Guest Inclusion Modulates Concentration and Persistence of Photogenerated Radicals in Assembled Triphenylamine Macrocycles

Article abstract of DOI:10.1021/jacs.9b11518

Substituted triphenylamine (TPA) radical cations show great potential as oxidants and as spin-containing units in polymer magnets. Their properties can be further tuned by supramolecular assembly. Here, we examine how the properties of photogenerated radical cations, intrinsic to TPA macrocycles, are altered upon their self-assembly into one-dimensional columns. These macrocycles consist of two TPAs and two methylene ureas, which drive the assembly into porous organic materials. Advantageously, upon activation the crystals can undergo guest exchange in a single-crystal-to-single-crystal transformation generating a series of isoskeletal host-guest complexes whose properties can be directly compared. Photoinduced electron transfer, initiated using 365 nm light-emitting diodes, affords radicals at room temperature as observed by electron paramagnetic resonance (EPR) spectroscopy. The line shape of the EPR spectra and the quantity of radicals can be modulated by both polarity and heavy atom inclusion of the encapsulated guest. These photogenerated radicals are persistent, with half-lives between 1 and 7 d and display no degradation upon radical decay. Re-irradiation of the samples can restore the radical concentration back to a similar maximum concentration, a feature that is reproducible over several cycles. EPR simulations of a representative spectrum indicate two species, one containing two N hyperfine interactions and an additional broad signal with no resolvable hyperfine interaction. Intriguingly, TPA analogues without bromine substitution also exhibit similar quantities of photogenerated radicals, suggesting that supramolecular strategies can enable more flexibility in stable TPA radical structures. These studies will help guide the development of new photoactive materials.

Full text of DOI:10.1021/jacs.9b11518

Subscriber access provided by EDINBURGH UNIVERSITY LIBRARY | @ http://www.lib.ed.ac.uk
Article
Guest Inclusion Modulates Concentration and Persistence of
Photogenerated Radicals in Assembled Triphenylamine Macrocycles.
Ammon J. Sindt, Baillie Ann DeHaven, Dustin W. Goodlett, Johannes O. Hartel,
Pooja J. Ayare, Yong Du, Mark D. Smith, Anil K. Mehta, Alexander M. Brugh,
Malcolm D. E. Forbes, Clifford R. Bowers, Aaron K. Vannucci, and Linda S. Shimizu
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b11518 • Publication Date (Web): 08 Dec 2019
Downloaded from pubs.acs.org on December 9, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 11
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Guest Inclusion Modulates Concentration and Persistence of Photo-
generated Radicals in Assembled Triphenylamine Macrocycles.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
†
†,
†
†
†
⊥
Ammon J. Sindt, Baillie A. DeHaven, Dustin W. Goodlett, Johannes O. Hartel, Pooja J. Ayare,
‡
†
§
∥
∥
Yong Du, Mark D. Smith, Anil K. Mehta, Alexander M. Brugh, Malcolm D. E. Forbes, Clifford
R. Bowers,‡ Aaron K. Vannucci,† and Linda S. Shimizu*,†
†Department of Chemistry and Biochemistry, University of South Carolina, Columbia, South Carolina 29208, United
States
‡Department of Chemistry, University of Florida, Gainesville, Florida 32611, United States
§National High Magnetic Field Laboratory and McKnight Brain Institute, University of Florida, Gainesville, Florida
32610, United States
∥
Department of Chemistry and Center for Photochemical Sciences, Bowling Green State University, Bowling Green,
Ohio 43403, United States
ABSTRACT: Substituted triphenylamine (TPA) radical cations show great potential as oxidants and as spin containing units
in polymer magnets. Their properties can be further tuned by supramolecular assembly. Here, we examine how the prop-
erties of photogenerated radical cations, intrinsic to TPA macrocycles, are altered upon their self-assembly into 1D columns.
These macrocycles consist of two TPAs and two methylene ureas which drive the assembly into porous organic materials.
Advantageously, upon activation the crystals can undergo guest exchange in a single-crystal-to-single-crystal transfor-
mation generating a series of isoskeletal host-guest complexes whose properties can be directly compared. Photoinduced
electron transfer, initiated using 365 nm LED’s, affords radicals at room temperature as observed by EPR spectroscopy. The
line shape of the EPR spectra and the quantity of radicals can be modulated by both polarity and heavy atom inclusion of
the encapsulated guest. These photogenerated radicals are persistent, with half-lives between 1-7 days and display no deg-
radation upon radical decay. Re-irradiation of the samples can restore the radical concentration back to a similar maximum
concentration, a feature that is reproducible over several cycles. EPR simulations of a representative spectrum indicate two
species, one containing two N hyperfine interactions and an additional broad signal with no resolvable hyperfine interac-
tion. Intriguingly, TPA analogs without bromine substitution also exhibit similar quantities of photogenerated radicals,
suggesting that supramolecular strategies can enable more flexibility in stable TPA radical structures. These studies will
help guide the development of new photoactive materials.
INTRODUCTION
Construction of hierarchical materials through supra-
molecular assembly of small molecules is an expedient
method for crafting materials with useful properties.1-4
These properties range from conductivity,5 magnetism,6 to
dichromism.7 In addition, supramolecular assembly can
also be used to make porous materials that encapsulate
small guests.8,9 This leads to materials of use in catalysis,10,11
storage,12,13 confinement,14 separation,15,16 and sensing.17,18
Furthermore, bound guests may alter the chemical and
physical properties of the host itself, for example, modulat-
ing the rotational speed of host-bound molecular rotors19
or expanding/contracting the host framework.20,21 Here, we
investigate how different guests encapsulated within a tri-
phenylamine (TPA) host affect its ability to form radicals
in the solid-state upon UV irradiation (Figure 1). The host
is robust and exchanges guests via single-crystal-to-single-
crystal (SC-SC) transformations while retaining its original
Figure 1. Self-assembly of TPA macrocycles results in the formation of a
columnar assembled host. Activation of this host by heating allows for the
introduction of new guests via SC-SC transformations. Each complex gen-
erates radicals upon irradiation with 365 nm LEDs, affording EPR spectra
with different line shapes and intensities. A comparison of four of these
EPR is shown above (benzene, 1,4-dioxanes, DME, and DMF). The insets
for each graph show structure of the guest, SC-XRD of the host-guest com-
plex, and the percent of molecules that generate a radical upon UV-
irradiation at the maximum radical concentration compared to the total
number of molecules in the bulk material.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 11
framework. This affords a series of isoskeletal host-guest assembled TPA systems that allow for stable radical for-
1
2
3
4
5
6
7
8
structures whose properties can be directly compared.
Upon UV irradiation of the host, long-lived radicals are
generated. Encapsulated guests within the host modulate
the line shape of their EPR spectra and the concentration
of radicals observed during irradiation. Additionally, after
decay of the photogenerated radicals, samples can be re-
irradiated to regenerate the observed radicals without
harming or degrading the TPA host. Understanding how
guest inclusion affects these radical properties will be use-
ful for designing the next generation of conductive and
magnetic materials.
mation. Our goal is to understand the effects different
guests have on radical formation and to determine what
characteristics are necessary for TPA-containing com-
pounds to exhibit these properties.
RESULTS AND DISCUSSION
Macrocycles (1 and 1a) and linear analogs (2 and 2a) were
synthesized in five steps following established methods
(Scheme S1).39,40 For macrocycle 1, its dibromide and pro-
tected macrocycle precursor were structurally character-
ized (see SI). Control compounds 3 and 3a were purchased
from commercial suppliers, while control 4 was synthe-
sized in two steps using a Vilsmeier–Haack reaction to
yield the aldehyde,41 followed by a reductive alkylation of
urea to give the desired product.42 Host-guest crystals of 1
and 1a were obtained from vapor diffusion of either water
or 1,2-dimethoxyethane (DME), respectively, into DMSO
solutions. Solvent free crystals for 2-4 were grown via slow
evaporation of acetonitrile (2 and 3a), ethyl acetate (2a), or
ethanol (3); or by vapor diffusion of water into a DMF so-
lution (4).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Organic radicals have many uses, ranging from MRI con-
trasting agents22 to multifunctional magnetic materials.23
One class of these radicals are TPA radical cations, which
have undergone a single electron oxidation from their neu-
tral state.24 Key to their stability is the presence of para sub-
stitution on all three phenyl rings of the TPA system, which
slows degradation reactions such as benzidine for-
mation.25,26 This is seen for Magic Blue, a commercial one-
electron oxidant,27,28 as it has bromine substitution on all
three para sites of the TPA. When stabilized, TPAs find use
as high-spin polymers,29,30 cathodes in batteries,31,32 and as
hole-transport layers in solar cells.33 Typically, these radi-
cals are formed through chemical or electrochemical oxi-
dation pathways,34 but they may also be generated by UV-
irradiation when an electron acceptor is present.35,36
Our group employs the three-centered urea hydrogen
bonding interaction to organize linear and macrocyclic
compounds in high fidelity. This affords nanoporous ma-
terials that are typically used as nanoreactors for photo-
chemical reactions.37 In addition to forming nanocham-
bers, these structures also exhibit markedly different pho-
tophysical properties when compared to their unassem-
bled monomers in solution. For example, assembly of ben-
zophenone containing urea macrocycles and linear analogs
display surprisingly stable radical formation upon UV-
irradiation; this behavior is not observed in solution.38 Sim-
ilarly, brominated TPA 2a (Figure 2) also shows enhanced
stability of photogenerated radicals within its assembled
structure, while radical formation in solution results in
complete degradation of the TPAs.39 Thus, demonstrating
the importance of supramolecular assembly on the overall
stability of the photogenerated radicals. Here, we investi-
gate if TPA containing macrocycle system, 1a, also exhibits
enhanced photogenerated radical stability upon self-as-
sembly.
Figure 2. Comparison of TPA structures investigated. Macrocycles 1 and
1a, linear analogs 2 and 2a, and control compounds 3, 3a, and 4 are shown.
For TPA 1, the X-ray structure revealed the desired mac-
rocycle in the space group P21/n of the monoclinic system.
The macrocycles were found in the syn conformation orga-
nized into tubular columns encapsulating disordered
DMSO in a 1:1 host-guest ratio. The syn conformation was
surprising as previous bis-urea macrocycles crystallize in
the anti-conformation, where the urea groups in a single
macrocycle are oppositely aligned to minimize dipole in-
teractions.37 Despite this, the macrocycles still organize
into columns through the characteristic three-centered
urea hydrogen bond (d(N···O) = 3.090(5), 3.078(5), 3.147(5),
and 3.063(5) Å). Additional stabilization occurs through
intracolumnar edge-to-face π-stacking of the TPA groups
(Figure S22). This stacking affords individual nanotubes
with a pore aperture of 6.4  4.3 Å after accounting for the
van der Waals (vdW) radii of the participating atoms (Fig-
ure 3A, top). The resulting nanotubes pack together to
form robust hexagonal arrays as seen in Figure S16.
TPA 1a self-assembles to form a porous host that can fa-
cilitate guest inclusion and removal via SC-SC crystal
transformations.40 By irradiating this host we can probe if
macrocycle 1a exhibits similar radical formation as its lin-
ear analog counterpart 2a and if guest inclusion alters the
process of radical generation within the host. Thus, the
photophysical properties and photogenerated radical for-
mation of this host were investigated. The guests in these
studies were selected to cover a range of polarities as well
as heavy atom inclusion. Control compounds 1, 2, and 3-4
were prepared to elucidate the structure of the photogen-
erated radicals and to identify the features of these
For comparison, the brominated macrocycle, 1a, crystal-
lizes as colorless needles in the space group P21/c of the
monoclinic system.40 This macrocycle adopts the typical
anti conformation and organizes into columnar tubes. The
channels display an interior cross-section diameter of 6.5 
4.3 Å and contain encapsulated, disordered DME (Figure
3B, right) in a host-guest ratio of 2:1. The tubes form along
the crystallographic b axis and are held together via
ACS Paragon Plus Environment

Page 3 of 11
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
Figure 3. Comparison of TPA hosts 1·DMSO and 1a·DME. (A) Compar-
ison of cross-sectional areas with 1 on top and 1a on bottom (subtracting
vdW radii). (B) Comparison of columnar structures with guests included
with 1·DMSO on the left and 1a·DME. Disorder in the guests has been
omitted for clarity.
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Guest inclusion complexes of 1a with their host-guest ratios. (A)
Heating of 1a·DME results in an activated host. (B) Guests can then be
added upon submersion of the activated crystals into a liquid of the new
desired guest, resulting in SC-SC transformation to afford a new host-guest
complex. The DME complex, activated host, and four benzene derivative
guest complexes have been previously reported,40 while the DMF and 1,4-
dioxane structures are new to these studies.
hydrogen bonds between the ureas (d(N···O) = 2.848(4)
and 2.929(4) Å) and π-stacking between neighboring TPAs.
Individual columns are held together with π-π and halo-
gen-π stacking interactions forming a hexagonal array sim-
ilar to 1·DMSO.
to-face π-stacking from the TPAs holding alternating
chains together (Figure S23).
To generate different host-guest complexes of host 1a,
crystals of 1a·DME were activated and loaded with new
guests via SC-SC transformations.40 Crystals of 1a·DME
were activated by heating under vacuum at 90 °C for 2.5 h
to remove the DME. Next, a series of host-guest complexes
(Figure 4) were prepared by immersing the activated crys-
tals in a guest solution for 24 h. Afterwards, the crystals
were filtered and air dried affording the new host-guest
complex. The activated host and benzene derivative loaded
hosts of 1a have been previously characterized via SC-XRD
and are all isoskeletal in regard to the macrocycle frame-
work of 1·DME. New to this work, DMF and 1,4-dioxanes
were also loaded into the nanotubes. Again, the host
framework remained isoskeletal.
In the benzene derivative loaded hosts of 1a, (Figure 4),
the guests were arranged in a planar tape-like manner in-
side the channels and were modeled on two crystallo-
graphic independent sites.40 These sites were located near
inversion centers giving four possible sites for guest loca-
tion. Each of these sites had similar occupancy giving an
approximate host-guest ratio of 1:0.5. The 1,4-dioxane and
DMF-loaded 1a complexes exhibited a very similar guest
disorder, although 1,4-dioxane was modeled on three inde-
pendent crystallographic sites instead of two. Overall, the
1,4-dioxanes and DMF guests exhibited host-guest ratios of
1:0.58(2) and 1:0.65(1), respectively. These loading ratios
were higher compared to all the other modeled complexes,
likely a result of the smaller size of these guests.
Figure 5. Comparison of triphenylamine linear analogs 2 (left) and 2a
(right). (A) Three-center urea hydrogen bonding interactions stack the tri-
phenyl amine groups on top of one another forming urea tape motifs. (B)
Top-down view of the urea tapes showing either a cruciform pattern (left)
or X-shaped pattern (right).
The brominated derivative 2a was similarly organized in
the orthorhombic system in the Pccn space group.39 TPA
2a was also found in a linear trans-trans arrangement with
both TPA groups extended out on either side of the meth-
ylene urea tether (Figure 5). The TPA units show minor dis-
order in this structure (5% to 9% depending on the chosen
crystal) resulting from an opposite TPA rotation relative to
the urea group. Overall, the urea groups still organize 2a
into chains along the crystallo-graphic c axis (d(N···O) =
2.823(3) and 2.70(2) Å), resulting in an X-shape pattern in
projection along the chain direction instead of the cruci-
form pattern observed for TPA 2.
For comparison to the macrocycles, linear analog 2 crys-
tallized in the orthorhombic system in the centrosymmet-
ric space group Pbcn. In this structure, both TPA groups
extend to either side of a disordered methylene urea tether
in a linear trans-trans arrangement (Figure 5). The disor-
der in the urea tether is oriented either up (50%) or down
(50%) relative to the c axis, forming three-centered urea
hydrogen bonded chains (d(N···O) = 2.749(8) and 2.716(8)
Å). The TPA groups, which are not disordered, organize in
a cruciform pattern in relation to the urea chain with edge-
Triphenylamines (3 and 3a) and urea derivative 4 were
also crystallized. Together with the linear analogs and
macrocycle 1, this provides a series of control compounds
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 11
for photophysical studies of macrocycle 1a. The structure
aggregation in planar dyes.44 However, Yang et al. observed
such shifts in a crystalline styrene derivatized TPAs45 and
we also observed a similar shift for the related linear analog
2a.39 Similar absorption spectra were also observed for
1a·DME, 1a·C6H6, and 1a·DMF. These host-guest complexes
were chosen because of their polarity differences since po-
larity played a large role in determining the photophysical
properties of linear analog 2a in solution. Interestingly, the
non-brominated TPAs of macrocycle 1 and linear analog 2
also exhibited a red-shift in absorption, although their
broad band at ~400 nm is comparatively more intense.
1
2
3
4
5
6
7
8
of TPA 3 has been previously reported (CCDC# 1319035),
and it crystallized in the monoclinic system in the Bb space
group with multiple edge-to-face π-stacking interactions
driving the assembly.43 For TPA 3a, two different poly-
morphs were found, crystallizing either as colorless nee-
dles (triclinic system, P-1 (No.2) space group) or as color-
less blocks (monoclinic system, P21/c space group). The
packing in both polymorphs was primarily driven through
edge-to-face π stacking interactions (Figures S24 and S25);
however, the bromines stacked much closer to each other
in the monoclinic polymorph (4.473 Å vs 6.230 Å). Accord-
ing to powder X-ray diffraction results, the bulk sample
most similarly matched the monoclinic polymorph (Figure
S30). Lastly, TPA 4 crystallized in the monoclinic space
group P21/c as colorless flakes. Assembly was organized
through the three-centered urea hydrogen bond forming
twisted chains along the crystallographic c axis similar to
2a (d(N···O) = 3.069(3) and 3.028(2) Å) with additional sta-
bilization coming via edge-to-face π stacking interactions
(Figure S26).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Assembly had less influence on the emission spectra and
photoluminescent lifetimes for macrocycle 1a. Indeed, ac-
tivated host 1a and its complexes (1a·DME, 1a·C6H6, and
1a·DMF) exhibited similar λmax of emissions at ~460 nm
upon excitation at the λmax of absorption with similar life-
times of ~1.2 ns. For comparison, activated 1, 2, and 2a39
show similar behavior with the non-halogen containing
compounds exhibiting marginally longer lifetimes (1.5 ns)
likely due to the lack of a heavy atom. Our hypothesis is
that these shortened lifetimes are a result of non-radiative
pathways which may include radical formation.
To probe how structure, self-assembly, and guest encap-
sulation effects the photophysics of macrocycle 1a, the ab-
sorption, emission, and photoluminescence lifetimes for
solution and solid-state samples were measured. Table 1
compares the photophysics of macrocycle 1a and linear an-
alog 2a as 10 M solutions in nitrogen purged DMSO to the
assembled crystals of 1-2a and some guest inclusion com-
plexes of host 1a. In solution, macrocycle 1a exhibits only
one band in its absorption spectra with a λmax at 303 nm and
molar absorptivity of ~5  104 M-1 cm-1 which is similar to its
linear analog counterpart 2a under comparable condi-
tions.39 In the emission spectra, macrocycle 1a shows two
bands, with a smaller one at 378 nm and the larger at 450
nm, also quite similar to linear analog 2a. The photolumi-
nescence lifetime of macrocycle 1a was quite short and ap-
proximately half of what was observed for linear analog 2a
(2.3 ns versus 4.1 ns).
Previously, we reported that UV-irradiation of linear an-
alog 2a generated radicals, which were unstable in solution
resulting in degradation of the material.39 Intriguingly, the
radicals were found to be stable and persistent when gen-
erated within the assembled structures. Thus, we irradi-
ated the activated host of 1a to see if similar radicals would
be observed. For this, freshly activated 1a was sealed under
Argon and was examined by X-band EPR spectroscopy pre
and post irradiation. For UV-irradiation, 365 nm LEDs
were employed instead of the medium pressure Hg lamp
used previously.38,39 The 365 nm LEDs are close to the λmax
of absorbance for these materials and show significant re-
duction in the background signal (see SI). Figure 6A com-
pares the EPR signal of 1a after 4 h of irradiation versus the
pre-UV sample. After exhibiting very little signal pre irra-
diation, the EPR signal exhibits a broad, axial powder pat-
tern shape with a g-value of 2.008 post irradiation.
In the solid-state, activated host 1a showed a 60 nm red-
shift of the absorption maxima versus dissolution in DMSO
going from 303 to 366 nm (Figure S38). The formation of
an additional broad absorption band at ~400 nm was also
observed. The shift in absorption is more typical for J-
Next, we measured the EPR signal of activated 1a (9.8
mg) under increasing irradiation time (1 to 24 h) to see if
more radicals would be generated with longer irradiation
times. To monitor the formation of radicals, we plotted the
Table 1. Measured photophysical properties for compounds 1, 1a, 2, and 2a with different conditions.
Compound
λabs (nm)a
ε
(

104 M-1 cm-1)
λems (nm)b
τ<avg> (ns)c
1
369, 414
303
--
4.65
--
478
378, 450*
465
1.6
2.3
1.1
1.4
1.2
1.3
1.5
4.1d
1.0d
10 μM 1a in DMSO
1a
366
1a·DME
377
--
456
1a·C6H6
370
--
455
1a·DMF
372
--
456
2
366
--
5.41d
487
10 μM 2a in DMSO
302d
358d
369, 452*d
447d
2a
--
a Peak position at largest absorption band. b Peak positions at largest emission bands in nm (largest denoted with * if applicable, excited at λ ). c Average
abs.
lifetime of the most intense emission peak. d Values taken from reference.39
ACS Paragon Plus Environment

Page 5 of 11
Journal of the American Chemical Society
framework, or because the filled channel slows down oxy-
1
2
gen quenching processes.
3
4
5
6
7
8
9
10
11
12
13
Figure 6. EPR studies for activated 1a. (A) EPR signal pre and post UV
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
irradiation. (B) Double integration over time of UV irradiation. (C) Dark
decay spectra for activated 1a after it was irradiated to its maximum radical
concentration. (D) Double integration versus time after irradiation.
Figure 7. EPR studies for guest inclusion complexes of 1a. EPR signal pre
and post UV irradiation is given for each complex. Additionally, the g-val-
ues are given.
double integration of the EPR spectra over time. To esti-
mate the number of radicals generated, the double integra-
tion of the EPR spectra were compared to a calibration of
Magic Blue standard solutions in dichloromethane (Figure
S68). A comparison of these can give an approximate con-
centration of the number of radicals generated in the solid-
state. As seen in Figure 6D, the concentration of radicals
grows with increasing irradiation time until it plateaus
around 20 h. At this point, ~ 0.69% of the molecules gen-
erated a radical (or ~ 1 in 150 molecules) which estimates to
the same number of radicals seen in 100 μL of a 0.83 mM
solution of Magic Blue.
Table 2. Approximate number of radicals generated during UV-
irradiation from 365 nm LEDs.
Compound
4 Hours
Max
1
0.69%
0.42%
0.20%
0.55%a
0.33%
0.19%
0.13%
--
1a
0.69%
0.28%
0.85%a
0.45%
0.24%
0.23%
1a·DME
1a·C6H6
1a·C6H5F
1a·C6H5Cl
1a·C6H5Br
Next, seven host-guest complexes of 1a were systemati-
cally investigated to quantify the maximum amount of rad-
ical they could generate upon irradiation. These complexes
were chosen as to vary the polarity and heavy atom substi-
tution of the guests to see if they had any effects on radical
formation. As seen in Figures S61-67, most of these com-
plexes reached their maximum EPR signals after ~ 20-24 h,
apart from 1a·C6H5F and 1a·C6H5Cl which reached a maxi-
mum after 12 h. Interestingly, the presence of encapsulated
guests altered both the line shape of the signal as well as
the quantity of radicals formed as seen in Figures 1 and 7;
and Table 2, respectively. This was quite surprising since
guest encapsulation had little effect on the absorption,
emission and lifetimes of host 1a. For the line shapes, the
least polar and least heavy atom substituted complex,
1a·C6H6, showed the line shape most similar to the empty
structure, activated 1a, while the more polar, heavy atom
complexes of 1a·DMF and 1a·C6H5Br showed a narrowing
of the line shape. Similarly, the least polar and least heavy
atom substituted complex, 1a·C6H6, generated the most
radicals (0.78%) while the more polar, heavy atom substi-
tuted complexes of 1a·DMF and 1a·C6H5Br exhibited the
least number of radicals (0.15% and 0.23%, respectively). In
fact, trends were observed along these lines with polarity
(1a·C6H6 > 1a·1,4-dioxanes > 1a·DME > 1a·DMF) and heavy
atom incorporation (1a·C6H6 > 1a·C6H5F > 1a·C6H5Cl >
1a·C6H5Br) clearly affecting the amount of radical ob-
served. Intriguingly, encapsulation of benzene within host
1a afforded an even higher radical concentration than acti-
vated 1a. This may be due to a stabilization of the
1a·1,4-dioxanes
0.24%
0.08%
0.38%
1a·DMF
0.15%
2
2a
3
0.16%
--
--
--
--
--
0.16%
~ 0%
3a
4
Very Smallb
Very Smallb
a Average of four trials. A standard deviation of 0.06% was found for
these trials. This is considered to be the error in all the measurements
taken. b Concentration below values of calibration curve given in Fig-
ure S68.
To further estimate the error of these measurements, we
repeated the 1a·C6H6 experiment three more times, by ir-
radiating the sample until radical formation plateaued
(Figure S61). On average 0.85% of the molecules generated
a radical at maximum concentration (or 1 in 120 molecules)
with a standard deviation of 0.06% which is considered to
be the error on all the other measurements. Overall, this
demonstrates the reproducibility of radical concentrations
between samples.
Next, the persistence of the photogenerated radicals
were examined with dark decay studies. For this, samples
which were irradiated to their maximum concentration,
were stored in the dark at room temperature. Then EPR
spectra were taken periodically afterwards to assess the
stabilities of the radicals. Figures 6C and 6D show the EPR
signal and the double integration of the EPR signal, respec-
tively, over time for the activated 1a host over three weeks.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Figure 6D shows that the number of radicals decays to
Page 6 of 11
to be introduced during later irradiation, since this is sus-
pected to be a charge separation-based process.39 Overall,
in each case radicals could be formed, decay, and then
formed again over multiple 2-day cycles, in a reproducible
manner.
1
2
3
4
5
6
7
8
about 20% of the maximum concentration after one week,
with an estimated half-life of ~ 24 h. The remaining radical
species display more stability, with their concentration re-
maining somewhat constant up to three weeks later. These
results in combination with the different line shapes
shown in the latter spectra (Figure 6C), suggest that at least
two types of radical decay processes are occurring. Simula-
tions of the EPR spectra (see below) also support this sug-
gestion.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For the host-guest complexes, dark decay studies were
performed on complexes 1a·C6H6, 1a·DME and 1a·DMF as
these guests afforded the greatest difference in radical con-
centrations (Table 2). Similar behavior and half-lives were
observed for 1a·C6H6 and 1a·DMF crystals when compared
to activated 1a (Figures S61 and S63). A longer half-life of ~
1 week was estimated for the 1a·DME complex (Figure S62).
Additionally, the radical concentration only dipped to
about 40% of its maximum after 3 weeks instead of 20%
which was observed for all the other cases for host 1a. This
enhanced stability may be a result of minimal processing,
as complex 1a·DME is formed during the initial crystalliza-
tion conditions for macrocycle 1a. Activated 1a is prepared
by heating complex 1a·DME to remove the DME guest.
Complexes 1a·C6H6 and 1a·DMF are formed by soaking ac-
tivated 1a crystals in neat C6H6 or DMF. The extra steps in
crystal processing may explain why these three materials
exhibit a shorter radical lifetime versus complex 1a·DME
during the dark decay of the radical signal. In fact, linear
analog 2a which also had minimal processing, displayed a
similar half-life (1 week), and decayed to only ~ 40% of its
initial signal similar to complex 1a·DME.39 Therefore, there
is significant similarity between freshly crystallized sam-
ples. Overall, the line shapes of the EPR spectra for each
case involving macrocycle 1a decayed to very similar signal
levels, suggesting that the longer-lived radical species
among these samples is similar.
Figure 8. Regeneration of radical signals for activated 1a, 1a·C6H6,
1a·DME, and 1a·DMF. After initial irradiation to maximum concentration,
each sample was allowed to decay for two days. Then they were re-irradi-
ated overnight to restore radical signal.
After seeing that the radicals could be repeatably lost
and regenerated, we wanted to monitor if the host material
could survive this process similar to linear analog 2a.39 For
this, each sample used from the regeneration studies (acti-
vated 1a, 1a·DME, 1a·C6H6, 1a·DMF) were dissolved in
1
DMSO-d6 and their H NMR recorded and compared to a
freshly synthesized sample of 1a. As seen in Figures S69-
S72, no degradation was observed for any of the samples.
Comparing these spectra, the only difference is the pres-
ence of guests in the NMR spectra (as is the case with ben-
zene). These studies suggest that radical formation of host
1a is reversible and does not lead to the degradation of the
material as a whole.
Next, we compared the EPR spectra for control com-
pounds 1 and 2-4 before and after irradiation (4 h, 365 nm
LEDs) to investigate the chemical structure that contrib-
utes to radical formation (Figure 9 and Table 2). The first
of these, TPA 3, which has no substitution on the TPA
group, displayed no significant radical formation after irra-
diation, suggesting that at least some substitution on the
TPA moiety is needed for radical formation in the solid-
state. The mono-brominated equivalent of TPA 3, com-
pound 3a, showed a very small signal after irradiation. This
signal was so small that it was below the detection limits of
our calibration curve to quantify the radical concentration.
This suggests that the signal from TPA 3a does not arise
from the same process as macrocycle 1a as many more rad-
icals were produced for macrocycle 1a. The radicals gener-
ated from TPA 3a may just be the result of photolysis of the
C-Br, bond which can be expected for halogenated aromat-
ics.46 Control 4, which has one methylene urea connected
to the TPA group, also exhibited a very small signal similar
to the singly brominated TPA of 3a suggesting that it may
also be susceptible to a small amount of photolysis.
Usually with chemical or electrochemical radical gener-
ation methods, the sample must be resynthesized once the
TPA radicals have decayed due to degradation. However,
no degradation was detected by NMR for linear analog 2a
after radical generation.39 Moreover, after radical decay, re-
exposure of linear analog 2a to UV-light led to regeneration
of the radical species. This led us to test four samples of
host 1a (activated 1a, 1a·DME, 1a·C6H6, 1a·DMF) to deter-
mine if similar stable and regenerable radicals would be
observed. For these experiments, each sample (5-10 mg)
was irradiated to its maximum concentration, its EPR spec-
trum measured, followed by storage in the dark for two
days. EPR signals were recorded at t = 8, 24, and 48 h. Then
the samples were re-irradiated (12 – 14 h) so the cycle could
be repeated. Figure 8 plots the area of the EPR signal versus
time over four cycles of this photo-regeneration process for
each sample. Remarkably, in all four cases, the radical sig-
nal could be fully restored in both intensity and line shape
(Figures S60-S63). For complex 1a·C6H6, the maximum rad-
ical concentration slightly increased over each cycle. This
unexpected result may be a consequence of some charge
equilibria during radical decay, allowing for more charge
Given these controls, we suspected that closely organiz-
ing two TPA units in a linear or macrocyclic system is
ACS Paragon Plus Environment

Page 7 of 11
Journal of the American Chemical Society
important for generating significant radical concentrations the T1 relaxation time of the protons in the vicinity of the
1
2
3
4
5
6
7
8
in the solid-state. Therefore, we tested the non-bromin-
ated analogs of macrocycle 1a and linear analog 2a to see if
they afforded significant amounts of radical. Indeed, non-
brominated linear analog 2 forms a similar number of rad-
icals as its brominated counterpart, 2a, after 4 h of irradia-
tion (both ~ 0.16%). Surprisingly, the non-brominated
macrocycle 1 exhibited the highest efficiency of radical
generation after four hours of irradiation (0.69% for the
non-brominated macrocycle 1 versus 0.42% for the bro-
minated macrocycle of 1a) and showed more hyperfine in-
teractions than its linear counterpart, 2. Considering that
macrocycle 1a also had more hyperfine splitting than its
linear analog counterpart 2a, this suggests that the addi-
tional tether between the TPAs (macrocycles having two
and linear analogs having one) increases the hyperfine
splitting in the EPR spectra.
carbonyl carbon were among the shortest in the sample,
where the recovery null occurs at approximately 3.5 s. The
proton T1 of the protons on the aromatic rings varied and
was shortest for broad resonance at ~120 ppm and in-
creased with increasing chemical de-shielding, consistent
with the observed variations in the line broadening. The
CP-MAS NMR results suggest localization of unpaired
electron density in the BrNPh3 moiety. Significant line
broadening on the carbonyl carbon but not the methylene
carbon suggests radical formation impacts the urea net-
work more strongly than the methylene bridge. To probe if
the structure of the radicals changes with time, the stabil-
ity of the UV-induced radicals were monitored over the
course of 16 h by acquiring a series of 13C CP-MAS spectra
every 53 minutes. As seen in the overlay of these spectra in
Figure S75, the spectra are all identical (within the noise),
demonstrating the stability of the photoinduced radicals
on the timescale of this experiment.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To gain more insight into structure of the photogener-
ated radicals we next investigated the more soluble TPAs
2-4 via electrochemical methods. Prior electrochemical
studies with linear analog 2a demonstrated oxidation at 1.0
V versus a saturated calomel electrode (SCE) as a 1 mM so-
lution in dichloromethane. This resulted in a dication with
both TPA groups containing a radical cation.39 Bulk elec-
trolysis at this potential exhibited a very similar EPR spec-
trum to the photogenerated radicals of linear analog 2a in
the solid-state indicating that a similar radical species (rad-
ical cation) was present in both samples.
For this study, a cathodic investigation of TPA and its
related derivatives were investigated with cyclic voltam-
metry. Voltammograms of TPAs 2-4, Figures S76-S80, all
exhibit irreversible reductions occurring between -1.2 and
-1.5 V versus SCE as 1 mM dichloromethane solutions. Re-
duction of TPA 3 occurred with an Ep,c = -1.28 V under the
standard experimental conditions. Reduction of the bro-
minated TPA, 3a, occurred at a less negative potential of
Ep,c = -1.18 V. This shift can be attributed to the electron
withdrawing Br moiety on TPA 3a. In addition, reduction
of methylene urea substituted TPA 4 occurred with a neg-
ative potential shift relative to TPA 3 likely due to the elec-
tron donating behavior of the methylene urea substituent
on the TPA core. These redox trends are consistent with
previously reported redox behavior for related TPA deriva-
tives.47 Furthermore, linear analogs 2 and 2a exhibit two
closely spaced 1e- reductions. These reductions are at-
tributed to reduction of the individual TPA units tethered
by the urea linkers and illustrates an ability to generate
closely spaced radicals.
Figure 9. EPR signals pre and post irradiation for control samples 1, 2, 2a,
3, 3a, and 4. Each sample was irradiated for four hours before the post-
irradiation EPR spectra were taken.
To investigate how UV-induced radical formation af-
fected the TPA materials and to probe insight into a possi-
ble mechanism of radical formation, both linear analog 2a
and the activated host of 1a were further investigated by
Cross-Polarized Magic Angle Spinning (CP-MAS) 13C NMR.
Overlays of the NMR spectra of the pre- and post-irradi-
ated solids are shown in Figure S73. While the methylene
carbon resonances of the urea linker (ca. 43 ppm) were
minimally affected, the overlays reveal significant UV-
induced broadening of the aromatic and carbonyl 13C lines,
particularly for linear analog 2a. Thus, the proton spin-lat-
tice relaxation times at different sites in crystals of the UV-
irradiated linear analog 2a were probed by application of
the inversion-recovery pulse sequence to the protons prior
to cross-polarization and detection on 13C (Figure S74). The
longest spin relaxation times were observed for protons at-
tached to the urea linker carbons, consistent with the min-
imal line-broadening observed at these sites. In contrast,
Using the EPR, NMR, and electrochemical data, we next
set out to simulate the EPR spectra of the UV-irradiated
activated 1a. Figure 10 shows our best effort to simulate the
EPR spectrum with the most fine structure. The fact that
two independent radicals are needed to get this reasonable
fit, with all lines accounted for, supports that two radical
decay processes may be at play. The discrepancies in inten-
sity on the high field side of the spectrum are almost cer-
tainly due to anisotropy in this solid-state spectrum.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 11
Furthermore, the presence of a signal carrier with two N complexes of 1a·C6H6, 1a·DMF, and 1a·C6H5Br, with the
1
2
3
4
5
6
7
8
hyperfine interactions and one without strongly supports
the suggestion above that more than one redox pathway is
available to these structures and that both mono-cations
and di-cations may be present. Future isotopic substitu-
tions and in situ electrochemical EPR experiments are
planned in order to explore this further.
first of these displaying the most radical formation after 24
h of irradiation with 0.78% of the molecules generating a
radical. The latter two complexes are more polar or contain
heavy atoms and show less radical formation after a similar
amount of UV-irradiation, 0.15% and 0.23%, respectively.
The increase in radical formation by supramolecular en-
capsulation of less polar guests was surprising and differs
from what is predicted in covalent radical systems that are
substituted with both electron donating and electron with-
drawing groups.49,50 In such push-pull radicals, the capto-
dative effect predicts that the radicals are more stable in
polar solvents.51 However, such effects appear to vary de-
pending on the structure of the radicals. For example,
alkoxy cyanomethyl radicals and cyclohexadienyl radicals
showed no solvent effect.52,53 Our observations, coupled
with the persistent and regenerable nature of radicals in
self-assembled TPA materials will help guide the develop-
ment of new photoactive materials in which the properties
can be tuned simply via guest exchange. Thus, future work
will focus on the mechanism of how the encapsulated guest
and framework affects radical formation and lifetime.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 10. Experimental steady–state X-band EPR spectrum (solid black
line) of activated host 1a post UV-irradiation, overlaid onto a best fit sim-
ulation (dashed red line) using EasySpin.48 The simulation was performed
using parameters for two independent radicals. The first has two nitrogen
hyperfine interactions with g = 2.0049, aN = 39 Gauss for two equivalent
nitrogens, line width (peak to peak) 5 Gauss, and a relative weighting of
0.1. The second radical has g = 2.0087, contains no hyperfine interactions,
line width = 20 G, and a relative weighting of 1.1.
Overall, the complexes reported here show similar radi-
cal half-lives of 24 hours and persist up to 3 weeks. Intri-
guingly, reirradiation of these materials with more UV-
light can regenerate radical quantities to similar amounts
to their pre-radical decays. In fact, this radical for-
mation/decay process can continue over several cycles
with reproducible results and occurs without degradation
of the host material. The key to forming these robust ma-
terials with regenerative radical properties is connecting
two TPAs together through a methylene urea bridge. Alt-
hough UV-irradiation of simple bromine or methylene
urea substituted TPAs resulted in a small amount of radical
signal, significant amounts of photogenerated radicals
were not observed until the two TPAs were connected.
Currently, we are planning Q-band EPR, ENDOR, ESEEM,
and SQUID experiments to try and pinpoint the mecha-
nism of radical formation upon UV-irradiation for these
materials.
Overall, UV-irradiation of self-assembled methylene
urea tethered TPA macrocycles gives rise to persistent or-
ganic radicals. Guests loaded within the TPA hosts can
modulate the concentration of radicals generated with
concentrations ranging from 0.15-0.85% of the molecules
generating a radical at maximum concentration. These
radicals display a half-life of about 24 hours, persist an up-
wards of 3 weeks, and can be regenerated with additional
UV-irradiation to their original maximum concentration
without causing degradation to the TPA host. Future work
with these TPA hosts could include investigating how dif-
ferent electron accepting guests affect the radical for-
mation of the TPA host and if it leads to any new conduc-
tive properties of the material.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at http://pubs.acs.org.
Synthetic procedures, product characterization, X-
ray, photophysical characterization, NMR, EPR, and
CV spectra
CONCLUSIONS
In summary, several TPA derivatives were synthesized
and were organized in the solid-state via crystalline assem-
blies. One of these derivatives, brominated macrocycle 1a,
organizes into robust columnar needles and displays 1D
porosity through small nanochannels. Inside these chan-
nels, small guests can be exchanged via SC-SC transfor-
mations without changing the host framework. This cre-
ates a series of isoskeletal complexes whose properties can
be directly compared. We have monitored how the photo-
physics and photo-induced radical formation change with
the inclusion of different guest molecules.
X-ray data for dibromide 1
X-ray data for protected 1·CHCl3
X-ray data for protected 1·MeOH
X-ray data for 1·DMSO
X-ray data for 1a·DMF
X-ray data for 1a·1,4-dioxanes
X-ray data for 2
X-ray data for monoclinic 3a
X-ray data for triclinic 3a
X-ray data for 4
Although the absorption, emission, and photolumines-
cent lifetime properties of each complex were somewhat
similar, the amount of photogenerated radicals produced
by each complex varied considerably. It was found that an
increase in polarity or heavy atom substitution decreases
the number of radicals observed. This is highlighted in the
ACS Paragon Plus Environment

Page 9 of 11
Journal of the American Chemical Society
(9) Lin, R.-B.; He, Y.; Li, P.; Wang, H.; Zhou, W.; Chen, B. Mul-
tifunctional Porous Hydrogen-Bonded Organic Framework Mate-
rials. Chem. Soc. Rev. 2019, 48, 1362-1389.
AUTHOR INFORMATION
1
Corresponding Author
2
3
4
5
6
7
8
9
(10) Dhakshinamoorthy, A.; Alvaro, M.; Garcia, H. Commercial
Metal-Organic Frameworks as Heterogeneous Catalysis. Chem.
Commun. 2012, 48, 11275-11288.
(11) Tan, Y.-X.; He, Y.-P.; Yuan, D.; Zhang, J. Use of Aligned Tri-
phenylamine-Based Radicals in a Porous Framework for Promot-
ing Photocatalysis. Appl. Catal. B-Environ. 2018, 221, 664-669.
(12) Lü, J.; Cao, R. Porous Organic Molecular Frameworks with
Extrinsic Porosity: A Platform for Carbon Storage and Separation.
Angew. Chem., Int. Ed. 2016, 55, 9474-9480.
(13) Kitagawa, S.; Kitaura, R.; Noro, S. Functional Porous Coor-
dination Polymers. Angew. Chem., Int. Ed. 2004, 43, 2334-2375.
(14) Tabacchi, G. Supramolecular Organization in Confined
Spaces. ChemPhysChem 2018, 19, 1249-1297.
(15) Sumida, K.; Rogow, D. L.; Mason, J. A.; McDonald, T. M.;
Bloch, E. D.; Herm, Z. R.; Bae, T.-H.; Long, J. R. Carbon Dioxide
Capture in Metal-Organic Frameworks. Chem. Rev. 2012, 112, 724-
781.
(16) Li, J.-R.; Sculley, J.; Zhou, H.-C. Metal-Organic Frameworks
for Separations. Chem. Rev. 2011, 112, 869-932.
(17) Kreno, L. E.; Leong, K.; Farha, O. K.; Allendorf, M.; Van
Duyne, R. P.; Hupp, J. T. Metal-Organic Framework Materials as
Chemical Sensors. Chem. Rev. 2012, 112, 1105-1125.
(18) Wales, D. J.; Grand, J.; Ting, V. P.; Burke, R. D.; Edler, K. J.;
Bowen, C. R.; Mintova, S.; Burrows, A. D. Gas Sensing Using Po-
rous Materials for Automotive Applications. Chem. Soc. Rev. 2015,
44, 4290-4321.
(19) Comotti, A.; Bracco, S.; Sozzani, P. Molecular Rotors Built
in Porous Materials. Acc. Chem. Res. 2016, 49, 1701-1710.
(20) Alhamami, M.; Doan, H.; Cheng, C.-H. A Review on
Breathing Behaviors of Metal-Organic-Frameworks (MOFs) for
Gas Adsorption. Materials 2014, 7, 3198-3250.
(21) Yanai, N.; Uemura, T.; Inoue, M.; Matsuda, R.; Fukushima,
T.; Tsujimoto, M.; Isoda, S.; Kitagawa, S. Guest-to-Host Transmis-
sion of Structural Changes for Stimuli-Responsive Adsorption
Property. J. Am. Chem. Soc. 2012, 134, 4501-4504.
(22) Akakuru, O. U.; Iqbal, M. Z.; Saeed, M.; Liu, C.; Paunesku,
T.; Woloschak, G.; Hosmane, N. S.; Wu, A. The Transition from
Metal-Based to Metal-Free Contrast Agents for T1 Magnetic Reso-
nance Imaging Enhancement. Bioconjugate Chem. 2019, 30, 2264-
2286.
(23) Ratera, I.; Veciana, J. Playing with Organic Radicals as
Building Blocks for Functional Molecular Materials. Chem. Soc.
Rev. 2012, 41, 303-309.
(24) Forrester, A. R.; Hay, J. M.; Thompson, R. H. Organic
Chemistry of Stable Free Radicals; Academic Press: New York,
1968.
(25) Ito, A.; Ino, H.; Tanaka, K.; Kanemoto, K.; Kato, T. Facile
Synthesis, Crystal Structures, and High-Spin Cationic States of
All-para-Brominated Oligo(N-phenyl-m-aniline)s. J. Org. Chem.
2002, 67, 491-498.
(26) Yano, M.; Sato, K.; Shiomi, D.; Ichimura, A.; Abe, K.; Takui,
T.; Itoh, K. Synthesis of 1,3-Bis(diarylamino)benzenes as Model
Precursors for One-Dimensional Organic Ferromagnetic Metals;
Characterization of the Dications by Cyclic Voltammetry and
Electron Spin Transient Nutation Spectroscopy. Tetrahedron Lett.
1996, 37, 9207-9210.
(27) Fukui, N.; Cha, W.; Shimizu, D.; Oh, J.; Furukawa, K.; Yo-
rimitsu, H.; Kim, D.; Osuka, A. Highly Planar Diarylamine-Fused
Porphyrins and their Remarkably Stable Radical Cations. Chem.
Sci. 2017, 8, 189-199.
(28) Bill, N. L.; Ishida, M.; Bähring, S.; Lim, J. M.; Lee, S.; Davis,
C. M.; Lynch, V. M.; Nielsen, K. A.; Jeppesen, J. O.; Ohkubo, K.;
Fukuzumi, S.; Kim, D.; Sessler, J. L. Porphyrins Fused with
Strongly Electron-Donating 1,3-Dithiol-2-ylidene Moieties: Redox
* shimizls@mailbox.sc.edu
Author Contributions
The manuscript was written through contributions of all au-
thors. / All authors have given approval to the final version of
the manuscript.
Present Address
⊥
Department of Chemistry, Northwestern University, Evans-
ton, Illinois 60208, United States
Notes
The authors declare no competing financial interest.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACKNOWLEDGMENT
This work was supported in part by the National Science
Foundation CHE-1904386 (L. S. S.) and CHE-1900541 (M. D. E.
F.). A portion of this work was performed in the McKnight
Brain Institute at the National High Magnetic Field Labora-
tory’s AMRIS Facility, which is supported by the National Sci-
ence Foundation Cooperative Agreement Number DMR-
1644779 and the State of Florida.
ABBREVIATIONS
SC-SC, single-crystal-to-single-crystal; TPA, triphenylamine;
SCE, saturated calomel electrode; LED, light emitting diode;
DME, 1,2-dimethoxyethane; DMF, N,N-dimethylformamide;
SC-XRD, single crystal x-ray diffraction; MRI, magnetic reso-
nance imaging, DMSO, dimethylsulfoxide; SCE, saturated cal-
omel electrode; EPR, electron paramagnetic resonance; MAS,
magic angle spinning; NMR, nuclear magnetic resonance;
ENDOR, electron nuclear double resonance; ESEEM, electron
spin echo envelope modulation; C6H6, benzene; C6H6F,
fluorobenzene; C6H6Cl, chlorobenzene; C6H6Br, bromoben-
zene; vdW, van der Waals.
REFERENCES
(1) Mattia, E.; Otto, S. Supramolecular Systems Chemistry. Nat.
Nanotechnol. 2015, 10, 111-119.
(2) Mahadevi, A. S.; Sastry, G. N. Cooperativity in Noncovalent
Interactions. Chem. Rev. 2016, 116, 2775-2825.
(3) Steed, J. W. 21st Century Developments in the Understand-
ing and Control of Molecular Solids. Chem. Commun. 2018, 13175-
13182.
(4) Ogoshi, T.; Kakuta, T.; Yamagishi, T. Applications of Pil-
lar[n]arene-Based Supramolecular Assemblies. Angew. Chem., Int.
Ed. 2019, 58, 2197-2206.
(5) Würthner, F.; Saha-Mꢀller, C. R.; Fimmel, B.; Ogi, S.; Le-
owanawat, P.; Schmidt, D. Perylene Bisimide Dye Assemblies as
Archetype Functional Supramolecular Materials. Chem. Rev. 2016,
116, 962-1052.
(6) Mills, M. B.; Wohlhauser, T.; Stein, B.; Verduyn, W. R.; Song,
E.; Dechambenoit, P.; Rouzières, M.; Clérac, R.; Preuss, K. E. Mag-
netic Bistability in Crystalline Organic Radicals: The Interplay of
H-Bonding, Pancake Bonding, and Electrostatics in 4-(2’-Benzim-
idazolyl)-1,2,3,5-dithiadiazolyl. J. Am. Chem. Soc. 2018, 140, 16904-
16908.
(7) Liu, M.; Zhang, L.; Wang, T. Supramolecular Chirality in
Self-Assembled Systems. Chem. Rev. 2015, 115, 7304-7397.
(8) Das, S.; Heasman, P.; Ben, T.; Qiu, S. Porous Organic Mate-
rials: Strategic Design and Structure-Function Correlation. Chem.
Rev. 2017, 117, 1515-1563.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 11
Control by Metal Cation Complexation and Anion Binding. J. Am.
Chem. Soc. 2013, 135, 10852-10862.
(29) Murata, H.; Miyajima, D.; Nishide, H. A High-Spin and
Helical Organic Polymer: Poly{[4-(dianisylaminium)phenyl]acet-
ylene}. Macromolecules 2006, 39, 6331-6335.
triindole derivatives to their 2,7,12-isomers. Org. Lett. 2010, 12,
5192-5195.
(42) Xu, D.; Ciszewski, L.; Li, T.; Repič, O.; Blacklock, T. J. Re-
ductive alkylation of urea: a practical route to substituted ureas. Tetra-
hedron Lett. 1998, 39, 1107-1110.
1
2
3
4
5
6
7
8
9
(43) Sobolev, A. N.; Belsky, V. K.; Romm, I. P.; Chermikova, N.
Y.; Guryanova, E. N. Structural Investigation of Triaryl Derivatives
of the Group V Elements. IX.* Structure of Triphenylamine,
C18H15N. Acta Crystallogr. C 1985, C41, 967-971.
(44) Würthner, F. Perylene Bisimide Dyes as versatile Building
Blocks for Functional Supramolecular Architectures. Chem. Com-
mun. 2004, 1564-1579.
(45) Yang, M.; Xu, D.; Xi, W.; Wang, L.; Zheng, J.; Huang, J.;
Zhang, J.; Zhou, H.; Wu, J.; Tian, Y. Aggregation-Induced Fluores-
cence Behavior of Triphenylamine-Based Schiff Bases: The Com-
bined Effect of Multiple Forces. J. Org. Chem. 2013, 78, 10344-
10359.
(46) Matsuura, T.; Omura, K. Photo-induced Reaction. III. The
Synthesis of Biphenyls from Aromatic Bromo-Compounds. Bull.
Chem. Soc. Jpn. 1966, 39, 944-947.
(47) Data, P.; Pander, P.; Zassowski, P.; Mimaite, V.; Karon, K.;
Lapkowski, M.; Grazulevicius, J. V.; Slepski, P.; Darowcki, K. Elec-
trochemically Induced Synthesis of Triphenylamine-based Poly-
hydrazones. Electrochim. Acta 2017, 230, 10-21.
(48) Stoll, S.; Schweiger, A. EasySpin a Comprehensive Software
Package for Spectral Simulation and Analysis in EPR. J. Magn. Re-
son. 2006, 178, 42-55.
(30) Skorka, L.; Kurzep, P.; Chauviré, T.; Dubois, L.; Mouesca,
J.-M.; Maurel, V.; Kulszewicz-Bajer, I. High-Spin Polymers: Ferro-
magnetic Coupling of S =1 Hexaazacyclophane Units up to a Pure
S = 2 Polycyclophane. J. Phys. Chem. B 2017, 121, 4293-4298.
(31) Su, C.; He, H.; Xu, L.; Zhao, K.; Zheng, C.; Zhang, C. A Mes-
oporous Conjugated Polymer Based on a High Free Radical Den-
sity Polytriphenylamine Derivative: Its Preparation and Electro-
chemical Performance as a Cathode Material for Li-Ion Batteries.
J. Mater. Chem. A 2017, 5, 2701-2709.
(32) Chen, Z.; Su, C.; Zhu, X.; Xu, R.; Xu, L.; Zhang, C. Micro-
/Mesoporous Conjugated Polymer Based on Star-Shaped Triazine
Functional Triphenylamine Framework as the Performance-Im-
proved Cathode of Li-Organic Battery. J. Polym. Sci. A: Polym.
Chem. 2018, 56, 2574-2583.
(33) Agrarwala, P.; Kabra, D. A Review on Triphenylamine
(TPA) Based Organic Hole Transport Materials (HTMs) for Dye
Sensitized Solar Cells (DSSCs) and Perovskite Solar Cells (PSCs):
Evolution and Molecular Engineering. J. Mater. Chem. A 2017, 5,
1348-1373.
(34) Quinton, C.; Alain-Rizzo, V.; Dumas-Verdes, C.; Mioman-
dre, F.; Clavier, G.; Audebert, P. Redox-Controlled Fluorescence
Modulation (Electrofluorochromism) in Triphenylamine Deriva-
tives. RSC Adv. 2014, 4, 34332-34342.
(35) Saha, A.; Chen, M.; Lederer, M.; Kahnt, A.; Lu, X.; Guldi, D.
M. Sulfur Rich Electron Donors – Formation of Singlet Versus Tri-
plet Radical Ion Pair States Featuring Different Lifetimes in the
Same Conjugate. Chem. Sci. 2017, 8, 1360-1368.
(36) Moulin, E.; Armao, J. J.; Giuseppone, N. Triarylamine-
Based Supramolecular Polymers: Structures, Dynamics, and Func-
tions. Acc. Chem. Res. 2019, 52, 975-983.
(37) Shimizu, L. S.; Salpage, S. R.; Korous, A. A. Functional Ma-
terials from Self-Assembled Bis-urea Macrocycles. Acc. Chem. Res.
2014, 47, 2116-2127.
(38) DeHaven, B. A.; Goodlett, D. W.; Sindt, A. J.; Noll, N.; De
Vetta, M.; Smith, M. D.; Martin, C. R.; González, L.; Shimizu, L. S.
Enhancing the Stability of Photogenerated Benzophenone Triplet
Radical Pairs through Supramolecular Assembly. J. Am. Chem.
Soc. 2018, 140, 13064-13070.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(49) Viehe, H. G.; Janousek, Z.; Merꢁnyi, R. Substituent Effects
in Radical Chemistry; Springer: Netherlands, 2012.
(50) Ito, O.; Matsuda, M. Solvent Effect on Rates of Free-Radical
Reactions. 2. Addition of the p-(Dimethylamino)benzenethiyl
Radical to α-Methylstyrene. J. Phys. Chem. 1984, 88, 1002-1005.
(51) Peterson, J. P.; Winter, A. H. Solvent Effects on the Stability
and Delocalization of Aryl Dicyanomethyl Radicals: The Captoda-
tive Effect Revisited. J. Am. Chem. Soc. 2019, 141, 12901-12906.
(52) Beckhaus, H.-D.; Rüchardt, C. The Effect of Solvent on the
Stability of mero-Substituted Alkyl Radicals. Angew. Chem., Int.
Ed. Engl. 1987, 26, 770-771.
(53) Rhodes, C. J.; Roduner, E. Muon Spin Rotation Studies of
the Role of Solvent in Captodative Interactions of Cyclohexadi-
enyl Radicals. Hyperfine Interact. 1991, 65, 975−978.
(39) Sindt, A. J.; DeHaven, B. A.; McEachern, D. F.; Dis-
sanayake, D. M. M. M.; Smith, M. D.; Vannucci, A. K.; Shimizu, L.
S. UV-irradiation of self-assembled triphenylamines affords per-
sistent and regenerable radicals. Chem. Sci. 2019, 10, 2670-2677.
(40) Sindt, A. J.; Smith, M. D.; Berens, S.; Vasenkov, S.; Bowers,
C. R.; Shimizu, L. S. Single-crystal-to-single-crystal guest ex-
change in columnar assembled brominated triphenylamine bis-
urea macrocycles. Chem. Comm. 2019, 55, 5619-5622.
(41) Ji, L.; Feng, M.-S.; Liu, Z.-Q.; Shen, Y.-X.; Chen, H.-F.
Switching high two-photon efficiency: from 3,8,13-substituted
ACS Paragon Plus Environment

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