Rapidly Reversible Organic Crystalline Switch for Conversion of Heat into Mechanical Energy

Article abstract of DOI:10.1021/jacs.1c01549

Solid-state thermoelastic behavior - a sudden exertion of an expansive or contractive physical force following a temperature change and phase transition in a solid-state compound - is rare in organic crystals, few are reversible systems, and most of these

Full text of DOI:10.1021/jacs.1c01549

pubs.acs.org/JACS
Article
Rapidly Reversible Organic Crystalline Switch for Conversion of Heat
into Mechanical Energy
Madushani Dharmarwardana, Srimanta Pakhira, Raymond P. Welch, Carlos Caicedo-Narvaez,
Michael A. Luzuriaga, Bhargav S. Arimilli, Gregory T. McCandless, Babak Fahimi,
Jose L. Mendoza-Cortes,* and Jeremiah J. Gassensmith*
Cite This: J. Am. Chem. Soc. 2021, 143, 5951−5957
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Solid-state thermoelastic behaviora sudden exer-
tion of an expansive or contractive physical force following a
temperature change and phase transition in a solid-state com-
poundis rare in organic crystals, few are reversible systems, and
most of these are limited to a dozen or so cycles before the crystal
degrades or they reverse slowly over the course of many minutes or
even hours. Comparable to thermosalience, wherein crystal phase
changes induce energetic jumping, thermomorphism produces
physical work via consistent and near-instantaneous predictable
directional force. In this work, we show a fully reversible
thermomorphic actuator that is stable at room temperature for
multiple years and is capable of actuation for more than 200 cycles at
near-ambient temperature. Specifically, the crystals shrink to 90% of their original length instantaneously upon heating beyond 45 °C
and expand back to their original length upon cooling below 35 °C. Furthermore, the phase transition occurs instantaneously, with
little obvious hysteresis, allowing us to create real-time actuating thermal fuses that cycle between on and off rapidly.
than the best pure metals and undergo phase transitions at
rates exceeding 105 times faster than non-thermosalient
transitions.9,11 This expansion is caused by a diffusionless
phase transition that is initiated at crystalline extremities and
rapidly propagates throughout the crystal.12,13 In some of these
systems, expansion can cause a sudden single-crystal-to-single-
crystal (SCSC) phase transition with substantial changes in
unit cell size, thus providing even greater material expan-
sion.14−23 In many of these SCSC thermosalient events, the
transitions are typically irreversible and result in their
destruction by splitting,24,25 jumping,26,27 or even dramatically
exploding.28 These irreversible dynamic crystals have found
applications as fuses,29 making use of their one-time ability to
actuate to break a circuit; however, this irreversibility greatly
limits their applications. In contrast to irreversible systems,
there are very few thermally actuated reversible systems,30,31
and most of these are limited to a dozen or so cycles before the
crystal degrades or they reverse slowly over the course of many
minutes or even hours. Designing a thermoelastic system that
INTRODUCTION
■
Materials that can transduce heat into well-defined and
directed motion are rare but have a wide range of applications
for a diverse array of technologies from microscopic pumps to
energy harvesting. For instance, transducing well-defined
thermally induced reciprocating actuation simply by taking
advantage of the natural temperature fluctuation between day
and night cycles is a possible way to harness ambient
temperature changes into useful pollution-free electricity. In
addition, the ever-increasing interest in the miniaturization of
mechanical devices demands new actuating materials that can
produce motion in the micrometer to millimeter regime. While
actuators that reversibly operate from electrochemical1,2 and
photochemical3 stimuli are known, the application of thermal
expansion (thermoelastic) actuators has been limited by poor
mechanical properties, material fatigue, or low coefficients of
thermal expansion (CTEs). For instance, metallic aluminum
and lead, which have the highest known CTEs in pure metals,
are still in the regime of 20 × 10−6 K−1. Thermoresponsive
actuating shape-memory polymers have been developed,4−7
but they typically require a significant change in the global
environmental temperature to actuate.8
Received: February 8, 2021
Published: April 6, 2021
Dynamic crystals are mechanically active single-crystalline
entities capable of amplifying small changes in molecular
packing into large-scale anisotropic macroscopic motion.9,10
These materials have CTEs over an order of magnitude higher
https://doi.org/10.1021/jacs.1c01549
J. Am. Chem. Soc. 2021, 143, 5951−5957
© 2021 American Chemical Society
5951

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 1. (a) Structural formula of the naphthalene diimide (NDI) systems, which all show thermoelastic behavior. When the alkyl chain is 4−9
carbons long, the crystals are thermosalient. This work shows that when the alkyl chain is 10 carbons long (DNDI), the fluctuation becomes
reversible and thermomorphic. Overlap of both crystal phases at high (red) and low (blue) temperatures and (b) down the crystallographic a-axis,
where the expansion along the b-axis and c-axis from packing is illustrated. (c) Zoom-in into the ring overlaps down the a-axis shows the NDI core
and phenoxyl ring overlap becomes slip stacked at high temperatures and (d) down the crystallographic b-axis, where the expansion at low
temperature is obviated. (e) Optical micrographs of a DNDI crystal at room temperature (top), heated (middle), and an overlay illustrating the
length difference (bottom). (f) Graph showing crystal length changes of a single crystal over 34 thermal cycles from cold (green squares) to hot
(orange circles) and back with percent length change (red triangles).
has the rapid expansive force of a thermosalient solid yet can
predictably actuate the same way at hysteresis free rates is
complicated by the fact that structure−function relationships
in molecular crystals can be difficult to derive, and thus a priori
synthesis of actuating materials is challenging.
chain length.32 We hypothesize that longer chains and smaller
ring overlaps might flatten the potential energy curve, thus
facilitating the transition and improving the reversibility.
Accordingly, we observe that this NDI family exhibits
thermosalience and/or thermochromism across its many
derivatives. Thermosalience is a special case of thermoelastic
behavior where the release of accrued elastic stress causes
dramatic and often irreversible changes in crystal dimensions.34
Upon reaching ten carbons (Decoxyphenyl-NDI, or DNDI for
short), we find the crystals are no longer thermosalient, yet
they remain thermoelastic, with exceptional rates of thermal
cycling and measurable expansion force. What is more, the
transition temperature is around 45 °C, a temperature range
better suited for near everyday temperature applications than
other reversibly thermosalient crystals, which actuate at >80
°C35,36 or lower than 0 °C.37
In this report, we introduce a solid state rapidly actuating
organic crystal that undergoes a significant and sudden change
in size following a SCSC transition with considerable and
measurable force. We find it is a fully reversible actuator
capable of more than 200 cycles at room temperature that
remains stable on the benchtop for at least 3 years. Like
thermosalient crystals, the phase transition occurs instanta-
neously, with little evident hysteresis, allowing us to create real-
time thermomechanical actuators that cycle an electrical circuit
on and off rapidly. We have previously discovered a family of
thermosalient molecular crystals based on a naphthalene
diimide (NDI) core with alkoxyphenyl substituents (Figure
1a).12 We have reported that the single-action thermosalient
behavior is tunable across temperature ranges by altering the
alkyl chain from four to nine carbons.32 Here, we report that
when the alkyl chain is ten linear carbons, the crystals become
thermomorphicthe crystals predictably, reversibly, and
instantaneously actuate much like a thermosalient system but
in a single direction via negative thermal expansion along one
axis. Specifically, the crystals shrink to 90% of their original
length upon heating beyond 45 °C and expand back to their
original length upon cooling below 35 °C. This temperature
regime is important because it occurs at just the minimum
temperature that can cause a burn in a finite amount of time,
giving rise to possible bioengineering applications.33
DNDI was synthesized based on literature protocols,12,32
substituting longer alkyl chains, and crystallized from the slow
evaporation of chloroform. A detailed synthesis is also available
in the Supporting Information. These crystals were examined
with single-crystal X-ray diffraction at temperatures above and
below the transition point, and the single-crystal structure was
solved across the operating temperature range. The crystallo-
graphic parameters are summarized in Table 1. Structural
analysis indicates that the molecular arrangement within the
lattice positions align the NDI cores parallel to each other with
the alkyl chains fully extended away, forming layers of ribbon-
like planes (Figure 1b−d). The centroid-to-centroid dis-
tancesdefined by the 16 atoms of the rylene center
between the NDI cores shorten considerably with heating
despite little change in the interplanar distances. This is
because the NDI centers slide closer together when the
crystal’s large axis of thermal contraction is arranged
horizontally (Figure 1c). This is enabled by a change in the
aromatic ring overlap, which causes a contraction of the linkers
RESULTS AND DISCUSSION
■
After our discoveries that butoxyphenyl-NDI crystals12 exhibit
irreversible thermosalience and reversible thermochromism, we
expanded our search to other NDI derivatives based on alkyl
5952
https://doi.org/10.1021/jacs.1c01549
J. Am. Chem. Soc. 2021, 143, 5951−5957

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Table 1. Crystallographic Parameters at 298 and 338 K
a
parameter
298 K
338 K
crystal system
space group
formula
triclinic
P1
C46H54N2O6
̅
unit cell
a (Å)
b (Å)
5.232 (4)
7.643 (5)
4.6739 (19)
7.602 (2)
c (Å)
25.120 (18)
86.887 (19)
87.346 (19)
80.28 (2)
29.218 (10)
98.780 (15)
88.17 (2)
92.12 (6)
1024.9 (6)
α (deg)
β (deg)
γ (deg)
volume (Å3)
crystal color
Z
987.8 (12)
yellow
1
1
a
To aid in direct comparison between these two polymorphs, the unit
cell parameters of the 338 K data have to be converted by applying a 3
× 3 transformation. See Table S1 for full details.
as temperature decreases. Upon heating past the transition
point of 45 °C, the crystal shrinks to ∼90% of its cold length in
the long direction (Figure 1e). Put differently, the DNDI
crystals were observed to have sudden negative thermal
expansion at 45 °C. The width increases slightly, and there
is no apparent change in thickness. The hot and cold lengths
remain relatively stable (Figure 1f) following multiple cycles
without loss in compressibility (Movie S1). Single-crystal
analysis shows that the cool and warm structures are two
different crystal habits, both of which are triclinic, and differ
only modestly in unit cell lengths and angles. This small
change in packing may explain why the crystals can cycle
without substantial degradation.
We conducted differential scanning calorimetry on bulk
DNDI crystals cycling the temperature from 20 to 60 °C and
back for 200 cycles (Figure 2a). The exothermic and
endothermic peaks show no shift in temperature, width, or
height over the cycles, indicating that the bulk molecular
rearrangement is not attenuated. The recovered crystals were
apparently thinner, having slightly delaminated in the DSC
though the recovered crystals continued to actuate at the same
temperatures with the same 10% decrease in length. One such
recovered crystal is shown in the inset of Figure 2a.
DSC analysis coupled with computational efforts at the
density functional theory (DFT) level of theory was employed
to give some insight into the driving forces behind the crystal
transformation. The endothermic contraction process, which is
at its most endothermic at 46.3 °C, occurs with a ΔG of −6.2
× 103 J/mol, a ΔH − 7.7 × 103 J, and an entropic penalty of
ΔS = −35 J/(mol C). Conversely, the “expansion” process is
ΔG of +6.0 × 103 J/mol at 33.3 °Cthe exothermic peak
maximum. The reverse process appears largely entropy driven,
with a ΔS = 59 J/(mol C) and a ΔH 8.0 × 103 J/mol. The
slight discrepancy in free energy between these transitions may
arise from the thermal delamination of the crystals.
The DFT calculations were performed considering the
periodic crystal structure and molecular model systems
(Figures S1 and S2).38,39 The crystal structure of the DNDI
crystal (Figures S1a and S2a) and the molecular clusters
models have been used to estimate the different local and
nonlocal effects. More specifically, we studied different
molecular models surrounded by 1−3 layers to study the
neighboring effect (Figures S1b−e and S2b−e). Thermochem-
Figure 2. (a) DSC traces of DNDI are shown for 50, 100, 150, and
200 cycles. The inset shows a single DNDI crystal from the DSC
experiment, which remains ∼800 μm long, after 207 cycles. Major
divisions of the scale are 100 μm. (b) Dynamic motions that DNDIs
undergo from the low-temperature phase to the high-temperature
phase involve a rotation of one of the two aromatic rings on each NDI
(red arrow) to allow the NDI cores to move over each other (blue
arrow). (c) While the rotation of the aromatic ring is energetically
uphill, the greater π overlap in the high-temperature structure is a
lower energy structure at these temperatures with the distances and
angles tabulated and color-coded from the illustration in panel b.
istry calculations were performed at 298 and 338 K in all these
crystal structures and molecular clusters. Our calculations
reveal that ΔHtheory = (7.2−8.7) × 103 J/mol when DNDI is
surrounded by mono-, bi-, and trilayer molecules. This is
excellent in agreement with the experimental value (ΔHexp
=
(7.7−8.0) × 103 J/mol). The entropy changes from our
theoretical calculations is ΔStheory = 53.0−54.3 J/(mol C) for
the same molecular models, which is in agreement with the
experimental values (ΔSexp = 35−59 J/(mol C)). To
understand the barriers to switch from one form to the
other, we hypothesize that the driving force for the phase
transition is a lower energy cofacial overlap of the NDI cores.
This better overlap is inhibited at low temperature by the steric
effects of the bulky aromatic rings. This assessment would be
consistent with literature that has shown rotational dynamics in
the solid state are important factors in a number of thermally
responsive systems.40 Only when there is enough thermal
energy for these rings to rotate out of the way can the NDI
cores move closer to each other. Only one ring rotates
significantly on each DNDI moleculethe other ring rotates
by only a few degrees, which we assumed to have a negligible
effect on the energetics. To test this hypothesis, a scan was
5953
https://doi.org/10.1021/jacs.1c01549
J. Am. Chem. Soc. 2021, 143, 5951−5957

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
performed on the rotation of the phenoxy ring per DNDI
structure (Figure S3) within repeating stacks modeled from the
solid-state data. The estimated electronic rotational barrier is
ΔE† = 13.6−14.4 kJ mol−1, which is not much higher than the
available thermal energy at the phases transition. Thus, we
postulate that a possible mechanism driving the phase
transition is the greater thermal energy to permit partial
rotation of the rings. In other words, a full rotation of the rings
is not necessary, but the available thermal energy allows for the
rings to be perturbed enough to allow for an energetically more
favorable π−π interaction between the NDI cores, which is
observed as a decrease in centroid distances. In summary, the
overall global motion and experimental thermodynamic data
support that an energetically penalizing rotation of every other
aromatic ring must occur to allow the NDI cores to slide over
each other as depicted in Figure 2b. This rotational dynamics
permits greater stabilization from improved overlap of the NDI
π-systems (Figure 2c). These results are consistent with the
fact that our n-phenoxyl-substituted NDI systems containing 4,
7, 8, and 9 carbons likewise undergo thermosalience. Indeed,
more than 820 solid-state structures of functionalized NDI
cores have been uploaded to the Cambridge Structural
Database and many experimentally tested as organic semi-
conductor systems, yet only the derivatives featuring an n-
phenoxyl substituent have demonstrated thermoelastic behav-
ior. Therefore, the importance of the aromatic ring and its
rotational dynamics cannot be understated.40
(Figure 3, yellow arrows). At crystal imperfections, cracks, or
dislocations, the wavefront can stall, and the stress lines
become apparent. The thermal gradient takes a bit longer to
affect the other side of the obstacle, during which the
wavefront above or below the disturbance may continue
propagating past, causing the crystal to split disjointly until the
wave fronts catch up and the sections merge again (Movie S2).
The most common method of crystal splitting in this manner is
along the plane parallel to the large [001] face, causing
delamination of the parallel sheets. Repeated thermal cycles
tend to cleave or shear the crystal into thinner crystals that
continue to actuate. Intriguingly, the phase transition can occur
at any point in the crystal and occasionally at multiple
locations.
To take advantage of this thermoelasticity, we fabricated a
thermoelectric actuator by coating the crystal in a conductive
silver layer29 for use in an electronic circuit. DNDI exhibits no
conductivity in its natural state, so a conductive coating is
required to allow current to pass through. Selected large DNDI
crystals were metallized via the Tollens’ silver mirror reaction
and examined by scanning electron microscopy (SEM) (Figure
4a). A thin uniform coating of silver nanoparticles was found
on the crystal surface, and the crystals were visually gray and
shiny instead of their usual yellow color. Attempts to close a
circuit with these crystals failed because the silver coating was
not thick enough, and the silver mirror reaction was redone on
the same crystals. As the silver coating grew thicker, the silver
particles merged into a solid sheet. Curiously, once the
electrothermal motion began, the wetted crystal could actuate
(Movie S3). However, these crystals successfully closed a
circuit and were deemed conductive enough to proceed.
The conductive crystals were connected in an electronic
circuit (Figure 4b) to bridge a gap of a width such that it is
smaller than a cold crystal but larger than a hot one. Initial
attempts using copper electrodes resulted in intermittent
conduction at best, as there was not very good contact between
the crystal and the copper electrodes. Filter paper soaked in
saturated sodium chloride as a liquid electrolyte solution was
found to work well. Before coating with silver, the DNDI
crystals show no electrical conductivity whatsoever, so any
electrical measurements are entirely based on the nature of the
coating. Upon passing a current through the coated crystal, the
silver coating visibly turned black moving from the negative to
the positive end of the crystal (Movie S4). Once the current
provided enough heat to the circuit patheither the silver
coating or the liquid electrolytesthe crystal contracted.
Indeed, when the crystal was placed in just the right location, a
pulsatile expansion and contraction is observedseveral times
a secondas the crystal cooled and heated (Movie S5 at
approximately the 12 s mark).
The single-crystal-to-single-crystal transition occurs with a
slight change in unit cell parameters, which contributes to a
difference in cellular birefringence. We were able to use this
change in birefringence to better understand the mechanism of
actuation. During a gradual thermal gradient, the transition
occurs slowly, nucleating at a certain point along the crystal
long axis while all of the molecules in a line rearrange. The
“wavefront” (Figure 3, red diagonal lines) then propagates
lengthwise as a solid line in either direction down the crystal
On applying or removing the voltage, the crystal rapidly
responded by contracting or expanding, respectively (Movies
S1−S5). If the crystal is placed properly, the heating and
cooling induces a pulsive effect as the crystal heats, shrinks, and
shorts the circuit. The crystal then cools, expands, and closes
the circuit. Analysis of the motion shows that we were able to
obtain expansion−re-expansion rates between 1 and 2 Hz,
which is exceptionally rapid for organic crystals. On turning off
the power supply, the voltage drops slowly at first and then
rapidly to the baseline, indicated by the dotted ellipses in
Figure 4c. This is because the capacitors in the power supply
are discharging briefly after turning the voltage off.
Figure 3. Still frames of DNDI crystal between crossed polarizers set
at the extinction position after 200 thermal cycles subjected to a slow
thermal gradient. Propagation (yellow arrows) of the molecular
rearrangement wavefronts (red lines) through the long axis of the
crystal upon (a1−a6) heating and (b1−b6) cooling past the
respective thermal transition points.
5954
https://doi.org/10.1021/jacs.1c01549
J. Am. Chem. Soc. 2021, 143, 5951−5957

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 4. (a) Scanning electron micrographs of silvered DNDI crystals at progressively higher magnifications. At higher magnification, left to right,
individual silver nanoparticles are visible. (b) Circuit schematic for electrical actuation. (c) Oscilloscope traces of the measured voltage and current
through the circuit over time. At the bottom, the red sections represent the time the crystal was expanded, and the blue sections represent the time
the crystal was contracted. The dotted black ellipses on the voltage trace indicate the capacitor discharge in the power supply upon turning off the
voltage. (d) Force−length comparison of four DNDI crystals. The inset shows a crystal (yellow) resting on the aluminum heatsink with the
backstop on the right and the sensor on the left.
Re-expansion force measurements were obtained by using a
piezoelectric sensor placed against a heated crystal and
measuring the voltage produced by the piezoelectric material
in response to the crystal pushing into it (Figure 4d). This
voltage can be converted to force once the physical parameters
of the sensor are determined. The crystals selected were
difficult to align and register any force, but the force output
was plotted against the crystal length for four crystals in a range
of ∼1.5 to ∼3.5 mm. The forces registered in a range of ∼4 to
∼5.5 mN show a loose inverse proportion with length.
However, the crystals tend to break apart under resistance to
expansion. This suggests that the internal forces holding the
[001] planes together are weaker than the outward expansion
force, as the energy required to push outward instead shears
the crystal along its plane.
EXPERIMENTAL SECTION
■
Synthesis of DNDI. DNDI was synthesized based on literature
protocols,32,41 purified by silica column chromatography in isocratic
chloroform and crystallized by evaporation from the fraction tubes.
Metallization of DNDI. Sufficiently large DNDI crystal
candidates were selected and added to a scintillation vial with 5 mL
of silver nitrate, 200 mg of glucose, and 5 drops of ammonium
hydroxide and gently heated while swirling to around 50 °C until the
silver mirror began to form. The vial was then left on a rotisserie at
room temperature for 2 days and then vacuum filtered while rinsing
with ultrapure water and drying.
Circuit Design. A circuit was set up and soldered to a breadboard
consisting of a terminal for power supply, a 270 Ω resistor, a red LED,
terminals for current measurement, electrodes for the crystal switch
contacts, and a terminal for return power supply. The crystal switch
contacts consisted of filter paper strips moistened by saturated sodium
chloride solution and mounted above acrylic blocks with an air gap for
the crystal to bridge without electrolytes, causing a short circuit. The
LED indicated a closed circuit, the resistor protected the LED from
overcurrent, and the current was monitored as the power supply
voltage was increased until the crystal actuated. The metallized crystal
actuation was monitored with a USB microscope.
Force Measurements. Prior to coating with silver, crystals were
placed between the fins of an aluminum heatsink with a machined
steel rod as an adjustable backstop on one end and a Honeywell
FSG005WNPB force sensor on the other and a thermocouple glued
to the outside of the fins. The system was heated by direct contact
with a soldering iron tip at 800 °C until the crystal contracted. The
force sensor and backstop were adjusted such that the crystal was
butted up against both and would press against the sensor upon re-
expansion. The system was then cooled by direct contact with an ice
cube on the underside of the heatsink until the crystal expanded into
the force sensor. The voltage change was recorded and converted to
CONCLUSION
■
In summary, we have shown that DNDI can be used as a
thermally activated actuator for use as an electrical switch or
indicator, reusable over many cycles in ambient and near
physiologically important temperatures. These crystals retain
their actuation capability even after breaking into smaller parts
and show promising potential for use embedded in an elastic
matrix that can align and magnify the expansive capabilities of
DNDI for use as a thermally actuated sensor, muscle,
extracting energy in the diurnal cycle of desert climates or in
thermomechanical automatic devices. These crystals have the
potential to impact the fields of electronics and micro-
mechanics in different directions.
5955
https://doi.org/10.1021/jacs.1c01549
J. Am. Chem. Soc. 2021, 143, 5951−5957

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
force according to the sensor properties listed in the manufacturer’s
datasheet.
Raymond P. Welch − Department of Chemistry and
Biochemistry, University of Texas at Dallas, Richardson,
Texas 75080, United States; orcid.org/0000-0003-3114-
8801
Computational Details. Because of the extended nature of the
bulk DNDI crystal structure, the periodic crystal structures (Figures
S1a and S2b) and four cluster modeling systems with 1−4 layers were
used (Figures S1b−d and S2b−d). The equilibrium structures,
geometries, calculations of energetics, and rotational barriers of the
model systems of the DNDI crystal were obtained with the hybrid
unrestricted dispersion-corrected density functional theory (DFT)
method noted by UB3LYP-D3. The correlation-consistent polar-
ization double ζ-quality basis sets was used for all atoms.38,39 The
vibrational frequencies at the optimized geometries were analyzed to
confirm ground states. The thermochemical calculations were
performed at 298 and 338 K corresponding to each crystal structure
configuration. The rigid scan was performed by considering the same
models (Figure S3). The threshold used for evaluating the
convergence of the energy, forces, and electron density was 10−7 au
for each parameter. All calculations were performed with the
Gaussian09 quantum chemistry code.42 The details of the
computations can be found in the Supporting Information.
Carlos Caicedo-Narvaez − Renewable Energy and Vehicular
Technology Laboratory, University of Texas at Dallas,
Richardson, Texas 75080, United States
Michael A. Luzuriaga − Department of Chemistry and
Biochemistry, University of Texas at Dallas, Richardson,
Texas 75080, United States; orcid.org/0000-0001-6128-
8800
Bhargav S. Arimilli − Department of Chemistry and
Biochemistry, University of Texas at Dallas, Richardson,
Texas 75080, United States
Gregory T. McCandless − Department of Chemistry and
Biochemistry, University of Texas at Dallas, Richardson,
Texas 75080, United States
Babak Fahimi − Renewable Energy and Vehicular Technology
Laboratory, University of Texas at Dallas, Richardson, Texas
75080, United States
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.1c01549.
■
sı
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.1c01549
Author Contributions
M.D., S.P., and R.P.W. contributed equally to this work.
Experimental and computational details (PDF)
Movie S1 (MP4)
Movie S2 (MP4)
Notes
Movie S3 (MP4)
The authors declare no competing financial interest.
Movie S4 (MP4)
Movie S5 (MP4)
ACKNOWLEDGMENTS
■
J.J.G. acknowledges the National Science Foundation
[CAREER DMR-1654405 and DMR-2003534], the Welch
Foundation [AT-1989-20190330], and the ACS Petroleum
Research Fund [57627-DNI10] S.P. acknowledges the support
from the SERB-DST, Government of India, for providing the
Ramanujan Faculty Fellowship under Grant SB/S2/RJN-067/
2017 and Early Career Research Award (ECRA) under Grant
ECR/2018/000255.
Accession Codes
CCDC 1992519−1992523 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Authors
■
REFERENCES
(1) Otero, T. F.; Sansien
for Artificial Muscles. Adv. Mater. 1998, 10 (6), 491−494.
(2) Okuzaki, H.; Osada, Y. Electro-driven chemomechanical
polymer gel as an intelligent soft material. J. Biomater. Sci., Polym.
Ed. 1994, 5 (5), 485−496.
(3) Seki, T.; Sakurada, K.; Muromoto, M.; Ito, H. Photoinduced
single-crystal-to-single-crystal phase transition and photosalient effect
of a gold(i) isocyanide complex with shortening of intermolecular
aurophilic bonds. Chem. Sci. 2015, 6 (2), 1491−1497.
(4) Peterson, G. I.; Dobrynin, A. V.; Becker, M. L. Biodegradable
Shape Memory Polymers in Medicine. Adv. Healthcare Mater. 2017, 6
(21), 1700694.
(5) Ge, F.; Zhao, Y. A new function for thermal phase transition-
based polymer actuators: autonomous motion on a surface of
constant temperature. Chem. Sci. 2017, 8 (9), 6307−6312.
(6) Zhao, Q.; Zou, W.; Luo, Y.; Xie, T. Shape memory polymer
network with thermally distinct elasticity and plasticity. Science
Advances 2016, 2 (1), e1501297.
(7) Wang, J.; Zhao, Q.; Cui, H.; Wang, Y.; Chen, H.; Du, X. Tunable
shape memory polymer mold for multiple microarray replications. J.
Mater. Chem. A 2018, 6 (48), 24748−24755.
■
̃
a, J. M. Soft and Wet Conducting Polymers
Jeremiah J. Gassensmith − Department of Chemistry and
Biochemistry and Department of Biomedical Engineering,
University of Texas at Dallas, Richardson, Texas 75080,
United States; orcid.org/0000-0001-6400-8106;
Email: gassensmith@utdallas.edu
Jose L. Mendoza-Cortes − Department of Chemical &
Biomedical Engineering, FAMU-FSU Joint College of
Engineering, Tallahassee, Florida 32310, United States;
Department of Chemical Engineering & Materials Science,
Michigan State University, East Lansing, Michigan 48824,
United States; orcid.org/0000-0001-5184-1406;
Email: jmendoza@msu.edu
Authors
Madushani Dharmarwardana − Department of Chemistry
and Biochemistry, University of Texas at Dallas, Richardson,
Texas 75080, United States; orcid.org/0000-0002-3086-
047X
Srimanta Pakhira − Discipline of Physics, Discipline of
Metallurgy Engineering and Materials Science (MEMS) &
Centre for Advanced Electronics (CAE), Indian Institute of
Technology Indore (IIT Indore), Indore 453552, Madhya
Pradesh (M.P.), India; orcid.org/0000-0002-2488-300X
(8) Zhao, Q.; Wang, Y.; Cui, H.; Du, X. Bio-inspired sensing and
actuating materials. J. Mater. Chem. C 2019, 7 (22), 6493−6511.
(9) Omoto, K.; Nakae, T.; Nishio, M.; Yamanoi, Y.; Kasai, H.;
Nishibori, E.; Mashimo, T.; Seki, T.; Ito, H.; Nakamura, K.;
5956
https://doi.org/10.1021/jacs.1c01549
J. Am. Chem. Soc. 2021, 143, 5951−5957

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Kobayashi, N.; Nakayama, N.; Goto, H.; Nishihara, H. Thermosa-
lience in Macrocycle-Based Soft Crystals via Anisotropic Deformation
of Disilanyl Architecture. J. Am. Chem. Soc. 2020, 142 (29), 12651−
12657.
(30) Jin, M.; Yamamoto, S.; Seki, T.; Ito, H.; Garcia-Garibay, M. A.
Anisotropic Thermal Expansion as the Source of Macroscopic and
Molecular Scale Motion in Phosphorescent Amphidynamic Crystals.
Angew. Chem., Int. Ed. 2019, 58 (50), 18003−18010.
(31) 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 and
function memory effect in organic semiconductors. Nat. Commun.
2018, 9 (1), 278.
(10) Ding, X.; Unruh, D. K.; Groeneman, R. H.; Hutchins, K. M.
Controlling thermal expansion within mixed cocrystals by tuning
molecular motion capability. Chem. Sci. 2020, 11 (29), 7701−7707.
̌
(11) Panda, M. K.; Runcevski, T.; Chandra Sahoo, S.; Belik, A. A.;
Nath, N. K.; Dinnebier, R. E.; Naumov, P. Colossal positive and
negative thermal expansion and thermosalient effect in a pentam-
orphic organometallic martensite. Nat. Commun. 2014, 5, 4811.
(12) Dharmarwardana, M.; Welch, R. P.; Kwon, S.; Nguyen, V. K.;
McCandless, G. T.; Omary, M. A.; Gassensmith, J. J. Thermome-
chanically responsive crystalline organic cantilever. Chem. Commun.
2017, 53 (71), 9890−9893.
(32) Dharmarwardana, M.; Arimilli, B. S.; Luzuriaga, M. A.; 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.
(33) Dewhirst, M. W.; Viglianti, B. L.; Lora-Michiels, M.; Hoopes, P.
J.; Hanson, M. Thermal Dose Requirement for Tissue Effect:
Experimental and Clinical Findings. Proc. SPIE 2003, 4954, 37.
(34) Park, S. K.; Diao, Y. Martensitic transition in molecular crystals
for dynamic functional materials. Chem. Soc. Rev. 2020, 49 (22),
8287−8314.
̀
(13) Panda, M. K.; Centore, R.; Causa, M.; Tuzi, A.; Borbone, F.;
Naumov, P. Strong and Anomalous Thermal Expansion Precedes the
Thermosalient Effect in Dynamic Molecular Crystals. Sci. Rep. 2016,
6, 29610.
(14) Chung, H.; Chen, S.; Sengar, N.; Davies, D. W.; Garbay, G.;
Geerts, Y. H.; Clancy, P.; Diao, Y. Single Atom Substitution Alters the
Polymorphic Transition Mechanism in Organic Electronic Crystals.
Chem. Mater. 2019, 31 (21), 9115−9126.
(15) Pathigoolla, A.; Sureshan, K. M. Synthesis of triazole-linked
homonucleoside polymers through topochemical azide-alkyne cyclo-
addition. Angew. Chem., Int. Ed. 2014, 53 (36), 9522−5.
(16) Lauher, J. W.; Fowler, F. W.; Goroff, N. S. Single-crystal-to-
single-crystal topochemical polymerizations by design. Acc. Chem. Res.
2008, 41 (9), 1215−29.
(17) Pathigoolla, A.; Gonnade, R. G.; Sureshan, K. M. Topochemical
click reaction: spontaneous self-stitching of a monosaccharide to
linear oligomers through lattice-controlled azide-alkyne cycloaddition.
Angew. Chem., Int. Ed. 2012, 51 (18), 4362−6.
(18) Barbour, L. J. Single Crystal to Single Crystal Transformations.
Aust. J. Chem. 2006, 59 (9), 595.
(35) Alimi, L. O.; van Heerden, D. P.; Lama, P.; Smith, V. J.;
Barbour, L. J. Reversible thermosalience of 4-aminobenzonitrile.
Chem. Commun. 2018, 54 (48), 6208−6211.
̌
́
́
̌
́
(36) Loncaric, I.; Popovic, J.; Despoja, V.; Burazer, S.; Grgicevic, I.;
̌
́
Popovic, D.; Skoko, Z. Reversible Thermosalient Effect of N′-2-
Propylidene-4-hydroxybenzohydrazide Accompanied by an Immense
Negative Compressibility: Structural and Theoretical Arguments
Aiming toward the Elucidation of Jumping Phenomenon. Cryst.
Growth Des. 2017, 17 (8), 4445−4453.
(37) Karothu, D. P.; Weston, J.; Desta, I. T.; Naumov, P. Shape-
Memory and Self-Healing Effects in Mechanosalient Molecular
Crystals. J. Am. Chem. Soc. 2016, 138 (40), 13298−13306.
(38) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A consistent and
accurate ab initio parametrization of density functional dispersion
correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 2010,
132 (15), 154104.
(39) Becke, A. D. Density-functional thermochemistry. III. The role
of exact exchange. J. Chem. Phys. 1993, 98 (7), 5648−5652.
(40) Hutchins, K. M.; Groeneman, R. H.; Reinheimer, E. W.;
Swenson, D. C.; MacGillivray, L. R. Achieving dynamic behaviour and
thermal expansion in the organic solid state via co-crystallization.
Chem. Sci. 2015, 6 (8), 4717−4722.
(41) Fomine, S.; Fomina, L.; Arreola, R.; Alonso, J. Bisimide-
lactamimide ring contraction in six-membered polyimides. Polymer
1999, 40 (8), 2051−2058.
(42) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson,
G. A.; Nakatsuji, H.; Li, X.; Caricato, M.; Marenich, A. V.; Bloino, J.;
Janesko, B. G.; Gomperts, R.; Mennucci, B.; Hratchian, H. P.; Ortiz, J.
V.; Izmaylov, A. F.; Sonnenberg, J. L.; Williams; Ding, F.; Lipparini,
F.; Egidi, F.; Goings, J.; Peng, B.; Petrone, A.; Henderson, T.;
Ranasinghe, D.; Zakrzewski, V. G.; Gao, J.; Rega, N.; Zheng, G.;
Liang, W.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.;
Throssell, K.; Montgomery, Jr., J. A.; Peralta, J. E.; Ogliaro, F.;
Bearpark, M. J.; Heyd, J. J.; Brothers, E. N.; Kudin, K. N.; Staroverov,
V. N.; Keith, T. A.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A. P.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.;
Millam, J. M.; Klene, M.; Adamo, C.; Cammi, R.; Ochterski, J. W.;
Martin, R. L.; Morokuma, K.; Farkas, O.; Foresman, J. B.; Fox, D. J.
Gaussian 09, Rev. B.01; Gaussian, Inc.: Wallingford, CT, 2016.
(19) Naumov, P.; Chizhik, S.; Panda, M. K.; Nath, N. K.; Boldyreva,
E. Mechanically Responsive Molecular Crystals. Chem. Rev. 2015, 115
(22), 12440−90.
(20) Burgert, I.; Fratzl, P. Actuation systems in plants as prototypes
for bioinspired devices. Philos. Trans. R. Soc., A 2009, 367 (1893),
1541−57.
(21) Balzani, V.; Credi, A.; Raymo, F. M.; Stoddart, J. F. Artificial
Molecular Machines. Angew. Chem., Int. Ed. 2000, 39 (19), 3348−
3391.
(22) Kobatake, S.; Takami, S.; Muto, H.; Ishikawa, T.; Irie, M. Rapid
and reversible shape changes of molecular crystals on photo-
irradiation. Nature 2007, 446 (7137), 778−81.
(23) Morimoto, M.; Irie, M. A diarylethene cocrystal that converts
light into mechanical work. J. Am. Chem. Soc. 2010, 132 (40), 14172−
8.
(24) Medishetty, R.; Sahoo, S. C.; Mulijanto, C. E.; Naumov, P.;
Vittal, J. J. Photosalient Behavior of Photoreactive Crystals. Chem.
Mater. 2015, 27 (5), 1821−1829.
(25) Liu, G.; Liu, J.; Ye, X.; Nie, L.; Gu, P.; Tao, X.; Zhang, Q. Self-
Healing Behavior in a Thermomechanically Responsive Cocrystal
during a Reversible Phase Transition. Angew. Chem., Int. Ed. 2017, 56
(1), 198−202.
(26) Lusi, M.; Bernstein, J. On the propulsion mechanism of
“jumping” crystals. Chem. Commun. (Cambridge, U. K.) 2013, 49 (81),
9293−5.
(27) Panda, M. K.; Runcevski, T.; Husain, A.; Dinnebier, R. E.;
Naumov, P. Perpetually self-propelling chiral single crystals. J. Am.
Chem. Soc. 2015, 137 (5), 1895−902.
(28) Nath, N. K.; Panda, M. K.; Sahoo, S. C.; Naumov, P. Thermally
induced and photoinduced mechanical effects in molecular single
crystalsa revival. CrystEngComm 2014, 16 (10), 1850.
(29) Khalil, A.; Ahmed, E.; Naumov, P. Metal-coated thermosalient
crystals as electrical fuses. Chem. Commun. 2017, 53 (60), 8470−
8473.
5957
https://doi.org/10.1021/jacs.1c01549
J. Am. Chem. Soc. 2021, 143, 5951−5957