Supramolecular Porous Organic Nanocomposites for Heterogeneous Photocatalysis of a Sulfur Mustard Simulant

Article abstract of DOI:10.1002/adma.202001592

Efficient heterogeneous photosensitizing materials require both large accessible surface areas and excitons of suitable energies and with well-defined spin structures. Confinement of the tetracationic cyclophane (ExBox⁴⁺) within a nonporous anionic polystyrene sulfonate (PSS) matrix leads to a surface area increase of up to 225 m² g^?1 in ExBox?PSS. Efficient intersystem crossing is achieved by combining the spin-orbit coupling associated to Br heavy atoms in 1,3,5,8-tetrabromopyrene (TBP), and the photoinduced electron transfer in a TBP?ExBox⁴⁺ supramolecular dyad. The TBP?ExBox⁴⁺ complex displays a charge transfer band at 450 nm and an exciplex emission at 520 nm, indicating the formation of new mixed-electronic states. The lowest triplet state (T₁, 1.89 eV) is localized on the TBP and is close in energy with the charge separated state (CT, 2.14 eV). The homogeneous and heterogeneous photocatalytic activities of the TBP?ExBox⁴⁺, for the elimination of a sulfur mustard simulant, has proved to be significantly more efficient than TBP and ExBox⁺⁴, confirming the importance of the newly formed excited-state manifold in TBP?ExBox⁴⁺ for the population of the low-lying T₁ state. The high stability, facile preparation, and high performance of the TBP?ExBox?PSS nanocomposites augur well for the future development of new supramolecular heterogeneous photosensitizers using host–guest chemistry.

Full text of DOI:10.1002/adma.202001592

CommuniCation
www.advmat.de
Supramolecular Porous Organic Nanocomposites
for Heterogeneous Photocatalysis of a Sulfur
Mustard Simulant
Yassine Beldjoudi, Ahmet Atilgan, Jacob A. Weber, Indranil Roy, Ryan M. Young,
Jierui Yu, Pravas Deria, Alan E. Enciso, Michael R. Wasielewski, Joseph T. Hupp,*
and J. Fraser Stoddart*
transformation is by using heavy atoms
Efficient heterogeneous photosensitizing materials require both large accessible
surface areas and excitons of suitable energies and with well-defined spin struc-
tures. Confinement of the tetracationic cyclophane (ExBox4+) within a nonpo-
rous anionic polystyrene sulfonate (PSS) matrix leads to a surface area increase
of up to 225 m2 g−1 in ExBox•PSS. Efficient intersystem crossing is achieved by
combining the spin-orbit coupling associated to Br heavy atoms in 1,3,5,8-tetra-
bromopyrene (TBP), and the photoinduced electron transfer in a TBP⊂ExBox4+
supramolecular dyad. The TBP⊂ExBox4+ complex displays a charge transfer
band at 450 nm and an exciplex emission at 520 nm, indicating the formation of
new mixed-electronic states. The lowest triplet state (T1, 1.89 eV) is localized on
the TBP and is close in energy with the charge separated state (CT, 2.14 eV). The
homogeneous and heterogeneous photocatalytic activities of the TBP⊂ExBox4+,
for the elimination of a sulfur mustard simulant, has proved to be significantly
more efficient than TBP and ExBox+4, confirming the importance of the newly
formed excited-state manifold in TBP⊂ExBox4+ for the population of the low-
lying T1 state. The high stability, facile preparation, and high performance of the
TBP⊂ExBox•PSS nanocomposites augur well for the future development of new
supramolecular heterogeneous photosensitizers using host–guest chemistry.
since spin-orbit (λ) coupling increases pro-
portionally with the increase in the atomic
4
number (λ ∝ Z). The lone-pair electrons
of heavy atoms additionally augment
intersystem crossing (ISC) as evident by
the El-Sayed rule[6] which states that 1S
3
(π, π*) → T (n, π*) ISCs are faster than
3
1
1S (π, π*) → T (π, π*) transitions, and S
(n, π*) → 3T (π, π*) are faster than 1S
(n, π*) → 3T (n, π*) transitions. To a lesser
extent, intramolecular charge transfer
(ICT) in donor–acceptor (D−A) dyads has
been employed in the design of heavy
atoms-free photosensitizers. For instance,
it has been reported[7] that the utilization of
an intramolecular D−A boron-dipyrrome-
thene−anthracene (BODIPY-anthracene)
dyad which undergoes a charge separation
between the two units, acts as an efficient
triplet sensitizer on account of the very
small S−T gap (ΔEST). More recently, other
investigations have led to the report[8]
that charge-transfer (CT) states formed
in the BODIPY-Pyrene D−A dyad, as a
In recent years, triplet excited-state chromophores have
attracted considerable attention on account of their applications
in photodynamic therapy,[1] optoelectronic devices,[2] photocatal-
ysis,[3] bioimaging and sensing,[4] and photon up-conversion.[5]
The most common strategy to enhance the singlet-triplet (S−T)
result of photoinduced electron transfer recombine to yield
the BODIPY triplet excited state. In the presence of molecular
oxygen, BODIPY-Pyrene dyads sensitize singlet oxygen (1O2)
with quantum yields of up to 0.75. In this context, the design
of an efficient photosensitizer requires an understanding of
Dr. Y. Beldjoudi, A. Atilgan, Dr. J. A. Weber, Dr. I. Roy, Dr. R. M. Young,
Dr. A. E. Enciso, Prof. M. R. Wasielewski, Prof. J. T. Hupp,
Prof. J. F. Stoddart
J. Yu, Prof. P. Deria
Department of Chemistry and Biochemistry
Southern Illinois University
1245 Lincoln Drive, Carbondale IL 62901, USA
Prof. J. F. Stoddart
Institute for Molecular Design and Synthesis
Tianjin University
Department of Chemistry
Northwestern University
Evanston, IL 60208, USA
E-mail: j-hupp@northwestern.edu; stoddart@northwestern.edu
Dr. R. M. Young, Prof. M. R. Wasielewski
Institute for Sustainability and Energy
Northwestern University
92 Weijin Road, Nankai District, Tianjin 300072, China
Prof. J. F. Stoddart
School of Chemistry
University of New South Wales
Sydney, NSW 2052, Australia
Evanston, IL 60208, USA
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adma.202001592.
DOI: 10.1002/adma.202001592
©
2001592 (1 of 13)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2020, 2001592

www.advancedsciencenews.com
www.advmat.de
the different parameters which promote exciton transforma-
tion. Theoretical investigations[9] of the S−T transformation
have shown that, besides the well-recognized importance of the
ΔEST and heteroatom participation for exciton transformation,
the two excited states should contain the same components of
the transition configurations to establish the transformation
channels in bridging the spin-forbidden transitions between
two electronic states with different spin multiplicities. Hence,
without the matching energy gap, the excitons are weak in
S−T transformation because of the lack of enough energy from
thermal vibrations or other suitable ways in dissipating excess
energy.
energy transfer from ExBox4+. Other investigations reported[17]
that the close interaction and the significant orbital overlap
between the PDI (perylene diimide) as a guest and ExBox4+
acting as a host, enables ultrafast energy transfer to proceed
by the electron exchange Dexter mechanism.[18] In addition,
incorporation of heavy atoms into the cyclophane leads to an
efficient quenching of the fluorescence as a result of efficient
spin-orbit ISC pathways leading to the generation of the triplet
state of the PDI guest.[16]
Here, we propose to utilize 1,3,6,8-tetrabromopyrene (TBP)
as an electron donor with ExBox4+ as the electron acceptor in
order to form a host–guest D−A supramolecular complex,
namely, TBP⊂ExBox4+. This complex is expected (Figure 1) to
promote the S−T exciton transformation between the excited
states of the two components, enhancing ISC to populate the
low-lying locally excited (LE) triplet state (T1) of TBP. This
design strategy requires (i) efficient CT between the guest (D)
and host (A), (ii) the two fluorophores absorb similar radiation
wavelengths in order to access the excited-states of both chromo-
phores, (iii) a small ΔEST (<0.37 eV) and small distance (≈3.5 Å)
between the D and A in order to facilitate spin-orbit charge-
transfer intersystem crossing[19] (SOCT-ISC), (iv) the energies of
both the CT and LE T1 states must be similar, and (v) incorpo-
ration of heteroatoms (N atoms) in the host (ExBox4+) and the
guest (Br atoms) which can, not only facilitate the S−T transfor-
mation, but also offer a low lying triplet state that can promote
energy transfer to molecular oxygen[20] (1Σ = 1.63 eV).
From a practical perspective, the very low solubility of the
TBP in organic and aqueous media at ambient temperatures is
necessary in order to enhance the stability of the supramolec-
ular photocatalyst since host–guest formation is not in equilib-
rium. It was previously reported[21] that water-soluble cobalt(III)
tetrahedral coordination capsules exhibit non-equilibrium guest
binding properties because of the hydrophobic effect which
is associated with the low solubility of the guest molecules
in aqueous media. Finally, incorporation of the tetracationic
TBP⊂ExBox4+ photosensitizer within the anionic matrix of PSS
leads to the formation of a stable and porous composite for the
heterogeneous photocatalysis of CEES. All compounds have
been characterized in solution and in the solid state by absorp-
tion, diffuse reflectance and fluorescence spectroscopies. Fur-
thermore, the electronic properties of the host–guest complex
have been unraveled using transient absorption spectroscopy
and time-dependent DFT calculations. Finally, we have inves-
tigated the photocatalytic performance of TBP⊂ExBox4+•4Cl,
and TBP⊂ExBox4+•PSS for the elimination of the sulfur mus-
tard simulant (CEES) in both homogeneous and heterogeneous
liquid-phase reactions.
In the past decade, considerable interest has been devoted
toward the photo-oxidation of the sulfur mustard (SM) and
1
2-chloroethyl ethyl sulfide (CEES) using O2 since the latter is
a mild oxidant and photocatalysis has been proven to involve
faster kinetics and to be more selective with the less harmful
sulfoxide derivative, 2-chloroethyl ethyl sulfoxide (CEESO), is
formed as a major product while the 2-chloroethyl ethyl sul-
fone derivative (CEESO2), is a minor product.[10] Several porous
materials, such as metal-organic frameworks (MOFs) and cova-
lent organic frameworks (COFs), with photosynthesizing prop-
erties[11] have been utilized for heterogeneous photocatalysis of
SM or CEES since the large surface areas of these porous mate-
rials facilitate accessibility of the reactants to the photoactive
sites. Processability of these crystalline powder materials for
use in military protective equipment (MPE), however, remains
challenging. Recently, Karwacki and co-workers[12] reported
efficient photocatalysis of CEES to CEESO using 4,4-difluoro-
4-bora-3a,4a-diaza-s-indacene (BODIPY) photosensitizers doped
into organic polymeric matrices. Nevertheless, the weak inter-
actions through dispersive forces of the BODIPY photosensi-
tizers to the polymer matrices hamper development of viable
MPEs because of the leaching of the photosensitizer under cat-
alytic conditions. In addition, a large amount of photocatalyst is
required to decrease the conversion lifetime to <1 min. In this
context, development of sustainable photosensitizing organic
materials for the heterogeneous catalysis of SM or CEES
requires that the material fulfill four main requirements: (i)
the material is porous, increasing the photoactive surface area
1
and facilitating the diffusion of reactant and products, (ii) O2
generation is efficient, (iii) the material is stable under photo-
catalytic conditions, and (iv) its preparation is easy, inexpensive,
scalable, and capable of being incorporated into MPEs.
In the present article, we describe (Scheme 1) the prepa-
ration of supramolecular porous organic composites using
an anionic polymeric matrix such as Polystyrene Sodium
Sulfonate (Na•PSS) and extended tetracationic cyclophanes
such as ExBox4+ and Ex2.2Box4+. The rigidly defined cavities of
these cyclophanes, when assembled within a polymeric matrix
relying on electrostatic interactions, offers porous properties
that are necessary in order to optimize the diffusion of reac-
tants and products within them and increase the active surface
area for 1O2 generation. Furthermore, tetracationic cyclophanes
such as ExBox4+ are attractive candidates for ultrafast inter-
molecular CT from an electron-rich guest,[13] intramolecular
through-bond CT from the p-xylylene bridges to the extended
bipyridinium units[14] and multielectron accumulation,[15]
leading to an array[16] of accessible mixed-valence states, and
The ExBox•4PF6 and Ex2.2Box•4PF6 cyclophanes were
synthesized following protocols already reported in the lit-
erature.[22] Although TBP is insoluble in the most common
organic solvents, at high temperatures it becomes soluble
in PhMe to afford a pale-yellow solution. The host–guest
complex TBP⊂ExBox•4PF6 can be formed (Scheme 1e) by
dropwise addition of TBP, solubilized in hot PhMe into a
solution of ExBox•4PF6 in hot dimethyl formamide (DMF).
After heating the mixture for 24 h at 80 °C, an intense yellow/
orange colored solution is formed. After evaporation of the
solvent and solubilization of TBP⊂ExBox•4PF6 in MeCN, the
©
Adv. Mater. 2020, 2001592
2001592 (2 of 13)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
Scheme 1. (top) The structural formulas of building blocks utilized in the design of supramolecular photosensitizing porous nanocomposites for
heterogeneous photocatalysis. a) ExBox4+, b) Ex2.2Box4+, c) 1,3,6,8-tetrabromopyrene (TBP), d) Sodium Polystytene Sulfonate (Na•PSS), e) Synthesis of
the TBP⊂ExBox•4PF6, TBP⊂ExBox•4Cl, and TBP⊂ExBox•PSS composites.
©
2001592 (3 of 13)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2020, 2001592

www.advancedsciencenews.com
www.advmat.de
Figure 1. The photoinduced electron exchange in a D−A dyad, acting as an efficient triplet photosensitizer.
insoluble excess of TBP can be removed by filtration. Tetrabu-
tylammonium chloride was added to the MeCN solution con-
taining TBP⊂ExBox•4PF6 in order to exchange the PF6 to
sizes.[24] The Brunauer–Emmett–Teller (BET) surface areas of
ExBox•PSS and Ex2.2Box•PSS at 195 K were found (Figure 2a)
to be 226 and 86 m2 g−1 respectively. Clearly, the inherent
cavities in the tetracationic cyclophanes, combined with their
distribution within an anionic polymeric matrix, leads to an
increase in the surface area of the PSS matrix. The differences
in the porosity between ExBox•PSS and Ex2.2Box•PSS, however,
can be attributed to the coexistence of intrinsic porosity asso-
ciated with the cyclophane’s cavity and the extrinsic porosity
associated with the morphology of the composite. It has been
−
Cl− anions, a process that renders the cyclophanes soluble in
aqueous media. After isolation of the TBP⊂ExBox•4Cl complex
as a yellow powder, it was dissolved in H2O and Na•PSS was
added dropwise under strong agitation to form (Scheme 1e)
a precipitate of TBP⊂ExBox•PSS composite of 5/3 w/w ratio.
The cyclophane has four positive charges and requires four
styrene sulfonate (C8H7O3S−) anions in order to calibrate the
charges leading to the dispersion of the TBP⊂ExBox4+ along
the polymer backbone. The very low solubility of the TBP,
combined with the trapping of TBP⊂ExBox4+ within the PSS
polymer matrix as a result of electrostatic interactions, is
essential in order to enhance the stability of the composite in
both aqueous and organic media for efficient heterogeneous
photocatalysis.
In order to ascertain the role of the host–guest D−A complex
in photocatalysis, the ExBox•PSS and Ex2.2Box•PSS compos-
ites have also been prepared (Schemes S1 and S2, Supporting
Information) quantitatively following similar protocols. After
dissolution of ExBox•4Cl and Ex2.2Box•4Cl in H2O, Na•PSS
was added dropwise to form the ExBox•PSS and Ex2.2Box•PSS
composites at 1/1 and 3/2 w/w ratios, respectively. These com-
posites are insoluble in both aqueous and non-aqueous media.
In order to study the optical properties of the composites in
aqueous solution, we prepared the two composites at 1/3 and
1/1 w/w ratios, respectively. See the Supporting Information for
more details.
The CO2 adsorption on the ExBox•PSS and Ex2.2Box•PSS
composites has been performed (Figure 2a) and compared
to the adsorption isotherm of pristine Na•PSS in order to
confirm the role of the tetracationic cyclophanes in forming
the porous nature of these composites. Figure 2a shows, as
expected, a negligible adsorption of CO2 into Na•PSS at 195 K
indicative of its nonporous nature. The rapid increase in CO2
uptake at low pressures for both ExBox•PSS and Ex2.2Box•PSS
indicates the presence of micropores, whilst the continuous
increase of the uptake confirms the presence of larger pores.
The pore volume plot revealed (Figure S6, Supporting Infor-
mation) the existence of several pores of different sizes, e.g.,
medium-sized micropores (7–9 Å) and ultra-micropores
(<7 Å). These pore sizes are like those of other porous mate-
rials, such as MIL-47 and TIF-1 which possess[23] pore sizes
in a range of 7–9 Å, while the MFI and MOR have pore sizes
of <7 Å. Other polymers of intrinsic microporosity (PIMs)
have been reported in the literature and exhibit similar pore
Figure 2. (a) CO2 sorption isotherms at 195 K for Na•PSS, ExBox•PSS,
and Ex2.2Box•PSS. SEM image showing the rough texture of the (b,c)
ExBox•PSS composite and (d,e) TBP⊂ExBox•PSS composite.
©
Adv. Mater. 2020, 2001592
2001592 (4 of 13)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
suggested[24] that microporous materials can be derived directly
from soluble polymers whose randomly contorted shapes pre-
vent an efficient packing of the macromolecules in the solid
state.
Supporting Information) the characteristic absorption fea-
tures of ExBox4+ and PSS. The emission of this composite in
aqueous solution exhibits a slight bathochromic shift of 47 nm
to become (Figure S15, Supporting Information) centered at
430 nm as a consequence of the change in the polarity and
viscosity of the medium.[27] Time-resolved photoluminescence
decay was monitored at 430 nm, using 374 nm as the excitation
wavelength. The decay curve was fitted (Figure S15, Supporting
Information) to the double-exponential function, resulting in a
slow component (τ1 = 1.43 ns) and a fast one (τ2 = 0.47 ns).[28]
In PhMe, TBP is weakly soluble and the absorption profile of
TBP shows (Figure S20 (left), Supporting Information) several
absorption bands at 378, 359, 341, and 293 nm characteristic of
the π → π∗ and n → π∗ transitions. The diffuse reflectance of
TBP reveals (Figure 3a) the existence of two maxima at 320 and
380 nm, similar (Figure S20 (right), Supporting Information) to
the solution absorption profile. Upon excitation at 380 nm, TBP
offers a featured (Figure S18a, Supporting Information) emis-
sion band in a range 400–470 nm (λmax = 439 nm, ΦF = 1.62%)
with a Stokes shift (Table S2, Supporting Information) of
0.22 eV. The singlet excited-state lifetime (Table 1) is rather long
(τ1 = 0.11 ns, τ2 = 0.60 ns, τ3 = 10.27 ns), which is associated
with the excimer emission as the result of the [π⋅⋅⋅π] interac-
tions. Whilst at 298 K the excitation band is centered on 377 nm
associated with the existence [π···π] interaction in the ground
state, at 77 K, an intense excitation band appears at 315 nm with
a smaller broad band in the range of 340−420 nm (Figure S18b,
Supporting Information). The incorporation of TBP inside the
cavity of ExBox4+ does not affect significantly the ground-state
electronic properties of either component. Indeed, the absorp-
tion spectra of TBP⊂ExBox•4Cl and TBP⊂ExBox•4PF6 are
characterized (Figure S20 (right), Supporting Information) by
the overlap of the absorption bands of both the host and guest
components, with a maximum absorption centered on 320 and
358 nm in MeCN and H2O, respectively. The diffuse reflectance
spectrum of the TBP⊂ExBox•4PF6 reveals (Figure 3a) the exist-
ence of a broad CT band centered on 455 nm. It follows that
the TBP⊂ExBox4+ supramolecular D−A dyads might exhibit a
SOCT-ISC in order to enhance the exciton transformation. Pre-
vious work indicates[29] that efficient ISC can be obtained with
intramolecular electron D−A dyads, displaying n−π* ↔π−π∗
systems because the electron transfer (charge separation or
recombination) will result in magnetic torque on the electron
spin which will induce a molecular orbital angular momentum
change, enhancing ISC. Steady-state fluorescence spectroscopy
revealed (Figure 3b and Table 1) the existence of two emis-
sion bands centered on 440 and 512 nm at 298 K. The band at
440 nm can be attributed to TBP monomer emission, while the
In order to test the diffusion of larger molecules, we inves-
tigated the adsorption of tetrathiafulvalene (TTF) inside the
ExBox•PSS composite. Previous studies have revealed[25] that
TTF has a relatively strong affinity for the tetracationic cyclo-
phanes, forming dark green host–guest complexes. Addition
of ExBox•PSS composite to a solution of the TTF of concen-
tration of 10−5 m led to the absorption of the TTF molecules,
affording (Figure S10, Supporting Information) a dark green
composite as a result of the CT interactions between TTF and
ExBox4+ in the TTF⊂ExBox4+ host–guest complex. The crystal
structure of TTF⊂ExBox4+ shows (Figure S72, Supporting
Information) one molecule of TTF trapped inside the cavity of
ExBox4+, while additional TTF molecules remain outside the
cavity. Similarly, in the ExBox•PSS matrix, some TTF molecules
are hosted inside the ExBox4+ cavities, while other TTF mole-
cules reside outside. We conclude that the composite possesses
large pores which allow diffusion of both the reactants (CEES,
O2) and product (CEESO) molecules for photocatalytic applica-
tions. In recent years, considerable interest has been focused[26]
toward the use of porous materials for catalytic applications on
account of their high active surface areas and low diffusion bar-
riers. Scanning electron microscopy (SEM) has revealed that,
while the Na•PSS (Figure S7, Supporting Information) has
a smooth texture, the composites ExBox•PSS, Ex2.2Box•PSS
and TBP⊂ExBox•PSS are all characterized (Figure 2 (bottom),
Figures S8 and S9, Supporting Information) by having a rough
and spongy texture indicative of their porous natures. Powder
XRD has revealed (Figures S4 and S5, Supporting Informa-
tion) that all the composites are amorphous, confirming the
distribution of the tetracationic cyclophanes in the PSS matrix
and the absence of phase separation between the amorphous
Na•PSS and the crystalline cyclophane phase. Notably, the
discrepancy in the porosity and the morphology between
ExBox•PSS and Ex2.2Box•PSS implies that the two cyclophanes
interact with the PSS matrix differently. The length of ExBox4+
and Ex2.2Box4+ are 12 and 17 Å, respectively, while their heights
are similar. The distance between the styrene sulfonate units
in the PSS backbone is ≈3 Å. In this situation, the discrepancy
between the morphologies and porosities of the ExBox•PSS and
Ex2.2Box•PSS composites could be associated with different dis-
tributions of the cyclophanes in relation to the charged units in
the PSS polymer.
Absorption and fluorescence investigations were carried out
in order to unravel the electronic properties of the host–guest
complex (TBP⊂ExBox•PF6) in solution and the polymer com-
posites in the solid-state. Na•PSS is colorless in H2O and its
UV–vis absorption profile is characterized (Figure S11, Sup-
porting Information) by the existence of two absorption bands
at 223 and 252 nm, while fluorescence spectroscopy has shown
(Figure S12, Supporting Information) that excitation at 254 nm
offers a single emission band at 308 nm. ExBox•4Cl in H2O
displays (Figure S13, Supporting Information) excitation and
emission bands at 358 and 383 nm, respectively, arising from
the lowest singlet excited state. The Exbox•PSS composite
of 1/3 w/w ratio is soluble in H2O and displays (Figure S14,
one at 512 nm is considered to be an exciplex emission (1S1CT
)
arising from the TBP⊂ExBox4+ host–guest complex.
The electronic properties of the TBP⊂ExBox•4PF6 complex
have also been investigated with transient absorption spectros-
copy. By exciting at either 414 or 450 nm, the kinetics of the
charge separation and recombination for TBP⊂ExBox4+ were
obtained. See Figure 3c and Figures S22–S24 (Supporting Infor-
mation). In addition, excitation at each of these wavelengths
allows deconvolution of the roles of the LE and CT states in the
overall electronic properties. Indeed, on excitation at 414 nm the
LE state TBP can be accessed, while at 450 nm only the lowest
©
2001592 (5 of 13)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2020, 2001592

www.advancedsciencenews.com
www.advmat.de
Figure 3. (a) Diffuse reflectance spectra recorded for the TBP⊂ExBox•4PF6 complex in MeTHF (blue), for the TBP in PhMe (black) and for the
ExBox4+•PSS (green) and TBP⊂ExBox4+•PSS composite (red) in the solid state. (b) Excitation and emission spectra of the TBP⊂ExBox•PF6 complex
in MeTHF, recorded at 77 and 298 K. (c) Kinetic analysis of the femtosecond transient absorption data of TBP⊂ExBox•PF6 at λex = 414 nm showing i)
Fits to the solution of a 4-state kinetic model (A → B → C → D). ii) Model populations as a function of time. iii) Evolution-associated spectra for each
species in the model. (d) Excitation and emission spectra of the TBP⊂ExBox•PSS composite. (e) Emission decay of the TBP⊂ExBox•PSS composite at
λ
ex = 405 nm fitted using a triple exponential fit.
CT state can be reached. Photoexcitation of TBP⊂ExBox4+ at
414 or 450 nm results (Figure 3c and Figure S24, Supporting
Information) in the appearance of strong peaks at 522, 985,
and 1140 nm as well as a radiative recombination band at
655 nm. Similar absorption bands have been observed[12] for the
Perylene⊂ExBox4+ CT complex without the radiative recombi-
nation band. This radiative recombination gives an estimate of
the energy of the CT state of 1.89 eV. The DFT-calculated energy
of the lowest triplet state (T1) state is also 1.89 eV, implying
that these states may interact via SOCT-ISC. When excited at
414 nm, these bands are formed (Figure 3c and Figure S22,
Supporting Information) immediately and rise over the next
≈7 ps, then decay in ≈54 and ≈300 ps. The immediate appear-
ance of the bands associated with ExV+• indicates that CT from
the LE state of TBP is ultrafast (<300 fs), as it is the case for
Perylene⊂ExBox4+. The ≈7 ps time constant is associated with
a structural relaxation of the charge-separated state,[13] and the
biexponential decay of the TBP+⊂ExBox3+ state is most likely a
consequence of the distribution of binding geometries in solu-
tion. Both recombination processes are slightly longer than the
≈ 40 ps charge recombination observed for Perylene⊂ExBox4+.
Notably, direct excitation of the CT band (λex = 450 nm) offers
a similar TA profile; however, with generally longer time con-
stants a similar rise occurs (Figure S24, Supporting Informa-
tion) with ≈8.6 ps, then a decay in ≈71 ps and a minor-compo-
nent decay in ≈900 ps.
Nanosecond transient absorption measurements leads
to the observation (Figure S23, Supporting Information) at
λ
ex = 414 nm of a long-lived triplet of >1.5 µs, implying that exci-
1
1
tation of the upper CT and LE states (S2, S3, S4 states, vide
infra) populates the T1 state of TBP following charge recombi-
1
nation, while excitation of the CT states at 450 nm does not
lead to a detectable triplet population. The lack of triplet for-
mation, following 450 nm excitation, is associated with the
lower amount of triplet character in the CT state populated by
absorption, which is also consistent with the discrepancy in the
decay time-constants at different excitation wavelengths. Whilst
excitation at 414 nm offers shorter time-constants, associated
with the more efficient SOCT-ISC between the upper states
(S2 → T6, S3 → T6 and S4 → T8 for example, Figure 5a) with
higher triplet character, excitation at 450 nm offers longer time-
constants associated with slower SOCT-ISC between the S1 and
T2 and T3 states. Thus, the triplet population observed upon
higher energy excitation (HLCT states) is a result of a rapid ISC
induced by both the heavy Br atoms and SOCT-ISC between
the D−A. These results are consistent with the efficient photo-
catalytic conversion of CEES at 395 nm, while photoexcitation
at 450 nm is very slow (Figure S63, Supporting Information).
Diffuse reflectance measurements on solid films of the
Table 1. Fluorescence parameters of the ExBox•PSS, TBP, and
TBP⊂ExBox•PSS at 298 K.
Sample
λem [nm]
τ1 [ns]
τ2 [ns]
τ3 [ns]
[Amplitude %] [Amplitude %] [Amplitude %]
440
515
480
439
0.137 [37.97]
0.3512 [13.39]
0.22 [20.27]
0.11 [40.09]
1.32 [27.7]
2.027 [56.8]
1.40 [42.89]
0.60 [8.55]
4.031 [34.33]
6.14 [29.81]
8.23 [36.84]
10.27 [51.36]
TBP⊂ExBox•PSS
ExBox•PSS
a)TBP
ExBox•PSS composite exhibit (Figure 3a)
a broad peak
in the range of 200−450 nm centered on 350 nm. The
ExBox•PSS emission (Figure S16, Supporting Information)
a)In MePh.
©
Adv. Mater. 2020, 2001592
2001592 (6 of 13)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
is excitation-dependent, ranging from 470 to 525 nm
(λex 380−450 nm). The emission is centered on 470 nm at
λex = 380 nm (Figure S17, Supporting Information), with a
Stokes shift of 0.54 eV (Table S2, Supporting Information)
and singlet excited-state lifetimes (ΦF = 1.32%, τ1 = 0.22 ns,
τ2 = 1.40 ns, τ3 = 8.23 ns at 298 K) (Table 2) slightly larger than
those in solution. The TBP⊂ExBox4+•PSS composites show
(Figure 3a) an intense broad reflectance peak in the range of
200−440 nm (centered on 360 nm) and a small CT band at
455 nm similar to that of TBP⊂ExBox•4PF6 in solution. The
broad absorption spectrum (Figure S21, Supporting Informa-
tion) extending up to 600 nm indicates the absence of a well-
defined band edge in the UV–vis energy range for all the mate-
rials. As in the case of TBP⊂ExBox•4PF6 in solution, the com-
posite TBP⊂ExBox•PSS displays (Figure 3d) a first emission
In order to understand better the electronic properties of the
TBP⊂ExBox4+ complex and obtain an estimation of the singlet-
triplet energy gap (ΔEST), we utilized both the APFD and the
B3LYP functionals in conjunction with the 6–31G(d) basis set
to calculate molecular geometries. Optimization of the super-
structure of the TBP⊂ExBox4+ at the B3LYP/6-31G(d) energy
level leads to a larger interplanar distance (≈4.2 Å) between the
TBP and Exbipy2+ units while utilization of the APFD functional
resulted in a superstructure (Figure S33, Supporting Informa-
tion) with interplanar distances between TBP and Exbipy2+
of 3.5 Å, similar to those reported[22] for the crystal structures
of polyaromatic compounds inside ExBox4+. The discrepancy
between these optimized superstructures is a result of incorpora-
tion of an empirical dispersion correction term within the APFD
formalism, while dispersion interactions are neglected within
the B3LYP functional.[31] In addition, these two geometries offer
the possibility to determine the energy of the LE states of the TBP
and ExBox4+ and, hence, unravel the role of the orbital overlap
between the D−A into the formation of mixed excited states. Sat-
isfied by the presence of zero negative frequencies, gas-phase
TD-DFT calculations have been subsequently carried out at the
B3LYP/6-31G(d) level of theory using Gaussian16 software.[32]
Here we discuss the electronic properties of TBP⊂ExBox4+
derived from the APFD/6-31G(d) optimized structure (Figure 5
and Table 2), while a detailed analysis of the electronic proper-
ties of the B3LYP/6-31G(d) geometry can be found in the Sup-
porting Information. The molecular electrostatic potential
difference map (CI–SCF) of the TBP⊂ExBox4+ complex revealed
(Figure S42a, Supporting Information) that the negative electron
density is localized on TBP, while the positive charge density is
localized on ExBox4+, consistent with the electron D−A nature of
the complex. The calculated absorption spectrum (Figure S42b,
Supporting Information) reproduces well the experimental
band at 440 nm and an exciplex emission band at 522 nm (λex
=
380 nm, ΦF = 2.18%) with a Stokes shift of 1.07 eV (Tables S1
and S2, Supporting Information), indicating of persistence of
the host–guest complex in the composite. The time-correlated
emission measurements at λex = 405 nm revealed the existence
of two components with different S1 lifetimes at 440 nm (τ1 =
0.14 ns, τ2 = 1.32 ns, τ3 = 4.03 ns) and 515 nm (τ1 = 0.35 ns, τ2 =
2.02 ns, τ3 = 6.14 ns)[30] (Figure 3e and Table 1) which are associ-
ated with TBP and TBP⊂ExBox4+, respectively. It is noteworthy
1
that the SCT lifetimes of TBP⊂ExBox4+ in the solid state are
1
slightly shorter than the LE states in the two separate compo-
nents, TBP and ExBox•PSS, indicating the existence of com-
peting decay pathways. Furthermore, this observation indicates
the formation of a new mixed dipole entity in TBP⊂ExBox4+
that exhibits faster relaxation from the CT state, consistent with
the TA studies which support efficient ISC associated with the
large SOC of the Br atoms but also the influence of SOCT-ISC
in the D−A dyad.
Table 2. Excitation energy (E in eV), oscillator strength (f), transition configuration of S0 →Sn and S0 →Tn for exciton transformation, and energy gap
of the singlet-triplet splitting (ΔEST). These calculation are based on the APFD-6-31G(d) geometry-optimized molecular structure.
Sn
1
Ea)
[eV]
Oscillator strength
Transition configuration
[%]
Tn
1
E
[eV]
Transition configuration [%]
ΔEST
ΔEST
[eV]
[fb)]
H→L+1(99.7%)
H→L+2 (88%),
H→L+3 (5.2%)
2,1ΔEST
2.24
0.0004
1.89
1.0971
1,2ΔEST
1,3ΔEST
3,4ΔEST
2
3
2.99
3.11
0.0097
0.0131
2
3
4
2.19
2.24
2.56
0.049
0.0076
0.5546
H→L+2(33.8%), H→L+3(62.6%)
H-1→L+1(14.1%)
H→L(99.6%)
H→L+1(98.4%)
H→L+2 (45.3%), H→L+3 (31.1%)
H-4→L+1(29.8%), H-1→L(47.5%)
4
5
6
3.20
3.41
3.46
0.2478
0.0017
0.0012
5
6
7
2.57
2.96
3.10
3.19
0.5462
0.1573
H-1→L+1(80.5%), H→L+2 (8.8%)
H-3→L(94.5%)
H-4→L+1(32.8%), H-1→L(43.9%)
H→L+3(80.2%)
3,5ΔEST
3,6ΔEST
H-1→L+2(28.7%)
H→L+4(97.3%)
8
0.0065
0.0064
0.098
H→L+5(37.6%), H→L+7(13.5%)
H-5→L(87.0%), H-4→L (6.0%)
H-6→L+1 (73.2%), H-4→L(17.3%)
H-6→L+1(18.5%)
H-9→L+1(19.2%)
4,8ΔEST
4,9ΔEST
3,10ΔEST
7
8
9
3.51
3.55
3.57
0.0346
0.0052
0.2771
H-7→L(29.1%), H-1→L+1(29.1%)
H-9→L(22.5%)
9
3.19
3.21
H-7→L+1(20.7%), H-1→L(12.4%)
H-1→L+2(10.6%), H→L+3(10.4%)
H→L+5(38.8%), H→L+7(17.8%)
10
H-7→L (30.2%), H-4→L (34.3%)
10
3.65
0.323
H-4→L(28.7%), H-7→L (53.7%),
H-9→L+1 (5.9%)
a)H and L represent, respectively, HOMOs and LUMOs; b)The most similar and energetically close ΔEST are highlighted with the same color.
©
2001592 (7 of 13)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2020, 2001592

www.advancedsciencenews.com
www.advmat.de
proposed[9] that the minimum requirement for realizing exciton
transformation is the matching of the energy levels of the two
states based on the thermal equilibrium between the singlet
and triplet excited states. Although the exciton transformation
Table 3. Homogeneous photocatalysis parameters using 1 mol% of
photocatalyst at 395-nm photoirradiation in CD3OD.
Catalyst
Irradiation
time
[min]
Total
conversion
[%]
Full conver-
sion time
[min]
t1/2
[min]
(CEESO)
Selectivity
3
channel S1→T3 (1CT → CT) has a very small ΔEST (≈0 eV), the
weak f of the CT transitions in the TBP⊂ExBox4+ leads to a low
population of the T1 state as observed by TA experiments, corre-
sponding to (Figure S63, Supporting Information) a weak pho-
tosensitizing efficiency at λex = 450 nm.
ExBox•4Cl
20
120
27
100
60
16
/
7
97
34
97
Ex2,2Box•4Cl
79
3.5
100
9
TBP⊂ExBox•4Cl
The photocatalytic performance of the TBP⊂ExBox4+ D−A
dyad is high in the excitation range 380−420 nm (λmax = 395 nm,
3.13 eV), and so the photocatalytic properties arise from the
1HLCT states, S2, S3, and S4 (band at 387 nm, Figure S42b, Sup-
porting Information). The S0→S2 (2.99 eV), S0→S3 (3.11 eV)
and S0→T1 (1.89 eV) transitions are similar (HOMO→LUMO+2
and HUMO→LUMO+3), and the ΔEST is very large (>1 eV),
hampering (Figure 5a) efficient ISC between S2→T1 and S3→T1
channels. From the TA experiments, the S−T transformation is
more efficient when higher energy excited states are accessed.
The S0→T6 transition configuration is similar to that of S0→S2
and S0→S3, both containing a high HOMO→LUMO+3 compo-
nent. The ΔEST between the S2 and S3 states with the T6 states
is small (ΔEST = 0.03 and ΔEST = 0.15 eV, respectively) and
implies a facile exciton transformation. It is noteworthy that the
S0→S4 state is characterized by a large oscillator strength (f =
0.24) and arises (Figure 4 and Table 2) predominantly from the
HOMO−1→LUMO+1 transition. From Table 3, showing S4→T8
and S4→T9, the ΔEST is very small and can serve as a potential
channel for ISC. Previous investigations have shown that, even
in the absence of heavy atoms, CT states can undergo efficient
ISC, through either radical-pair intersystem crossing[34] (RP-
ISC) for long-lived charge separated states or recombination, to
form local triplet excited states of either the D or A units using
SOCT-ISC.[19] Recently, dyads combining BODIPY as an elec-
tron acceptor and pyrene or perylene as electron donor subu-
nits were shown to display[7] CT states formed as a result of
photoinduced electron transfer and were found to yield triplet
excited states of BODIPY.
In order to decipher further the contribution of the
LE, CT and HLCT states to the overall ISC process in the
TBP⊂ExBox4+ complex, natural transition orbital (NTO) anal-
ysis, based on the singular value decomposition of a 1-particle
transition density matrix, was performed. The NTO analysis
shows a compact representation of the orbital transforma-
tion composition for a given transition. The highest occupied
natural transition orbital (HONTO) and the lowest unoccupied
natural transition orbital (LUNTO) represent the most domi-
nant electronic transition and excitation amplitude obtained by
diagonalizing generalized hole- and particle-density matrices
and by applying additional terms that represent the correlation
effects. The NTOs offer the best possible particle/hole picture
of an excited state and allow unravelling of the dominant role
of the one electronic transition for the generation of the cor-
responding excited state from the ground state (S0).[35] The
HONTOs and LUNTOs of all the hybridized singlet (S2, S3
and S4) and triplet states (T6, T8, T9, and T10) were investigated.
Within the singlet/triplet excited-state pairs that can undergo
exciton transformation (Figure 5c) according to the energy
gap law (|ΔEST|<0.37eV), very similar HONTO and LUNTO
absorption profile with dominant bands at 420 and 387 nm. As
expected, the HOMO is localized on TBP while the LUMO is
localized on ExBox4+ with an ΔEH-L (HOMO-LUMO energy gap)
of 2.74 eV (452 nm) (Table S13, Supporting Information).
The singlet and triplet excited states of the TBP⊂ExBox4+
complex consist of (Figure 5a and Table 3) LE states in the TBP
or ExBox4+ components, CT excited states, and a hybridized
locally charge transfer excited state (HLCT) which is a mixed
state between the LE and CT states.[32] The first CT transition
(Table 3), which is the S0→S1 transition (552 nm, f = 0.0004),
is associated with the HOMO→LUMO+1 (99%) transition. The
energy of the S1 state is consistent with the observation of exci-
plex emission at 520 nm associated with the radiative charge
recombination in the D−A dyad. The S0→S2 transition at 2.99 eV
(415 nm, f = 0.0097) corresponds to the 1HLCT state in involving
26
36
1
minor LE transitions on the TBP (HOMO→LUMO+2, 34%)
and
a
major 1CT component (HOMO→LUMO+3, 63%)
(Table 3). These results are consistent with the weak broad
absorption band observed experimentally at 455 nm. Figure 5a
presents the excited state energy diagram and transition con-
figurations of the singlet (Sn) and triplet (Tn) excited states of
TBP⊂ExBox4+. Previous investigations have revealed[9] that the
S−T transformation is rather facile when the two excited states
contain the same components of transition configurations to
establish the transformation channels in bridging the spin-for-
bidden transitions between two electronic states with different
spin multiplicities. The S0→S1 transition possesses a very weak
oscillator strength and involves only a CT transition from TBP
to ExBox4+. Notably, the lowest triplet state (T1, 1.89 eV) contains
two components namely, the TBP → TBP (HOMO→LUMO+2,
88%) and CT (HOMO→LUMO+3, 5%) relaxation processes,
3
which can be considered essentially as a LE state. Computed
singlet and triplet excited states of only TBP (Figures S31 and
S32, Supporting Information) revealed that the T1 (1.91 eV)
state has similar energies as those found for TBP⊂ExBox4+,
and is significantly lower in energy than the S1 and T2 states,
hampering, therefore, its population either through ISC or
internal conversion (IC) excited-state relaxation mechanisms.
The S0→T3 (2.24 eV) transition is identical to the S0→S1 transi-
tion and is associated with a CT transition in the TBP⊂ExBox4+
host–guest complex. The S−T transformation between the
S1 and T3 states occurs through the SOCT-ISC (Table 2) since
13
the ΔEST = 0.0076 eV (<<0.37 eV). The extent of the HOMO−
LUMO orbital overlap is small (15%), consistent with the
interplanar distances between the TBP and Exbipy2+ units
of ≈3.5 Å and similar to van der Waals radii (3.5 Å) between
carbon atoms. These distances are similar to those reported[22]
from crystal structures of Pyrene⊂ExBox4+. It was previously
©
Adv. Mater. 2020, 2001592
2001592 (8 of 13)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
SM stockpiles, but also in the design of MPEs. Compared to
other oxidants, the reaction of 1O2 with CEES leads to the selec-
tive formation of less toxic CEESO as a major product, while
CEESO2 is formed as a minor product (Figure 6a) in MeOH
solution.[10] In this study, the photocatalytic activity of the supra-
molecular photosensitizer TBP⊂ExBox4+ has been explored in
both homogeneous and heterogeneous media (see the Sup-
porting Information for more details). In order to confirm the
role of exciton transformation in the D−A dyad, we also investi-
gated the photocatalytic performances of the TBP, ExBox4+ and
Na•PSS separately in order to unravel the contribution of each
component to the overall catalytic activity of the composites.
In addition, we explored the photocatalytic activity of the light-
sensitive Ex2.2Box4+ (see the Supporting Information for more
details on its optical properties) in order to emphasize the role
of incorporating tetracationic cyclophanes into a PSS matrix so
as to increase photostability, resulting in enhancement of the
photocatalytic selectivity. All the catalytic results are summa-
rized in Tables 3 and 4.
The photocatalysis of CEES with 1% mol catalyst of
ExBox•4Cl, Ex2.2Box•4Cl or TBP⊂ExBox•4Cl relative to
the molar charge of CEES was carried out (Figure 6b and
Figure S50, Supporting Information) in CD3OD under photo-
irradiation at 395 nm (Figure S45, Supporting Information).
1
The kinetics of CEES conversion were monitored by H NMR
and decoupled 13C NMR. The ExBox•4Cl photocatalyst led
(Figure 6b) to 100% conversion of CEES to CEESO after 16 min
irradiation. The half-life time of the reaction was observed to
be 7 min. The 1H NMR spectrum (Figure S48, Supporting
Information) of CEES contains two triplets centered on 3.7 and
2.9 ppm. After irradiation, these peaks disappear and two multi-
ples appear at 4.0 and 3.3 ppm, corresponding to the chemical
shifts of CEESO (Figures S46 and S47, Supporting Informa-
tion). Formation of this sulfoxide product was also confirmed
(Figure S49, Supporting Information) by 13C NMR spectroscopy.
After 20 min photoirradiation, 1H NMR spectroscopy shows the
formation of 2% CEESO2 based on the appearance of a peak
at 3.57 ppm and negligible amount of vinyl derivatives. X-ray
crystallography revealed (Figure S71, Supporting Information)
that the molecular structure of ExBox4+ remained unchanged,
indicative of its high stability under the photocatalysis condi-
tions. We tested also the photoactivity of the Ex2.2Box•4Cl as a
photosensitizer for the oxidization of CEES; however, we found
that only 45% of CEES was converted in 60 min. It is note-
worthy at Ex2.2Box•4Cl leads to the formation of different major
products, such as CEESO, MeOEES and MeOEESO (Figure 6a
and Figures S50 and S51, Supporting Information). Methanol
can stabilize sulfide-sulfoxide intermediates[36] while CEES can
undergo methanolysis to form MeOEES if exposed to MeOH
for long period of time. The low selectivity (34%) of CEESO
formed after 120 min in the presence of (Table 3) Ex2.2Box•4Cl
is associated with the decomposition of Ex2.2Box4+ under
Figure 4. Frontier orbitals of the TBP⊂ExBox4+ complex calculated from
the APFD/6-31G(d) geometry-optimized molecular structure. H and L
represent, respectively, HOMOs and LUMOs.
distributions at both singlet and triplet excited states were
observed in TBP⊂ExBox4+ where the D dominates HONTO
and the A determines LUNTO with very small overlap (IS
and IT) between HONTO and LUNTO. The almost identical
HONTO and LUNTO distributions for S0 →S2 (IS = 37%) and
S0 →T6 (IT = 40%) transitions, and the small ΔEST (<0.37 eV)
combined with non-negligible orbital overlap, provide a facile
exciton transformation channel for efficient ISC processes
between S2 and T6. The NTOs of the S3 and T10 excited states
reveal that they have predominantly a LE character associated
with the TBP, with a small contribution from the CT states,
supporting the role of both the Br atom and CT for in the
overall ISC process. The S4→T8 and S4→T9 channels are asso-
ciated with a CT from TBP to the ExBox4+ with a small con-
tribution from the ExBox4+↔ExBox4+ ISC. At higher excitation
energy (>3.4 eV), CT from the p-xylylene unit of the Exbox4+ is
triggered as reported[13] previously. In this context, utilization
of D−A complexes with HLCT character (Figure 5c) can shed
important light on the fundamental S−T exciton transforma-
tion mechanism in host–guest supramolecular organic com-
plexes, stimulating further the research into purely organic
materials capable of facile exciton transformation.
1
photoirradiation as confirmed by UV–vis absorption and H
NMR spectroscopies (Figures S29 and S30, Supporting Infor-
mation). Since TBP is insoluble in MeOH, its photocatalytic
performance in homogeneous media could not be obtained.
The TBP⊂ExBox•4Cl D-A dyad is soluble in MeOH and leads
to significant increase in the CEES oxidation rate (Figure 6b
and Table 3), reaching 100% conversion in less than 10 min
Generation of 1O2 by stable microporous organic photo-
catalysts in both aqueous and organic media provide countless
opportunities, not only for the development of environmen-
tally and economically viable materials for the elimination of
©
2001592 (9 of 13)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2020, 2001592

www.advancedsciencenews.com
www.advmat.de
Figure 5. a) Energy diagrams and transition configurations of singlet (Sn) and triplet (Tn) excited states of the TBP⊂ExBox4+ complex. Feasible excited
states for exciton transformation are highlighted in colors. H and L represent, respectively, HOMOs and LUMOs. b) Schematic representation of the
energy levels of the TBP⊂ExBox4+ complex highlighting the photosensitized singlet oxygen production. c) Calculated HONTO and LUNTO distributions
with population percentages are shown in red and blue colors for the singlet and triplet transitions, respectively. The extents of the orbital overlap in
the singlet (Sn) and triplet (Tn) states for exciton transformation in TBP⊂ExBox4+ are given in purple color.
(t1/2 = 3.5 min). Compared to ExBox4+ alone, inclusion of TBP
significantly enhances the CEES conversion rate. In addi-
tion, 1H and 13C NMR spectra (Figures S55 and S56, Sup-
porting Information) of photo-oxidation processes catalyzed by
TBP⊂ExBox4+•4Cl showed full conversion of CEES to its sul-
foxide form with no detectable toxic sulfone formation.
The design of protective equipment against chemical war-
fare agents such as SM requires the development of efficient
heterogeneous photocatalysts. The fulfilment of this goal
requires taking into account multiple parameters, namely (i)
transparency of the polymer matrix, allowing the photocatalyst
to absorb a maximum of light irradiation, (ii) porosity of the
polymer to allow facile transport of species to and from the
active sites, (iii) photostability of the photosensitizer in relation
to photobleaching, (iv) transition-state stabilization for optimi-
zation of selectivity, and finally (v) insolubility of the different
components so as to avoid chemical leaching. The blending
of a commercially available anionic polymer matrix with cati-
onic cyclophanes leads to the formation of insoluble compos-
ites with relatively large surface areas, a characteristic that can
©
Adv. Mater. 2020, 2001592
2001592 (10 of 13)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
Figure 6. a) Structural formulas of the possible products of the photocatalysis of CEES with 1O2. b) Homogeneous photocatalysis of CEES in CD3OD
using ExBox4+•4Cl, Ex2.2Box•4Cl and the TBP⊂ExBox•4Cl complex. c) Heterogeneous catalysis in liquid phase of CEES with 1% mole ExBox4+•PSS,
Ex2.2Box•PSS and TBP⊂ExBox•PSS composites.
enhance the transport of reactants (3O2, CEES) and products
(1O2 and CEESO) to and from the photocatalytic sites. Het-
erogeneous catalysis has been achieved with 1 mol% catalyst
of ExBox•PSS, Ex2.2Box•PSS, and TBP⊂ExBox•PSS relative to
the molar charge of CEES, under photoirradiation at 395 nm
(Figure 6c). Control reactions have been conducted with PSS
and TBP•PSS composites and TBP using identical amounts of
catalyst (1 mol%) to estimate the contribution of the PSS and
TBP components in the overall photocatalytic performance
(Figure S57, Supporting Information). All data are summarized
in Table 4.
The Na•PSS did not show significant photocatalytic perfor-
mance except for a slight conversion because of the decompo-
sition of CEES to MeOEES in MeOH (Figure 6a and Figure S69,
Supporting Information). The CEES conversion is
significantly slower with the ExBox•PSS composite in het-
erogeneous media (Figure 6c and Table 4) compared to the
photocatalytic performance of the ExBox•4Cl in homogeneous
media. After 60 min of photoirradiation, the conversion of
1
CEES did not exceed 40%. H and 13C NMR spectra revealed
Table 4. Heterogeneous photocatalysis parameters using 1 mol% of
photocatalyst obtained at 395 nm photoexcitation in MeOH.
(Figures S58 and S59, Supporting Information) the selective
formation of CEESO. The slow photoactivity of the ExBox4+ is
consistent with the longer S1 lifetime (≈ 8 ns) in the solid state
than in solution (≈1.4 ns), indicating less-efficient ISC in the
solid state for ExBox4+. As in the case of ExBox•PSS, the TBP-
Na•PSS composite showed (Figures S57 and S68, Supporting
Information) less than 50% conversion of CEES, implying
Catalyst
Total Photoir-
Total
Time to full Half-life % Selectivity
radiation conversion conversion t1/2 [min]
(CEESO)
time [min]
[%]
50
[min]
ExBox•PSS
Ex2,2Box•PSS
TBP⊂ExBox•PSS
TBP•PSS
60
60
25
60
60
60
/
54
20
/
60
18
5
>99
>99
>99
97
1
100
100
60
the low O2 generation. This observation is consistent with
DFT calculations which have shown that the T1 state of TBP
cannot be populated efficiently through an ISC or IC mecha-
nism because of the large energy barriers (Figure S32, Sup-
porting Information). Surprisingly, the Ex2.2Box•PSS com-
posite showed a catalytic conversion in heterogeneous media
52
60
/
TBP
50
/
58
Na•PSS
10
/
0
©
2001592 (11 of 13)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2020, 2001592

www.advancedsciencenews.com
www.advmat.de
of 100% in 60 min and t1/2 = 18 min, whilst in homogeneous
media, the conversion did not exceed 45% within a similar
photoirradiation timeframe. More importantly, the formation
of CEESO was fully selective and 13C NMR spectroscopy did
not show (Figure S60, Supporting Information) the appear-
ance of MeOEES, as observed in homogeneous catalysis as a
result of the enhanced stability of the Ex2.2Box4+ cyclophane
within the PSS matrix. Previous investigations[37] have shown
an increase in the stability of air-sensitive S/N radicals when
incorporated into polymers matrices. The TBP⊂ExBox•PSS
composite has registered a 12-fold increase in the kinetics of
conversion of CEES to CEESO compared to ExBox•PSS and
TBP-Na•PSS at λex = 395 nm (Figure 6c and Figure S57, Sup-
porting Information). Within 20 min of photoirradiation, the
conversion reached 100% with t1/2 = 5 min and 100% selec-
tivity for the formation (Table 4) of CEESO. These results are
consistent with DFT calculations and spectroscopic investiga-
tions showing that excitation of the HLCT excited states trig-
gers an efficient S−T transformation that leads to population
of the T1 state. The energy of the T1 state (1.89 eV) is close
singlet excited state. Other cyclophanes, such as Ex2.2Box•4Cl,
are not stable under photoirradiation, and the photocatalysis
of CEES is neither fast nor selective. Notably, Ex2.2Box•PSS
is stable under photoirradiation and the conversion of CEES
to CEESO is 100% selective. Although the lowest triplet state
(T1) of 1,3,5,8-tetrabropyrene is low in energy, it is inaccessible
on account of the large energy barrier separating the T1 states
from the S1 and T2 states. The efficiency of the singlet to triplet
(S−T) transformation in the TBP⊂ExBox4+ host–guest complex
is associated with a combination of both a large SOC of the Br
atoms and SOCT-ISC of the D−A dyad. In addition, DFT calcu-
lations revealed the existence of a manifold of excited states in
the TBP⊂ExBox4+ complex that can enhance the IC relaxation
mechanism from the upper triplet states to populate the low-
lying T1 excited state. The efficient S−T transformation and IC
relaxation mechansims play a central role in the enhancement
of 1O2 generation and subsequent increase in photocatalytic
performance. The high stability, facile preparation, processa-
bility and high performance of the TBP⊂ExBox•PSS composite
augur well for the future development of supramolecular het-
erogeneous photosensitizers using host–guest chemistry. More
broadly, these results reveal a number of other opportunities
for facile fine-tuning of the S−T transformation in D−A dyads
using host–guest chemistry which can unleash fundamental
and technological advances for the future design of triplet
excited-state chromophores.
1
to that of the first excited state of O2 (1.63 eV),[20] offering
(Figure 5b), therefore, an energy gap of ≈ 0.26 eV which is
ideal for efficient Dexter energy transfer.[18,20] Notably, photo-
catalysis of CEES at λex = 450 nm (Figures S63 and S64,
Supporting Information) is very slow, confirming the weak
population of the T1 states upon excitation of the low-energy
1CT states. The catalytic performance of TBP⊂ExBox•PSS at
395 nm is comparable (Table S14, Supporting Information)
with that of other MOF photocatalysts, such as NU-1000,[38]
UMCM-313[38] and Al-PMOF-fiber.[39] Photocatalysis of CEES
or SM using porous organic polymers is less common.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
Recently,
a
carbazole-based conjugated microporous
polymer was reported,[40] which performs a selective photo-
catalysis of CEES over a period of 60 min using 8 mole %
of the photocatalyst. To the best of our knowledge, the
TBP⊂ExBox•PSS composite is among the most efficient
organic polymer photocatalysts that has been tested for the
elimination of Chemical Warfare Agent simulants. In order
to confirm the stability of the TBP⊂ExBox•PSS photocatalyst,
we performed a leaching test which consists (Figure S70,
Supporting Information) of the removal of the catalyst upon
reaching 50% CEES conversion. Further irradiation of the
solution showed no significant change in the concentration
of CEES, indicating that (i) the TBP⊂ExBox•PSS composite is
responsible for the transformation of CEES to CEESO under
photoirradiation and (ii) the composite is stable under the
photocatalytic conditions because of the absence of a chromo-
phores in solution. Further photoirradiation does not lead to
a change in CEES conversion confirming the stability of the
supramolecular TBP⊂ExBox•PSS photocatalyst.
Acknowledgements
Y.B. and A.A. contributed equally to this work. The authors thank
Northwestern University (NU) for their generous support of this
research. The authors thank the personnel in the Integrated Molecular
Structure Education and Research Center (IMSERC) at NU for their
assistance in the collection of the data. This project was also supported
by the Defense Threat Reduction Agency under grant number HDTRA1-
19-1-0010 (J.T.H.). P.D. thanks Southern Illinois University - Carbondale
(SIUC) for start up support. It was also supported in the Wasielewski
laboratory by the National Science Foundation under grant number
DMR-1710104 (M.R.W.).
Conflict of Interest
The authors declare no conflict of interest.
In conclusion, supramolecular porous organic composites
based on tetracationic cyclophanes (ExBox4+ and Ex2.2Box4+) and
an anionic polymer matrix such as polystyrene sulfonate have
been prepared. These materials were found to be microporous
as evidenced by CO2 adsorption isotherms. In addition, larger
molecules such as TTF can diffuse in/out of the ExBox•PSS
composite. While the photocatalysis of CEES by ExBox•4Cl in
solution is fast and selective, in the solid state the conversion of
the CEES to CEESO is very slow as a result of stabilization of the
Keywords
cyclophanes, nanocomposites, photocatalysis, photosensitizers,
polymers
Received: March 5, 2020
Revised: June 1, 2020
Published online:
©
Adv. Mater. 2020, 2001592
2001592 (12 of 13)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
[1] S. B. Brown, E. A. Brown, I. Walker, Lancet Oncol. 2004, 5, 497.
[2] R. C. Evans, P. Douglas, C. J. Winscom, Coord. Chem. Rev. 2006,
250, 2093.
[22] J. C. Barnes, M. Juríček, N. L. Strutt, M. Frasconi, S. Sampath,
M. A. Giesener, P. L. McGrier, C. J. Bruns, C. L. Stern, A. A. Sarjeant,
J. F. Stoddart, J. Am. Chem. Soc. 2013, 135, 183.
[3] D. M. Arias-Rotondoa, J. K. McCusker, Chem. Soc. Rev. 2016, 45,
5803.
[23] K. C. Kim, T.-U. Yoon, Y.-S. Bae, Microporous Mesoporous Mater.
2016, 224, 294.
[4] X.-D. Wang, O. S. Wolfbeis, Chem. Soc. Rev. 2014, 43, 3666.
[5] J. Zhou, Q. Liu, W. Feng, Y. Sun, F. Li, Chem. Rev. 2015, 115, 395.
[6] a) M. A. El-Sayed, J. Chem. Phys. 1974, 60, 4502; b) C. M. Marian,
Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2012, 2, 187.
[24] a) M. M. Khan, G. Bengtson, S. Neumann, M. Rahman, V. Abetz,
V. Filiz, RSC Adv. 2014, 4, 32148; b) P. M. Budd, B. S. Ghanem,
S. Makhseed, N. B. McKeown, K. J. Msayiba, C. E. Tattershalla,
Chem. Commun. 2004, 230.
[7] M. A. Filatov, S. Karuthedath, P. M. Polestshuk, H. Savoie,
K. J. Flanagan, C. Sy, E. Sitte, M. Telitchko, F. Laquai, R. W. Boyle,
M. O. Senge, J. Am. Chem. Soc. 2017, 139, 6282.
[8] M. A. Filatov, S. Karuthedath, P. M. Polestshuk, S. Callaghan,
K. J. Flanagan, T. Wiesner, F. Laquai, M. O. Senge, ChemPhotoChem
2018, 2, 606.
[9] R. Chen, Y. Tang, Y. Wan, T. Chen, C. Zheng, Y. Qi, Y. Cheng,
W. Huang, Sci. Rep. 2017, 7, 6225.
[10] a) Y. Liu, A. J. Howarth, J. T. Hupp, O. K. Farha, Angew. Chem.,
Int. Ed. 2015, 54, 9001; b) F. Jensen, A. Greer, E. L. Clennan, J. Am.
Chem. Soc. 1998, 120, 4439.
[11] a) A. Atilgan, T. Islamoglu, A. J. Howarth, J. T. Hupp,
O. K. Farha, ACS Appl. Mater. Interfaces 2017, 9, 24555; b) H. Wang,
G. W. Wagner, A. X. Lu, D. L. Nguyen, J. H. Buchanan, P. M. McNutt,
C. J. Karwacki, ACS Appl. Mater. Interfaces 2018, 10, 18771;
c) A. J. Howarth, C. T. Buru, Y. Liu, A. M. Ploskonka, K. J. Hartlieb,
M. McEntee, J. J. Mahle, J. H. Buchanan, E. M. Durke, S. S. Al-Juaid,
J. F. Stoddart, J. B. DeCoste, J. T. Hupp, O. K. Farha, Chem. - Eur.
J. 2017, 23, 214; d) C. F. Pereira, Y. Liu, A. Howarth, F. Figueira,
J. Rocha, J. T. Hupp, O. K. Farha, J. P. C. Tomé, F. A. A. Paz, ACS
Appl. Nano Mater. 2019, 2, 465; e) Y. Liu, C. T. Buru, A. J. Howarth,
J. J. Mahle, J. H. Buchanan, J. B. DeCoste, J. T. Hupp, O. K. Farha,
J. Mater. Chem. A 2016, 4, 13809.
[25] a) M. M. Cetin, Y. Beldjoudi, I. Roy, O. Anamimoghadam, Y.-J Bae,
R. M. Young, M. D. Krzyaniak, C. L. Stern, D. Philp, F. M. Alsubaie,
M. R. Wasielewski, J. F. Stoddart, J. Am. Chem. Soc. 2019, 141, 18727;
b) K. Cai, M. C. Lipke, Z. Liu, J. Nelson, T. Cheng, Y. Shi, C. Cheng,
D. Shen, J.-M. Han, S. Vemuri, Y. Feng, C. L. Stern, W. A. Goddard III,
M. R. Wasielewski, J. F. Stoddart, Nat. Commun. 2018, 9, 5275.
[26] a) Y. Luan, N. Zheng, Y. Qi, J. Tang, G. Wang, Catal. Sci. Technol.
2014, 4, 925; b) W. Zhang, B. Zheng, W. Shi, X. Chen, Z. Xu, S. Li,
Y. R. Chi, Y. Yang, J. Lu, W. Huang, F. Huo, Adv. Mater. 2018, 30,
1800643; c) J. Duan, S. Chen, C. Zhao, Nat. Commun. 2017, 8,
15341; d) S. Zhao, Y. Wang, J. Dong, C.-T. He, H. Yin, P. An, K. Zhao,
X. Zhang, C. Gao, L. Zhang, J. Lv, J. Wang, J. Zhang, A. M. Khattak,
N. A. Khan, Z. Wei, J. Zhang, S. Liu, H. Zhao, Z. Tang, Nat. Energy
2016, 1, 16184.
[27] a) R. A. Marcus, J. Chem. Phys. 1956, 24, 966; b) P. Vath,
M. B. Zimmt, J. Phys. Chem. A 2000, 104, 2626; c) J. T. Hupp,
Y. Dong, R. L. Blackbourn, H. Lu, J. Phys. Chem. 1993, 97, 3278.
[28] Y. Wu, J. Zhou, B. T. Phelan, C. M. Mauck, J. F. Stoddart,
R. M. Young, M. R. Wasielewski, J. Am. Chem. Soc. 2017, 139, 14265.
[29] a) Z. E. X. Dance, S. M. Mickley, T. M. Wilson, A. B. Ricks,
A. M. Scott, M. A. Ratner, M. R. Wasielewski, J. Phys. Chem. A 2008,
112, 4194; b) C. V. Suneesh, K. R. Gopidas, J. Phys. Chem. C 2010, 114,
18725; c) Y. Dong, A. A. Sukhanov, J. Zhao, A. Elmali, X. Li, B. Dick,
A. Karatay, V. K. Voronkova, J. Phys. Chem. C 2019, 123, 22793.
[30] J. R. Lakowicz, Anal. Biochem. 2001, 298, 1.
[12] H. Wang, G. W. Wagner, A. X. Lu, D. L. Nguyen, J. H. Buchanan,
P. M. McNutt, C. J. Karwacki, ACS Appl. Mater. Interfaces 2018, 10,
18771.
[31] A. Austin, G. A. Petersson, M. J. Frisch, F. J. Dobek, G. Scalmani,
K. Throssell, J. Chem. Theory Comput. 2012, 8, 4989.
[13] a) R. M. Young, S. M. Dyar, J. C. Barnes, M. Juríček, J. F. Stoddart,
D. T. Co, M. R. Wasielewski, J. Phys. Chem. A 2013, 117, 12438;
b) I. Roy, S. Bobbala, R. M. Young, Y. Beldjoudi, M. T. Nguyen,
M. M. Cetin, J. A. Cooper, S. Allen, O. Anamimoghadam, E. A. Scott,
M. R. Wasielewski, J. F. Stoddart, J. Am. Chem. Soc. 2019, 141, 12296.
[14] S. M. Dyar, J. C. Barnes, M. Juríček, J. F. Stoddart, D. T. Co,
R. M. Young, M. R. Wasielewski, Angew. Chem., Int. Ed. 2014, 53, 5371.
[15] R. M. Young, S. C. Jensen, K. Edme, Y. Wu, M. D. Krzyaniak,
N. A. Vermeulen, E. J. Dale, J. F. Stoddart, E. A. Weiss,
M. R. Wasielewski, D. T. Co, J. Am. Chem. Soc. 2016, 138, 6163.
[16] S. T. J. Ryan, J. Del Barrio, I. Ghosh, F. Biedermann, A. I. Lazar,
Y. Lan, R. J. Coulston, W. M. Nau, O. A. Scherman, J. Am. Chem.
Soc. 2014, 136, 9053.
[32] M. J Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson,
H. Nakatsuji, X. Li, M. Caricato, A. V. Marenich, J. Bloino,
B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian,
J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, W. F. Ding,
F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson,
D. Ranasinghe, V. G. Zakrzewski, et al., Gaussian 16 Rev. C.01,
Wallingford, CT 2016.
[33] Y. Y. Pan, J. Huang, Z. M. Wang, S. T. Zhang, D. W. Yu, B. Yang,
Y. G. Ma, RSC Adv. 2016, 6, 108404.
[34] K. Hasharoni, H. Levanon, S. R. Greenfield, D. J. Gosztola,
W. A. Svec, M. R. Wasielewski, J. Am. Chem. Soc. 1996, 118, 10228.
[35] R. L. Martin, J. Chem. Phys. 2003, 118, 4775.
[36] a) J. J. Liang, C. L. Gu, M. L. Kacher, C. S. Foote, J. Am. Chem. Soc.
1983, 105, 4717; b) R. Limvorapitux, H. Chen, M. L. Mendonca,
M. Liu, R. Q. Snurr, S. T. Nguyen, Catal. Sci. Technol. 2019, 9, 327.
[37] Y. Beldjoudi, I. Osorio-Román, M. A. Nascimento, J. M. Rawson,
J. Mater. Chem. C 2017, 5, 2794.
[38] C. T. Buru, Cassandra, T. Buru, M. B. Majewski, A. J. Howarth,
R. H. Lavroff, C.-W. Kung, A. W. Peters, S. Goswami, O. K. Farha,
ACS Appl. Mater. Interfaces 2018, 10, 23802.
[17] S. T. J. Ryan, R. M. Young, J. J. Henkelis, N. Hafezi, N. A. Vermeulen,
A. Hennig, E. J. Dale, Y. Wu, M. D. Krzyaniak, A. Fox, W. M. Nau,
M. R. Wasielewski, J. F. Stoddart, O. A. Scherman, J. Am. Chem. Soc.
2015, 137, 15299.
[18] D. L. Dexter, J. Chem. Phys. 1953, 21, 836.
[19] a) Y. Hou, X. Zhang, K. Chen, D. Liu, Z. Wang, Q. Liu, J. Zhao,
A. Barbon, J. Mater. Chem. C 2019, 7, 12048; b) Z. Wang, M. Ivanov,
Y. Gao, L. Bussotti, P. Foggi, H. Zhang, N. Russo, B. Dick, J. Zhao,
M. Di Donato, G. Mazzone, L. Luo, M. Fedin, Chem. Eur. J. 2020,
26, 1091.
[39] D. T. Lee, J. D. Jamir, G. W. Peterson, G. N. Parsons, Matter 2020,
2, 404.
[20] D. Kovalev, M. Fujii, Adv. Mater. 2005, 17, 2531.
[21] P. R. Symmers, M. J. Burke, D. P. August, P. I. T. Thomson, G. S. Nichol,
M. R. Warren, C. J. Campbell, P. J. Lusby, Chem. Sci. 2015, 6, 756.
[40] Y. Zhi, Z. Yao, W. Jiang, H. Xia, Z. Shi, Y. M, X. Liu, ACS Appl. Mater.
Interfaces 2019, 11, 37578.
©
2001592 (13 of 13)
2020 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Mater. 2020, 2001592