Redox-Active Macrocycles for Organic Rechargeable Batteries

Article abstract of DOI:10.1021/jacs.7b01209

Organic rechargeable batteries, composed of redox-active molecules, are emerging as candidates for the next generation of energy storage materials because of their large specific capacities, cost effectiveness, and the abundance of organic precursors, whe

Full text of DOI:10.1021/jacs.7b01209

Subscriber access provided by UNIV OF NEW ENGLAND
Article
Redox-Active Macrocycles for Organic Rechargeable Batteries
Dong Jun Kim, Keith R. Hermann, Aleksandrs Prokofjevs, Michael T
Otley, Cristian Pezzato, Magdalena Owczarek, and J. Fraser Stoddart
J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 24 Apr 2017
Downloaded from http://pubs.acs.org on April 24, 2017
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 free 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 accessible to all
readers and 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.
Journal of the American Chemical Society 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 28
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
RedoxꢀActive Macrocycles for
Organic Rechargeable Batteries
Dong Jun Kim, Keith R. Hermann, Aleksandrs Prokofjevs, Michael T. Otley, Cristian Pezzato,
*
Magdalena Owczarek and J. Fraser Stoddart
Department of Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, Illinois
6
0208-3113, USA
*
Eꢀmail: stoddart@northwestern.edu
MAIN TEXT
*
Correspondence Address
Professor J. Fraser Stoddart
Department of Chemistry
Northwestern University
2145 Sheridan Road
Evanston, IL 60208 (USA)
Email: stoddart@northwestern.edu
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 28
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
■
ABSTRACT
Organic rechargeable batteries, composed of redoxꢀactive molecules, are emerging as candidates for the
next generation of energy storage materials because of their large specific capacities, cost effectiveness,
and the abundance of organic precursors, when compared with conventional lithiumꢀion batteries.
Although redoxꢀactive molecules often display multiple redox states, precise control of a molecule’s
redox potential, leading to a single output voltage in a battery, remains a fundamental challenge in this
popular field of research. By combining macrocyclic chemistry with density functional theory
calculations (DFT), we have identified a structural motif that more effectively delocalizes electrons
during lithiation events in battery operations—namely, throughꢀspace electron delocalization in
triangular macrocyclic molecules that exhibit a single wellꢀdefined voltage profile—compared to the
discrete multiple voltage plateaus observed for a homologous macrocyclic dimer and an acyclic
derivative of pyromellitic diimide (PMDI). The triangular macrocycle, incorporating three PMDI units
in close proximity to one another, exhibits a single output voltage at 2.33 V, compared with two peaks at
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
(i) 2.2 and 1.95–1.60 V for reduction and (ii) 1.60–1.95 and 2.37 V for oxidation of the acyclic PMDI
derivative. By investigating the two cyclic derivatives with different conformational dispositions of their
PMDI units and the acyclic PMDI derivative, we identified noticeable changes in interactions between
the PMDI units in the two cyclic derivatives under reducing conditions, as determined by differential
pulse voltammetry (DPV), solutionꢀstate spectroelectrochemistry, and variable temperature UVꢀVis
spectra. The numbers and relative geometries of the PMDI units are found to alter the voltage profile of
the active materials significantly during galvanostatic measurements, resulting in a desirable single
plateau for the triangular macrocycle. The present investigation reveals that understanding and
controlling the relative conformational dispositions of redoxꢀactive units in macrocycles are key to
achieving high energy density and long cycleꢀlife electrodes for organic rechargeable batteries.
1
ACS Paragon Plus Environment

Page 3 of 28
Journal of the American Chemical Society
1
2
3
■
INTRODUCTION
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
Rechargeable batteries impact our everyday lives and play a critical role in meeting modern energy
demands, while also holding out promise for a more sustainable future by facilitating the utilization of
1
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
energy from renewable sources. Amongst many different rechargeable ones, lithiumꢀion batteries
(
LIBs) have become ubiquitous in today's world, delivering spectacular performance in applications,
2
ranging from portable electronic devices to electric vehicles. Nevertheless, conventional materials
employed in commercial LIBs rely on nonꢀrenewable sources of rare metals, a situation which is not
3
desirable from a longꢀterm environmental perspective. Hence, new strategies are needed in order to
4
develop next generation energy storage systems based on more sustainable and lowꢀcost materials.
Organic rechargeable batteries (ORBs) utilizing redoxꢀactive small molecules or polymers have
5
received a significant amount of attention on account of their versatility, sustainability, flexibility, and
low environmental impact, all of which reflect the wide availability of electrode materials from
abundant natural sources. In order to design an active material for ORBs, the compounds need to satisfy
the following criteria. They must (i) display redox reversibility in the solidꢀstate in a wet electrolyte
environment, (ii) intercalate cations or anions reversibly, and (iii) exhibit the lack of solubility toward
6
the electrolyte during electrochemical cycling. While recent investigations into ORBs have greatly
improved electrochemical performance, major challenges in organic electrode design have still not been
addressed in a wholly satisfactory manner. When compared to their conventional transitionꢀmetal based
counterparts, cathodes based on redoxꢀactive organic molecules featuring numerous redox states often
deliver larger capacities, yet they typically exhibit unfavorable voltage profiles. Thus, multiple voltage
plateaus originating from discrete redox potentials are typically observed, and substantial voltage
decline during discharge cannot be avoided, even when neighboring redox units are forced to interact
7
with each other. Steadily declining voltage profiles pose a substantial problem for designing fullꢀcell
2
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 28
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
batteries, where consistent output voltages are desirable. Furthermore, small redoxꢀactive organic
2
0
molecules are often soluble in electrolyte solutions, or participate in irreversible side reactions during
8
, 9
repeated battery cycling, causing severe capacity drops over time. Achieving (i) a single wellꢀdefined
voltage profile, in combination with (ii) practical cyclability characteristics in small moleculeꢀbased
organic rechargeable batteries, therefore, still present considerable challenges.
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
In an effort to address these challenges, we chose to explore redoxꢀactive macrocycles, which have
received much less attention than their corresponding smallꢀmolecule and polymerꢀbased counterparts.
Rigid macrocycles exhibit unique physical and chemical properties, which result primarily from the
1
0,
presence of enforced cavities, and are tunable through controlling the conformations of the molecules.
11
For our part, we have been focusing on gaining a better understanding of the material properties of
rigid macrocycles, some of which result from electronic communication between redoxꢀactive units
1
2, 13, 14
10
within the structure.
Despite the long history of macrocyclic chemistry, direct observation of its
impact on performance of energy storage devices, such as batteries and supercapacitors, has limited
14, 15
precedent.
Herein, we demonstrate that constraining three redoxꢀactive units exhibiting multiple discrete
potentials—and thus giving rise to twoꢀplateau voltage profiles in batteries—into a rigid triangular
macrocycle has allowed us to access cathode materials which exhibit single wellꢀdefined voltage
profiles. Our investigation suggests that the intrinsic properties of the redox unit itself and the mutual
arrangement of individual redox units within the active material are equally critical to designing organic
electrodes for rechargeable batteries.
■
RESULTS AND DISCUSSION
3
ACS Paragon Plus Environment

Page 5 of 28
Journal of the American Chemical Society
1
2
3
4
5
Design and Synthesis. In order to probe the macrocyclic effect on battery performance, we
synthesized a series of three compounds—namely, bis(cyclohexyl)pyromellitic diimide (PMDIꢀRef),
the cyclic dimeric pyromellitic diimide ((–)ꢀ2PMDI), and the cyclic trimeric pyromellitic diimide ((–
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
)
ꢀ3PMDIꢀꢁ). The first compound PMDIꢀRef is a simple pyromellitic diimide featuring one PMDI
redox unit (Figure 1a), bearing cyclohexyl rings attached to both nitrogen atoms and so experiences
relatively facile rotation around exocyclic C–N bonds. The second compound (–)ꢀ2PMDI is a strained
cyclic dimer (Figure 1b) with two cofacially oriented redoxꢀactive PMDI units held in close proximity
by two (RR)ꢀ1,2ꢀdiaminocyclohexano linkers. The third compound in the series, (–)ꢀ3PMDIꢀꢁ, a larger
triangular macrocycle (Figure 1c) composed of alternating PMDI and (RR)ꢀ1,2ꢀdiaminocyclohexano
units, is less strained than the dimer.
1
6
While PMDIꢀRef and (–)ꢀ2PMDI were synthesized according to previously reported procedures, the
synthesis of (–)ꢀ3PMDIꢀꢁ has been optimized for multiꢀgram scale production without the use of
column chromatography. XꢀRay crystal structures of the PMDI macrocycles exhibit minor differences
16
from those reported in the previous study, apparently reflecting the influence of the solvents used in
crystallizations. We also attempted many different crystallizations of the radical anion along with
numerous lithium salts – lithium cations from lithium naphthalenide, lithium hexafluorophosphate,
lithium bis(trifuoromethane)sulfonimide – but all attempts were unsuccessful. Detailed experimental
1
13
procedures and full H and C NMR spectroscopic characterization, as well as singleꢀcrystal Xꢀray
diffraction analysis, are described in detail in the Supporting Information.
Redox Properties and UV−Vis−NIR Spectrophotometry. The redox properties of all three
compounds were studied in 1 mM solutions in DMF, using 0.1 M tetrabutylammonium
hexafluorophosphate (nꢀBu NPF ) as the supporting electrolyte. Since cyclic voltammetry (CV, Figure
4
6
4
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 28
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
S11) did not afford good resolution of individual redox peaks, differential pulse voltammetry (DPV,
Figure 2a) was used as the electrochemical technique of choice.
Upon reduction, PMDIꢀRef shows two peaks at –0.69 and –1.38 V vs Ag/AgCl, resulting from 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
•–
2–
formation of [PMDIꢀRef] and [PMDIꢀRef] , respectively. In contrast, the DPV plot of (–)ꢀ2PMDI
shows four distinct peaks clustered in two wellꢀseparated groups (–0.41, –0.70 and –1.73, –1.92 V vs
Ag/AgCl), originating from a series of oneꢀelectron reduction processes, featuring electronic interactions
between the two diimide units. A similar pattern is observed in the case of (–)ꢀ3PMDIꢀꢁ, where the total
number of reduction steps is six (–0.61, –0.70, –0.77 and –1.33, –1.55, –1.72 V vs Ag/AgCl), confirming
the fact that the triangle also allows for some throughꢀspace electron delocalization in the reduced state.
Spectroelectrochemistry (SEC) experiments were performed (Figure S13) in order to gather additional
information about the reduced forms of the PMDI derivatives. At the reducing potential of –1.25 V, all
the PMDI derivatives exhibit absorptions around 650 and 720 nm, indicating the formation of radical
anionic species. Reduced (–)ꢀ2PMDI, however, also showed two broad absorption bands, one around
5
50–1100 nm, and the other above 1200 nm, qualitatively resembling observations in a closely related
1
3
naphthalene diimide dimer reported previously, lending further support for intramolecular interactions
between the diimide units in the dimer. The reduced states of all PMDI derivatives were also
characterized by temperatureꢀ (Figure 3a−c) and concentrationꢀdependent (Figure 3d−f) UV–Vis–NIR
spectrophotometry. We employed 1 equiv of cobaltocene (CoCp ) per PMDI unit in the particular
2
1
8
+
molecule as reducing agent, on account of its reduction potential of E = −1.91 V vs Fc/Fc , which is
o
sufficient to deliver one electron per PMDI unit in each of the three compounds in this investigation. On
increasing the temperature from 20 to 80 °C, the intensity of the peak at 720 nm in the UV−Vis
spectrum (Figure 3a) of PMDIꢀRef decreases considerably. No new signals appear in the spectrum, and
the initial absorption intensity is restored upon returning the sample to 20 °C. While the UV−Vis
5
ACS Paragon Plus Environment

Page 7 of 28
Journal of the American Chemical Society
1
2
3
spectrum (Figure 3b) of (–)ꢀ2PMDI at 20 °C differs considerably from that of PMDIꢀRef, the
differences become progressively less apparent on increasing the temperature, and at 80 °C, the UV−Vis
spectra of the reduced states of all three PMDI derivatives exhibit considerable similarities. The
temperatureꢀdependent changes in the UV−Vis spectrum (Figure 3c) of (–)ꢀ3PMDIꢀꢁ, reduced on
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
adding 3 equiv CoCp are unremarkable, and are best described as an incremental change in the
2
normalized absorbance with temperature. Concentrationꢀdependent (0.05−0.5 mM) experiments (Figure
3
d–f) involving chemically reduced PMDIꢀRef and (–)ꢀ3PMDIꢀꢁ did not present any evidence of
13, 18
intermolecular radicalꢀanion pairing.
The interpretation of the concentrationꢀdependent UV−Vis
spectrophotometric data (Figure 2e) for the reduced (–)ꢀ2PMDI is less straightforward, as changes in the
intensity of the 1600 nm peak may potentially indicate some contribution from intermolecular
interactions between reduced (–)ꢀ2PMDI molecules. The chemical stability of the reduced forms of all
three compounds was confirmed by NMR experiments involving reduction of the imides by adding
CoCp in DMFꢀd solution, followed by reoxidation with nitrosonium hexafluorophosphate (NOPF )
2
7
6
after 24 h at ambient temperature. (Figure S5–7) In all cases, proton resonances corresponding to
starting imides reappeared upon addition of the oxidant, indicating that the compounds suffered little to
no degradation in their reduced states.
Battery Measurements of Macrocyclic Compounds. Since our initial observations indicated a strong
dependence of the electronic properties of the cyclic PMDI derivatives on the mutual arrangement of
redoxꢀactive units within the macrocycle, we have performed galvanostatic measurements on all three
compounds in order to elucidate the effect of changes in conformation on LIB performance. Coinꢀtype
cells, which were fabricated inside an Arꢀfilled glovebox, were employed in the galvanostatic
measurements. Each PMDI redox unit in these derivatives undergoes two sequential oneꢀelectron redox
processes. In other words, PMDIꢀRef, (–)ꢀ2PMDI and (–)ꢀ3PMDIꢀꢁ are capable of accepting two, four
6
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 28
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
and six lithium atoms, respectively, as depicted in Figure S12. Hereafter, all of the potentials listed are
+
based on the Li/Li redox couple, while charging and discharging processes refer to the lithiation
(reduction) and delithiation (oxidation), respectively. Current rates were calculated based on the
theoretical capacity of each PMDI derivative, and not on the actual charge/discharge cycle durations.
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
At the current rate of C/20, PMDIꢀRef exhibits (Figure 3a) a voltage profile consisting of two
1
9
plateaus, — a distinct one at 2.11 V, followed by a less wellꢀdefined one in the range of ca. 1.95−1.60
•–
2–
V, corresponding to the formation of [PMDIꢀRef] and [PMDIꢀRef] , respectively. Similarly, during
2–
the discharge process, oxidation of [PMDIꢀRef] resulted in a distorted plateau in the range of
•
–
ca. 1.60−1.95 V, while further oxidation of [PMDIꢀRef] delivered a clear plateau at 2.37 V. The
–1
specific capacities observed during the first charge and discharge were 94 and 172 mAh g ,
–1
respectively. The capacity decayed rapidly in the subsequent cycles, resulting in only 52 mAh g during
the third discharge. The observed low values for both the Coulombic efficiency (CE) and specific
capacity of the first cycle are most likely a result of the dissolution of the PMDIꢀRef (or its lithiated
2
0
forms) in the electrolyte. A similar voltage profile was also observed (Figure 3b) for (–)ꢀ2PMDI,
2
–
where the reduction to [(–)ꢀ2PMDI] gives rise to a plateau at around 2.47 V, and the subsequent
reductions to [(–)ꢀ2PMDI]3•– and [(–)ꢀ2PMDI] occurred between 2.22 and 1.60 V. During the
discharge process, a sloped voltage profile was observed in the range of 1.60−2.31 V, corresponding to
4–
4
–
the oxidation of [(–)ꢀ2PMDI] , while a more discrete plateau at 2.58 V stems from the oxidation of [(–
2–
)
ꢀ2PMDI] . The charging profile of (–)ꢀ2PMDI also showed some deviation during the first and
subsequent cycles. When compared with PMDIꢀRef, (–)ꢀ2PMDI revealed substantial CE improvement
in the first cycle, as well as better capacity retention in the subsequent cycles, demonstrating a specific
–1
capacity of 126 mAh g after three cycles. Significantly, the voltage profile of (–)ꢀ3PMDIꢀꢁ (Figure
4
c) is substantially different from those of PMDIꢀRef and (–)ꢀ2PMDI, featuring a single wellꢀdefined
7
ACS Paragon Plus Environment

Page 9 of 28
Journal of the American Chemical Society
1
2
3
6–
plateau at 2.33 V, corresponding to reduction of (–)ꢀ3PMDIꢀꢁ to [(–)ꢀ3PMDIꢀꢁ] . This triangular
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
–
1
+
macrocycle also showed a specific capacity of 163 mAh g , indicating acceptance of 5.8 Li per
2
1
molecule. The highly desirable singleꢀplateau voltage profile and good capacity retention of (–
ꢀ3PMDIꢀꢁ were also preserved in galvanostatic rate capability tests (Figure 4e and f) performed at
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
)
current rates of up to 2C. Both macrocycles demonstrated substantially higher first cycle CE compared
to the reference compound. Thus, while the first cycle CEs of (–)ꢀ2PMDI and (–)ꢀ3PMDIꢀꢁ were found
to be 96% and 94%, respectively, that of PMDIꢀRef was only 55%. In galvanostatic cycling tests
performed at a C/5 current rate, the (–)ꢀ3PMDIꢀꢁ showed substantially better capacity retention than the
–1
other two compounds, delivering 86 mAh g after 50 cycles, a value which corresponds to 64% of the
initial capacity. At the same time, both PMDIꢀRef and (–)ꢀ2PMDI displayed rapid capacity decay, with
–1
less than 10 mAh g remaining after 50 cycles. High solubility of the neutral state of PMDIꢀRef in the
carbonateꢀbased electrolyte was confirmed experimentally, thus providing a convenient rationale for the
unstable cycling behavior of the reference compound. Both (–)ꢀ3PMDIꢀꢁ and (–)ꢀ2PMDI, however,
were found to be insoluble in the electrolyte, and so we performed additional investigations to explain
the different cycling performance of these two compounds. Postulating that more prominent differences
in the physical properties may be observed for the lithiated states of these compounds, we attempted
reducing both macrocycles (Figure S14) with metallic lithium in DMF. Upon reduction, the solution of
(
–)ꢀ3PMDIꢀꢁ was rapidly transformed into a gel, which reverted back to a liquid upon reꢀoxidation. At
the same time, reduction had much less influence on the viscosity of the (–)ꢀ2PMDI solution. It is
possible that agglomeration via intermolecular O–Li•••O interactions may be more favorable in the case
of (–)ꢀ3PMDIꢀꢁ on account of its triangular geometry, therefore preventing dissolution of the active
9
material during battery operation.
8
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 28
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
Computational Studies. The marked differences which exist between the three PMDI derivatives
explored experimentally motivated us to perform density functional theory (DFT) calculations in order
to gain a better understanding of how the molecular structure of these compounds translates into their
performance in ORBs. Towards this end, we performed structural optimization of all PMDI derivatives,
as well as their respective O–lithiated states containing up to two Li atoms per PMDI unit, at
BPꢀ86(D3BJ)/def2ꢀTZVP level of theory. Computational details are described in the Supporting
Information. Even although the computational approach employed ignored the condensed phase effects
beyond solvation, resulting in the systematic overestimation of absolute values of reduction potentials by
ca. 1 V in all cases, nevertheless it was possible to reproduce major trends observed in the experimental
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
+
studies. Thus, a 0.43 eV separation between two calculated Li ꢀcoupled reduction potentials (−3.15 and
−2.72 eV) for PMDIꢀRef fits (Figure 4a) well within ca. 0.16‒0.77 eV range obtained from
galvanostatic measurements. For (–)ꢀ2PMDI, calculated sequential redox potentials (−3.80, −3.73 and
−3.00, −2.96 eV) parallel the marginally distorted twoꢀplateau voltage profile determined experimentally
(Figure 4b), while also predicting correctly the higher operational voltages of LIBs based on the cyclic
+
dimer. More importantly, the theoretical studies (Figure 4c) predict that the Li ꢀcoupled redox behavior
of (–)ꢀ3PMDIꢀꢁ differs substantially from those of PMDIꢀRef and (–)ꢀ2PMDI, mirroring all major
features of the experimental data. Thus, the calculated difference between the lowest and the highest
reduction potentials of (–)ꢀ3PMDIꢀꢁ is only 0.33 eV, a value which is particularly noteworthy in view
of the fact that six distinct redox states (−3.53, −3.48, −3.52 and −3.25, −3.26, −3.20 eV) are contained
within this narrow region, leading to the unique oneꢀplateau discharge profile.
Atoms in Molecules (AIM) charge and Shannon aromaticity (see the Supporting Information)
calculations were used to trace changes occurring during the reduction of different PMDI derivatives.
The results (Figure 5 and Figure S15–17) show representative bond lengths, and changes in AIM
9
ACS Paragon Plus Environment

Page 11 of 28
Journal of the American Chemical Society
1
2
3
charges of the PMDIꢀRef core occurring upon reduction with one and two Li atoms, respectively. In
good agreement with chemical intuition, reduction of PMDIꢀRef with 1 equiv of Li results in the
formation of Oꢀlithiated species (Figure 5b) with substantially decreased AIM charges at the C (by 0.23)
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
and O (by 0.18) atoms of the carbonyl group coordinated to Li . Some electron density is also
transferred to the other three carbonyl carbon atoms in the molecule, yet to a substantially smaller
extent, while the remaining nonꢀhydrogen atoms play a relatively small role in accommodating the
electron density transferred from lithium. The second Oꢀlithiation event (Figure 5c) shows little
+
regiochemical preference, although the structure with both Li coordinating to different carbonyls of the
same imide unit is of marginally lower energy. Both lithiated carbonyl groups of the doubly reduced
PMDIꢀRef feature longer C–O distances (1.28 Å vs. 1.25 Å) and lower AIM charges (by 0.22 at O and
0
.16 at C, see the Supporting Information) than the nonꢀlithiated carbonyl groups in the same molecule.
+
In contrast, reduction of the (–)ꢀ2PMDI leads to species (Figure 5d–f) in which every Li cation is
simultaneously coordinated by two PMDI units, and each electron transferred from the lithium is
accommodated by both participating carbonyl groups in a nearly equal manner. This situation is most
apparent in the fully lithiated (–)ꢀ2PMDI (Figure 5f), where all carbonyl C and O charges, as well as C–
O distances (1.27Å) are nearly equal throughout the molecule. A very similar picture is observed (Figure
+
5
. g–i) in (–)ꢀ3PMDIꢀꢁ, where each Li cation is again bonded to two oxygen atoms belonging to
different sides of the (–)ꢀ3PMDIꢀꢁ, and both carbonyl groups involved accept roughly equal amounts of
electron density, implying that the third side of the (–)ꢀ3PMDIꢀꢁ plays no role in the monolithiation
event (Figure 5h). In the sixꢀfold lithiated state of (–)ꢀ3PMDIꢀꢁ (Figure 5i), atomic charges, as well as
C–O distances once again show little spread within the molecule, and are almost equal to the
corresponding values in the fully lithiated (–)ꢀ2PMDI (Figure 5f). Despite the similarities between
charge distribution in both exhaustively lithiated macrocycles, electron repulsion between the reduced
1
0
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 28
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
PMDI units is diminished considerably in the case of the molecular triangle on account of larger spacing
between the redox units. In particular, while differences in shape between the (–)ꢀ2PMDI and the (–
)ꢀ3PMDIꢀꢁ have relatively little influence on O•••O separation within each C–O–Li–O–C subunit (3.02
Å in (–)ꢀ2PMDIꢀLi , and 3.15 Å in (–)ꢀ3PMDIꢀꢁꢀLi ), and almost no influence on the lengths of Li–O
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
4
6
bonds, the carbonyl C•••C separations are considerably larger in the latter species, i.e., 3.19 Å in (–
ꢀ2PMDIꢀLi , and 3.63 Å in (–)ꢀ3PMDIꢀꢁꢀLi . Therefore, at high levels of the lithiation process,
)
4
6
electrons experience more stabilization in the case of (–)ꢀ3PMDIꢀꢁ, consequently leading to a
substantially lessꢀsloped voltage profile.
The dramatic changes in voltage profiles observed across the series of PMDI derivatives can thus be
attributed by interplay of two factors—namely, the size of the delocalization surface involved in each
reduction event, as well as spacing between PMDI units in the structure. Specifically, the PMDIꢀRef
produces a more localized reduced state, leading to the familiar twoꢀplateau voltage profile. In turn, both
the (–)ꢀ2PMDI and (–)ꢀ3PMDIꢀꢁ feature more complete utilization of carbonyl groups in the
stabilization of the fully lithiated state. As the degree of lithiation increases, however, so does the
electron repulsion between the reduced PMDI units in the (–)ꢀ2PMDI. In the (–)ꢀ3PMDIꢀꢁ, the
geometrical arrangement of the PMDI units is such that it reduces unfavorable interactions in highly
+
lithiated states, without sacrificing the efficiency of Li chelation.
■
CONCLUSIONS
We have synthesized a series of PMDIꢀbased redoxꢀactive compounds, including two rigid
macrocycles, and measured their performance as active materials in batteries. The macrocycles (–
ꢀ2PMDI and (–)ꢀ3PMDIꢀꢁ exhibit throughꢀspace electron delocalization on account of the forced
proximity between redox units. We have also observed substantial differences in the electrochemical
)
1
1
ACS Paragon Plus Environment

Page 13 of 28
Journal of the American Chemical Society
1
2
3
performance between (–)ꢀ2PMDI and (–)ꢀ3PMDIꢀꢁ, with only the latter compound producing the
desired oneꢀplateau voltage profile on account of dense clustering of all six redox potentials of the (–
)ꢀ3PMDIꢀꢁ within a narrow voltage range. We believe that this difference arises primarily from
repulsive interactions between the reduced PMDI units, which are particularly strong in (–)ꢀ2PMDI
because of their cofacial orientation and the strain present in the cyclic dimer, both of which are lessened
substantially in the triangular macrocycle. We conclude that controlling the delocalization of the reduced
states and minimizing the electron repulsion between redoxꢀactive units are important considerations in
designing active materials for organic rechargeable batteries.
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
■
EXPERIMENTAL SECTION
Preparation of PMDI derivatives. Bis(cyclohexyl)pyromellitic diimide (PMDIꢀRef), the cyclic
dimeric pyromellitic diimide ((–)ꢀ2PMDI), and the cyclic trimeric pyromellitic diimide ((–)ꢀ3PMDIꢀꢁ))
1
6
were synthesized as previously described in the literature. The first two compounds were purified
using automated column chromatography on a Teledyne Isco Combiflash RF 200 system, while the
triangular macrocycle was purified according to the procedures described in Section B without the use
of column chromatography.
Material characterization. Degassed N,Nꢀdimethylformamideꢀd (DMFꢀd ) or DMF was used in all
7
7
investigations in order to ensure the use of a single solvent for the acquisition of all the experimental
data. A threeꢀcycle freezeꢀpumpꢀthaw method was employed in order to degas DMFꢀd for the NMR
7
titrations and UV–Vis spectroscopy. Liquid N was used as the coolant, and a 0.1 mtorr vacuum was
2
applied for 30 min in each cycle. Thin layer chromatography (TLC) was performed on silica gel 60 F254
(E. Merck). UV–Vis–NIR spectra were recorded between 20 and 80 °C on a Shimadzu UVꢀ3600
spectrophotometer, equipped with an external temperature controller S1700. Nuclear magnetic
resonance (NMR) spectra were recorded on Bruker Avance 500 spectrometers, with working
1
13
frequencies of 500 MHz ( H nuclei) and 125 MHz ( C nuclei). Chemical shifts are reported in ppm
relative to the signals corresponding to the residual nonꢀdeuterated solvents ((CD ) NCDO δ = 8.03,
3
2
H
1
2
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 28
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
2
.92, 2.75 and δ = 163.2, 34.9, 29.8 ppm). Single crystals of (–)ꢀ2PMDI and (–)ꢀ3PMDIꢀꢁ were grown
C
by slow vapor diffusion of H O into a solution of (–)ꢀ2PMDI in DMF and (–)ꢀ3PMDIꢀꢁ in NMP,
2
respectively. Details of the crystallization procedures are given in Supporting Information.
Electrochemical measurements. Cyclic voltammetry (CV) and differential pulse voltammetry (DPV)
were carried out using a threeꢀelectrode beaker cell, with Ag/AgCl as a reference electrode. Glassy
carbon and platinum mesh was used for the working and counter electrodes, respectively.
Spectroelectrochemistry (SEC) was performed using an EFꢀ1362 BASi spectroelectrochemical cell with
Pt mesh as the working electrode. A Gamry Reference 600 potentiostat interfaced to a PC was employed
in the collection of CV, DPV and SEC data.
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
Battery electrode fabrication and testing. Coinꢀtype cells were fabricated inside an Arꢀfilled
glovebox. The battery electrode slurry was prepared by mortar and pestle, mixing each PMDI derivative
with carbon black and polyvinylidene fluoride (PVDF) binder in a weight ratio of 60:30:10 using in N-
methylpyrrolidone (NMP) as the solvent. The mixed slurry was cast onto copper foil using the Doctorꢀ
o
Blade technique and dried in vacuum oven at 70 C overnight. Lithium hexaflurophosphate (LiPF ),
6
dissolved in a mixture of ethylene carbonate and dimethyl carbonate (EC:DMC, 1:1 v/v, Sigma
Aldrich), was employed as the electrolyte, while metallic lithium was used as both counter and reference
electrodes. Celgard 2320 microporous film was used as a separator. All coinꢀtype cells were subjected to
+
galvanostatic cycling using automated tester with 1.6–3.2 V vs Li/Li potential limitation condition. The
current rates from galvanostatic measurements were calculated based on the theoretical specific capacity
of each compound.
Computational details. All geometry optimizations and frequency calculations were performed using
2
2
the ORCA 3.0.2 suite of computational programs. Tight SCF convergence criteria (ORCA: TightSCF)
and increased resolution integration grids (ORCA: Grid6) were employed throughout calculations.
23
Geometry optimization was performed at BPꢀ86(D3BJ)/def2ꢀTZVP level of theory using COSMO
24
solvation model. Additionally, resolution of identity (RIꢀJ) approximation and D3BJ empirical
2
5
dispersion correction were used. For solution phase calculations, ε = 38 was assumed based on the
1
3
ACS Paragon Plus Environment

Page 15 of 28
Journal of the American Chemical Society
1
2
3
26
dielectric constant estimated for the actual solvent mixture used in battery fabrication. The fact that the
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
resulting superstructures represent true minima on the potential energy surface was confirmed by gas
2
7
phase reꢀoptimization at HFꢀ3c level of theory, and subsequent frequency calculations. A large number
+
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
of lithiated superstructures were explored for the PMDIꢀRef to identify general patterns in Li ion
binding to the reduced species. It was established that Oꢀlithiated superstructures were consistently of
+
lower energy compared to potential alternatives, e.g., superstructures with Li ion located above the
PMDI ring. In the ease of the larger cyclic dimer (–)ꢀ2PMDI and triangle (–)ꢀ3PMDIꢀꢁ attention was
focused on the systematic evaluation of all conceivable Oꢀlithiated superstructures for each reduced
2
8
state, and only the lowest energy superstructures were considered. AIM and Shannon aromaticity (SA)
calculations were performed as follows. Starting from optimized geometries, wave function files were
generated in singleꢀpoint gas phase calculations at B3LYP/6ꢀ31+G(d,p) level of theory using Gaussian
2
9
0
9W package. Subsequently, AIM charges and electron densities at bond critical points were calculated
3
0
by processing wave function files with MultiWFN 3.3.8 program. SA calculations were performed
using charge densities at all C–C bond critical points within the ring.
■
ASSOCIATED CONTENT
Supporting Information
Detail experimental details, synthetic procedures, spectroscopic and crystallographic characterization,
computational details, and supportive figures. This material is available free of charge via the Internet at
http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
1
4
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 28
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
*
stoddart@northwestern.edu
Notes
The authors declare no competing financial interest.
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
■
ACKNOWLEDGEMENTS
This research was conducted as part of the Joint Center of Excellence in Integrated Nanosystems at King
Abdulaziz City for Science and Technology (KACST) and Northwestern University (NU). The authors
acknowledge both KACST and NU for their financial support of this research. The Integrated Molecular
Structure Education and Research Center (IMSERC) at NU is acknowledged for the use of its facilities.
■
REFERENCES
(1) Huggins, R., Advanced Batteries: Materials Science Aspects. Springer Science & Business
Media: 2008.
(2) Goodenough, J. B.; Kim, Y., Chem. Mater. 2009, 22, 587ꢀ603; Armand, M.; Tarascon, J.ꢀM.,
Nature 2008, 451, 652ꢀ657.
(3) Choi, J. W.; Aurbach, D., Nat. Rev. Mater. 2016, 1, 16013.
(
4) (a) Yang, Y.; Zheng, G.; Cui, Y., Chem. Soc. Rev. 2013, 42, 3018ꢀ3032. (b) Ji, X.; Lee, K. T.;
Nazar, L. F., Nat. Mater. 2009, 8, 500ꢀ506. (c) Wang, C.; Wu, H.; Chen, Z.; McDowell, M. T.;
Cui, Y.; Bao, Z., Nat. Chem. 2013, 5, 1042ꢀ1048. (d) Bruce, P. G.; Freunberger, S. A.; Hardwick,
L. J.; Tarascon, J.ꢀM., Nat. Mater. 2012, 11, 19ꢀ29. (e) Sun, Y.; Lee, H.ꢀW.; Seh, Z. W.; Liu, N.;
Sun, J.; Li, Y.; Cui, Y., Nat. Energy 2016, 1, 15008.
(
5) (a) Liang, Y.; Tao, Z.; Chen, J., Adv. Energy Mater. 2012, 2, 742ꢀ769. (b) Häupler, B.; Wild, A.;
Schubert, U. S., Adv. Energy Mater. 2015, 5, 1402034. (c) Song, Z.; Zhou, H., Energy Environ.
Sci. 2013, 6, 2280ꢀ2301; Armand, M.; Grugeon, S.; Vezin, H.; Laruelle, S.; Ribière, P.; Poizot, P.;
1
5
ACS Paragon Plus Environment

Page 17 of 28
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Tarascon, J.ꢀM., Nat. Mater. 2009, 8, 120ꢀ125. (d) Novák, P.; Müller, K.; Santhanam, K.; Haas,
O., Chem. Rev. 1997, 97, 207ꢀ282. (e) Kim, H.; Kwon, J. E.; Lee, B.; Hong, J.; Lee, M.; Park, S.
Y.; Kang, K., Chem. Mater. 2015, 27, 7258ꢀ7264. (f) Liang, Y.; Zhang, P.; Chen, J., Chem. Sci.
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
2
013, 4, 1330ꢀ1337. (g) Hong, J.; Lee, M.; Lee, B.; Seo, D.ꢀH.; Park, C. B.; Kang, K., Nat.
Commun. 2014, 5, 5355.
(
6) (a) Suga, T.; Ohshiro, H.; Sugita, S.; Oyaizu, K.; Nishide, H., Adv. Mater. 2009, 21, 1627ꢀ1630.
(b) Morita, Y.; Nishida, S.; Murata, T.; Moriguchi, M.; Ueda, A.; Satoh, M.; Arifuku, K.; Sato,
K.; Takui, T., Nat. Mater. 2011, 10, 947ꢀ951. (c) Nokami, T.; Matsuo, T.; Inatomi, Y.; Hojo, N.;
Tsukagoshi, T.; Yoshizawa, H.; Shimizu, A.; Kuramoto, H.; Komae, K.; Tsuyama, H., J. Am.
Chem. Soc. 2012, 134, 19694ꢀ19700. (d) DeBlase, C. R.; Hernández‐Burgos, K.; Rotter, J. M.;
Fortman, D. J.; dos S Abreu, D.; Timm, R. A.; Diógenes, I. C.; Kubota, L. T.; Abruña, H. D.;
Dichtel, W. R., Angew. Chem. Int. Ed. 2015, 54, 13225ꢀ13229.
(7) (a) Song, Z.; Qian, Y.; Zhang, T.; Otani, M.; Zhou, H., Adv. Sci. 2015, 2, 1500124. (b) Song, Z.;
Zhan, H.; Zhou, Y., Angew. Chem. Int. Ed. 2010, 122, 8622ꢀ8626.
(
8) (a) Kwon, M.ꢀS.; Choi, A.; Park, Y.; Cheon, J. Y.; Kang, H.; Jo, Y. N.; Kim, Y.ꢀJ.; Hong, S. Y.;
Joo, S. H.; Yang, C., Sci. Rep. 2014, 4, 7404. (b) Geng, J.; Bonnet, J.ꢀP.; Renault, S.; Dolhem, F.;
Poizot, P., Energy Environ. Sci. 2010, 3, 1929ꢀ1933.
(
9) Senoh, H.; Yao, M.; Sakaebe, H.; Yasuda, K.; Siroma, Z., Electrochim. Acta 2011, 56, 10145ꢀ
1
0150.
(10) Cram, D. J., Angew. Chem. Int. Ed. 1988, 27, 1009ꢀ1020.
(
11) Barnes, J. C.; Dale, E. J.; Prokofjevs, A.; Narayanan, A.; GibbsꢀHall, I. C.; Juríček, M.; Stern, C.
L.; Sarjeant, A. A.; Botros, Y. Y.; Stupp, S. I., J. Am. Chem. Soc. 2015, 137, 2392ꢀ2399.
1
6
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 28
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
(
12) (a) Wu, Y.; Young, R. M.; Frasconi, M.; Schneebeli, S. T.; Spenst, P.; Gardner, D. M.; Brown, K.
E.; Würthner, F.; Stoddart, J. F.; Wasielewski, M. R., J. Am. Chem. Soc. 2015, 137, 13236ꢀ13239.
(b) Schneebeli, S. T.; Frasconi, M.; Liu, Z.; Wu, Y.; Gardner, D. M.; Strutt, N. L.; Cheng, C.;
Carmieli, R.; Wasielewski, M. R.; Stoddart, J. F., Angew. Chem. Int. Ed. 2013, 125, 13338ꢀ13342.
13) Wu, Y.; Frasconi, M.; Gardner, D. M.; McGonigal, P. R.; Schneebeli, S. T.; Wasielewski, M. R.;
Stoddart, J. F., Angew. Chem. Int. Ed. 2014, 53, 9476ꢀ9481.
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
(
(
(
14) Chen, D.; Avestro, A. J.; Chen, Z.; Sun, J.; Wang, S.; Xiao, M.; Erno, Z.; Algaradah, M. M.;
Nassar, M. S.; Amine, K., Adv. Mater. 2015, 27, 2907ꢀ2912.
15) (a) Huang, W.; Zhu, Z.; Wang, L.; Wang, S.; Li, H.; Tao, Z.; Shi, J.; Guan, L.; Chen, J., Angew.
Chem. Int. Ed. 2013, 125, 9332ꢀ9336. (b) Zhu, Z.; Hong, M.; Guo, D.; Shi, J.; Tao, Z.; Chen, J., J.
Am. Chem. Soc. 2014, 136, 16461ꢀ16464.
(16) Gawroński, J.; Brzostowska, M.; Gawrońska, K.; Koput, J.; Rychlewska, U.; Skowronek, P.;
Nordén, B., Chem. Eur. J. 2002, 8, 2484ꢀ2494.
(17) Connelly, N. G.; Geiger, W. E., Chem. Rev. 1996, 96, 877ꢀ910.
(
18) (a) Sun, D.; Rosokha, S. V.; Kochi, J. K., J. Am. Chem. Soc. 2004, 126, 1388ꢀ1401. (b) Lü, J.ꢀM.;
Rosokha, S. V.; Kochi, J. K., J. Am. Chem. Soc. 2003, 125, 12161ꢀ12171. (c) Würthner, F.; Chen,
Z.; Dehm, V.; Stepanenko, V., Chem. Commun. 2006, 1188ꢀ1190. (d) Shao, C.; Grüne, M.; Stolte,
M.; Würthner, F., Chem. Eur. J. 2012, 18, 13665ꢀ13677.
(
19) (a) Kato, S.ꢀi.; Matsumoto, T.; Ideta, K.; Shimasaki, T.; Goto, K.; Shinmyozu, T., J. Org. Chem.
2006, 71, 4723ꢀ4733. (b) Kato, S.ꢀi.; Nonaka, Y.; Shimasaki, T.; Goto, K.; Shinmyozu, T., J. Org.
Chem. 2008, 73, 4063ꢀ4075. (c) Renault, S.; Geng, J.; Dolhem, F.; Poizot, P., Chem. Commun.
2
011, 47, 2414ꢀ2416.
(20) Williams, D.; Byrne, J.; Driscoll, J., J. Electrochem. Soc. 1969, 116, 2ꢀ4.
1
7
ACS Paragon Plus Environment

Page 19 of 28
Journal of the American Chemical Society
1
2
3
(
21) Following previously reported work (ref •16), we have synthesized a linear PMDI trimer, linked
by 1,2ꢀdiaminocyclohexane units and terminated by cyclohexylamine residues. Galvanostatic
measurements were conducted. We could not achieve galvanostatic cycling data, however,
because of the high solubility of this compound in the electrolyte.
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
(22) Neese, F., Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2012, 2, 73ꢀ78.
(23) Sinnecker, S.; Rajendran, A.; Klamt, A.; Diedenhofen, M.; Neese, F., J. Phys. Chem. A 2006, 110,
235ꢀ2245.
24) Neese, F., J. Comput. Chem. 2003, 24, 1740ꢀ1747.
(25) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H., J. Chem. Phys. 2010, 132, 154104.
2
(
(
(
(
26) Wang, P.; Anderko, A., Fluid Phase Equilib. 2001, 186, 103ꢀ122.
27) Sure, R.; Grimme, S., J. Comput. Chem. 2013, 34, 1672ꢀ1685.
28) Noorizadeh, S.; Shakerzadeh, E., Phys. Chem. Chem. Phys. 2010, 12, 4742ꢀ4749.
(29) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.;
Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.;
Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.;
Vreven, T.; Montgomery Jr., J. A.; Peralta, J. E.; Ogliaro, F.; Bearpark, M. J.; Heyd, J.; Brothers,
E. N.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell,
A. P.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.;
Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.;
Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma,
K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.;
1
8
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 28
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
Farkas, Ö.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Gaussian, Inc.:
Wallingford, CT, USA, 2009.
(30) Lu, T.; Chen, F., J. Comput. Chem. 2012, 33, 580ꢀ592.
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
9
ACS Paragon Plus Environment

Page 21 of 28
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Captions of Figures
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 1. Series of synthesized pyromellitic diimide (PMDI) molecules. The structural formulas,
schematic illustration, and spaceꢀfilling representation of the superstructures of PMDI derivatives. a)
Bis(cyclohexyl)pyromellitic diimide (PMDIꢀRef), b) cyclic dimeric pyromellitic diimide ((–)ꢀ2PMDI),
and c) cyclic trimeric pyromellitic diimide ((–)ꢀ3PMDIꢀꢁ).
ꢀ1
Figure 2. Differential pulse voltammetry (DPV) of PMDI derivatives. DPV (10 mV s scan rate / 25
mV pulse / 0.1 msec pulse time / 1 mV step size / 1 sec sample period / 1 mM of PMDI derivatives /
DMF / 298 K / 0.1 M nꢀBu NPF ) results of three PMDI compounds. Peak separations during reduction
4
6
indicate the presence of throughꢀspace electron communication of the PMDI units in the macrocycle.
Figure 3. UV–Vis–NIR spectrophotometry of reduced PMDI derivatives. (aꢀc) Temperatureꢀdependent
o
o
o
UV–Vis spectrophotometry from 20 C (blue) to 80 C (red) in 10 C increments (0.5 mM of PMDI
derivatives / DMF / 1 equiv. CoCp per PMDI unit). (dꢀf) Concentrationꢀdependent UVꢀVisꢀNIR
2
spectrophotometry (0.05–0.5 mM of PMDI derivatives / DMF / 1 equiv. CoCp per PMDI unit), while
2
the concentration dependent relationship for (–)ꢀ2PMDI at 1600 nm is shown in the inset of e)
Compound (–)ꢀ2PMDI is observed to show a large absorption in the nearꢀIR range, which is indicative
of intramolecular interactions. Based on our experiments, (–)ꢀ3PMDIꢀꢁ does not appear to have
noticeable intramolecular interactions.
2
0
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 28
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
Figure 4. Galvanostatic measurements of PMDI derivatives. The galvanostatic voltage profile for the
first three cycles of a) PMDIꢀRef (red), b) (–)ꢀ2PMDI (green), and c) (–)ꢀ3PMDIꢀꢁ (blue) at C/20.
Calculated voltage profile of each compounds are described (orange) in right axis. d) The cycling
performance at the current rate of C/5 and, e) rate capability test of each PMDI derivatives. f) Charge
and discharge voltage profiles of (–)ꢀPMDIꢀꢁ at different current rates. In all the galvanostatic
measurements, the same current rate was applied during both charge and discharge.
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 5. Changes of Atoms in Molecules (AIM) and bond lengths during redox process. Illustration of
(aꢀc) PMDIꢀref, (dꢀf), (–)ꢀ2PMDI, and (gꢀi), (–)ꢀ3PMDIꢀꢁ at neutralꢀ (a, d, g), monoꢀ (b, e, h) and
exhaustive (c, f, i, – two Li atoms per each PMDI in the molecule) lithium redox process. During the
lithiation processes of PMDIꢀRef, major electron density is transferred to the lithiated carbonyl group
carbons, producing a more localized reduced state. In case of (–)ꢀ2PMDI and (–)ꢀ3PMDIꢀꢁ, lithiumꢀion
is stabilized by each carbonyl oxygens belonging to different sides of the macrocycles. Both carbonyl
groups accept approximately equal amounts of electron density, providing higher stabilization of the
fully lithiated state. Changes in atomic charges of carbon (black font), oxygen (red font) and selected C–
O bond lengths (grey font) are calculated relative to the corresponding neutral (a, d, g – nonꢀlithiated)
structures. Atom color code–carbon (black), nitrogen (blue), oxygen (red), and lithium (magenta).
2
1
ACS Paragon Plus Environment

Page 23 of 28
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 1
2
2
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 28
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
2
3
ACS Paragon Plus Environment

Page 25 of 28
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
2
4
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 26 of 28
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
2
5
ACS Paragon Plus Environment

Page 27 of 28
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
2
6
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 28 of 28
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
TOC Image
2
7
ACS Paragon Plus Environment