ExBox: A polycyclic aromatic hydrocarbon scavenger

Article abstract of DOI:10.1021/ja307360n

A template-directed protocol, which capitalizes on donor-acceptor interactions, is employed to synthesize a semi-rigid cyclophane (ExBox ⁴⁺) that adopts a box-like geometry and is comprised of π-electron-poor 1,4-phenylene-bridged ("extended") bipyridinium units (ExBIPY²⁺). ExBox⁴⁺ functions as a high-affinity scavenger of an array of different polycyclic aromatic hydrocarbons (PAHs), ranging from two to seven fused rings, as a result of its large, accommodating cavity (approximately 3.5 A? in width and 11.2 A? in length when considering the van der Waals radii) and its ability to form strong non-covalent bonding interactions with π-electron-rich PAHs in either organic or aqueous media. In all, 11 PAH guests were observed to form inclusion complexes with ExBox⁴⁺, with coronene being the largest included guest. Single-crystal X-ray diffraction data for the 11 inclusion complexes ExBox ⁴⁺?PAH as well as UV/vis spectroscopic data for 10 of the complexes provide evidence of the promiscuity of ExBox⁴⁺ for the various PAHs. Nuclear magnetic resonance spectroscopy and isothermal titration calorimetric analyses of 10 of the inclusion complexes are employed to further characterize the host-guest interactions in solution and determine the degree with which ExBox⁴⁺ binds each PAH compound. As a proof-of-concept, a batch of crude oil from Saudi Arabia was subjected to extraction with the water-soluble form of the PAH receptor, ExBox·4Cl, resulting in the isolation of different aromatic compounds after ExBox·4Cl was regenerated.

Full text of DOI:10.1021/ja307360n

Article
pubs.acs.org/JACS
ExBox: A Polycyclic Aromatic Hydrocarbon Scavenger
†
,‡
†,‡
†
†
†
̌
Jonathan C. Barnes, Michal Jurícek, Nathan L. Strutt, Marco Frasconi, Srinivasan Sampath,
†
†
†
†
†
Marc A. Giesener, Psaras L. McGrier, Carson J. Bruns, Charlotte L. Stern, Amy A. Sarjeant,
,†
and J. Fraser Stoddart*
†Department of Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208, United States
*
S Supporting Information
ABSTRACT: A template-directed protocol, which capitalizes on donor−
acceptor interactions, is employed to synthesize a semi-rigid cyclophane
4+
(
ExBox ) that adopts a box-like geometry and is comprised of π-electron-
2+
poor 1,4-phenylene-bridged (“extended”) bipyridinium units (ExBIPY ).
ExBox4 functions as a high-affinity scavenger of an array of different
polycyclic aromatic hydrocarbons (PAHs), ranging from two to seven fused
rings, as a result of its large, accommodating cavity (approximately 3.5 Å in
width and 11.2 Å in length when considering the van der Waals radii) and its
ability to form strong non-covalent bonding interactions with π-electron-rich
PAHs in either organic or aqueous media. In all, 11 PAH guests were
+
4
+
observed to form inclusion complexes with ExBox , with coronene being the
largest included guest. Single-crystal X-ray diffraction data for the 11 inclusion
4
+
complexes ExBox ⊂PAH as well as UV/vis spectroscopic data for 10 of the
complexes provide evidence of the promiscuity of ExBox4+ for the various PAHs. Nuclear magnetic resonance spectroscopy and
isothermal titration calorimetric analyses of 10 of the inclusion complexes are employed to further characterize the host−guest
interactions in solution and determine the degree with which ExBox4+ binds each PAH compound. As a proof-of-concept, a
batch of crude oil from Saudi Arabia was subjected to extraction with the water-soluble form of the PAH receptor, ExBox·4Cl,
resulting in the isolation of different aromatic compounds after ExBox·4Cl was regenerated.
3
1
. INTRODUCTION
different ratios between Clar’s aromatic sextets and double
bonds not included in the Clar’s sextet within a homologous
series of molecular compounds. Figure 1 illustrates the Clar’s
structures of the 11 PAH molecules that were investigated in
this work and demonstrates how the overall electronic structure
undergoes change within PAH isomers. For example, fully
benzenoid triphenylene (containing three aromatic sextets) is
extremely stable even under drastic conditions, while its isomer
tetracene (containing one aromatic sextet and six non-sextet
double bonds) is much more reactive and, compared with
A period of 136 years has now passed since the smallest
possible polycyclic aromatic hydrocarbon (PAH), naphthalene,
was synthesized for the first time. Somewhat less time has
1
passed since the synthesis of the largest well-defined PAH
(giant graphene disk containing 222 carbon atoms and 37 fused
2
benzenoid units) was reported a decade ago. During the more
than a century-long period in between, PAHs, also known as
polyaromatic hydrocarbons, which consist of fused aromatic
rings that do not contain heteroatoms or carry substituents,
have become an important class of compounds when
considering different aspects of life. Not only have they
posed questions about the origin of life, as the spectral
signatures of anthracene and pyrene have recently been found
in the ultraviolet light emitted by the red rectangle nebula, but
also somewhere along the way they have become one of the
main pollutants and harmful chemical species known on the
planet. Yet, without a doubt, they remain among the most
beautiful and fascinating classes of organic compounds.
3
6
triphenylene, possesses a lower band gap.
PAHs can be found naturally in oil, coal, and tar. They are
produced as byproducts in combustion processes of fuel, be it
fossil fuel or biomass. Being lipophilic, meaning they mix more
7
4
easily with oil than water, PAHs dispersed into the environment
are primarily deposited in soil, sediment, and oily substances, as
particulate matter suspended in air, and in water, although to a
lesser extent. As a pollutant, they are of concern because some
of the smaller (partially water-soluble) PAHs, such as
benzo[a]pyrene, have been identified as carcinogenic, muta-
genic, and teratogenic, and they are widely believed to make a
substantial contribution to the overall burden of cancer in
5
3
,6
The highly characteristic (“fingerprint”) optical absorption
and emission behavior of PAHs makes it possible to identify
unambiguously a compound because the optical as well as the
chemical properties of PAHs depend largely on their size and
8
humans. The major route of exposure to PAHs for people
6
geometry. There are many different combinations of PAHs
comprised of the same number of rings, where the number of
combinations increases as more rings are added, resulting in
Received: July 26, 2012
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/ja307360n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. Single-crystal XRD data were obtained for all 11 of the inclusion complexes of ExBox⊂guest·4PF (1−11), along with that of ExBox·4PF
6
6
alone (top). The counterions and solvent molecules are removed for the sake of clarity. In each case, the crystals that were used to obtain the XRD
data were photographed under an optical microscope. It is important to note the lack of color in the crystal of ExBox·4PF by itself and the presence
6
of different colors for each inclusion complex. The colors arise from charge-transfer interactions between the host and guests, with the exception of 1
because no charge-transfer band was observed in the optical absorption spectrum. The structural formulas depicted with bolded π-bonds highlight
the Clar’s sextet in each π-electron-rich guest.
nonsmokersis consumption of food, which can be con-
taminated from environmental sources as well as industrial or
home food processing. The European Food Safety Authority
not be ideal from environmental and economical perspectives.
Ideally, the PAH receptor should (1) be robust, so it can be
used multiple times to extract PAHs, (2) bind a vast range of
PAHs which possess a different number of fused benzenoid
rings, and (3) be soluble in both organic and aqueous media,
whereby the receptor cyclophane can be regenerated easily by
way of solvent extraction.
(
EFSA), documenting that PAHs are potentially genotoxic and
9
carcinogenic to humans in food, reported that some of the
potentially harmful PAHs are present even in fresh fruit.
Considering the potential health risks associated with the
release of PAHs into the environment, it is of fundamental
importance to understand the supramolecular chemistry
surrounding PAHs in order to be able to detect and sequester
these harmful molecules. To date, cyclodextrins (CDs) have
been used most often to isolate PAHs from crude mixtures,
although more recently derivatives of calix[n]arenes, cholic
In this article, we describe the synthesis and properties of a
10
4+
macrocyclic organic salt ExBox , which serves as a high-affinity
scavenger of an array of PAHs, up to the size of coronene, in
both aqueous and organic media (Figure 1). We demonstrate
the ability of ExBox4+ to bind PAHs of different size and shape
by means of donor−acceptor interactions on 11 examples of
common PAHs ranging in size from two to seven aromatic
11
12
1
3
14
acid, and metal−ligand diazapyrenium-based metallocycles
9
have been explored as potential suitors for PAHs of different
rings, two of which (tetraphene and chrysene) are known for
shapes and sizes. It should be noted, however, that the
their mutagenicity/genotoxicity in somatic cells in experimental
animals in vivo. We present the full solid-state characterization
of the inclusion complexes of ExBox4+ with all 11 PAHs and
envisage its potential in the extraction and detection of harmful
14
metallocycles described by Quintela et al. incorporate heavy
Pd and Pt ligand−metal coordination in order to form
supramolecular architectures, which are expensive and would
B
dx.doi.org/10.1021/ja307360n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
17
Scheme 1. Reaction Scheme for the Synthesis of ExBox·4PF , Starting with the Reported Extended Bipyridine (ExBIPY)
6
PAHs from the environment. Finally, the potential for the
ExBox−PAH donor−acceptor superstructures to be used in
organic field-effect transistors is considered.
g, 92%) that was collected by filtration as a white solid. HRMS-ESI for
+
DB·2PF : calcd for C H Br F N P , m/z = 743.0256 [M − PF ] ,
6
32 28
2
12
2
2
6
2
+
+
5
2
97.0536 [M − 2PF ] ; found, 743.0262 [M − PF ] , 597.0538 [M −
6
6
2+ 1
PF ] . H NMR (500 MHz, CD CN, ppm): δ 8.66 (AA′ of
6
3
H
AA′XX′, J = 6.9 Hz, 4H), 8.19 (XX′ of AA′XX′, J = 6.9 Hz, 4H), 7.97
s, 4H), 7.40 (AA′ of AA′BB′, J = 8.2 Hz, 4H), 7.31 (BB′ of AA′BB′, J
8.2 Hz, 4H), 5.58 (s, 4H), 4.46 (s, 4H). 13C NMR (125 MHz,
CD CN, ppm): δC 155.0, 144.3, 139.9, 136.7, 132.9, 129.8, 129.1,
2
. EXPERIMENTAL SECTION
(
=
The full experimental details are provided in the Supporting
Information (SI). Below, the most important information is briefly
summarized.
3
1
29.1, 125.6, 63.1, 32.3.
Cyclobis(4,4′-(1,4-phenylene)bispyridine-p-phenylene)-
Single-Crystal X-ray Diffraction (XRD). Single crystals of
ExBox·4PF6 and its complexes 1−11 were grown by slow vapor
diffusion of iPr O into solutions of ExBox·4PF (or ExBox·4PF and
tetrakis(hexafluorophosphate) (ExBox·4PF ). Two synthetic
routes (Scheme 1), methods A (no template) and B (templated),
6
2
6
6
were employed to prepare the tetracationic cyclophane. Method A:
PAH guests (1:1.1)) in MeCN (3.0 mM) over the course of 1 week.
Data were collected at 100 K on a Bruker Kappa APEX CCD
diffractometer equipped with a Cu Kα microsource with Quazar
optics.
Isothermal Titration Calorimetry (ITC). All ITC measurements
were performed in dry, degassed MeCN at 298 K. A solution of
DB·2PF (312 mg, 0.350 mmol) and ExBIPY (81.3 mg, 0.350 mmol)
6
were added to dry MeCN (200 mL) and stirred at room temperature.
After 17 days, the reaction was stopped by adding concentrated HCl
(
2−3 mL), which caused the crude product to precipitate from
solution. The precipitate was collected by filtration, dissolved in H O,
2
and precipitated again by adding NH PF (∼100−200 mg). The solid
ExBox·4PF in MeCN was used as the host solution in a 1.8 mL cell.
4
6
6
material was collected by filtration and then subjected to column
Solutions of aromatic guests in MeCN were added by injecting
successively 10 μL of titrant over 20 s (25×) with a 300 s interval
between each injection. Experiments were repeated three times.
Thermodynamic information was calculated using a one-site binding
model utilizing data from which the heat of dilution of the guest was
subtracted, with the average of three runs reported.
chromatography using silica gel and 1% NH PF in MeCN (w/v) as
4
6
the eluent. The final product was purified further through
recrystallization in MeCN on slow vapor diffusion of iPrO , yielding
2
pure ExBox·4PF (83 mg, 19%) as a white solid. Method B: DB·2PF
6
6
(
312 mg, 0.350 mmol), ExBIPY (81.3 mg, 0.350 mmol), and the
1
template pyrene (212 mg, 1.05 mmol) were added to dry MeCN (200
mL) and stirred at room temperature. After 17 days, the reaction was
stopped by adding concentrated HCl (2−3 mL), causing the crude
product to precipitate from solution. The yellowish-orange precipitate
Nuclear Magnetic Resonance (NMR) Spectroscopy. H NMR
298 K, 500 MHz) titrations were performed by adding small volumes
(
of a concentrated guest solution/suspension in CDCl to a solution of
3
ExBox·4PF in CD CN. Tetramethylsilane was used as a reference.
6
3
1
was collected by filtration and then dissolved in H O, followed by
Significant upfield shifts of the H resonances for γ protons were
2
extracting 5× with CH Cl (200 mL per extraction). This solvent
observed and used to determine the association constants (K ). The Ka
2
2
a
extraction step changes the organic layer from an orange solution to a
white suspension, indicating that most, if not all, of the template had
been removed. The extracted material was precipitated from solution
values were calculated using Dynafit, a program that employs
nonlinear least-squares regression on ligand−receptor binding data.
In the case of 10 and 11, the low solubility of perylene and coronene
prohibited the calculation of reliable K values. The estimated K value
by adding NH PF (∼100−200 mg) before being subjected to column
4
6
a
a
chromatography using silica gel and CH Cl /MeCN (1:1) and 1%
for 10 is reported, however, the error is substantial (76%).
2
2
NH PF in MeCN (w/v) as the eluents, followed by recrystallization
Bisbromomethyl(bis-p-benzyl-4,4′-(1,4-phenylene)-
4
6
bispyridine)bis(hexafluorophosphate) (DB·2PF ). α,α′-Dibromo-
as in the case of nontemplated reaction to yield the pure product (184
mg, 42%) as a white solid. HRMS-ESI for ExBox·4PF : calcd for
6
p-xylene (4.58 g, 17.2 mmol) was added to CH Cl (30 mL) in a 250
6
2
2
+
mL round-bottomed three-neck flask, and the resulting mixture was
refluxed while stirring until all of the solid material dissolved. Next, the
temperature of the oil bath was raised to 90 °C, allowing the reaction
mixture to reflux, while a suspension of ExBIPY (400 mg, 1.72 mmol)
in MeCN (60 mL) was added in five aliquots slowly over 1 h. After
only 30 min, however, a yellow precipitate, indicative of DB·2Br
formation, began to appear. After heating under reflux for 48 h, the
reaction mixture was cooled to room temperature, and the yellow
precipitate was collected by filtration and washed with CH Cl . The
C
2PF
48
H
40
F
24
N
4
P
4
, m/z = 1107.2173 [M − PF
6
] , 481.1263 [M −
2
+
+
2+ 1
] ; found, 1107.2184 [M − PF
] , 481.1278 [M − 2PF ] . H
6
6
6
NMR (500 MHz, CD
CN, ppm): δ 8.78 (AA′ of AA′XX′, J = 7.0 Hz,
3
H
8H), 8.18 (XX′ of AA′XX′, J = 6.9 Hz, 8H), 7.93 (s, 8H), 7.61 (s, 8H),
5.68 (s, 8H). 13C NMR (125 MHz, CD
CN, ppm): δ 155.5, 145.1,
3
C
1
137.4, 137.1, 131.1, 130.2, 126.7, 64.8. H NMR (500 MHz,
CD COCD , ppm): δ 9.28 (AA′ of AA′XX′, J = 7.1 Hz, 8H), 8.51
3
3
H
(XX′ of AA′XX′, J = 7.1 Hz, 8H), 8.15 (s, 8H), 7.86 (s, 8H), 6.08 (s,
8H). 13C NMR (125 MHz, CD COCD , ppm): δ 155.2, 145.5,
2
2
3
3
C
yellow solid was dissolved in cold (≤25 °C) MeOH (∼500−750 mL)
137.4, 137.3, 131.3, 130.2, 126.6, 64.7.
followed by the addition of NH PF (∼100−200 mg) and cold (≤25
Crystal Parameters. ExBox·4PF : [C H N ·(PF ) ]·(MeCN) ;
4
6
6
48 40
4
6
4
3
°
C) H O (∼2 L), resulting in the precipitation of pure DB·2PF (1.41
colorless block, 0.13 × 0.12 × 0.11 mm; monoclinic, C2/c; a =
2
6
C
dx.doi.org/10.1021/ja307360n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
2
9
1.920(6), b = 16.300(5), and c = 36.904(11) Å; α = 90.000, β =
“extended viologens”. The synthesis begins by treating the
3
17
8.513(2), and γ = 90.000°; V = 13 040.3(7) Å ; Z = 8; T = 100(2) K;
extended bipyridine (ExBIPY) with an excess of α,α′-
−3
−1
ρcalc = 1.402 g cm ; μ = 2.059 mm . CCDC no.: 893549.
dibromo-p-xylene in a refluxing MeCN/CH Cl solvent
2
2
E x Bo x⊂Az ule ne · 4PF (az ulene) (1 ) : [C 4 8 H4 0 N ⊂
6
4
mixture (2:1), which initially produces the yellow, water-
soluble bromide salt (DB·2Br, not shown) after 2 days. In order
to render the initial viologen salt soluble in organic media for
the next step, the bromide salt is converted to the
C H ·(PF ) ]·(C H )·(MeCN) ; gold block, 0.33 × 0.24 × 0.14
mm; triclinic, P1
Å; α = 71.694(10), β = 79.448(10), and γ = 65.810(10)°; V =
1
0
8
6
4
10
8
3
̅
; a = 10.090(3), b = 11.382(3), and c = 18.082(5)
3
−3
1
794.77(9) Å ; Z = 1; T = 100(2) K; ρ = 1.510 g cm ; μ = 1.971
calc
−1
mm . CCDC no.: 893550.
hexafluorophosphate salt (DB·2PF ) by first dissolving
6
ExBox⊂Anthracene·4PF (anthracene) (2): [C H N ⊂
DB·2Br in MeOH, as opposed to H O as a consequence of
6
48 40
4
2
C H ·(PF ) ]·(C H )·(MeCN) ; red block, 0.11 × 0.05 × 0.05
mm; monoclinic, P2/c; a = 11.174(3), b = 18.965(5), and c =
3
7
the poor solubility of DB·2Br and potential hydrolysis at the
1
4
10
6
4
14 10
2
benzylic positions, followed by the addition of solid NH PF
4
6
3.876(8) Å; α = 90.000, β = 90.818(10), and γ = 90.000°; V =
and subsequent precipitation of DB·2PF6 after copious
amounts of H O is added (approximately a few liters for
3
−3
177.8(3) Å ; Z = 4; T = 100(2) K; ρ = 1.565 g cm ; μ = 1.992
calc
−1
2
mm . CCDC no.: 893551.
every 500 mL of MeOH used). In the final step, two synthetic
pathways lead to the final viologen-based tetracationic cyclo-
ExBox⊂Phenanthrene·4PF (phenanthrene) (3): [C H N ⊂
6
48 40
4
C H ·(PF ) ]·(C H )·(MeCN) ; yellow block, 0.37 × 0.12 × 0.09
mm; monoclinic, C2/c, a = 11.247(7), b = 19.121(13), and c =
3
7
1
4
10
6
4
14 10
2
phane (ExBox·4PF ). The first pathway consists of simply
6
3.700(2) Å; α = 90.000, β = 90.897(14), and γ = 90.000°; V =
mixing DB·2PF and ExBIPY in a 1:1 ratio in MeCN and
6
3
−3
246(8) Å ; Z = 4; T = 100(2) K; ρ = 1.550 g cm ; μ = 1.973
allowing the reaction to proceed for a little over 2 weeks. The
yield from this reaction, however, is low (19%). Therefore, a
calc
−1
mm . CCDC no.: 893552.
ExBox⊂Pyrene·4PF (pyrene) (4): [C H N ⊂C H ·(PF ) ]·
18
6
48 40
4
16 10
6
4
method involving templation was adopted wherein 3 equiv of
(
P1
C H )·(MeCN) ; orange block, 0.30 × 0.14 × 0.14 mm; triclinic,
16 10 2
pyrene is used to help stabilize the transition state during
cyclophane closure as a result of donor−acceptor interactions
between the π-electron-rich pyrene and the π-electron-poor
̅
; a = 9.926(3), b = 10.725(4), and c = 18.056(6) Å; α = 97.803(10),
3
β = 91.899(10), and γ = 103.705(10)°; V = 1845.73(11) Å ; Z = 1; T
=
8
−3
−1
100(2) K; ρcalc = 1.565 g cm ; μ = 1.956 mm . CCDC no.:
4+
extended viologen units of ExBox . Moreover, the pyrene
template is removed easily by solvent extraction with CH Cl ,
2
93553.
ExBox⊂Tetracene·4PF6 (5): [C H N ⊂(C H ) ·(PF ) ]·
2
48
40
4
16 10 0.5
6 4
(
=
C H ) ·(MeCN) ; red block, 0.47 × 0.16 × 0.07 mm; triclinic, P1
̅
which changes the organic layer from an orange solution to a
white suspension, after the desired cyclophane has been
converted to the chloride salt (ExBox·4Cl, not shown). The
final step of the synthesis involves another counterion exchange
wherein the chloride salt (ExBox·4Cl) is converted to the
hexafluorophosphate salt by adding solid NH PF , an action
7
8
2
2
10.140(3), b = 11.252(3), and c = 17.944(5) Å; α = 72.280(17), β =
6.535(17), and γ = 65.818(15)°; V = 1774.21(9) Å ; Z = 1; T =
05.4(2) K; ρcalc = 1.529 g cm ; μ = 1.989 mm . CCDC no.:
93554.
3
8
1
8
−3
−1
ExBox⊂Tetraphene·4PF6 (6): [C H N ⊂(C H ) ·(PF ) ]·
48
40
4
18 12 0.5
6 4
4
6
(
MeCN) ; orange block, 0.76 × 0.22 × 0.12 mm; triclinic, P1
̅
3
which precipitates white ExBox·4PF from the water layer. This
1
8
1
0.143(4), b = 11.110(4), and c = 17.953(7) Å; α = 74.054(2), β =
4.440(2), and γ = 64.486(2)°; V = 1754.90(12) Å ; Z = 1; T =
6
3
method affords the desired product in 42% yield after silica
−3
−1
00(2) K; ρcalc = 1.410 g cm ; μ = 1.959 mm . CCDC no.: 893555.
ExBox⊂Chrysene·4PF (7): [C H N ⊂(C H )·(PF ) ]·
column chromatography, 1% NH PF in MeCN (w/v), and
4 6
6
48 40
4
18 12
6
4
several recrystallizations. A more detailed synthetic procedure
and spectroscopic analysis of each compound can be found in
(
MeCN) ; orange block, 0.58 × 0.33 × 0.17 mm; monoclinic, P2/n;
6
a = 11.017(5), b = 14.579(7), and c = 24.259(7) Å; α = 90.000, β =
the SI.
3
9
8.998(2), and γ = 90.000°; V = 3848.5(3) Å ; Z = 2; T = 100(2) K;
From a photophysical perspective, ExBox4+ demonstrates
−3
−1
ρcalc = 1.491 g cm ; μ = 0.209 mm . CCDC no.: 893556.
(
Figure 2a) a strong absorbance in the ultraviolet region at 319
ExBox⊂[4]Helicene·4PF (8): [C H N ⊂(C H )·(PF ) ]·
6
48 40
4
18 12
6 4
nm. Upon addition of a π-electron-rich guest and formation of
(
1
2
̅
MeCN) ; yellow block, 0.11 × 0.08 × 0.02 mm; triclinic, P1; a =
4+
a 1:1 complex (ExBox ⊂guest), the optical absorption data
show (Figure S13) the emergence of a charge-transfer band
(400−650 nm) in the visible region of the spectrum. All of the
electron-donating guests assessed (Figure 1), with the
exception of azulene and tetracene, exhibit this behavior. In
the case of the former, there appears to be no change in the
0.561(6), b = 11.101(6), and c = 18.025(11) Å; α = 72.955, β =
3
8
2.376(2), and γ = 63.569°; V = 1809.2(18) Å ; Z = 1; T = 220(2) K;
−3
−1
ρcalc = 1.417 g cm ; μ = 1.918 mm . CCDC no.: 893557.
ExBox⊂Triphenylene·4PF (9): [C H N ⊂(C H )·(PF ) ]·
6
48 40
4
18 12
6 4
(
1
3
̅
MeCN) ; yellow block, 0.35 × 0.21 × 0.04 mm; triclinic, P1; a =
0.019(12), b = 10.770(13), and c = 18.071(2) Å; α = 73.537(6), β =
3
7
4.926(6), and γ = 70.374(4)°; V = 1731.9(4) Å ; Z = 1; T = 100(2)
4+
absorption spectra (Figure S13a) when ExBox is mixed with
−3
−1
K; ρcalc = 1.538 g cm ; μ = 0.224 mm . CCDC no.: 893558.
ExBox⊂Perylene·4PF (10): [C H N ⊂(C H )·(PF ) ]·
azulene in a 1:1 ratio, as compared to the superposition of the
individual molecular spectra. No absorption data were obtained
6
48 40
4
20 12
6 4
(
2
MeCN) ; red block, 0.59 × 0.44 × 0.33 mm; monoclinic, P2/c; a =
7
4+
for the ExBox ⊂tetracene inclusion complex because of
0.957(5), b = 17.985(5), and c = 22.883(6) Å; α = 90.000, β =
3
1
07.919(10), and γ = 90.000°; V = 8206.5(4) Å ; Z = 4; T = 100(2) K;
solubility issues associated with tetracene. It is important to
−3
−1
ρcalc = 1.451 g cm ; μ = 1.793 mm . CCDC no.: 893559.
note that the ExBox⊂tetracene·4PF single crystals (5), shown
6
ExBox⊂Coronene·4PF (11): [C H N ⊂(C H )·(PF ) ]·
in Figure 1, were obtained by first heating the 1:1 mixture of
6
48 40
4
24 12
6 4
(
MeCN) ; orange block, 0.46 × 0.30 × 0.09 mm; triclinic, P1
̅
4+
7
the π-electron-rich guest and ExBox in PhMe/MeCN (1:1)
1
9
1
0.573(6), b = 14.317(7), and c = 14.733(8) Å; α = 102.508(3), β =
3.376(3), and γ = 109.336(3)°; V = 2033.67(19) Å ; Z = 1; T =
00(2) K; ρcalc = 1.503 g cm ; μ = 0.203 mm . CCDC no.: 893560.
prior to dilution with an additional MeCN and crystallization.
3
Analysis of the cyclic voltammogram (CV, Figure 2b) of
−3
−1
4+
2+
ExBox and the phenylene-bridged methyl viologen (MPV )
provides some insight into the electrochemical behavior of the
extended viologen units on an individual basis as well as when
more than one unit is part of a rigid cyclophane. The CV (blue
3
. RESULTS AND DISCUSSION
The chemical constitution of the high-affinity PAHs receptor
15
2+
(Scheme 1) is based on the well-known family of π-electron-
trace, Figure 2b) of MPV consists essentially of one broad
poor viologen salts, more specifically the phenylene-bridged
redox couple where the two one-electron reduction waves
overlap completely at −0.82 V, an observation which is an
16
4,4′-bipyridines, which are known throughout the literature as
D
dx.doi.org/10.1021/ja307360n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
in size (from azulene up to coronene) and shape (from linear
tetracene to star-like triphenylene). Looking at the manner in
which the locally flexible, yet globally rigid ExBox4+ cyclophane
adapts to different guests by adopting small and subtle changes
in its conformation, one senses that the pool of PAH molecules
that can be bound inside ExBox4+ is not limited only to these
1
1 examples but, most likely, many morepotentially
harmfulsmaller PAHs can be trapped inside this promiscuous
receptor.
High-quality single crystals of ExBox4+ and its inclusion
complexes were obtained by slow vapor diffusion of iPr O into
2
4+
the solution of ExBox and the 1:1.1 stoichiometric mixtures
of ExBox4+ with the PAHs in MeCN over a period of a few
days. The color of these crystals (Figure 1) corresponds to the
absorption wavelength of the charge-transfer band of the
corresponding complex in solution (see the SI) and makes it
possible for one to detect visually when a PAH molecule binds
4
+
4+
inside ExBox . This feature makes ExBox a suitable
candidate for visual PAH sensing.
The dimensions of the cavity of ExBox4+ are 6.9 Å (width) ×
4.6 Å (length) as deduced from the distances between the
1
aromatic planes on opposite sides of the empty cyclophane in
the X-ray crystal structure (see the SI) and approximately 3.5 Å
(
width) × 11.2 Å (length) when considering the van der Waals
radii. The 6.9 Å value is similar to that reported for an
analogous tetracationic cyclophane, cyclobis(paraquat-p-phe-
20
4+
nylene) (CBPQT ), which is known to bind smaller
aromatic guests by means of donor−acceptor charge-transfer
interactions, where the typical dihydroxynaphthalene-derived
4
+
Figure 2. UV/vis absorption spectrum (a) of ExBox4+ displaying a
strong absorbance band centered around 319 nm and sharing no
guests reside in the CBPQT ring in a perpendicular
18
manner. Both CBPQT and ExBox4+ bind π-electron-rich
guests effectively as a result of their inherent π−π stacking
distances (3.5 Å) between the π-electron-poor units of the host
and the guest. The ability of ExBox to bind larger PAH guests
relative to CBPQT4 is a consequence of the inside length
parameter of ExBox4+ (approximately 7.8 Å). Coronene is the
largest guest (in one dimension) we have investgated. In
principle, thin graphene nanoribbons that would be “coronene-
wide” in one dimension (i.e., approximately 7.3 Å) could
potentially serve as a guest for multiple host molecules of
4+
bands in the visible region. The CV (b) of the phenylene-bridged
2+
methyl viologen (MPV ) shows one broad redox couple, which
consists of two one-electron redox processes and occurs as a result of a
lack of communication between the two redox centers of the molecule.
In contrast, the CV of ExBox4 demonstrates a splitting of the two
redox couples (by 140 mV), presumably a consequence of the steric
and electronic constraints associated with the rigid geometry of the
cyclophane. A glassy carbon working electrode, a platinum counter
4+
+
+
electrode, and a Ag/AgCl reference electrode were used to
characterize a 1 mM DMF solution of ExBox4+ at 298 K with 0.1
4+
−1
ExBox .
The general trend, which becomes evident when comparing
M TBAPF serving as the electrolyte. A scan rate of 200 mV s was
6
used for this analysis.
the crystal superstructures of 1−11, is that each PAH guest
indication that the two bipyridinium redox centers are not
adopts the greatest possible π−π overlap with the extended
1
9
2+
4+
communicating and behave quite independently of one
bipyridinium (ExBIPY ) units of ExBox . This principle can
be demonstrated quite nicely by comparing the two linear PAH
molecules, anthracene and tetracene. While the shorter
4
+
another. Conversely, the CV of ExBox consists of two closely
overlapping two-electron redox couples, where the potentials of
the two cathodic peaks are separated by 140 mV. This
separation reflects the level of electronic communication
between the two redox centers of each extended bipyridinium
unit insofar as that, after formation of the radical cationic
ExBox2 , the second two-electron reduction, which leads to
the formation of the neutral species, is shifted to lower
potentials and is therefore more difficult to reduce. This direct
comparison demonstrates how the electronic properties of the
extended viologen units can be altered by controlling the steric
and electronic environment as well as incorporating them into a
rigid cyclophane.
4+
anthracene (7.3 Å) is fully aligned inside ExBox , the length
4
+
of tetracene (9.7 Å) overcomes the size limit of ExBox and is
forced out of colinearity by approximately 44°, and the two
terminal benzenoid rings protrude outside the cavity (Figure
1). Also, as the size of the PAH guest is increased, the torsion
angles between the pyridinium and the phenylene rings of
ExBox4+ tend to decrease, reaching a lower limit in the case of
perylene.
(•+)
The size and shape of the PAH guests play an important role
also on the overall packing (Figure 5 and the SI) in the crystal.
Generally, the smaller PAHs (1−4) crystallize in a 2:1
4+
Figure 1 illustrates the superstructures of host−guest
(PAH:ExBox ) manner, where one PAH molecule binds
4
+
4+
complexes of ExBox with 11 PAH guests viewed from
different angles, as obtained from the XRD analysis of single
crystals of these complexes. They demonstrate the remarkable
ability of ExBox4 to scavenge a large number of PAHs varying
inside ExBox and the second molecule is located in the outer
4
+
space interacting through C−H···π interactions with ExBox
or the complexed PAH molecules, while the larger PAHs (7−
11) crystallize in a 1:1 (PAH:ExBox ) fashion, where all of the
+
4+
E
dx.doi.org/10.1021/ja307360n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
4
+
Table 1. Binding Parameters and H -Proton Shifts of the Studied 1:1 ExBox ⊂PAH Complexes 1−11
γ
binding parameters
a
K /10 (M−1)
3
ΔH (kcal mol−1
ΔS (cal mol−1 K−1
ΔG (kcal mol−1
b
1
:1 complex
π-electron count
)
)
)
δ (ppm)
a
c
1
2
3
4
5
6
7
8
9
10
14
14
16
18
18
18
18
18
20
24
0.286 ± 0.011
0.883 ± 0.11
1.38 ± 0.016
7.16 ± 0.50
−
−
−
−
−3.35 ± 0.022
−4.01 ± 0.073
−4.28 ± 0. 0068
−5.25 ± 0.041
−
−4.03 ± 0.0065
−4.59 ± 0.038
−5.12 ± 0.0056
−5.85 ± 0.15
−6.74 ± 0.45
7.86
7.38
7.26
6.76
c
d
−5.72 ± 0.024
−5.84 ± 0.076
−4.83 ± 0.084
−1.97 ± 0.29
,
de
f
g
−
−
−
−
d
0.914 ± 0.010
2.32 ± 0.15
5.71 ± 0.054
19.7 ± 5.0
−5.79 ± 0.028
−5.91 ± 0.096
7.19
7.10
6.93
6.91
6.90
5.40
c
−
−
d
−7.66 ± 0.018
−8.52 ± 0.063
c
−
−
−
−
−
−
c
1
1
0
1
88.1 ± 67
f
−
−
c
a
b
1
Number of π-electrons in the PAH aromatic system. Chemical shifts of the H -protons of the 1:1 complex. Determined by H NMR
γ
d
e
1
3
−1
f
spectroscopy. Determined by ITC. Binding constant determined by H NMR spectroscopy for complex 4 is 10.1 ± 1.1 × 10 M . Binding
g
parameters of these complexes could not be determined by either ITC or NMR spectroscopy on account of solubility restrictions. Chemical shift of
the H -protons of this complex could not be determined because of solubility restrictions.
γ
4
+
21
PAH molecules are inside ExBox . The two exceptions (5 and
) to this experimental observation, which are, sizewise,
somewhere in between the smaller and larger studied PAHs,
of 4 could be caused by the presence of dimeric forms of
6
pyrene in MeCN, decreasing the entropy of the initial (prior-to-
complexation) state.
The H NMR spectra (Figure 3 and the SI) of the 1:1
ExBox ⊂PAH complexes display significant upfield shifts for
the signals corresponding to β and γ protons of ExBox and all
4
+
4+
1
crystallize in a 1:2 (PAH:ExBox ) manner; half of the ExBox
4+
hosts in the superstructure being empty. This phenomenon
could play an important role in the tuning of the overall
properties of such “superstacks” when considering their
possible functions as organic semiconductors. Additionally, a
certain degree of disorder of PAH molecules, either inside or
outside of ExBox4 or both, in the superstructure was observed
in some cases (1, 3, 5, 6, 8, and 9; see the SI).
It should be noted that the least planar guest [4]helicene
possesses a twisted helical conformation, demonstrating the
ability of ExBox to bind PAHs is not strictly limited to the
planar guests. The high-affinity diverse binding capability of
ExBox4 arises, presumably, from the locally flexible character
of this globally rigid cyclophane.
4+
of the signals for the protons of the PAH guests as well as a
4
+
downfield shift of the phenylene protons of ExBox . This
pattern is caused by π-electron shielding of the face-to-face
oriented aromatic rings, which occurs upon complexation. The
+
4+
α and CH protons of ExBox display only very slight shifts in
2
1
their H NMR spectra since they are located in the “corners” of
ExBox4+ and are therefore not affected significantly by the
shielding effect of PAH guests. This behavior is in excellent
4+
4
+
agreement with the inclusion of PAH guests inside ExBox in
the solution state. Additionally, it is important to note that
there are only one set of proton signals observed for ExBox
+
4
+
The thermodynamic parameters (ΔH, ΔS, and ΔG) and the
and the guest, as a result of the various species that exist in
equilibrium undergoing fast exchange on the H NMR time
4+
1
association constants (K ) for the 1:1 complexes of ExBox
a
with PAH guests (1−11) in solution (MeCN) are summarized
in Table 1. The thermodynamic parameters for the binding
events of ExBox4 with the PAHs (complexes 3, 4, 6, and 8)
were obtained by ITC in dry, degassed MeCN at 298 K. The
restricted solubility of the remainder (complexes 1, 2, 5, 7, and
scale.
A general trend associated with the binding affinity of
ExBox toward PAH guests exists, whereby the stability
+
4+
constant of the complex increases exponentially (Figure 4a) as
the PAH guests become more π-electron-rich resulting in
greater π−π interactions between the host and guest. This
trend is exhibited by all of the complexes, except for 6−8,
which cannot adopt maximum π−π overlap on account of the
unfavorable size/shape of the PAH guest. The side-on view
(Figure 1) of the crystal superstructure of 6−9 reveals that the
9
−11) of the PAHs in MeCN prohibited us from using ITC to
1
monitor the thermodynamic parameters of binding. H NMR
spectroscopy was used to obtain K values for the association
a
between ExBox4 and the PAHs which are, at least partially,
+
soluble in CDCl (complexes 1, 2, 7, 9, and 10). In the case of
3
4
+
complexes 5 and 11, the extremely low solubility of tetracene
and coronene, respectively, in most organic solvents render it
face-to-face π−π overlaps between ExBox and the PAH
molecules increase with higher and higher K values. A similar
a
impossible for us to obtain K values for these complexes.
Details on both methods and determination of the binding
parameters are provided in the SI.
Comparison of the full thermodynamic parameters available
for the four complexes 3, 4, 6, and 8 leads to the conclusion
that the host−guest binding process is mainly enthalpy driven
in the case of 3, 6, and 8. With 4, however, a surprisingly small
(although linear) trend is observed (Figure 4b) when the
difference between the chemical shifts (in ppm) of the γ
protons of the 1:1 complexes and those for the unbound
ExBox4+ is plotted against the number of π-electrons in the
PAH molecule. In this case, the chemical shift not only depends
a
on the K value, which correlates to the number of π-electrons,
a
but also directly on the number of π-electrons as they dictate
change in entropy (ΔS), which contributes to the large K value
the shielding effect of the PAH guest.
a
for this complex, was observed. Although the complexes of
Four PAH guests (varying between three and seven rings)
were explored (Figure 5) in a solid-state setting on SiO2
platforms in the form of 2:1 or 1:1 inclusion complexes with
ExBox4 with 4 and 6 have very similar values of ΔH, the K
+
a
values for the complex with 6 differs by approximately a factor
of 7 compared with that for 4. This low value of ΔS in the case
4+
ExBox . By lowering the concentration of the host and guest
F
dx.doi.org/10.1021/ja307360n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. K values (a) and chemical shift differences of the γ protons
a
of ExBox4 (7.93 ppm) and its 1:1 complex (Δδ of H ) (b) plotted
+
γ
against the number of π-electrons present in the respective PAH. Both
K (exponential behavior; logarithmic scale was used for the y-axis)
a
and Δδ values for H (linear behavior) increase as the number of π-
γ
electrons becomes larger and larger and show that the binding affinity
of ExBox4 toward PAHs is related directly to the increase of the size
of the PAH core.
+
Figure 5d). Their scanning electron microscopy (SEM) images,
displayed in Figure 5, provide a sense of the crystal morphology
of each of the complexes that were grown on the SiO surface.
2
The superstructures of each complex, shown to the right of
each SEM image in Figure 5, demonstrate the packing of the
individual inclusion complexes. In the case of anthracene
Figure 3. Aromatic-region insets of the 1H NMR spectra of
ExBox⊂4PF6 (a) and its 1:1 complexes with azulene (b),
phenanthrene (c), pyrene (d), perylene (e), and coronene (f)
(
Figure 5a) and pyrene (Figure 5b), the molecules are ordered
in an alternating fashion, whereby the inclusion complexes are
intercalated with continuous sheets of the electron-rich PAHs, a
feature which could lead potentially to these materials
functioning in an organic field-effect transistor. In the perylene
recorded in CD CN at 298 K on a 500 MHz instrument. Upfield
3
4+
shifts of the β and γ protons of ExBox as well as downfield shifts of
the phenylene protons of ExBox4 are evident. Both downfield and
upfield shifts increase as the size of the PAH core becomes larger and
larger, indicating the binding of a PAH molecule inside ExBox4 in
solution.
+
(
Figure 5c) and coronene (Figure 5d) superstructures,
+
however, the individual complexes do not appear to pack into
discrete donor−acceptor domains but instead are organized
through intermolecular steric interactions between each
complex.
to 0.625 and 0.750 mM in MeCN, respectively, micrometer-
4
+
sized single crystals were obtained for ExBox⊂anthracene·
In order to demonstrate the high affinity that ExBox has for
PAHs in a mixture of hydrocarbons, a crude oil sample from
Saudi Arabia containing an unspecified array of aromatic
compounds was tested against ExBox·4Cl in water. The results
4
PF (anthracene) (three rings, Figure 5a), ExBox⊂pyrene·
6
4
PF (pyrene) (four rings, Figure 5b), ExBox⊂perylene·4PF
6
6
(
five rings, Figure 5c), and ExBox⊂coronene·4PF (seven rings,
6
G
dx.doi.org/10.1021/ja307360n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 5. SEM images of the single crystals of the four inclusion complexes ExBox⊂anthracene·4PF (anthracene) (a), ExBox⊂pyrene·4PF (pyrene)
6
6
(
b), ExBox⊂perylene·4PF (c), and ExBox⊂coronene·4PF (d) showing the micrometer-sized crystals that formed on SiO wafers after drop-casting
6
6
2
mixed solutions of host and guest (0.625 and 0.750 mM, respectively) in MeCN onto the surface and carrying out slow vapor diffusion using iPrO
2
over the course of 1−3 days. The superstructures to the right of each SEM image were generated from X-ray crystallography carried out on single
crystals that were grown at a higher concentration (3.0 mM) and used to obtain the single-crystal XRD data shown in Figure 1. In the case of the
superstructures where a smaller guest was employed, i.e., three and four rings, it should be noted that the sheets of inclusion complexes are separated
by sheets of the free guest (a,b), whereas in the case of the larger guests, the packing in the crystal consists of only the 1:1 inclusion complexes and
−
no intercalated PAHs are observed (c,d). The PF6 counterions and the solvent molecules in the superstructures are omitted for clarity.
of this investigation are summarized in Figure 6. The starting
ducted on four standards (trans-stilbene, phenanthrene,
anthracene, and pyrene; Figure 6c) as well as the mixture of
aromatic compounds that were extracted from the crude oil in
steps iii and iv. This general MS characterization makes it
possible to generate mass ranges and molecular formulas for the
extracted aromatic material and gives a sense of the sizes of
compounds ExBox4+ is capable of scavenging. Overall, the oil
extraction study demonstrates clearly the ability of ExBox to
pull out potentially harmful PAHs from a crude mixture of oil,
which contains a large number of different aromatic hydro-
carbons.
crude oil mixture (Figure 6a) was first diluted (i) with CD Cl2
2
(150 μL of oil added to 4 mL) before the addition (ii) of
ExBox·4Cl in 4 mL of D O (8 mM) on top of the oil, taking
2
care not to disturb the oily layer. The vial labeled “6-C” in
Figure 6a is the control mixture where 4 mL of D O only was
2
added to the oily layer, and it was later found to contain no
4
+
PAHs. After shaking the oil−water mixture (which contains
4+
ExBox ) by hand for approximately 2 min (iii), the color of
the aqueous phase turned to an orange-yellow hue, which is
different from the previous lighter-colored suspension before
the two phases had been mixed. After separation of the two
phases, the aqueous layer was siphoned off and the newly
acquired aromatic guests were isolated (iv) by solvent
4. CONCLUSIONS
In summary, we have reported the synthesis and full
characterization of a robust tetracationic organic cyclophane,
extraction using a few iterations of CH Cl , while simulta-
2
2
4
+
4+
neously regenerating the ExBox in aqueous solution. Figure
ExBox , which serves as a promiscuous high-affinity PAH
scavenger in both organic and aqueous media. This PAH
receptor was obtained starting from easily accessible precursors
using pyrene to template the cyclophane-closing step, affording
6
a (far right) shows the final isolated PAHs (1.2 mg from 150
1
μL of crude oil) after CH Cl had been evaporated. H NMR
2
2
spectra recorded at each stage of the PAHs isolation procedure
are presented in Figure 6b. The spectrum at the top of Figure
ultimately ExBox·4PF in 42% yield. In order to demonstrate
6
1
4+
6
b shows H NMR peaks in the aromatic region present in the
the high flexibility of ExBox to bind practically any small
1
starting oily layer. The H NMR spectrum directly below the
crude oil spectrum shows that ExBox4+ has been successful in
extracting a range of different aromatic hydrocarbons in the
potentially harmfulPAHs, 11 representative PAHs, ranging in
size from two to seven fused aromatic rings, were trapped
4
+
inside ExBox , and the resulting complexes were fully
characterized, including the single-crystal superstructure
determinations of all 11 of them. Considering the known
harmful effects of many of the smaller PAHs on the
1
4+
crude oil. The H NMR spectrum of ExBox before the two
phases were mixed shows the starting chemical shifts of each
proton in the aromatic region in D O, whereas the spectrum
2
4
+
recorded after extraction of aromatics from the crude oil
solution shows small changes in the chemical shifts of the α, β,
and phenylene protons and a more dramatic shifting of the
resonances for the γ protons. In an effort to characterize the
extracted aromatic compounds, GC-MS studies were con-
environment and people, we envisage that this ExBox PAH
scavenger will play a key role in the extraction and detection of
harmful PAHs from the environment as well as in the
purification processes relating to both water and crude oil.
Furthermore, the well-defined donor−acceptor superstacks of
H
dx.doi.org/10.1021/ja307360n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 6. (a) A crude oil sample (“6”) from Saudi Arabia, containing an unknown amount of unspecified aromatic compounds, was first (i) diluted
with CD Cl (150 μL of oil added to 4 mL) and subsequently (ii) extracted with a solution of ExBox·4Cl in D O (8.0 mM, 4 mL) (intensive shaking
2
2
2
of the mixture for approximately 2 min), upon which the aqueous phase turned to an orange-yellow hue indicating supramolecular trapping of PAHs
4
+
inside ExBox . The orange-yellow solution was then (iii) extracted 3× with CH Cl (4 mL each time), and upon (iv) removal of the solvent, the
2
2
1
crude-oil extract was analyzed by H NMR spectroscopy and found to contain aromatic compounds. The same oil extraction experiment was carried
out using a blank (control; “6-C”) solution of D O without ExBox·4Cl, which did not undergo a color change upon extraction (ii) and did not
2
1
1
contain any aromatic species (control; H NMR in CD Cl ). (b) H NMR spectra (in order from top to bottom) for the crude oil sample (in
2
2
CD Cl ), crude oil extract (in CD Cl ), the control experiment (in CD Cl ), and ExBox·4Cl before and after the extraction (both in D O), recorded
2
2
2
2
2
2
2
at 298 K on a 500 MHz instrument. (c) GC-MS data for four PAH standards (top) and compared to the material (bottom) extracted from the crude
oil in steps iii and iv. From the MS data, it is possible to generate molecular formulas, corresponding to various PAH derivatives that fall within the
mass range of the tested standards.
ExBox4+ and PAHs, with potential semiconducting properties,
could serve as new alternative composite materials in organic
field-effect transistors.
Research Center (NERC), which is an Energy Frontier
Research Center (EFRC) funded by the U.S. Department of
Energy, Office of Basic Energy Sciences (DOE-BES) under
award DESC0000989. J.C.B. is supported by a National
Defense Science and Engineering Graduate (NDSEG) Fellow-
ship from the Department of Defense (DOD) and gratefully
acknowledges support from the Ryan Fellowship awarded
under the auspices of the Northwestern University Interna-
tional Institute for Nanotechnology. M.J. is supported by The
Netherlands Organisation for Scientific Research (NWO) and
the Marie Curie Cofund Action (Rubicon Fellowship). N.L.S. is
supported by a Graduate Research Fellowship (GRF) from the
National Science Foundation (NSF). P.L.M. acknowledges the
NSF and the Georgia Tech Facilitating Academic Careers in
Engineering and Science (FACES) committee for a postdoc-
toral fellowship. C.J.B. is supported by a GRF from the NSF.
ASSOCIATED CONTENT
Supporting Information
■
*
S
Detailed synthetic procedures and characterization (NMR and
HRMS) data for all compounds, crystallographic and
spectroscopic (NMR, UV/vis) characterization for ExBox·4PF6
and inclusion complexes 1−11, ITC data for 3, 4, 6, and 8. This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
stoddart@northwestern.edu
■
Author Contributions
These authors contributed equally.
Notes
‡
REFERENCES
■
(1) Radziszewski, B. Ber. Dtsch. Chem. Ges. 1876, 9, 260−262.
(2) Simpson, C. D.; Brand, J. D.; Berresheim, A. J.; Przybilla, L.;
̈
er, H. J.; Mullen, K. Chem.−Eur. J. 2002, 8, 1424−1429.
̈
3) (a) Clar, E. Polycyclic Hydrocarbons; Academic Press: London,
964; Vols. 1−2. (b) Clar, E. The Aromatic Sextet; John Wiley & Sons:
London, 1972. (c) Harvey, R. G. Polycyclic Aromatic Hydrocarbons;
Wiley-VCH: New York, 1997.
4) (a) Vijh, U. P.; Witt, A. N.; Gordon, K. D. Astrophys. J. 2005, 619,
The authors declare no competing financial interest.
Rad
(
1
ACKNOWLEDGMENTS
■
Data are tabulated, and the crystallographic parameters of each
single crystal were deposited into the Cambridge Crystallo-
graphic Data Centre (CCDC). We thank Anastasiya Vinokur
for her help with solving and refining the crystal structure of
complex 2. J.F.S. is supported by the Non-Equilibrium Energy
(
368−378. (b) Vijh, U. P.; Witt, A. N.; Gordon, K. D. Astrophys. J.
2004, 606, L65−L68.
I
dx.doi.org/10.1021/ja307360n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(
5) (a) Tang, J.; Carroquino, M. J.; Robertson, B. K.; Alexander, M.
Shveyd, A. K.; Dey, S. K.; Botros, Y. Y.; Stoddart, J. F. Chem. Sci. 2010,
1, 119−125.
Environ. Sci. Technol. 1998, 32, 3586−3590. (b) Ghosh, U.; Gillette, J.
S.; Luthy, R. G.; Zare, R. N. Environ. Sci. Technol. 2000, 34, 1729−
(21) Fo
̈
rster, T.; Kasper, K. Z. Elektrochem. 1955, 59, 976−980.
1
̈ ̈
736. (c) Bostrom, C.-E.; Gerde, P.; Hanberg, A.; Jernstrom, B.;
Johansson, C.; Kyrklund, T.; Rannug, A.; Tornqvist, M.; Victorin, K.;
Westerholm, R. Environ. Health Perspect. 2002, 110, 451−488.
(
d) Werner, D.; Higgins, C. P.; Luthy, R. G. Water Res. 2005, 39,
2
(
2
105−2113.
6) (a) Watson, M. D.; Fechtenkotter, A.; Mullen, K. Chem. Rev.
̈
̈
001, 101, 1267−1300. (b) Watanabe, M.; Chang, Y. J.; Liu, S.-W.;
Chao, T.-H.; Goto, K.; Islam, M. M.; Yuan, C.-H.; Tao, Y.-T.;
Shinmyozu, T.; Chow, T. Nat. Chem. 2012, 4, 574−578. (c) Rieger,
̈
R.; Mullen, K. J. Phys. Org. Chem. 2010, 23, 315−325.
(
7) Marinov, N. M.; Pitz, W. J.; Westbrook, C. K.; Castaldi, M. J.;
Senkan, S. M. Combust. Sci. Technol. 1996, 116, 211−287.
8) Luch, A. The Carcinogenic Effects of Polycyclic Aromatic
Hydrocarbons; Imperial College Press: London, 2005.
9) Scientific Opinion of the Panel on Contaminants in the Food
(
(
Chain on a request from the European Commission on Polycyclic
Aromatic Hydrocarbons in Food. EFSA J. 2008, 724, 1−114.
(
10) (a) Lehn, J.-M. Supramolecular Chemistry Concepts and
Perspectives; Wiley-VCH: Weinheim, Germany, 1995. (b) Beer, P.
D.; Gale, P. A.; Smith, D. K. Supramolecular Chemistry; Oxford
University Press: Oxford, U.K., 1999. (c) Steed, J. W.; Atwood, J. L.
Supramolecular Chemistry; Wiley-VCH: Weinheim, Germany, 2009.
(
d) Stoddart, J. F. Nat. Chem. 2009, 1, 14−15.
(
11) (a) Blyshak, L. A.; Dodson, K. Y.; Patonay, G.; Warner, I. M.
Anal. Chem. 1989, 61, 955−960. (b) Mohseni, R. M.; Hurtubise, R. J. J.
Chromatogr., A 1990, 499, 395−410. (c) Lamparczyk, H.; Zarzycki, P.;
Ochocka, R. J.; Sybilska, D. Chromatographia 1990, 30, 91−94.
(
6
d) Butterfield, M. T.; Agbaria, R. A.; Warner, I. M. Anal. Chem. 1996,
8, 1187−1190. (e) Shixiang, G.; Liansheng, W.; Qingguo, H.; Sukui,
H. Chemosphere 1998, 37, 1299−1305. (f) Ravelet, C.; Ravel, A.;
Grosset, C.; Villet, A.; Geze, A.; Wouessidjewe, D.; Peyrin, E. J. Liq.
Chromatogr. Relat. Technol. 2002, 25, 421−432.
(
12) (a) Li, H.; Qu, F. J. Mater. Chem. 2007, 17, 3536−3544. (b) Yin,
J.; Wang, L.; Wei, X.; Yang, G.; Wang, H. J. Chromatogr., A 2008, 1188,
99−207. (c) Bandela, A.; Chinta, J. P.; Hinge, V. K.; Dikundwar, A.
G.; Row, T. N. G.; Rao, C. P. J. Org. Chem. 2011, 76, 1742−1750.
13) Chen, Y.; Luo, J.; Zhu, X. X. J. Phys. Chem. B 2008, 112, 3402−
409.
14) (a) Peinador, C.; Pía, E.; Blanco, V.; García, M. D.; Quintela, J.
M. Org. Lett. 2010, 12, 1380−1383. (b) Blanco, V.; García, M. D.;
Terenzi, A.; Pía, E.; Fernandez-Mato, A.; Peinador, C.; Quintela, J. M.
Chem.−Eur. J. 2010, 16, 12373−12380.
1
(
3
(
́
(
15) (a) Michaelis, L.; Hill, E. S. J. Gen. Physiol. 1933, 16, 859−873.
(
(
b) Bird, C. L.; Kuhn, A. T. Chem. Soc. Rev. 1981, 10, 49−82.
c) Monk, P. M. S. The Viologens: Physicochemical Properties, Synthesis
and Applications of the Salts of 4,4′-Bipyridine; Wiley: New York, 1998.
16) (a) Takahashi, K.; Nihira, T.; Akiyama, K.; Ikegami, Y.; Fukuyo,
(
E. J. Chem. Soc., Chem. Commun. 1992, 620−622. (b) Porter, W. W.;
Vaid, T. P.; Rheingold, A. L. J. Am. Chem. Soc. 2005, 127, 16559−
1
6566. (c) Kolivosk
̌ ́ ̌ ́ ̌
a, V.; Gal, M.; Pospísil, L.; Valasek, M.;
Hromadova,
(
(
́
M. Phys. Chem. Chem. Phys. 2011, 13, 11422−11429.
17) Su, Y.-S.; Chen, C. F. Org. Lett. 2010, 12, 1888−1891.
18) (a) Brown, C. L.; Philp, D.; Stoddart, J. F. Synlett 1991, 7, 462−
er, R.; Vogtle, F. Angew. Chem., Int. Ed. Engl. 1997, 36,
30−944. (c) Arico, F.; Badjic, J. D.; Cantrill, S. J.; Flood, A. H.;
4
64. (b) Jag
̈
̈
9
́
Leung, K. C.-F.; Liu, Y.; Stoddart, J. F. Top. Curr. Chem. 2005, 249,
̌
́
O. S.; Spruell, J. M.; Heath, J.
2
03−259. (d) Dichtel, W. R.; Miljanic,
R.; Stoddart, J. F. J. Am. Chem. Soc. 2006, 128, 10388−10390.
e) Stoddart, J. F. Chem. Soc. Rev. 2009, 38, 1802−1820.
19) Flanagan, J. B.; Margel, S.; Bard, A. J.; Anson, F. C. J. Am. Chem.
Soc. 1978, 100, 4248−4253.
20) (a) Odell, B.; Reddington, M. V.; Slawin, A. M. Z.; Spencer, N.;
Stoddart, J. F.; Williams, D. J. Angew. Chem., Int. Ed. Engl. 1988, 27,
547−1550. (b) Asakawa, M.; Dehaen, W.; L’abbe, G.; Menzer, S.;
Nouwen, J.; Raymo, F. M.; Stoddart, J. F.; Williams, D. J. J. Org. Chem.
996, 61, 9591−9595. (c) Sue, C.-H.; Basu, S.; Fahrenbach, A. C.;
(
(
(
1
́
1
J
dx.doi.org/10.1021/ja307360n | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX