Mechanistic Analysis of Solid-State Colorimetric Switching: Monoalkoxynaphthalene-Naphthalimide Donor-Acceptor Dyads

Article abstract of DOI:10.1021/jacs.0c08137

There is growing interest in creating solids that are responsive to various stimuli. Herein we report the first molecular-level mechanistic picture of the thermochromic polymorphic transition in a series of MAN-NI dyad crystals that turn from orange to yellow upon heating with minimal changes to the microscopic morphology following the transition. Detailed structural analyses revealed that the dyads assemble to create an alternating bilayer type structure, with horizontal alternating alkyl and stacked aromatic layers in both the orange and yellow forms. The observed dynamic behavior in the solid state moves as a yellow wavefront through the orange crystal. The overall process is critically dependent on a complex interplay between the layered structure of the starting crystal, the thermodynamics of the two differently colored forms, and similar densities of the two polymorphs. Upon heating, the orange form alkyl chain layers become disordered, allowing for some lateral diffusion of dyads within their own layer. Moving to either adjacent stack in the same layer allows a dyad to exchange a head-to-head stacking geometry (orange) for a head-to-tail stacking geometry (yellow). This transition is unique in that it involves a nucleation and growth mechanism that converts to a faster cooperative wavefront mechanism during the transition. The fastest moving of the wavefronts have an approximately 38° angle with respect to the long axis of the crystal, corresponding to a nonconventional C-H···O hydrogen bond network of dyad molecules in adjacent stacks that enables a transition with cooperative character to proceed within layers of orange crystals. The orange-to-yellow transition is triggered at a temperature that is very close to the temperature at which the orange and yellow forms exchange as the more stable, while being lower than the melting temperature of the original orange, or final yellow, solids.

Full text of DOI:10.1021/jacs.0c08137

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
Mechanistic Analysis of Solid-State Colorimetric Switching:
Monoalkoxynaphthalene-Naphthalimide Donor-Acceptor Dyads
Christopher D Wight, Qifan Xiao, Holden R Wagner, Eduardo
A Hernandez, Vincent M Lynch, and Brent L. Iverson
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.0c08137 • Publication Date (Web): 08 Sep 2020
Downloaded from pubs.acs.org on September 8, 2020
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 14
Journal of the American Chemical Society
1
2
3
4
5
6
Mechanistic Analysis of Solid-State Colorimetric Switching: Mono-
alkoxynaphthalene-Naphthalimide Donor-Acceptor Dyads
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Christopher D. Wight, Qifan Xiao, Holden R. Wagner, Eduardo A. Hernandez, Vincent M. Lynch, and
Brent L. Iverson*
Department of Chemistry, The University of Texas at Austin, Austin, Texas 78712, United States
ABSTRACT: There is growing interest in creating solids that are responsive to various stimuli. Herein we report the first molecu-
lar-level mechanistic picture of the thermochromic polymorphic transition of a series of MAN-NI dyad crystals that turn from or-
ange to yellow upon heating with minimal changes to the microscopic morphology following the transition. Detailed structural
analyses revealed that the dyads assemble to create an alternating bilayer type structure, with horizontal alternating alkyl and
stacked aromatic layers in both the orange and yellow forms. The observed dynamic behavior in the solid state moves as a yellow
wavefront through the orange crystal. The overall process is critically dependent on a complex interplay between the layered struc-
ture of the starting crystal, the thermodynamics of the two differently colored forms, and similar densities of the two polymorphs.
Upon heating, the orange form alkyl chain layers become disordered, allowing for some lateral diffusion of dyads within their own
layer. Moving to either adjacent stack in the same layer allows a dyad to exchange a head-to-head stacking geometry (orange) for a
head-to-tail stacking geometry (yellow). This transition is unique in that it involves a nucleation and growth mechanism that con-
verts to a faster cooperative wavefront mechanism during the transition. The fastest moving of the wavefronts have an approximate-
ly 38° angle with respect to the long axis of the crystal, corresponding to a non-conventional C-H×××O hydrogen bond network of
dyad molecules in adjacent stacks that enables a transition with cooperative character to proceed within layers of orange crystals.
The orange-to-yellow transition is triggered at a temperature that is very close to the temperature at which the orange and yellow
forms exchange as the more stable, while being lower than the melting temperature of the original orange, or final yellow, solids.
INTRODUCTION
1,5-dialkoxylnaphthalene (DAN) and relatively electron poor 1,4,5,8-
naphthalenetetracarboxylic diimide (NDI) pioneered by our group in
There is still a great deal to learn about the mechanisms underlying
dynamic behaviors of molecules in the solid state. The higher order
structure of solids imposes severe limitations on molecular motion
and dynamics. Yet, there is growing interest in creating solids that are
responsive to various stimuli. An understanding of the interplay be-
tween all of the relevant non-covalent interactions is fundamental for
the rational design of solids that respond with useful properties and
behaviors. Electrostatic interactions of all types, including hydrogen
bonding and van der Waals interactions, combine to determine the
exact intermolecular geometries adopted by organic molecules and
assemblies in solids, liquids, and mesophases. Over the last few dec-
ades, our understanding of the interactions between aromatic mole-
2
1-29
the context of aqueous foldamers.
However, contrary to what the
polar/π model would predict but consistent with the local, direct inter-
action model, NDI has been shown to exhibit highly favorable NDI-
2
4
NDI associations in solution, as well as adopt NDI offset self-
29-36
stacking in a variety of contexts (Figure S1).
Previous work with
alkyl substituted DAN and NDI derivatives revealed thermochromic
changes that accompany the reorganization of some DAN and NDI
30-32
mixtures.
Cooling of certain DAN and NDI mixtures from the
isotropic phase produced a deep red colored mesophase, with DAN-
NDI adopting an alternating face-centered stacking geometry. Even
more interesting, a dramatic thermochromic change occurred in some
derivatives when the red mesophase cooled to form a yellow crystal-
line solid. Various experiments indicated this color change to be the
result of the self-sorting of NDI and DAN, with off-set parallel NDI
self-stacking and DAN self-stacking in a herringbone geometry. Thus,
NDI can be thought of as having a “split personality” in terms of its
favorable stacking modes: alternating face-centered stacking with
electron rich aromatics like DAN, and offset NDI self-stacking.
1
cules has increased significantly resulting in the increased exploita-
tion of organized aromatic molecules in numerous fields including
2-3
4-5
crystal engineering,
catalyst design,
supramolecular self-
assembly, and molecular sensing.8
6-7
An older rationale for understanding the relative geometries adopt-
ed by interacting aromatic units, known as the “polar/π model”, ra-
tionalizes aromatic stacking geometries as maximizing favorable
9
-10
As part of our comprehensive supramolecular program focused on
NDI and its analogs, a series of molecules were synthesized incorpo-
rating naphthalene monoimides (NI) covalently linked via an alkyne
to monoalkoxynaphthalenes (MAN). Interestingly, several of these
MAN-NI donor-acceptor dyads showed a dramatic difference in sol-
id-state color between solids created from relatively faster (yellow)
quadrupole-quadrupole interactions between aromatic units.
Later
1
1
work by Rashkin and Waters, as well as computational work by
12-13
and more recently Wheeler,14-17 has provided
Wheeler and Houk,
strong evidence in favor of a newer model, known as the “local, direct
1
4
interaction” model which postulates that the substituents on aro-
matic rings dictate aromatic stacking geometries as a result of local,
through-space electrostatic interactions from polarized bonds. Excel-
lent experimental work by Shimizu and co-workers utilizing molecu-
lar balances have further informed our understanding of aromatic
33
and slower (yellow-orange or orange) evaporation from solution. In
the solid-state, one of these dyads was found to change color in re-
sponse to grinding (orange to yellow), heating (orange to yellow), and
vapor fuming (yellow to orange), with repeatable cycles of color
18-20
interactions.
3
4
changing. Similar to the thermochromic behavior of independent
DAN and NDI mixtures previously described, the change in color was
determined to result from a conformational change of the aromatic
The polar/π model does provide an intuitive rationale for the face-
centered stacking geometry adopted by the relatively electron rich
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 14
units. Based upon single crystal data, X-ray diffraction (XRD) pat-
challenging because many experimental techniques employed only
42
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
terns, and modeling, the stimuli-responsive behavior was reported to
result from a conformational switch between two stable polymorphs.
When ground or heated, the orange crystalline form, possessing a
head-to-head (NI-NI) stacking geometry, interconverted to a yellow
soft crystalline mesophase with a head-to-tail (NI-MAN) stacking
geometry that persisted at room temperature in the absence of solvent
vapor, an indication that both packing geometries have kinetic stabil-
ity.
provide a time and space average of structural information. Applica-
tion of external stimuli to polymorphic systems containing flexible
substituents (e.g. alky chains) often involve switching between crys-
talline and non-crystalline states, with structures of the latter often
34, 46
being inferred from structural information like powder XRD.
A
subset of reports on polymorphic transitions have been able to corre-
59, 61-70
late macroscopic changes to specific nanoscopic reorganization.
For example, Ito and co-workers correlated molecular rearrangements
to macroscopic mechanical motion from the anisotropic strain release
Aromatic lumiphores, including the aforementioned MAN-NI dy-
ads, are often employed in solid-state sensing. Changes in the optical
properties of these materials can be achieved through synthetic modi-
fications that alter the energy of frontier orbitals as well as changes in
solid-state packing. In the solid-state, slight changes in molecular
packing can dramatically alter the optical and electronic properties of
conjugated materials through various mechanisms that can operate in
70
in thermosalient crystals. This thermosalient effect illustrates one of
the two general polymorphic transition mechanisms seen in organic
crystals, involving a rapid, cooperative displacement of molecules,
that is referred to by multiple names including martensitic, displacive,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
42, 69
or cooperative transitions
The other general polymorphism mech-
anism seen in organics is a nucleation and growth type mechanism
that, as the name implies, operates analogous to the mechanism of
3
7-39
isolation or parallel.
For example, long-range electronic coupling
4
2
crystal growth from solution and is noted to often disrupt the integri-
between molecules, that is dependent on the alignment of transition
dipole moments, can alter the spectroscopic properties of solid-state
systems compared to free molecules.39 In addition, short-range inter-
molecular interactions that manifest through direct orbital overlap,
especially through LUMO-LUMO and HOMO-HOMO overlap, can
6
9
ty of the initial crystal.
Herein we report the first molecular-level mechanistic picture of
the thermochromic polymorphic transition of the MAN-NI dyads. In
contrast to previously reported polymorphic systems that exploit alkyl
64, 68
facilitate the delocalization of locally excited states into delocalized
chains,
the stimuli-induced thermochromic transition of our
38-39
excitonic charge-transfer (CT) states.
While these intermolecular
MAN-NI dyads proceeded without altering the microscopic morphol-
ogy of the original crystal while simultaneously exhibiting a dramatic
visible color change. Comprehensive standard and in situ heating
XRD, grazing-incidence wide-angle X-ray scattering (GIWAXS),
non-contact atomic force microscopy (AFM), polarized optical mi-
croscopy (POM), spectroscopy, and differential scanning calorimetry
(DSC) analyses were used to develop a full picture of the thermo-
chromic process. Our analysis revealed that the MAN-NI dyads ap-
pear to be the first example of a polymorphic system involving a
nucleation and growth mechanism that converts to a faster coopera-
tive mechanism during a transition. The transition critically depends
upon a complex interplay between the overall layered structure of the
starting crystal, the thermodynamics of the two differently colored
forms, and similar densities of the interconverting polymorphs. These
insights will expand applications of polymorphic switching systems,
allowing for a new level of control over programmed dynamic mate-
rial properties without morphological changes.
excitonic interactions are distinct, the net electronic coupling experi-
enced is a product of the interference between short- and long-range
3
9
coupling.
Many materials exploit polymorphism as a functional property that
allows some form of reporting.40 Polymorphic phase transitions have
41
42
been reviewed in experimental, computational, and theoretical
contexts.
4
3
Stimuli induced polymorphic color switching has been
achieved in the solid-state following the application of various types
of external stimuli including heat,34 pressure, mechanical force,
and exposure to various analytes including salts and solvent. Cova-
44
45
46
47
lent modifications of polymorphic systems have been effectively used
30, 48
to impart tunability in various thermochromic systems.
Fang and
colleagues have shown that self-assembled thermochromic supramo-
lecular systems can be formed that allow for tunability of transition
temperature and colors based on competing CT interactions in both
gels and the solid-state.49-51
Although it is the aromatic p-system that directly influences the
electronic and photophysical properties of the system, substituents
appended to aromatic molecules are a critical and determining factor
in the solid-state electronic,52 spectroscopic and stimuli-responsive
properties of solid-state stimuli-responsive systems.53 An example of
utilizing substituents on aromatic systems to influence photophysical
properties in the solid-state came from Ito and co-workers, presenting
a design principle for stimuli-responsive color switching materials
that exploit polymorphism through competing interactions between
aromatic lumiphores and appended side-chains.46 By constructing
amphiphilic and dipolar characteristics into a single molecule, differ-
ent metastable polymorphic states were stabilized through competing
intermolecular interactions of the dipolar aromatic core, a hydrophilic
sidechain, and a hydrophobic sidechain. It is worth noting that this
system self-assembled through the formation of alternating aromatic,
hydrophilic, and hydrophobic layers, analogous to what is reported
here. The same group has also reported stimuli-responsive systems
that exploit the switching between multiple accessible polymorphs
MATERIALS AND METHODS
Synthesis of all dyads followed known procedures.33 Experimental
procedures and methods, crystallization conditions, synthetic proto-
cols, and characterization of all dyads and intermediates can be found
in the Supporting Information.
RESULTS
Stimuli-Responsive Behavior. Consistent with our previous re-
ports,33-34 most of the dyads in this study (Figure 1) could crystalize
as orange (O) or yellow (Y) polymorphs (Figure S2) depending on
crystallization conditions (Specific crystallization conditions are de-
tailed in Scheme S2).
When orange crystals of 6, 7, and 8, referred to here as 6O, 7O,
and 8O, respectively, are heated beyond 100˚C, the orange crystals
transition to a soft-crystalline yellow material that persists indefinitely
upon cooling. Reversal back to the orange crystalline phase occurs in
less than 24 hours when exposed to DCM vapor in a fuming cham-
ber.34
41, 45,
through mechanically induced crystal-to-crystal transformations.
5
4-55
Despite improvements in our understanding of polymorphism, tar-
geted polymorphism still remains a highly desirable goal.56 Computa-
tional efforts have successfully modeled polymorphic transfor-
5
7-59
mations, but have been limited to small molecules
or larger, rigid
60
systems. Experimental insights into the mechanisms by which pol-
ymorphs interconvert would better inform the future design of novel
polymorphic systems. While numerous examples for color changing
materials that rely on polymorphism exist, elucidating the molecular
mechanisms by which these materials interconvert is experimentally
Figure 1. Structure of the MAN-NI dyads used in this study.
ACS Paragon Plus Environment

Page 3 of 14
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Visualized under a microscope equipped with a heating stage, orange crystals are thermally converted to a yellow material that
persists indefinitely after cooling. Time and temperature for each image are provided below each crystal. (a) A single 7O orange crystal is
-
1
-1
heated at 2˚C min to form 7Y*. (b) A single 6O crystal is heated under a cross polarized microscope at 5˚C min to form 6Y*. (c) Car-
toon depiction of thermally induced changes occurring during the 6O to 6Y* transition in the corresponding images above, with yellow
coloring denoting complete transition to the yellow form through all layers of the crystal. Inside the cartoon in white, arrows show the
-1
direction of the wavefront movement and adjacent values representing the speed of the wavefront in µm sec . The angle of the wavefront
relative to the indicated long axis of the crystal is shown in blue.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 14
Uniquely, 1 was the only derivative studied that could only be
311+G**). A summary of dyad spectroscopic properties in solution,
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
grown as a single polymorph, and crystallized as bright orange nee-
dles (BO). When heated, 1BO crystals do not form a yellow material,
and sublime at 284˚C (Figure S3).
including quantum yields and fluorescence lifetimes, can be found in
Table S1 and Figure S4.
Color Switching Moves Most Quickly as a Wavefront Through
the Crystal. When orange crystals of 6, 7, and 8 are heated, they
transform to a yellow material (designated as Y* in the following
discussion) prior to melting. Examination of this thermochromic or-
ange-to-yellow conversion under a POM equipped with a heating
stage revealed that thermally generated yellow material retains the
overall microscopic morphology of the starting orange crystal (Figure
2
, Movie S1, and Movie S2).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Remarkably, numerous heating experiments of orange crystals un-
der a POM verified that the thermochromic transition proceeds as one
or more wavefronts moving through the orange crystal. It is assumed
that a structural transition is occurring at the orange-yellow wavefront
interface. The transition generally is observed to start at a crystal edge
or a small crack/defect in the crystals. The fastest moving orange-to-
yellow wavefronts proceed parallel to the long axis of the crystal, as
shown in Figure 2. The widest and fastest moving of these wave-
fronts move at speeds on the order of 12 µm s-1 and are generally
diagonal in nature, exhibiting an average angle of 38 ± 4° relative to
the long edge of the crystal (Figure 2c). When viewed from above the
crystal, horizontal wavefronts move parallel to either crystal edge and
proceed slower through the crystal, with speeds on the order of 0.2 –
Figure 4. Normalized absorbance and emission spectra for 7 in a
variety of solvents.
-1
0.5 µm s . Importantly, the slower horizontal waves have been ob-
served to “turn”, and as the angle becomes more pronounced the
wavefront speed also increases (bottom of crystal in Figure 2c, Mov-
ie S2). As the wavefront passes, the surface features and overall di-
mensions of the material only change slightly. For the 6O crystal
shown in Figure 2c, after heating from orange to yellow the average
dimensions of the crystal grew 2.038 µm (0.21%) in length, 2.049 µm
(
1.1%) in width, and 0.213 µm (2.8%) in height, all while maintaining
similar microscopic features.
Figure 5. Solid-state absorbance and emission for dyad deriva-
tives grown as crystals.
Figure 3. (a) Calculated dyad electrostatic potential, molecular
dipole, HOMO, and LUMO of 7 (wB97X-D, 6-311+G**). (b) 7
in solvents of increasing polarity from left to right: Hexanes, Tol-
uene, Ethyl Acetate, THF, DCM, Acetone, MeCN. Bottom: UV
irradiated samples, showing positive solvatochromism.
In thicker crystals, multiple wavefronts were generally seen, appar-
ently caused by multiple wavefronts moving through different hori-
zontal layers within the crystal. Thicker crystals exhibited more of
these waves. Thinner crystals appear to have only one or a few transi-
tion layers. Thicker crystals were also noted to deform or bend less
during the thermal transition as compared to thinner crystals, like the
7O crystal shown in Figure 2a that begins to bend upward in the
second-half of the transition.
Spectroscopic Characterization. In solution, all dyads had nearly
identical spectroscopic properties, as the alkyl side chains do not
affect electronic properties within the aromatic core. Dyads were
shown to exhibit significant solvatochromism (Figure 3), with a
strong bathochromic shift with solvents of increasing polarity (hex-
anes lEm-max = 444 nm, acetonitrile, lEm-max = 593 nm, Figure 4). This
shift is consistent with a highly charge-separated excited state due to
electron transfer from the electron rich MAN core to the electron
deficient NI core. This is further supported by 7 having a calculated
excited state dipole of 14.64 Debye in the first electronically excited
state, compared to just 7.21 Debye in the ground state (wB97X-D, 6-
Figure 6. Solid-state absorbance and emission comparing grown
yellow (Y) to thermally produced yellow (Y*) derivatives of 6
(top) and 7 (bottom).
ACS Paragon Plus Environment

Page 5 of 14
Journal of the American Chemical Society
Table 1. Solid-state absorbance and emission for dyad de-
isotropic phase generates the yellow material that is similar to Y* as
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
rivatives.
shown by successive heating-and-cooling cycles (Figure S6). Once
heated to Y*, this form persists indefinitely for each derivative, even
when cooled extremely slowly. For example, when 7O was heated to
produce 7Y*, the 7Y* form persisted even when cooled at the slowest
Compound
λ
Abs-Max (nm)
λEm-Max (nm)
-1
obtainable rate of 0.1˚C min (Figure S7).
1
6
7
BO
O
371
371
360
296
392
390
395
393
392
392
590
625
617
613
569
554
554
561
566
563
Orange-to-yellow and yellow-to-melt transition temperatures and
enthalpies for the dyads 6, 7, and 8 are presented in Table S2. In
general, when the respective orange crystals of similar size are heated
for the different derivatives, the longer the side chains, the larger the
enthalpy of transition for the O to Y* conversion. In addition, similar-
ly-sized yellow crystals grown from derivatives with longer side
chains melt to the isotropic phase with increasing enthalpies of transi-
tion ranging from 24.10 to 34.69 kJ mol- for 6Y-8Y, respectively
O
8O
6
7
8
6
Y
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Y
1
Y
(
Table S2). Importantly, compared to the grown yellow crystals (Y),
thermally generated Y* in general melts to the isotopic phase with
less energy as shown by comparatively lower transition enthalpies.
However, care must be taken when interpreting the differences be-
tween derivatives, as we observed that crystal size influences the
transition temperature and transition enthalpies measured for both the
O to Y transition and the Y* to the isotropic phase transition. For
example, as shown in Figure 7, larger 6O crystals (6OLarger) trans-
formed to Y* at higher temperatures, with larger enthalpies of transi-
tion, compared to smaller 6O crystals (6Osmaller). Melting of 6Y* to
the isotropic phase followed the same size dependent trend.
Y*
7Y*
Y*
8
For 6, 7, and 8 in the solid-state, orange and yellow crystals had
very similar absorbance and emission profiles. Orange crystals had
red-shifted absorbance and emission profiles relative to their yellow
counterparts (Figure 5). Note that the exact emission maxima of
orange crystals showed variability that appears correlated to crystal
size, with some observed red-shifting as crystal size increases (Table
Atomic Force Microscopy (AFM). Morphological changes ac-
companying the thermal conversion of O to Y* were investigated
using AFM. Non-contact AFM of a single 7O orange crystal reveals
clean, sharp edges (Figure 8). Thermal conversion of this single crys-
tal to 7Y* and subsequent imaging shows the general crystalline
shape is maintained following conversion, while the surface has nano-
scopic feature changes including the formation of numerous
ridges/steps.
1
).
For yellow crystals, 7Y and 8Y showed nearly identical absorbance
and emission spectra (Table 1). In comparison, the emission of 6Y
has a small bathochromic shift relative to 7Y and 8Y.
A comparison of solid-state spectroscopic properties between the
O, Y, and Y* forms for each 6 & 7, and 8 is shown in Figure 6 and
Figure S5, respectively. In general, Y and Y* have closely matching
absorption and emission spectra for all three of these derivatives. For
Non-contact AFM measurements of 7O microcrystals indicated the
presence of lamellar structures present with a step height of ~1.3 nm
(Figure 9), roughly the height of the aromatic core in the dyad mole-
cule. The general dimensions and lamellar structure are maintained
following heating, however, nanoscopic changes to the shape of indi-
vidual layers were observed (Figure 9). The differences in layer
shapes after heating are interpreted to indicate that the layers under-
went structural transitions somewhat independently from each other.
7
and 8, Y* emission profiles have a small bathochromic shift com-
pared to their Y counterparts, while 6Y* shows a small hypsochromic
shift in comparison to 6Y (Figure 6).
Structural Characterization. In addition to the single-crystal
34
structure of 8O previously reported, eight unique crystal structures
were solved for the four dyads in this study by single-crystal X-ray
diffraction. A summary of the crystallographic data for all dyad crys-
tal structures is present in Table S3.
Different crystallization methods were employed to grow either the
orange or yellow crystals as detailed in the Scheme S2. In general, for
a given dyad, orange crystals grew shorter and wider compared to
their longer, more needle-like yellow counterparts (Figure S2). Or-
ange polymorphs formed in solution were better diffracting, higher
quality crystals with smaller R-factors and no atoms with multiple site
occupancies. Yellow polymorphs had more disorder, especially in
alkyl side chains.
-1
Figure 7. Heated at 5˚C min , DSC curves show orange crystals
O) being thermally converted to a yellow material (Y*) prior to
(
melting to the isotropic phase (Iso). Values near peaks show the
corresponding peak temperature (˚C), bracketed values represent
integrated transition enthalpy [kJ mol ]. Larger 6O crystals are
seen to transition from orange-to-yellow at higher temperatures
and require more energy.
-1
Figure 8. Non-contact AFM generated 3D profiles of the same
crystal before and after heating, showing retention of overall
shape with nanoscopic surface changes. (a) Image of 7O crystal
before heating. (b) Image of the same crystal following thermal
conversion to 7Y*.
DSC Thermal Transition Characterization. DSC was used to
further characterize the orange-to-yellow transition for 6, 7 and 8.
The DSC traces of each derivative shows two endothermic peaks
corresponding to the O to Y* phase transition and the subsequent
melting of Y* to the isotropic phase (Figure 7). Cooling from the
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 14
Orange is Head-to-Head, Yellow is Head-to-Tail. For the deriva-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
tives 6, 7, and 8, multiple polymorphs were found. For these three
molecules, two general packing motifs were consistently observed:
orange crystals (O) were always observed to be stacked with the dyad
cores in a head-to-head geometry and yellow crystals (Y) were al-
ways observed to be stacked with the dyad cores in a head-to-tail
geometry (Figure 10). Two structures were solved for yellow hexyl
3
4
(
6Y and 6Y’) and heptyl (7Y and 7Y’) polymorphs, with each set of
polymorphs having similar packing of aromatic cores and the largest
differences being manifest in side-chain packing. Uniquely, and de-
spite examining dozens of growth conditions, 1 only crystallized as a
single polymorph (1BO), forming bright orange needles (Figure S2).
Detailed crystal packing figures for O, Y, and 1BO crystals are
shown in Figures S9, S10, and S11, respectively. It is important to
note, all of the solved crystal structures only contain the pure dyad;
solvent was never observed within the crystalline lattice.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Crystal Packing. In all crystals, the dyads pack to form 1-
dimensional columns formed from stacking of the aromatic dyad
cores. The spacing between aromatic cores is similar in all crystals,
with an average distance of 3.38 ± 0.02Å. The relative position of an
adjacent dyad within the same column can be described by a molecu-
lar slip about the long and short axes of the dyad (inset, Figure 11a).
As shown in Figure 11a, the O and Y polymorphs have distinctly
different slip-stacking geometries, with clustering among all of the O
and Y polymorphs, respectively.
All three O polymorphs have very similar slip-stacking, with a
long axis displacement of 3.07 ± 0.03Å and a short axis displacement
of 1.0 ± 0.1Å as illustrated by the overlay of crystal structure trimers
of 6O, 7O, and 8O in Figure 11b. In head-to-tail Y polymorphs, slip-
stacking can be calculated relative to either one of the two molecules
in the asymmetric unit, and hence, two values can be calculated that
describe the slip-stacking geometry for the Y polymorphs. In both
cases, there is similar slip-stacking within all Y polymorphs. While
1
BO stacks in a head-to-head geometry, it has a unique slip-stacking
Figure 9. Non-contact 3D AFM image of single 7 crystal before
and after heating showing nanoscopic surface changes following
heating. (a) 3D image of a single 7O crystal. (b) Step-height pro-
file of the blue trace (top) and the inset green trace (bottom) in the
adjacent 7O AFM image. (c) 3D image of the same region as
panel a) after conversion to 7Y*. (d) Step-height profile of the
black trace (top) and the inset yellow trace (bottom) in the adja-
cent 7Y* AFM image.
geometry with its largest offset along the short axis of the dyad, in
contrast to the head-to-head stacked O dyads (6O, 7O, 8O).
Figure 11. (a) Calculated dyad slip for O (squares), BO (trian-
gle), and Y (circles) dyad polymorphs. Inset: Adjacent dyads
within a column are offset with displacement along the long and
short axis of the dyad. (b) Overlay of 6O, 7O, and 8O single-
crystal structures shows the similarity in packing for all O crys-
tals. Hydrogen atoms omitted for clarity and crystallographic axes
inset.
In the three O polymorph crystals, aromatic cores in adjacent col-
umns are nearly perpendicular about the long axis of the dyad, with
an average intercolumnar angle of 87.4 ± 0.4˚ (Figure S9). In con-
trast, aromatic cores in adjacent columns of Y crystals are either par-
allel (6Y and 6Y’) or rotated about the short axis of the dyad between
4
8 – 49˚ (7Y, 7Y’, 8Y) (Figure S10).
The overall packing of molecules within the representative crystal
structures is shown in Figure 12. For both the orange and yellow
polymorphs, the long axis of the crystals correspond to what is most
likely to be the strongest intermolecular interactions, the electrostatic
attractions between aromatic dyad cores. The short axis of the crystal
is formed from alternating aromatic and aliphatic layers.
Figure 10. Single-crystal X-ray diffraction structures and cartoon
depiction of dyad polymorph packing. (a) Crystal structure of
1BO, and representative orange (7O) and yellow (7Y) crystal
structures shown as displacement ellipsoids set to the 50% proba-
bility level. (b) Cartoon representation of dyad head-to-head and
head-to-tail packing geometry.
ACS Paragon Plus Environment

Page 7 of 14
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 12. (a) Representation of 7 as shown in bilayer packing schematics. (b) Cartoon illustrating preferred orientation for both O and Y
polymorphs, where alternating alkyl and aromatic layers are parallel to the substrate. (c) Schematic of 7O packing. (d) Crystal structure of
7
O showing head-to-head dyad core stacking geometry, perpendicular columns relative to the long axis of the dyad molecule, and interdig-
itated side chains. (e) Schematic of 7Y packing. (f) Crystal structure of 7Y showing head-to-tail dyad core stacking geometry, adjacent
columns tilted 49˚ relative to the short axis of the dyad molecule, and segregated side chains. Select crystallographic planes of 7Y are
shown in magenta.
Aliphatic Layer Packing Differences. In all polymorphs, the dy-
ads assemble to create an alternating bilayer type structure, with al-
ternating alkyl and aromatic layers (Figure 12). However, the nature
of alkyl chain packing differs markedly between the O and Y poly-
morphs. Crystal structures of all O polymorphs have alkyl chains in
anti-staggered conformations, with side chains from adjacent layers
interdigitating, forming a close-packed zipper-like structure (Figure
12d). Alkyl side chains in Y crystals show more disorder and adopt
mixtures of staggered-anti and gauche conformations and have segre-
gated (i.e. not interdigitated) side chains (Figure 12f). Disorder in Y
side chains is seen with larger thermal displacement ellipsoids associ-
ated with side chains of Y crystals and multiple site occupancies for
one side chain in the asymmetric unit of 6Y, 7Y, and 7Y’ (Figure
S10, Table S3).
Thin Films. Dyad 7 was deposited onto quartz substrates using
physical vapor deposition and produced a yellow thin film. Vapor
annealing with DCM in a fuming chamber afforded a thin film of 7O
as discrete microcrystals that were characterized by structural and
spectroscopic measurements (Figure S12). GIWAXS of 7O thin
films (Figure 14) corroborated sample crystallinity and preferred
orientation seen in AFM measurements (Figure 9).
Importantly, in O crystals, the orientation of the dyads of adjacent
columns alternate as shown in Figure 12d. In other words, if a given
aromatic stack in an O crystal has the MAN units of the dyads orient-
ed “upwards” in the crystal, both adjacent columns within the same
layer will have the NI units oriented “upwards”, and vice versa.
The nearly perpendicular relationship between columns in all O
crystals appear to be stabilized through a combination of favorable
side-chain packing and electrostatic interactions between the most
polarized bonds in the dyads, namely the imide carbonyl C-O bonds
2
and the C-H bonds of alkoxy O-CH groups of the MAN units as
Figure 13. (a) Single-crystal structure of 7O illustrating intermo-
lecular C-H×××O hydrogen bonds (dashed green lines) between
adjacent columns. Terminal side-chain atoms removed for clarity.
shown by dashed green lines in Figure 13a. These proposed C-H×××O
hydrogen bonds create a zig-zag network of connected dyads in the
same layer of the crystal, and as seen in Figure 13, the molecules
linked in this way trace out an angle between 37 – 38°, relative to the
long axis of the aromatic stacks, in 6O, 7O, and 8O.
(
b) Cartoon depiction of typical orange crystal dimensions with
preferred orientation: crystal long-axis is made from dyads
stacked in columns, while the height of the crystal is made from
alternating aromatic and alkyl layers. Projecting out of the cartoon
is the 7O crystal structure as viewed from each unique crystal
face, with H-atoms omitted on the line structures for clarity.
Shown in space filling mode, six dyad molecules connected
through the intercolumnar C-H×××O hydrogen bonding in panel a.
Top Right in Magenta: When viewed from above the crystal, this
intermolecular H-bonding network creates an angle relative to the
long axis of the crystal of 37.5˚in 7O (shown in pink), and 37.2˚
in 6O.
Crystal Size Changes. The calculated density of 7O and both 7Y
polymorphs, as determined from single-crystal structures, are 1.25
-
3
and 1.22 g cm , respectively. Based on the differences in density of
O and either 7Y polymorph, a single-crystal-to-single-crystal transi-
7
tion from the more dense 7O to the less dense 7Y would theoretically
cause an increase in volume of ~2.5%. Optical profilometer meas-
urements of individual O crystals before and after heating to Y* were
used to quantify volume changes after the transition. Crystals of 7O
adsorbed onto a silicon substrate showed a 4.4 ± 0.5% increase in
crystal volume after heating 7O crystals to 7Y* (Table S4), similar to
that expected based on 7Y crystal structures.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 14
Molecular Orientation in the Crystals. The 7O polycrystalline
films were characterized by a dominant orientation preference for the
Structural Changes During Polymorphic Transitions in Poly-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
crystalline Samples. In an effort to elucidate the structural changes
that accompany the transition from O to Y, in-situ heating XRD ex-
periments were conducted. Numerous 6O and 7O crystals were
placed on a heating stage inside of a diffractometer and heated while
short transmission X-ray exposures (e.g. 30 seconds) were taken dur-
ing heating. The cross section of the X-ray beam is much larger than
all transition wavefronts seen under a POM, meaning that these expo-
sures are more of a bulk measurement that includes the O to Y* tran-
sition wavefront, as well as regions of growing Y* and shrinking O.
Despite these challenges, these measurements revealed that the transi-
tion from O to Y* proceed through an amorphous transition state. As
71
(
(
010) crystallographic axis to be normal to the quartz substrates
Figure 12d, 14a). The combination of GIWAXS scattering data and
non-contact AFM indicate the dyad formed a bilayer type lamellar
structure with alternating alkyl and aromatic layers (Figure 12b, 12c,
12e).
In agreement with the thermochromic conversion of a single crystal
of 7O to 7Y*, heating of a 7O polycrystalline thin film generated the
corresponding 7Y* thin film as confirmed by spectroscopic (Figure
S12) and GIWAXS measurements (Figure 14b). GIWAXS of the
7
Y* thin film reveals that the structure of 7Y* is not crystalline fol-
7
O is heated, the few spots that satisfy the Bragg condition are seen to
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
lowing thermal conversion, but is best described as a liquid-
crystalline mesophase. Figure 14b shows scattering of the 7Y* thin
film with in-plane diffractions that correspond to the (200) crystallo-
graphic axis perpendicular to the substrate. The other diffuse diffrac-
tion spots seen in the 7Y* scattering data match the expected location
weaken in intensity before disappearing. Quickly following this grad-
ual disappearance of 7O diffraction spots, diffraction spots from 7Y*
appear in rapid succession, with scattering from 7Y* lamellar spacing
being observed before scattering from dyad stacks is observed (Movie
S3, Figure S13).
72
calculated for the (20-2) and (40-2) crystallographic planes of 7Y.
As shown in Figure 12f, the (20-2) and (40-2) planes pass through
columns of the aromatic cores and observed scattering in these loca-
tions is interpreted to represent periodic order within columns. Taken
together, this structural data reveals that the 7Y* thin film has the
same lamellar structure as 7O, and that there is periodic order within
columns of dyads. An increase in amorphous scattering at larger scat-
tering angles in 7Y* thin films indicates that some disorder is present.
In-plane scattering from 7Y* thin films in the (h00) plane occurs at a
smaller scattering angle than would occur for 7Y, based on calculated
diffraction spots from 7Y crystal structure lattice parameters. This
smaller observed scattering angle of the (h00) planes in the 7Y* thin
film is indicative of a slightly larger (~3Å) lamellar spacing in 7Y* as
DISCUSSION
Polymorph Color Differences. Absorbance of visible light by dy-
ad molecules results in the intramolecular electron transfer from the
HOMO, centered largely on the electron rich MAN unit (see Figure
3
) to the LUMO, centered largely on the electron deficient NI unit.
This electron transfer is responsible for the striking solvatochromic
behavior of the dyads in solvents of different polarities and is ex-
pected for such highly polarized excited states.
The key structural difference in solid-state between O and Y poly-
morphs is the orientation of adjacent dyad molecules within a given
aromatic stack. Adjacent dyads in O polymorphs have offset head-to-
head stacking geometries, while Y polymorphs exhibit head-to-tail
stacking geometries. In other words, in the dyad stacks of O poly-
morphs, a given MAN unit is situated above and below MAN units of
adjacent dyads, and NI units are adjacent to NI units on adjacent mol-
ecules. In this head-to-head packing arrangement, there is a small
short-axis offset and a larger ~3Å offset along the long axis of the
molecule, where dyads shift unidirectionally in a column, much like
the offset of stairs in a staircase (Figure 11, S9). The offset head-to-
head stacking geometry in the O form will lead to mixing of LUMO
orbitals since orbital lobes of similar phase are in close proximity
6
1
compared to 7Y and is attributed to positive thermal expansion.
(
Figure S14). This LUMO mixing can facilitate the dissociation of an
initial locally excited state into charge-transfer states, where electrons
and holes are delocalized onto multiple molecules. This is believed to
be the basis for the red-shift seen in all O crystals. At the quantitative
level, differences in the precise geometry of LUMO orbital overlap
due to slightly different displacements along the short and long mo-
lecular axes can be used to explain subtle differences between the
optical properties of O polymorphs.
While dipolar coupling may be in play, it is not expected to be the
basis for the red-shift seen with orange crystals. A comparison of the
crystal structure of 1BO to all O polymorphs (Figure S11, S9, re-
spectively) shows that 1BO are also stacked in a head-to-head fash-
ion, however they are offset along the short molecular axis (Figure
1
1a). Compared to all O polymorphs, the smaller red-shift in 1BO is
attributed to a decrease in short-range coupling derived from orbital
overlap. The small observed red-shift in 1BO compared to all head-
to-tail Y polymorphs is believed to result from dipolar coupling with-
in head-to-head stacked 1BO crystals, while the short axis offset is
believed to prevent the favorable orbital overlap that resulted in the
larger red-shift observed in O polymorphs. All Y polymorphs exhibit
head-to-tail stacking geometries and show greater variation in slip-
stacking geometry compared to the O polymorphs (Figure 11a). The
absence of a red shift in Y polymorphs is interpreted to indicate that
MAN-NI dyads do not form CT complexes in head-to-tail packing
geometries, possibly due to energetic differences that minimize
HOMO-LUMO orbital mixing.
Figure 14. GIWAXS scattering pattern for thin films of 7 on a
quartz substrate before and after heating. (a) Scattering pattern for
7
O polycrystalline thin film. (b) Scattering pattern for 7Y* thin
film produced from heating the 7O thin film shown in panel a).
Inside each diffraction pattern, white circles show the calculated
diffraction spots (transmission) and corresponding Miller indices
based on lattice parameters from single-crystal structure data for
7O (a) and 7Y (b), with preferred orientation with the crystallo-
graphic long axis perpendicular to the substrate.
Orientations of Dyad Molecules in the Crystals. Through a
combination of transmission and GIWAXS experiments, done on
both single and polycrystalline samples, as well as prediction of crys-
tal morphology using Bravais, Friedel, Donnay, and Harker (BFDH)
8
ACS Paragon Plus Environment

Page 9 of 14
Journal of the American Chemical Society
calculations from single-crystal data, the orientation of the dyad mol-
dyads. Note in the structure of the orange material in Figure 12d, the
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ecule long axis in each polymorph was determined (Figure 12, 13).
The orange materials (6O, 7O, 8O) are structurally analogous to a
nice lasagna, with the relatively thick “noodle” layers comprised of
highly ordered, close-packed alkyl chains entirely in the staggered-
anti conformation, interdigitated with alkyl chains from both adjacent
layers. The “meat/cheese/sauce” layers have the head-to-head aro-
matic dyad units stacked with the direction of stacking parallel to the
long axis of the crystal (Figure 12c, 12d).
polarities of the adjacent head-to-head stacks of the dyads are exactly
inverted with respect to each other in alternating fashion, so that if a
given dyad molecule moves laterally into either adjacent stack in the
solid, it will essentially be switching from the head-to-head orienta-
tion in its own stack to a head-to-tail orientation on the adjacent stack!
Both Faster and Slower Wavefronts are Seen Slower transition
wavefronts move in variable directions, however all share the same
feature that they begin at an edge or crack and propagate towards the
interior of the crystal. The slowest transition wavefronts move parallel
to horizontal crystal boundaries (Figure 2c) while the fastest wave-
fronts move at an angle. Initiation of all thermal transitions and the
subsequent propagation of the slower transition wavefronts is be-
lieved to operate through a similar nucleation and growth mechanism.
A crystal edge provides opportunity for dyad translational movement
that does not exist in the packed crystal interior. Local melting of the
alkyl chain layer allows for mobility within layers, and lateral motion
allows for conformational sampling through translational movements
without disrupting the crystal morphology. Conformational sampling
on the crystal edges leads to nucleation of Y* that then templates
subsequent Y* growth within the same layer. As multiple nucleation
sites can emerge at different locations around single crystals, and
nucleation sights can have small differences in orientation, the transi-
tion creates a soft, polycrystalline Y* and contributes to the observa-
tion that no SCSC transition was observed, despite numerous at-
tempts. The darker “pixelated” region that makes up the majority of
Y* in the first four images shown Figure 2b is attributed to soft poly-
crystalline domains that formed from a nucleation and growth mecha-
nism, the major transition observed next to crystal edges and cracks
when dyads are not confined in all three crystallographic dimensions.
The yellow material is also structured like a molecular lasagna, but
with important differences compared to the orange polymorph. In the
yellow polymorph, the alkyl chains are conformationally less ordered
(no longer all staggered-anti) and are no longer interdigitated between
layers (Figure 12e, 12f). Similar to the orange material, however, the
head-to-tail stacked aromatic dyad units of the yellow material are
also stacked with the direction of stacking parallel to the long axis of
the crystal.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Note, that while the general features of yellow material produced in
various ways are consistent, especially with respect to dyad head-to-
tail packing geometries, there are some differences. Yellow crystals
grown directly from solution (Y) exhibit a high level of short- and
long-range order as indicated from single-crystal structures. Howev-
er, thermally generated yellow material derived from orange crystals
(
Y*) is best described as a soft-crystalline mesophase owing to some
long-range disorder not seen in the directly grown yellow crystals.
Transition Moves Through the Orange Crystal as a Wavefront.
We always observed the orange-to-yellow transition starting on either
end or at a crystal defect, with the majority of a given orange-to-
yellow transition sweeping along the long axis of the crystal (direc-
tion of stacking) as a wavefront, and different individual layers mov-
ing at slightly different rates (Movie S1, Movie S2, and Figure 2c).
The wavefronts from each end meet in the interior of the solid to
complete the conversion of the entire crystal. Using in situ heating
transmission XRD, we could not detect any intermediate states during
Why Do the Fastest Wavefronts Move at an Angle? Interesting-
ly, the fastest wavefronts move at an angle, and slower moving wave-
fronts speed up as an angled wavefront emerges. Because these fastest
wavefronts move in a single direction through the crystal at speeds 1
–
2 orders of magnitude faster than the slower wavefronts, the molec-
the transition process. In agreement with many proposed polymorphic
transition mechanisms,42 we believe the transition wavefront repre-
ular rearrangements resulting in the transition along the wavefront are
likely to be more or less concerted. A reasonable mechanistic hy-
pothesis for the orange to yellow transition also needs to explain the
observed 38 ± 4° of the wavefront relative to the long axis of the
crystal, noting that this angle is the same as the proposed strongest
intermolecular interactions between adjacent aromatic stacks in or-
ange crystals, the proposed C-H×××O hydrogen bonds (see Figure 13).
We therefore propose that once one of the molecules in the C-H×××O
hydrogen bonding network moves, the hydrogen bond to an adjacent
molecule is disrupted, allowing for more motion of that adjacent mol-
ecule so it moves, which disrupts the hydrogen bond to its adjacent
molecule, and so on right down the entire 38° line of hydrogen bond-
ed molecules at the wavefront. This wavefront then moves down the
crystal long axis, with a cooperative motion that turns the perpendicu-
lar dyad stacks parallel through a concerted “push”, where successive
dyads turn like individual pages when quickly flipping through a
book. To the best of our knowledge, the observation of both thermally
induced transition mechanisms operating in tandem and the switch
from a nucleation and growth mechanism on crystal edges to a coop-
erative mechanism in the crystal interior during a transition has not
previously been reported.
sents an amorphous phase boundary between O and Y* that enables
the conformational switching from O to Y* through this amorphous
wavefront interface. Interestingly, the fastest wavefronts move at an
approximately 38 ± 4° angle relative to the crystal long axis in what
behaves like a highly cooperative or concerted process.
Structural Transition “Trigger”. So, what triggers the structural
transition from the orange to the yellow polymorph prior to the crystal
melting entirely? We propose that upon heating of the orange crystal,
the structural transition “trigger” involves the local melting of the
interdigitated alkyl chain layer, which therefore becomes less ordered,
allowing for some motion of the alkyl chain layer, analogous to the
fluidity exhibited within biological membranes. This flexibility ena-
bles translational and/or rotational motion that has been seen in other
alkyl chain containing polymorphic materials6
4, 67-69
. This alkyl chain
layer fluidity that is the proposed triggering event does not disrupt the
overall crystal morphology (which is apparently still held together by
the stacked aromatics), but, importantly, we are assuming that this
fluidity does allow the dyad molecules to move laterally within the
solid to at least a limited extent.
Even with fluidity in the alkyl chain layer, how do all of the mole-
cules reorganize from head-to-head (O) to head-to-tail (Y*) as the
wavefront passes through the material without disrupting the overall
crystal morphology? All 6, 7, and 8 polymorphs have lamellar struc-
tures with consistent phase separation of alkyl and aromatic layers. A
vertical “flip” of a single dyad within an O stack would convert the
head-to-head packing into the head-to-tail packing seen in Y*. How-
ever, this vertical “flip” of the molecules is highly improbable as the
phase separation would be lost during a “flipping” transition. The
lowest energy transition pathway that maintains the microscopic mor-
phology of the crystal is thus expected to maintain this bilayer struc-
ture, with the majority of the reorganization happening within 2D
aromatic layers.
Thermodynamic Balancing Act. DSC measurements confirmed
that the orange material is the more stable at room temperature (and
below) presumably because of the combination of a reasonably stable
head-to-head stacking geometry, along with maximum van der Waals
interactions between the perfectly ordered and close-packed stag-
gered-anti, interdigitated alkyl chains. In the yellow form, stabilized
by slightly more favorable electrostatic interactions on the periphery
of the aromatic units, a more stable head-to-tail stacking geometry is
present, but the alkyl chains are no longer all staggered-anti, highly
ordered, close-packed, or interdigitated. Importantly, modeling indi-
cates that for geometric reasons, the dyad molecules cannot assemble
in the more stable head-to-tail aromatic stacking orientation of the
yellow solid and simultaneously accommodate the close-packed,
staggered-anti, interdigitated alkyl chain layer seen in the orange
crystal. The bottom line is that there is an energetic and structural
The head-to-head to head-to-tail stacking reorganization can readi-
ly occur with a limited lateral motion between adjacent layers of the
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 14
trade-off between the two forms; more stable side chain packing but
for this process appears to be that at lower temperature, the more
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
less stable aromatic stacking geometry in the orange form, yet more
stable aromatic stacking geometry with less favorable alkyl chain
packing in the yellow form. Affecting this balance, there is a domi-
nating influence of entropy so that temperature becomes determinant
of stability. As mentioned above, the more highly ordered orange
material is more stable at lower temperature because of larger surface
area of contact and therefore an overall increased van der Waals at-
traction between adjacent molecules. However, at higher temperature,
where DS becomes more dominant, the yellow material is now more
stable because of the more robust aromatic stacking geometry accom-
panied by a decrease in overall order (so an increase in intrinsic DS)
of the alkyl chains. Serendipitously, the temperature at which the
yellow form becomes more stable than the orange form occurs for our
dyad derivatives before the melting temperature of either the orange
or the yellow material is reached.
highly ordered orange form is lower in energy due to maximized van
der Waals contacts, while at higher temperature, the less ordered
yellow form is lower in energy dictated in part by the increased im-
portance of entropy at higher temperature. Once formed, the yellow
form becomes kinetically trapped and will persist in the absence of
solvent vapor. Remarkably, the orange-to-yellow transition is appar-
ently triggered at a temperature that is very close to the temperature at
which the orange and yellow forms exchange as the most stable form.
Just as remarkably, the transition occurs at a temperature that is lower
than the melting point of the original orange, or final yellow, solid.
Conclusion. Engineering molecular motion in the solid-state is an
emerging, but under investigated, area of supramolecular chemistry.
The payoff could be a new generation of smart solids, capable of a
range of easily visible responses to various stimuli. We now have a
clear understanding of the dynamic solid-state thermochromic behav-
ior of our MAN-NI dyads, and the complex interplay of intermolecu-
lar interactions, solid state structural architecture, and thermodynam-
ics that come together to enable the remarkable observed orange-to-
yellow color change without melting. Our MAN-NI system appears to
be the first example of a polymorphic system involving a nucleation
and growth mechanism that converts to a faster cooperative mecha-
nism within the crystal interior during a transition. The MAN-NI
system is also unique because it combines a dramatic visible color
change with only a minimal change in microscopic crystal morpholo-
gy, both of which will be important for potential practical applica-
tions. Finally, from a scientific curiosity perspective, it is also inter-
esting to understand how such a unique head-to-head to head-to-tail
stacking rearrangement could occur in a solid. We are using our
newfound understanding to design next-generation stimuli-responsive
solid-state systems that will incorporate tailored alkyl chain layers
with optimized C-H×××O hydrogen bonding networks to modulate both
transition temperatures and cooperativity in a defined way.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Significantly, once the transition wavefront has gone through the
heated crystal, cooling the solid does not bring back the orange mate-
rial because the alkyl chains do not have the opportunity to repack
and the material stays yellow. On the other hand, adding organic
solvent vapor allows the surface alkyl chains to reorder at room tem-
perature, regenerating the orange material only at the solid surface.
Once reorganized on the surface back to the orange structure, the
reorganization is propagated through the entire solid in a functional,
or possibly mechanistically, reverse of the process observed upon
heating.
Additional Evidence for Mechanism. One expected consequence
of the proposed mechanism is that each layer reorganizes more or less
independently. As the alkyl bilayer becomes more fluid, associations
between molecules in different layers are lost. In other words, once
alkyl chains are no longer interdigitated, the proposed mechanism
would predict that lateral diffusion of dyads and realignment to give
head-to-tail stacking can occur at different rates in adjacent layers.
Close inspection using optical analysis of extremely thin crystals
undergoing the orange-to-yellow transition indicated that indeed the
lower layers, closer to the heated surface, transitioned to yellow first,
apparently independently from layers farther from the heated surface,
that soon followed.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website.
The proposed mechanism would predict that the number of mole-
cules in each layer would remain the same before and after the orange
to yellow transition, but the morphology of each layer might change
slightly as the molecules can diffuse laterally to some extent. Such
minor changes in layer morphology, while maintaining overall layer
volume, was indeed seen in the high-resolution AFM images of Fig-
ure 9.
Experimental details and methods, Figures S1 – S14,
Tables S1 – S4, and other information (PDF)
Crystallographic data for 1BO (CIF)
Crystallographic data for 6O (CIF)
Crystallographic data for 7O (CIF)
Crystallographic data for 6Y (CIF)
Crystallographic data for 6Y’ (CIF)
Crystallographic data for 7Y (CIF)
Crystallographic data for 7Y’ (CIF)
Crystallographic data for 8Y (CIF)
Movie S1 (QT)
Finally, dyad 1 was synthesized before our detailed transition mod-
el was developed. Upon prolonged heating, crystals of dyad 1 remain
bright orange and do not undergo a thermally induced phase transition
prior to sublimation (Figure S3). In light of the proposed mechanism,
this is as predicted. The methyl groups of 1 do not form a highly or-
dered, interdigitated alky chain bilayer (Figure S11), so there can be
no triggering of a transition that would allow lateral motion of dyads
in the solid state, as required in the other derivatives as the orange-to-
yellow transition is triggered.
Movie S2 (QT)
Movie S3 (QT)
AUTHOR INFORMATION
Summary of Proposed Mechanism. To summarize, the orange
crystals of 6-8 adopt a lamellar, layered structure, alternating between
a highly ordered, interdigitated alkyl chain layer and stacked aromatic
dyad cores, with the stacks aligned along the long axis of the crystals.
Upon heating, the transition initiates when the alkyl chain layers be-
come disordered, allowing for some degree of lateral diffusion and
conformational sampling for dyads at a crystal phase boundary, with-
in their own layer. Moving to either adjacent stack in the same layer
allows a dyad to exchange a head-to-head stacking geometry (orange
form) for a head-to-tail stacking geometry (yellow form). The transi-
tion often initiates as a nucleation and growth mechanism that can
transform to a cooperative transition wavefront that propagates
through the interior of the crystal. The fastest moving wavefronts
have an approximately 38° angle with respect to the long axis of the
crystal, corresponding to a C-H×××O hydrogen bond network of dyad
molecules in adjacent stacks of the orange crystals. The driving force
Corresponding Author
*E-mail: iversonb@austin.utexas.edu
ORCID
Christopher D. Wight: 0000-0003-3389-1762
Qifan Xiao: 0000-0001-9867-553X
Holden R. Wagner: 0000-0001-8718-6474
Eduardo A. Hernandez: 0000-0002-6780-2302
Vincent M. Lynch: 0000-0002-5260-9913
Brent L. Iverson: 0000-0001-7974-3605
Author Contributions
The manuscript was written through contributions of all authors.
All authors have given approval to the final version of the manu-
script.
1
0
ACS Paragon Plus Environment

Page 11 of 14
Funding Sources
Journal of the American Chemical Society
Angewandte Chemie International Edition 2019, 58 (43), 15273-
15277.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Welch Foundation Grant #F-1188
9
.
Cozzi, F.; Cinquini, M.; Annuziata, R.; Siegel, J. S.,
Dominance of polar/.pi. over charge-transfer effects in stacked phenyl
interactions. Journal of the American Chemical Society 1993, 115
Notes
The authors declare no competing financial interest.
(
12), 5330-5331.
10. Hunter, C. A.; Sanders, J. K. M., The nature of .pi.-.pi.
interactions. Journal of the American Chemical Society 1990, 112
14), 5525-5534.
1. Rashkin, M. J.; Waters, M. L., Unexpected Substituent
ACKNOWLEDGMENT
(
We would like to thank Mr. Jon Bender for spectroscopic insight
and help calculating aromatic slip-stacking displacement, Ms.
Emily Raulerson for helpful conversations, and Mr. Daniel Man-
gel for help with excited-state TD-DFT calculations. Additional
special thanks to staff at the Texas Materials Institute (TMI), es-
pecially Dr. Raluca Gearba, Dr. Steve Swinnea, and Dr. Andrei
Dolocan as well as acknowledge access to the SAXSLAB Gane-
sha, supported by NSF MRI Grant No. CBET-1624659. CDW
would like to say a very special thanks to Dr. Richard Piner for
his expertise and help with AFM imaging. We would like to fur-
ther thank the UT NMR facility for the Bruker AVIII HD 500
1
Effects in Offset π−π Stacked Interactions in Water. Journal of the
American Chemical Society 2002, 124 (9), 1860-1861.
1
2.
Wheeler, S. E.; Houk, K. N., Substituent Effects in the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Benzene Dimer are Due to Direct Interactions of the Substituents with
the Unsubstituted Benzene. Journal of the American Chemical
Society 2008, 130 (33), 10854-10855.
13.
Wheeler, S. E.; Houk, K. N., Substituent Effects in
Cation/π Interactions and Electrostatic Potentials above the Centers of
Substituted Benzenes Are Due Primarily to Through-Space Effects of
the Substituents. Journal of the American Chemical Society 2009, 131
(9), 3126-3127.
(
NIH Grant No. 1 S10 OD021508-01) and the UT Mass Spec
facility for their valuable help. A special thanks to Prof. Nathaniel
Lynd for use of his group’s POM and DSC, as well as the incred-
ibly valuable technical help from Ms. Caitlin L. Bentley and Mr.
Aaron A. Burkey, as well as Prof. Sean T. Roberts for his useful
advice as well as the use of his Shimadzu UV-Vis spectrometer.
14.
Wheeler, S. E., Local Nature of Substituent Effects in
Stacking Interactions. Journal of the American Chemical Society
2011, 133 (26), 10262-10274.
15.
Bloom, J. W. G.; Wheeler, S. E., Taking the Aromaticity
out of Aromatic Interactions. Angewandte Chemie International
Edition 2011, 50 (34), 7847-7849.
16.
Wheeler, S. E., Understanding Substituent Effects in
ABBREVIATIONS
Noncovalent Interactions Involving Aromatic Rings. Accounts of
Chemical Research 2013, 46 (4), 1029-1038.
DAN,
1,5-dialkoxylnaphthalene;
NDI,
1,4,5,8-
naphthalenetetracarboxylic diimide; NI, naphthalene monoimide;
MAN, monoalkoxynaphthalene; LUMO, lowest unoccupied mo-
lecular orbital; HOMO, highest occupied molecular orbital; CT,
charge-transfer; XRD, X-ray diffraction; AFM, Atomic Force
Microscopy; POM, Polarized Optical Microscopy; DSC, Differ-
ential Scanning Calorimetry; PhMe, Toluene; EtOAc, Ethyl Ace-
tate; DCM, Dichloromethane; MeCN, Acetonitrile; THF, Tetra-
hydrofuran; GIWAXS, grazing-incidence wide-angle X-ray scat-
tering; Bravais, Friedel, Donnay, and Harker (BFDH).
1
7.
Wheeler, S. E.; Bloom, J. W. G., Toward a More Complete
Understanding of Noncovalent Interactions Involving Aromatic
Rings. The Journal of Physical Chemistry A 2014, 118 (32), 6133-
6147.
18.
Carroll, W. R.; Pellechia, P.; Shimizu, K. D., A Rigid
Molecular Balance for Measuring Face-to-Face Arene−Arene
Interactions. Organic Letters 2008, 10 (16), 3547-3550.
1
9.
Hwang, J.; Li, P.; Carroll, W. R.; Smith, M. D.; Pellechia,
P. J.; Shimizu, K. D., Additivity of Substituent Effects in Aromatic
Stacking Interactions. Journal of the American Chemical Society
2
014, 136 (40), 14060-14067.
0. Hwang, J. w.; Li, P.; Shimizu, K. D., Synergy between
2
REFERENCES
1.
Rings in Chemical and Biological Recognition: Energetics and
Structures. Angewandte Chemie International Edition 2011, 50 (21),
experimental and computational studies of aromatic stacking
interactions. Organic & Biomolecular Chemistry 2017, 15 (7), 1554-
1564.
Salonen, L. M.; Ellermann, M.; Diederich, F., Aromatic
2
1.
into a pleated secondary structure in solution. Nature 1995, 375
6529), 303.
2. Nguyen, J. Q.; Iverson, B. L., An Amphiphilic Folding
Lokey, R. S.; Iverson, B. L., Synthetic molecules that fold
4
808-4842.
2. Chen, T.; Li, M.; Liu, J., π–π Stacking Interaction: A
(
Nondestructive and Facile Means in Material Engineering for
Bioapplications. Crystal Growth & Design 2018, 18 (5), 2765-2783.
2
Molecule That Undergoes an Irreversible Conformational Change.
Journal of the American Chemical Society 1999, 121 (11), 2639-
3
.
Thakuria, R.; Nath, N. K.; Saha, B. K., The Nature and
Applications of π–π Interactions: A Perspective. Crystal Growth &
Design 2019, 19 (2), 523-528.
2
640.
3.
Characterization
Dialkoxynaphthalene/1,4,5,8-Naphthalenetetracarboxylic
2
Zych, A. J.; Iverson, B. L., Synthesis and Conformational
4.
Exploiting non-covalent π interactions for catalyst design. Nature
017, 543 (7647), 637-646.
Jung, H.; Schrader, M.; Kim, D.; Baik, M.-H.; Park, Y.;
Neel, A. J.; Hilton, M. J.; Sigman, M. S.; Toste, F. D.,
of
Tethered,
Self-Complexing
1,5-
Diimide
2
Systems. Journal of the American Chemical Society 2000, 122 (37),
898-8909.
4. Cubberley, M. S.; Iverson, B. L., 1H NMR Investigation of
5
.
8
Chang, S., Harnessing Secondary Coordination Sphere Interactions
That Enable the Selective Amidation of Benzylic C–H Bonds.
Journal of the American Chemical Society 2019, 141 (38), 15356-
2
Solvent Effects in Aromatic Stacking Interactions. Journal of the
American Chemical Society 2001, 123 (31), 7560-7563.
1
5366.
25.
Gabriel, G. J.; Iverson, B. L., Aromatic Oligomers that
6
.
Yuan, C.; Ji, W.; Xing, R.; Li, J.; Gazit, E.; Yan, X.,
Form Hetero Duplexes in Aqueous Solution. Journal of the American
Chemical Society 2002, 124 (51), 15174-15175.
Hierarchically oriented organization in supramolecular peptide
crystals. Nature Reviews Chemistry 2019, 3 (10), 567-588.
7.
Q.; Li, X.; Xu, L.; Ding, H.-M.; Yang, H.-B., Radical-Induced
Hierarchical Self-Assembly Involving Supramolecular Coordination
Complexes in Both Solution and Solid States. Journal of the
American Chemical Society 2019, 141 (40), 16014-16023.
2
6.
Zych, A. J.; Iverson, B. L., Conformational Modularity of
Huo, G.-F.; Shi, X.; Tu, Q.; Hu, Y.-X.; Wu, G.-Y.; Yin, G.-
an Abiotic Secondary-Structure Motif in Aqueous Solution. Helvetica
Chimica Acta 2002, 85 (10), 3294-3300.
2
7.
Gabriel, G. J.; Sorey, S.; Iverson, B. L., Altering the
Folding Patterns of Naphthyl Trimers. Journal of the American
Chemical Society 2005, 127 (8), 2637-2640.
8.
Su, F.; Chen, G.; Korevaar, P. A.; Pan, F.; Liu, H.; Guo, Z.;
2
8.
Bradford, V. J.; Iverson, B. L., Amyloid-like Behavior in
Schenning, A. P. H. J.; Zhang, H.-J.; Lin, J.; Jiang, Y.-B., Discrete π-
Stacks
Abiotic, Amphiphilic Foldamers. Journal of the American Chemical
Society 2008, 130 (4), 1517-1524.
from
Self-Assembled
Perylenediimide
Analogues.
1
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 14
2
9.
Peebles, C.; Piland, R.; Iverson, B. L., More than Meets the
induced reversible solid-state colour change of an intramolecular
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Eye: Conformational Switching of a Stacked Dialkoxynaphthalene–
Naphthalenetetracarboxylic diimide (DAN–NDI) Foldamer to an
NDI–NDI Fibril Aggregate. Chemistry - A European Journal 2013,
charge-transfer complex. Chemical Communications 2015, 51 (79),
14809-14812.
48.
Dharmarwardana, M.; Arimilli, B. S.; Luzuriaga, M. A.;
1
9 (35), 11598-11602.
0. Reczek, J. J.; Villazor, K. R.; Lynch, V.; Swager, T. M.;
Kwon, S.; Lee, H.; Appuhamillage, G. A.; McCandless, G. T.;
Smaldone, R. A.; Gassensmith, J. J., The thermo-responsive behavior
in molecular crystals of naphthalene diimides and their 3D printed
thermochromic composites. CrystEngComm 2018, 20 (39), 6054-
6060.
3
Iverson, B. L., Tunable Columnar Mesophases Utilizing C2
Symmetric Aromatic Donor−Acceptor Complexes. Journal of the
American Chemical Society 2006, 128 (24), 7995-8002.
3
1.
Reczek, J. J.; Iverson, B. L., Using Aromatic Donor
49.
Yuan, T.; Vazquez, M.; Goldner, A. N.; Xu, Y.; Contrucci,
Acceptor Interactions to Affect Macromolecular Assembly.
Macromolecules 2006, 39 (17), 5601-5603.
R.; Firestone, M. A.; Olson, M. A.; Fang, L., Versatile
Thermochromic Supramolecular Materials Based on Competing
Charge Transfer Interactions. Advanced Functional Materials 2016,
26 (47), 8604-8612.
3
2.
Alvey, P. M.; Reczek, J. J.; Lynch, V.; Iverson, B. L., A
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Systematic Study of Thermochromic Aromatic Donor−Acceptor
Materials. The Journal of Organic Chemistry 2010, 75 (22), 7682-
50.
Yuan, T.; Xu, Y.; Zhu, C.; Jiang, Z.; Sue, H.-J.; Fang, L.;
7
690.
33.
Olson, M. A., Tunable Thermochromism of Multifunctional Charge-
Transfer-Based Supramolecular Materials Assembled in Water.
Chemistry of Materials 2017, 29 (23), 9937-9945.
Peebles, C.; Alvey, P. M.; Lynch, V.; Iverson, B. L., Time-
Dependent Solid-State Polymorphism of a Series of Donor–Acceptor
Dyads. Crystal Growth & Design 2014, 14 (1), 290-299.
51.
Xu, Y.; Yuan, T.; Nour, H. F.; Fang, L.; Olson, M. A., Bis-
3
4.
Peebles, C.; Wight, C. D.; Iverson, B. L., Solution- and
Bipyridinium Gemini Surfactant-Based Supramolecular Helical
Fibers and Solid State Thermochromism. Chemistry – A European
Journal 2018, 24 (62), 16558-16569.
solid-state photophysical and stimuli-responsive behavior in
conjugated monoalkoxynaphthalene-naphthalimide donor-acceptor
dyads. Journal of Materials Chemistry C 2015, 3 (46), 12156-12163.
52.
Le, A. K.; Bender, J. A.; Arias, D. H.; Cotton, D. E.;
3
5.
Shao, H.; Nguyen, T.; Romano, N. C.; Modarelli, D. A.;
Johnson, J. C.; Roberts, S. T., Singlet Fission Involves an Interplay
between Energetic Driving Force and Electronic Coupling in
Perylenediimide Films. Journal of the American Chemical Society
2018, 140 (2), 814-826.
Parquette, J. R., Self-Assembly of 1-D n-Type Nanostructures Based
on Naphthalene Diimide-Appended Dipeptides. Journal of the
American Chemical Society 2009, 131 (45), 16374-16376.
36.
Shao, H.; Parquette, J. R.,
A
[small pi]-conjugated
53.
Xue, S.; Qiu, X.; Sun, Q.; Yang, W., Alkyl length effects
hydrogel based on an Fmoc-dipeptide naphthalene diimide
semiconductor. Chemical Communications 2010, 46 (24), 4285-4287.
on solid-state fluorescence and mechanochromic behavior of small
organic luminophores. Journal of Materials Chemistry C 2016, 4 (8),
1568-1578.
3
7.
H- and J-aggregates.(Report). Accounts of Chemical Research 2010,
3 (3), 429-439.
38. Varghese, S.; Das, S., Role of Molecular Packing in
Spano, F. C., The spectral signatures of Frenkel polarons in
54.
Jin, M.; Seki, T.; Ito, H., Mechano-Responsive
4
Luminescence via Crystal-to-Crystal Phase Transitions between
Chiral and Non-Chiral Space Groups. Journal of the American
Chemical Society 2017, 139 (22), 7452-7455.
Determining Solid-State Optical Properties of π-Conjugated
Materials. The Journal of Physical Chemistry Letters 2011, 2 (8),
55.
Jin, M.; Sumitani, T.; Sato, H.; Seki, T.; Ito, H.,
and Solvent-Vapor-Induced
8
63-873.
9.
Mechanical-Stimulation-Triggered
3
Hestand, N. J.; Spano, F. C., Molecular Aggregate
Reverse Single-Crystal-to-Single-Crystal Phase Transitions with
Alterations of the Luminescence Color. Journal of the American
Chemical Society 2018, 140 (8), 2875-2879.
Photophysics beyond the Kasha Model: Novel Design Principles for
Organic Materials. Accounts of Chemical Research 2017, 50 (2), 341-
3
50.
0.
56.
Cruz-Cabeza, A. J.; Reutzel-Edens, S. M.; Bernstein, J.,
4
Gentili, D.; Gazzano, M.; Melucci, M.; Jones, D.;
Facts and fictions about polymorphism. Chemical Society Reviews
2015, 44 (23), 8619-8635.
Cavallini, M., Polymorphism as an additional functionality of
materials for technological applications at surfaces and interfaces.
Chemical Society Reviews 2019, 48 (9), 2502-2517.
57.
Anwar, J.; Tuble, S. C.; Kendrick, J., Concerted Molecular
Displacements in a Thermally-Induced Solid-State Transformation in
Crystals of DL-Norleucine. Journal of the American Chemical
Society 2007, 129 (9), 2542-2547.
4
1.
Seki, T.; Ito, H., Molecular-Level Understanding of
Structural Changes of Organic Crystals Induced by Macroscopic
Mechanical Stimulation. Chemistry – A European Journal 2016, 22
58.
Beckham, G. T.; Peters, B.; Starbuck, C.; Variankaval, N.;
(
13), 4322-4329.
42. Anwar, J.; Zahn, D., Polymorphic phase transitions:
Trout, B. L., Surface-Mediated Nucleation in the Solid-State
Polymorph Transformation of Terephthalic Acid. Journal of the
American Chemical Society 2007, 129 (15), 4714-4723.
Macroscopic theory and molecular simulation. Advanced Drug
Delivery Reviews 2017, 117, 47-70.
59.
Yao, Z.-S.; Guan, H.; Shiota, Y.; He, C.-T.; Wang, X.-L.;
4
3.
Herbstein, F., On the mechanism of some first-order
Wu, S.-Q.; Zheng, X.; Su, S.-Q.; Yoshizawa, K.; Kong, X.; Sato, O.;
Tao, J., Giant anisotropic thermal expansion actuated by
thermodynamically assisted reorientation of imidazoliums in a single
crystal. Nature Communications 2019, 10 (1), 4805.
enantiotropic solid-state phase transitions: from Simon through
Ubbelohde to Mnyukh. Acta Crystallographica Section B 2006, 62
(3), 341-383.
4
4.
Zhao, W.; He, Z.; Peng, Q.; Lam, J. W. Y.; Ma, H.; Qiu,
60.
Duan, Y.; Semin, S.; Tinnemans, P.; Cuppen, H.; Xu, J.;
Z.; Chen, Y.; Zhao, Z.; Shuai, Z.; Dong, Y.; Tang, B. Z., Highly
sensitive switching of solid-state luminescence by controlling
intersystem crossing. Nature Communications 2018, 9 (1), 3044.
Rasing, T., Robust thermoelastic microactuator based on an organic
molecular crystal. Nature Communications 2019, 10 (1), 4573.
61.
Das, D.; Jacobs, T.; Barbour, L. J., Exceptionally large
45.
Yagai, S.; Seki, T.; Aonuma, H.; Kawaguchi, K.; Karatsu,
positive and negative anisotropic thermal expansion of an organic
crystalline material. Nature Materials 2010, 9 (1), 36-39.
T.; Okura, T.; Sakon, A.; Uekusa, H.; Ito, H., Mechanochromic
Luminescence Based on Crystal-to-Crystal Transformation Mediated
by a Transient Amorphous State. Chemistry of Materials 2016, 28 (1),
62.
Takamizawa, S.; Miyamoto, Y., Superelastic Organic
Crystals. Angewandte Chemie International Edition 2014, 53 (27),
6970-6973.
2
34-241.
46.
Yagai, S.; Okamura, S.; Nakano, Y.; Yamauchi, M.;
63.
Yao, Z.-S.; Mito, M.; Kamachi, T.; Shiota, Y.; Yoshizawa,
Kishikawa, K.; Karatsu, T.; Kitamura, A.; Ueno, A.; Kuzuhara, D.;
Yamada, H.; Seki, T.; Ito, H., Design amphiphilic dipolar π-systems
for stimuli-responsive luminescent materials using metastable states.
Nature Communications 2014, 5, 4013.
K.; Azuma, N.; Miyazaki, Y.; Takahashi, K.; Zhang, K.; Nakanishi,
T.; Kang, S.; Kanegawa, S.; Sato, O., Molecular motor-driven abrupt
anisotropic shape change in a single crystal of a Ni complex. Nature
Chemistry 2014, 6 (12), 1079-1083.
47.
Li, P.; Maier, J. M.; Hwang, J.; Smith, M. D.; Krause, J.
64.
Su, S.-Q.; Kamachi, T.; Yao, Z.-S.; Huang, Y.-G.; Shiota,
A.; Mullis, B. T.; Strickland, S. M. S.; Shimizu, K. D., Solvent-
Y.; Yoshizawa, K.; Azuma, N.; Miyazaki, Y.; Nakano, M.; Maruta,
1
2
ACS Paragon Plus Environment

Page 13 of 14
Journal of the American Chemical Society
G.; Takeda, S.; Kang, S.; Kanegawa, S.; Sato, O., Assembling an
alkyl rotor to access abrupt and reversible crystalline deformation of a
and function memory effect in organic semiconductors. Nature
Communications 2018, 9 (1), 278.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
cobalt(II) complex. Nature Communications 2015, 6 (1), 8810.
69.
Chung, H.; Chen, S.; Sengar, N.; Davies, D. W.; Garbay,
6
5.
Thirumalai, R.; Mukhopadhyay, R. D.; Praveen, V. K.;
G.; Geerts, Y. H.; Clancy, P.; Diao, Y., Single Atom Substitution
Alters the Polymorphic Transition Mechanism in Organic Electronic
Crystals. Chemistry of Materials 2019, 31 (21), 9115-9126.
Ajayaghosh, A., A slippery molecular assembly allows water as a
self-erasable security marker. Scientific Reports 2015, 5 (1), 9842.
66.
Karothu, D. P.; Weston, J.; Desta, I. T.; Naumov, P.,
70.
Seki, T.; Mashimo, T.; Ito, H., Anisotropic strain release in
Shape-Memory and Self-Healing Effects in Mechanosalient
Molecular Crystals. Journal of the American Chemical Society 2016,
a thermosalient crystal: correlation between the microscopic
orientation of molecular rearrangements and the macroscopic
mechanical motion. Chemical Science 2019, 10 (15), 4185-4191.
71.
L. J.; Toney, M. F., Molecular Characterization of Organic Electronic
Films. Advanced Materials 2011, 23 (3), 319-337.
72.
incidence X-ray scattering data visualization and reduction, and
indexing of buried three-dimensional periodic nanostructured films.
Journal of Applied Crystallography 2015, 48 (3), 917-926.
1
38 (40), 13298-13306.
7. Abe, Y.; Savikhin, V.; Yin, J.; Grimsdale, A. C.; Soci, C.;
6
DeLongchamp, D. M.; Kline, R. J.; Fischer, D. A.; Richter,
Toney, M. F.; Lam, Y. M., Unique Reversible Crystal-to-Crystal
Phase Transition—Structural and Functional Properties of Fused
Ladder Thienoarenes. Chemistry of Materials 2017, 29 (18), 7686-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Jiang, Z., GIXSGUI: a MATLAB toolbox for grazing-
7
696.
8.
6
Chung, H.; Dudenko, D.; Zhang, F.; D’Avino, G.; Ruzié,
C.; Richard, A.; Schweicher, G.; Cornil, J.; Beljonne, D.; Geerts, Y.;
Diao, Y., Rotator side chains trigger cooperative transition for shape
1
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 14
Insert Table of Contents artwork here
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
4
ACS Paragon Plus Environment