am. (84) Bertault, M.; Canceill, J.; Collet, A.; Toupet, L. J. Chem.
1
Chem. Soc. 1970, 92, 6682–6683.
Soc. Chem. Commun. 1988, 163–166.
(56) Curtis, S. M.; Le, N.; Fowler, F. W.; Lauher, J. W. Cryst.
Growth Des. 2005, 5, 2313–2321.
(85) Hwang, I. H.; Lee, S. J.; Chang, J. Y. J. Polym. Sci. Part
A: Polym. Chem. 2003, 41, 1881–1891.
(86) Greenfield, M. A.; Palmer, L. C.; Vernizzi, G.; Cruz, M.
O. de la; Stupp, S. I. J. Am. Chem. Soc. 2009, 131, 12030–
12031.
(87) Fujita, N.; Sakamoto, Y.; Shirakawa, M.; Ojima, M.; Fujii,
A.; Ozaki, M.; Shinkai, S. J. Am. Chem. Soc. 2007, 129,
4134–4135.
(88) Nagasawa, J.; Kudo, M.; Hayashi, S.; Tamaoki, N. Lang-
muir 2004, 20, 7907–7916.
2
3
4
5
6
7
8
9
(57) Enkelmann, V.; Graf, H. J. Acta Crystallogr., Sect. B:
Struct. Sci., Cryst. Eng. Mater. 1978, 34, 3715–3719.
(58) Steiner, T. Angew. Chem. Int. Ed. 2002, 41, 48–76.
(59) Venkataraman, D.; Lee, S.; Zhang, J.; Moore, J. S. Nature
1994, 371, 591–593.
(60) Etter, M. C.; MacDonald, J. C.; Bernstein, J. Acta Crystal-
logr., Sect. B: Struct. Sci., Cryst. Eng. Mater. 1990, 46,
256–262.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(61) Calculations were performed at the BHandHLYP/6-
31G(d) level of theory using Gaussian 09, Revision A.02.
See the SI for the full citation of Gaussian 09 and compu-
tational details.
(62) The crystal obtained from MeOH/THF has a similar, but
slightly smaller distortion within the dehydro[24]annulene
framework. The corresponding side-to-side distances are
7.617(2) and 8.100(2) Å.
(63) Scheiner, S. Hydrogen Bondingꢀ: A Theoretical Perspec-
tive: A Theoretical Perspective; Oxford University Press,
USA, 1997.
(64) Gilli, G.; Gilli, P. J. Mol. Struct. 2000, 552, 1–15.
(65) Rubin, Y.; Diederich, F. J. Am. Chem. Soc. 1989, 111,
6870–6871.
(66) Li, Y.; Rubin, Y.; Diederich, F.; Houk, K. N. J. Am. Chem.
Soc. 1990, 112, 1618–1623.
(67) Yamamoto, K.; Shibutani, M.; Kuroda, S.; Ejiri, M.; Oji-
ma, J. Tetrahedron Lett. 1986, 27, 975–976.
(68) Hisaki, I.; Sonoda, M.; Tobe, Y. Eur. J. Org. Chem. 2006,
833–847.
(89) Spagnoli, S.; Schott, M.; Johnson, M.; Toupet, L. Chem.
Phys. 2007, 333, 236–245.
(90) Brouty, C.; Spinat, P.; Whuler, A. Acta Crystallogr., Sect.
C: Struct. Chem. 1984, 40, 1619–1624.
(91) Whuler, A.; Spinat, P.; Brouty, C. Acta Crystallogr., Sect.
C: Struct. Chem. 1984, 40, 693–696.
(92) Naddo, T.; Che, Y.; Zhang, W.; Balakrishnan, K.; Yang,
X.; Yen, M.; Zhao, J.; Moore, J. S.; Zang, L. J. Am. Chem.
Soc. 2007, 129, 6978–6979.
(93) Laschat, S.; Baro, A.; Steinke, N.; Giesselmann, F.;
Hägele, C.; Scalia, G.; Judele, R.; Kapatsina, E.; Sauer, S.;
Schreivogel, A.; Tosoni, M. Angew. Chem. Int. Ed. 2007,
46, 4832–4887.
(94) Feng, X.; Marcon, V.; Pisula, W.; Hansen, M. R.; Kirkpat-
rick, J.; Grozema, F.; Andrienko, D.; Kremer, K.; Müllen,
K. Nat. Mater. 2009, 8, 421–426.
(95) An, Z.; Yu, J.; Domercq, B.; Jones, S. C.; Barlow, S.;
Kippelen, B.; Marder, S. R. J. Mater. Chem. 2009, 19,
6688–6698.
(96) Pisula, W.; Zorn, M.; Chang, J. Y.; Müllen, K.; Zentel, R.
Macromol. Rapid Commun. 2009, 30, 1179–1202.
(97) Zhao, K.-Q.; Chen, C.; Monobe, H.; Hu, P.; Wang, B.-Q.;
Shimizu, Y. Chem. Commun. 2011, 47, 6290–6292.
(98) Hayashi, H.; Nihashi, W.; Umeyama, T.; Matano, Y.; Seki,
S.; Shimizu, Y.; Imahori, H. J. Am. Chem. Soc. 2011, 133,
10736–10739.
(99) García-Frutos, E. M.; Pandey, U. K.; Termine, R.; Omenat,
A.; Barberá, J.; Serrano, J. L.; Golemme, A.; Gómez-Lor,
B. Angew. Chem. Int. Ed. 2011, 50, 7399–7402.
(100) Zhang, Y.; Hanifi, D.; Alvarez, S.; Antonio, F.; Pun, A.;
Klivansky, L. M.; Hexemer, A.; Ma, B.; Liu, Y. Org. Lett.
2011, 13, 6528–6531.
(101) Shimizu, Y.; Miyake, Y.; Yoshida, H.; Monobe, H.; Cook,
M. J.; Fujii, A.; Ozaki, M. Mol. Cryst. Liq. Cryst. 2011,
549, 127–132.
(102) Ni, H.-L.; Monobe, H.; Hu, P.; Wang, B.-Q.; Shimizu, Y.;
Zhao, K.-Q. Liq. Cryst. 2013, 40, 411–420.
(103) He, B.; Pun, A. B.; Klivansky, L. M.; McGough, A. M.;
Ye, Y.; Zhu, J.; Guo, J.; Teat, S. J.; Liu, Y. Chem. Mater.
2014, 26, 3920–3927.
(104) Delgado, M. C. R.; Kim, E.-G.; Filho, D. A. da S.; Bredas,
J.-L. J. Am. Chem. Soc. 2010, 132, 3375–3387.
(105) Calculations were performed using the molecular ADF
program (ver. 2013). ADF2013, SCM, Theoretical Chem-
istry, Vrije Universiteit, Amsterdam, The Netherlands.
See Supporting Information for the full citation of ADF
and computational details.
(106) Brédas, J. L.; Calbert, J. P.; Filho, D. A. da S.; Cornil, J.
Proc. Natl. Acad. Sci. 2002, 99, 5804–5809.
(107) Solooki, D.; Bradshaw, J. D.; Tessier, C. A.; Youngs, W.
J.; See, R. F.; Churchill, M.; Ferrara, J. D. J. Organomet.
Chem. 1994, 470, 231–236.
(69) Awwadi, F. F.; Willett, R. D.; Peterson, K. A.; Twamley,
B. Chem. – Eur. J. 2006, 12, 8952–8960.
(70) Holmes, B. T.; Deb, P.; Pennington, W. T.; Hanks, T. W.
J. Polym. Res. 2006, 13, 133–144.
(71) Crihfield, A.; Hartwell, J.; Phelps, D.; Walsh, R. B.; Har-
ris, J. L.; Payne, J. F.; Pennington, W. T.; Hanks, T. W.
Cryst. Growth Des. 2003, 3, 313–320.
(72) Feng, X.; Pisula, W.; Zhi, L.; Takase, M.; Müllen, K.
Angew. Chem. Int. Ed. 2008, 47, 1703–1706.
(73) Guo, L.; Bradshaw, J. D.; Tessier, C. A.; Youngs, W. J. J.
Chem. Soc. Chem. Commun. 1994, 243–244.
(74) Hanson, A. W. Acta Crystallogr., Sect. B: Struct. Sci.,
Cryst. Eng. Mater. 1975, 31, 831–834.
(75) Wilhelm, C.; Boyd, S. A.; Chawda, S.; Fowler, F. W.;
Goroff, N. S.; Halada, G. P.; Grey, C. P.; Lauher, J. W.;
Luo, L.; Martin, C. D.; Parise, J. B.; Tarabrella, C.; Webb,
J. A. J. Am. Chem. Soc. 2008, 130, 4415–4420.
(76) Ouyang, X.; Fowler, F. W.; Lauher, J. W. J. Am. Chem.
Soc. 2003, 125 (41), 12400–12401.
(77) Hohenstein, E. G.; Sherrill, C. D. J. Phys. Chem. A 2009,
113, 878–886.
(78) Nguyen, T. L.; Scott, A.; Dinkelmeyer, B.; Fowler, F. W.;
Lauher, J. W. New J. Chem. 1998, 22, 129–135.
(79) Etter, M. C. Acc. Chem. Res. 1990, 23, 120–126.
(80) Tachibana, H.; Kumai, R.; Hosaka, N.; Tokura, Y. Chem.
Mater. 2001, 13, 155–158.
(81) Wilson, R. B.; Duesler, E. N.; Curtin, D. Y.; Paul, I. C.;
Baughman, R. H.; Preziosi, A. F. J. Am. Chem. Soc. 1982,
104, 509–516.
(82) Sarkar, A.; Okada, S.; Matsuzawa, H.; Matsuda, H.;
Nakanishi, H. J. Mater. Chem. 2000, 10, 819–828.
(83) Patel, G. N.; Duesler, E. N.; Curtin, D. Y.; Paul, I. C. J.
Am. Chem. Soc. 1980, 102, 461–466.
(108) O’Connor, M. J.; Yelle, R. B.; Zakharov, L. N.; Haley, M.
M. J. Org. Chem. 2008, 73, 4424–4432.
ACS Paragon Plus Environment

Page 19 of 24
Journal of the American Chemical Society
SYNOPSIS TOC
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 24
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
175x127mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 21 of 24
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
177x127mm (300 x 300 DPI)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 24
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
84x178mm (300 x 300 DPI)
ACS Paragon Plus Environment

Page 23 of 24
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
84x188mm (300 x 300 DPI)
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 24
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
68x150mm (300 x 300 DPI)
ACS Paragon Plus Environment