The Topology of Molecules with Twelve Fused Phenyl Rings ([12]Circulenes):

Rings, Infinitenes, and Möbius Infinitenes

Steven M. Bachrach

Artis College of Science and Technology, Radford University, Radford VA 24142 USA

[email protected]

Abstract

Following the recent preparation of infinitene (J. Am. Chem. Soc. 2022, 144, 862-871), a

computational (ωB97XD/6-311G(d)) exploration of 42 isomeric compounds with 12 fused phenyl rings

identified structures with linking number of zero (ring, saddle, and ribbon shapes), two (infinitene-like

shape), and one (Möbius infinitene shape) is reported. An infinitene isomer composed of two

[5]helicene fragments connected to two stacked phenyl rings and a Möbius infinitene isomer are

identified that are more stable than the known infinitene. The energies of the structures are examined

by assessing their strain energies, π-stacking, and possible aromaticity. Examples of fused phenyl

molecules with linking numbers of 3, 4, 5, and 6 are shown, indicating the potential topological range

that these molecules can possess.

1
Introduction

With the recent synthesis, isolation, and characterization of large molecules composed of fused

aromatic rings, exemplified by graphene sheets, nanorods, and fullerenes, chemists and physicists were

inspired to create and examine smaller fragment molecules.1 These have included planar and curved

graphenes,2 buckybowls,3 and nanobelts.4 Interest has expanded beyond fused systems to include

molecules like the cycloparaphenylenes,5 where the aromatic rings are linked but not fused. All of these

novel small polycyclic aromatic molecules may provide interesting optical and electrical properties,6

especially as semiconductors.

Members of the family of molecules composed of 12 fused phenyl rings have garnered interest

for decades. Kekulene 1 (Figure 1) was first prepared by Diederich in 1978.7 Its aromaticity has been

much discussed, and a recent atomic force microscopy study suggests that 1 consists of alternating

benzene and non-aromatic rings.8 Itami synthesized the nanobelt 2.4a It is a cyclophenacene where the

rings alternate in an ortho and a para fusion pattern. The analogue with the all ortho-fusion is [12]-

circulene 3. Quantum computations indicates that 3 has a saddle geometry.9 The cycloacene 4 has the

twelve rings fused in a para fashion. This molecule, like 3, has not yet been prepared, though this family

of molecules was proposed as far back as 1954.10 These four molecules have very different geometries:

planar (1), ribbon (2 and 4) and saddle (3).

2
 Figure 1. Structures of some [12]circulenes.

In 2021, Itami reported the synthesis of infinitene 5, consisting of twelve fused phenyl rings that

form a figure eight, or an infinity symbol.11 This molecule is topologically distinct from 1-4. These four

molecules have linking number12 Lk of zero, indicating a molecule whose graph can be written within a

plane. The linking number of infinitene 5 is two, indicating that the molecule makes one full twist. It can

be thought of as a ribbon that is cut, one end twisted one full rotation, and then stitched back together.

The graph of 5 cannot be drawn in a plane without some edges crossing over each other.

3
 The synthesis of 5 inspired our pursuit of a Möbius form of twelve fused phenyl rings. Many

studies have explored a variety of aromatic molecules with a Möbius topology, including polyenes,

porphyrins, and cycloparaphenylenes.13 The key distinction of the Möbius topology is its half twist: a

ribbon is cut, one end is twisted one-half rotation, and then the ends of the ribbon are stitched

together. A Möbius system has linking number of one. More generally, the linking number of Möbius

systems is odd while Hückel systems have Lk values that are even. The molecular graphs of Möbius

systems cannot be drawn in a plane without edges that cross.

For simplicity, we adopt the term circulene to describe any poly-fused-phenyl system

irrespective of whether the fusion is ortho or para. It allows for a molecule to contain a mix of these

connections. This terminology allows for 1-5 to all be incorporated into the category of [12]circulene.

This article reports a DFT computational survey of a variety of [12]circulenes that demonstrate a

range of topologies. This category spans linking numbers of 0, 1, and 2, with shapes that range from

planar, to ribbon, to saddle, to half-twisted to full-twisted. Examples of some even more twisted

circulenes conclude this report.

Computational Methods

The structures of all compounds were fully optimized, with symmetry imposed where possible,

at ωB97XD/6-311G(d).14 Analytical frequencies were computed to confirm that every structure is a local

energy minimum. The unscaled zero-point vibrational frequencies were used to compute enthalpies at

298 K. NMR chemical shifts were computed at B3LYP/6-31+G(d,p) with solvent (chloroform) modeled

using SMD and the ωB97XD/6-311G(d) optimized geometries. All computations were performed using

Gaussian-16.15 Linking numbers were computed using the LinkTW and LinkTWM programs.16

4
Results and Discussion

Designating [12]circulene configuration

In order to aid in differentiating the many different [12]circulenes, we propose the following

nomenclature scheme. The circulenes will be considered as composed of a series of three fused-phenyl

rings. A symbol will be applied to each ring of the circulene based on its type, such that every unique

[12]circulene will have a unique 12-character designation.

There are two molecules composed of three fused phenyl rings: anthracene and phenanthrene.

The ring in the central position of an anthracene unit will be designated by “A” (Figure 2). There are two

orientations for phenanthrene. If moving left to right through the centers of the three rings is clockwise,

that central ring will be designated “C”, while the counterclockwise connection will be designated “D”

(Figure 2).

Application of this rule to two simple [5]acenes 6 and 7 is shown in Figure 2. The designation for

the three central rings is DAD for 6 and DAC for 7 when viewing left to right. When viewing right to left,

the designation for 7 is unchanged, but is CAC for 6. Furthermore, if the orientations of the two

molecules is flipped, as in the bottom of Figure 2, the designations for 6 are unchanged, but for 7 it is

CAD. The multiple designations for each molecule is a result of the combinatorics: the arbitrary

orientation of the molecule should not matter. We opt to choose the designation that has more C rings

and has them earlier in the designation.

5
 Figure 2. Definition of A, B, and C connectivity, with application to 6 and 7.

Combinatorial equivalencies are prevalent in the [12]circulenes. The starting ring position is

arbitrary, and so for any random [12]circulenes, one could start at any of the rings and then proceed

arbitrarily clockwise or counterclockwise about the macrocycle. This leads to 24 different designations

for the same molecule. This is a well-known problem, often contextualized as asking for the number of

unique necklaces formed of 12 beads, where each bead can be of one of three colors.17 The

redundancies here reduce the number of unique configurations from 312=531,441 to 44,368. Symmetry

considerations further reduce the number of unique configurations, but this number remains above

20,000.

6
 The designations of 1-5 are listed in Table 1. Examination of all possible [12]circulenes is a

daunting prospect, and so a small representative survey of [12]circulenes is reported here,

demonstrating the topological range of this family. Since these [12]circulenes are very similar in

structure, the InChIs and InChIKeys18 of all reported compounds were also determined and helped to

confirm that each compound is unique. These InChIs and InChIKeys are given in the Supporting

Materials.

Table 1. Nomenclature of 1-5.

Strain energy

Simple examination of the 2-D graphs of circulenes 1-5 suggest that many of these molecules

will exhibit significant strain, along with various degrees of aromaticity. Disentangling these two

countervailing effects is impossible since one cannot choose a reference system that uniquely accounts

for strain without also affecting the aromaticity; closure of reference compound to the macrocyle

simultaneously affects both aromaticity and strain.19

The group equivalent reaction20 preserves chemical groups, is homodesmotic,21 and attempts to

conserve significant bond relationships. The group equivalent reaction to assess the strain energy of 5 is

shown as Reaction 1. This reaction conserves the number of ortho acene fusions, as opposed to

Reaction 2 employed by Itami.11 The energy estimated by these two reactions is quite different: 55.1

kcal mol-1 using Reaction 1 and 42.2 kcal mol-1 with Reaction 2. This 13 kcal mol-1 difference is solely

7
attributable to inconsistencies in the reference compounds and has nothing to do with the strain of the

macrocyle itself. Reaction 1, which conserves more structural features of the macrocyle than does

Reaction 2, provides the better metric.

12 5 + 12 Rxn 1

6 5 + 6 Rxn 2

n + m [12]circulene + 12 Rxn 3

where n+m = 12

Reaction 1 can be generalized for any [12]circulene by Reaction 3. Reaction 3 conserves the

number of anthracenyl and phenanthrenyl (combining both types C and D) subunits. It is important to

keep in mind that comparisons of energies provided by Reaction 3 reflect a combination of differences

in strain energies and differences in aromatic stabilizations, though for simplicity we refer to them as

simply strain energies.

[12]Circulenes with linking number of zero: rings

The first set of [12]circulene isomers all have linking number of zero. These are molecules with a

ring geometry. They fall into three broad generic groupings. Molecules in the first group have a roughly

planar structure. The second group has a saddle-like geometry. The last group has a ribbon-like

geometry. A total of 19 different [12]circulenes with a ring geometry have been computed, shown in

Figure 3.

8
Figure 3. [12]circulene ring isomers

9
 The most stable [12]circulene isomers have planar-like structures. The most stable isomer

overall is R-1, known as kekulene. It is nearly 8 kcal mol-1 more stable than the next lowest energy

member R-2. R-1 has been controversially called “superaromatic”.8, 22 It is the only example of a

[12]circulene whose strain energy is negative; it is more stable that the reference of six anthracene

molecules and six phenanthrene molecules by 16 kcal mol-1 (Table 2). While the term “superaromatic”

has been discounted, R-1 certainly expresses an unusual stability. The extensive π-delocalization about

this perfectly planar molecule (see Figure 4) is unique to this class of [12]circulenes. Any change to the

alternating “CAC” pattern results in a non-planar structure that must have less than ideal π-overlaps,

and a higher relative energy.

10
Table 2. Relative energy and strain energy (kcal mol-1) of the ring [12]circulenes (R-1 – R-19)
Cmpd Designation Erel SE

R-1 CACACACACACA 0.00 -16.08

R-2 CACACACACCDC 7.79 4.41

R-3 CACACCDCCCDC 15.03 24.36

R-4a CACCDCCACCDC 18.27 27.59

R-4b CACCDCCACCDC 18.36 27.68

R-5 CAACACCAACAC 30.02 13.94

R-6 CCACAACCAACA 35.27 19.19

R-7 CCACCACCACCA 45.83 42.45

R-8 CCCDCCCDCCCD 57.20 79.23

R-9 CDACDADCADCA 104.02 100.64

R-10 CCCCCACCCCCA 104.16 113.48

R-11 CADACADACADA 104.81 88.73

R-12 CCCDACDCDCAD 107.80 117.13

R-13 CCCCCCCCCCCC 107.89 129.91

R-14 CADCAACADAAD 112.09 96.01

R-15 CAACAADAADAA 117.13 88.35

R-16 CAAAACDCAAAA 126.07 97.29

R-17 CCDDCCDDCCDD 130.52 152.54

R-18 CDAAAAAAAAAA 190.98 149.50

R-19 AAAAAAAAAAAA 213.93 159.75

11
 R-1 R-2

R-10 R-13

R-9 R-19

Figure 4. Structures of representative ring-like [12]circulene isomers

The next three lowest energy [12]circulene isomers (Table 2) exchange some “A” connections

with “C” or “D” connections. This results in a having at least two regions of four rings fused in a helicene

arrangement that must adopt a non-planar arrangement to avoid hydrogen clashes. R-2 has two such

helicene regions (Figure 4); it is non-planar and is about 8 kcal mol-1 higher in energy than R-1. R-3 has a

five-ring helicene and two four-ring helicene regions and is 15 kcal mol-1 less stable than R-1. R-4 has

12
four helicene regions and its two conformations (of C2v or C2h symmetry) both are about 18 kcal mol-1

higher in energy than R-1.

The helicene regions and the overall non-planar structure manifest in increased strain energy. R-

2 has a strain energy of 4 kcal mol-1, while R-3 and R-4 are strained by more than 24 kcal mol-1. While the

rough trend is for increasing RSE to track with increasing relative energy of the isomers, it is not a linear

relationship. For example, while R-2 is 7.8 kcal mol-1 higher in energy than R-1, its RSE is 20 kcal mol-1

more than that of R-1. This can be attributed to differences in the number of anthracenes and

phenanthrenes as the reference. Phenanthrene is lower in energy than anthracene, and so a circulene

with more “C” and “D” connections will have a lower reference energy. R-2 has two more “C/D”

connections than does R-1, and all things being equal, it should be more stable than R-1. In fact, R-2 is

less stable than R-1 so it must be more strained.

The second set of [12]circulenes with Lk=0 have a saddle structure. The saddle geometry is

favored by isomers that have a preponderance of “C” connections. The computational study by Hopf9 of

circulenes with only “C” connections indicated a saddle geometry when the compound had seven or

more fused rings. R-13 is this example for the [12]circulenes, made up of 12 “C” connections (Figure 4).

It is much higher in energy than R-1 (108 kcal mol-1) with a RSE of nearly 130 kcal mol-1. Replacing two

“C” connections with “A” connections (R-10) maintains the saddle geometry, and while it does reduce

both the relative and strain energies, it remains quite high in energy.

The last set of ring circulenes represent ribbon geometries. In general, these ribbon isomers are

among the highest energy [12]circulenes. The lowest energy isomer is R-9, lying 104 kcal mol-1 above R-

1. The isomer R-11 has been prepared, again by the Itami group, though our estimate of its strain

energy, 89 kcal mol-1, is substantially lower than their estimate of 120 kcal mol-1.4a The ribbon isomer

having all “A” connections R-19 is the highest energy [12]circulene: it is 213 kcal mol-1 higher in energy

than R-1. It is also the most strained isomer, with an RSE of 160 kcal mol-1. Every phenyl ring must be

13
significantly distorted from planarity in order to close the ribbon. Linear acene fragments of R-19 were

compared to their planar analogues to estimate this non-planar distortion. Extrapolating the energy

difference to linear [12]acene suggests a strain energy of about 135 kcal mol-1 simply due to distortion

from non-planarity (see Supporting Materials for details).

[12]Circulenes with linking number of two: infinitenes

The first [12]circulene with a linking number of two was prepared by the Itami group and

reported in 2022.11 I-2 (referred to in the Introduction as 5) was named infinitene for its structural

resemblance to the infinity symbol. The Itami group employed Reaction 2 and PBE0/6-311+G(d,p)

energies to estimate the strain energy of I-2 at 60.2 kcal mol-1 (see Table 3). Reaction 2 underestimates

the strain energy principally by using benzene as a reference compound. Reaction 1 conserves the

number of bridgehead and non-bridgehead aromatic carbon atoms, along with the number of

phenanthrenyl units (and anthracenyl units with Reaction 3). The strain energy is about 13 kcal mol-1

larger with Reaction 1 than with Reaction 2.

The figure-eight (or infinity sign) structure of I-2 enables a π-stacking arrangement at the

crossover region. The distance between the center of the stacked phenyl rings in the experimental

structure of I-2 is 3.152−3.192 Å. Itami and coworkers selected the PBE0/6-311+G(d,p) method in part

for its prediction of this distance of 3.192 Å, which is at the outer edge of the experimental range.

Grimme23 has noted that π-stacking compounds require a density functional that accounts for

dispersion, something that PBE0 lacks. Both B3LYP/6-311G(d) and ωB97X/6-311G(d) predict (Table 3)

long π-stacking distances that are dramatically reduced when a dispersion correction is included. Both of

these corrected methods predict a shorter stacking distance, and the distance in the ωB97XD/6-311G(d)

optimized structure is 3.163 Å, close to the low end of the experimental range.

Computation of the strain energies of R-1, R-7, and R-9 at B3LYP/6-311G(d), B3LYP-D3/6-

311G(d), ωB97X/6-311G(d), and ωB97XD/6-311G(d) (See Table 3) shows a difference of less than 4 kcal

14
mol-1 with and without the dispersion correction. This is opposed the large difference (15-17 kcal mol-1)

with the inclusion of a dispersion correction for I-2. The ring structures have no π-stacking and thus only

a small effect is seen with the dispersion correction. When a molecule (like I-2) possesses π-stacking, the

dispersion correction is necessary. Thus, the structural and strain energy arguments supports the

selection of the ωB97XD/6-311G(d) method for this study of the [12]acenes. So, the best estimate of the

strain energy of I-2, using Reaction 1 and ωB97XD/6-311G(d) energies, is 55.1 kcal mol-1.

Table 5. Strain energy (kcal mol-1) and π-stacking distance (Å) of I-2 computed with various methods.
Method SE (Rxn 1) SE (Rxn 2) r(π-stack)

ωB97X/6-311G(d) 72.0 58.4 3.198

ωB97X-D/6-311G(d) 55.1 42.2 3.163

PBE0/6-311G(d) 76.0 65.7 3.138

PBE0/6-311+G(d,p)11 60.2 3.192

B3LYP/6-311G(d) 74.7 68.8 3.248

B3LYP-D3/6-311G(d) 49.3 39.8 3.134

Expt11 3.152−3.192

Eight different configurations of [12]circulene (see Figure 5) were identified that have an

infinitene-like structure, i.e. the figure-eight arrangement with a linking number of 2. Their geometries

are shown in Figure 6, and their relative energies are listed in Table 4. All of these infinitenes possess a

set of 4-6 consecutive “C” connections (with I-8 having a “CCACC” stretch) and a set of 4-6 consecutive

“D” connections (again with I-8 having a “DDADD” stretch). These two regions create the turns of their

figure-eight structures.

15
 Figure 5. [12]circulene infinitene isomers

Table 4. Relative energy and strain energy (kcal mol-1) of the infinitene-like [12]circulenes (I-1 – I-8)
Cmpd Designation Erela SE
I-1 CCCCCADDDDDA 32.09 41.41
I-2 CCCCCCDDDDDD 33.11 55.13
I-3 CCCCCADDDADD 38.30 47.62
I-4 CCCCAADDDDAA 38.80 35.42
I-5 CCCCCDDDDDAA 45.33 54.65
I-6 CCCCCDDADDDD 54.14 69.81
I-7 CCCCCCADDDDD 60.45 76.12
I-8 CCACCADDADDA 94.01 90.64
a
Energy relative to that of R-1.

16
 I-1 I-2

I-3 I-4

I-5 I-6

I-7 I-8

Figure 6. Structures of [12]circulene infinitene isomers

17
 Interestingly, the infinitene that the Itami group prepared (I-2) is not the lowest energy

infinitene isomer. Rather, I-1 lies 1 kcal mol-1 below I-2. It is also significantly less strained than I-2: 41.4

kcal vs. 55.1 kcal mol-1. I-1 is characterized by two stacked phenyl rings at the crossing point of the figure

eight. These two phenyl rings are the central ring of an anthracenyl fragment (“A” type configuration).

The loops are closed by five consecutive “C”-type phenyl rings on one side and five consecutive “D”

types on the other side. The structural features are in contrast to those of I-2, namely, two stacked

naphthalenyl groups in the crossing region, part of six consecutive “C”-type and six consecutive “D” type

phenyl rings.

The distance between the centers of the two stacked anthracenyl fragments in I-1 is 3.061 Å. As

discussed with I-2, this distance increases to 3.081 Å when the dispersion correction is removed. More

important than that structural effect is that the strain energy of I-1 is about 14 kcal mol-1 larger at

ωB97X/6-311G(d) than when the dispersion correction is included.

Another interesting infinitene is I-4, characterized by having two stacked naphthalenyl groups,

each part of crossing tetracenyl moieties. While I-4 is 6.7 kcal mol-1 less stable than I-1, it has the lowest

strain energy of the eight computed infinitenes, only 35.4 kcal mol-1.

The strain energies of I-1, I-2, and I-4 were furthered analyzed by assessing one source of strain,

due to the distortion of the helicene component, and a source of stabilization, π-stacking. For I-2 and 1-

4, the energy of the hexahelicene fragment fixed to its geometry in the infinitene is compared to the

energy of fully optimized hexahelicene. For 1-1, the same energy comparison was made for the

pentahelicene fragment. (More details can be found in the Supporting Information.) The helicene

distortion energies are listed in Table 5. The two helicene components in I-4 are each 7 or 12 kcal mol-1

less strained than those in I-1 or I-2, respectively.

18
 Table 5. Helicene distortion energy and π-stacking energy in I-1, I-2, and I-4.a
Helicene distortion π-stacking
I-1b 42.8 -9.6
I-2c 53.4 +3.5
I-4c 29.4 -8.4
a
See Supporting Information for further details. bHelicene with five phenyl rings and anthracene
dimer. cHelicene with six phenyl rings and naphthalene dimer.

To assess the π-stacking energy, the dimers of naphthalene (for the crossing in I-2 and I-4) and

anthracene (for the crossing in I-1) were fully optimized. The two molecules were then separated to the

π-stacking distance found in the respective infinitenes, and a rotational potential energy surface was

computed. Using this PES, the dimer energy was estimated based on the rotational angle in each

infinitene. (Again, more details can be found in the Supporting Information.) The estimated π-stacking

energies are listed in Table 5. The π-stacking is stabilizing by about 9 kcal mol-1 for both I-1 and I-4, but

destabilizing (i.e. less stable than two isolated naphthalene molecules) by 3.5 kcal mol-1 for I-2.

From Table 4, the strain energy these three infinitenes are ordered: I-4 < I-1 < 1-2. The

components of the strain energy listed in Table 5 mimic these strain energies nicely. I-4, the least

strained infinitene, has the lowest helicene distortion energy and significant π-stacking stabilization. I-1,

experiences slightly more π-stacking stabilization, offset by more helicene distortion than I-4. The most

strained of these thee infinitenes, I-2, has the greatest helicene distortion energy and its π-stacking is

destabilizing.

The upshot is that in addition to the already prepared infinitene I-2, a number of other infinitene

isomers should be viable synthetic targets.

19
[12]Circulenes with linking number of one: Möbius infinitenes

The last set of [12]circulenes that were explored are those that have a Möbius topology. Möbius

infinitenes will have linking number of one, indicating that a half-twist occurs as one traverses across the

face of the molecule, coming out on the opposite face. A second traverse returns to the original starting

point. A total of fifteen different configurations of [12]circulene were identified that have a Möbius

topology, M-1 through M-15, see Figure 7. The structures of the eight lowest energy Möbius infinitenes

are displayed in Figure 8. The relative energies and strain energies of the Möbius infinitenes are listed in

Table 6. For all but two of these Möbius infinitenes, the connections of the rings are of the C-type, with

a small number of “A” connections providing the means for the half-twist.

20
Figure 7. [12]circulene Möbius infinitene isomers

21
Figure 8. Structure of the eight lowest energy Möbius infinitene isomers

22
 Table 6. Relative energy and strain energy (kcal mol-1) of the Möbius [12]circulenes (M-1 – I-12)
Cmpd Designation Erel SE
M-1 CCCACACACCCA 31.33 27.95
M-2 CCCCACCACCCC 33.02 42.34
M-3 CCCCACACCCCA 45.41 48.38
M-4 CCCCACCACCCA 47.32 50.29
M-5 CCCACCCACCCA 63.37 66.34
M-6 CCCCCDCCCCCC 66.08 88.11
M-7 CCCCCAACCCCC 75.48 84.80
M-8 CCCCCADDDDDC 79.11 94.78
M-9 CCCCAACCCCAA 80.21 76.83
M-10 CCCCCACCCCCC 81.53 97.21
M-11 CCACCACCACCA 82.61 79.23
M-12 CCCCCACCCCCA 97.32 106.65
M-13 CCCCCCDCCCCC 100.72 122.74
M-14 CAACAACAACAA 112.92 84.14
M-15 CCCCCCCCCCCA 138.61 154.29

The most compelling result pertaining to the Möbius infinitenes is that two of these structures,

M-1 and M-2, are lower in energy than I-2, the infinitene prepared by Itami.11 In fact, M-1 is lower in

energy than I-1 as well, though by less than 1 kcal mol-1. The stability of these two Möbius infinitenes

results from their low strain energy. In fact, M-1 has the smallest strain energy of all of the infinitenes

and Möbius infinitenes, and it is less strained than all but the five most stable ring analogues.

A component of the strain energy of a circulene is the energy required for distorting the

geometry into some curved shape. For M-1, a number of schemes for assessing the distortion energy

were examined. In each scheme, M-1 is decomposed into fragments, and the energy difference for each

fragment between its fully optimized geometry and its geometry in M-1 was computed. The four

different decomposition schemes are shown in the Supporting Materials (along with further details of

how these computations were performed), and one example is displayed in Scheme 1. In Scheme 1, M-1

is divided into two fragments, each of which optimize to planar structures. The top fragment has a

23
distortion energy of 22.8 kcal mol-1, while the bottom fragment is more seriously deformed, having a

distortion energy of 34.9 kcal mol-1. The total distortion energy of M-1 is thus estimated at 57.7 kcal

mol-1. The other three schemes predict a larger distortion energy, ranging from 59 to 74 kcal mol-1. As a

comparison, the total distortion energy of I-2 can be estimated as twice the distortion energy of its

[6]helicene fragment (Scheme 1); the computed value is 53.4 kcal mol-1.

Scheme 1. Distortion energy computation of M-1 and I-2

For I-2, this distortion energy is very similar to its strain energy, 53.4 vs. 55.1 kcal mol-1 (Table 1).

On the other hand, the distortion energy of M-1 is significantly larger than its strain energy, 57.7 vs. 28.0

kcal mol-1. This result suggests that M-1 possesses some stabilization. Though the structure of M-1

24
positions three hydrogen atoms in the interior that point towards the lower phenyl ring, a model of this

interaction suggests that it is slightly destabilizing (see Supporting Materials). The possible stabilization

from delocalization or aromaticity are addressed in the next section.

Another consideration is the difference in twisting of the infinitenes and Möbius infinitenes.

Molecules such as the infinitenes have a shape defined by twist Tw and writhe Wr, the sum of which

provides the linking number Lk.12 Twist is rotation about the imaginary central axis lying within the plane

of the phenyl rings. Writhe is rotation of the ribbon that creates crossings. Listed in Table 7 are the

values of the twist and writhe for the six lowest energy infinitenes and Möbius infinitenes. There is no

simple trend that relates twist or writhe to the overall stability or strain energy of these circulenes.

However, since M-1, M-2, I-1, and I-2 all lie within two kcal mol-1 of each other, the overall half twist or

full twist is not affecting their stability.

Table 4. Values of twist (Tw), writhe (Wr) and linking numbers (Lk) of the three lowest energy ring,
infinitene and Möbius infinitene isomers
cmpd Tw Wr Lk
I-1 1.22 0.78 2
I-2 1.17 0.83 2
I-3 1.40 0.60 2
I-4 1.15 0.85 2
I-5 1.19 0.81 2
I-6 1.44 0.56 2
M-1 0.61 0.39 1
M-2 0.55 0.45 1
M-3 0.40 0.60 1
M-4 0.54 0.46 1
M-5 0.86 0.14 1
M-6 0.53 0.47 1

25
NMR, HOMA and aromaticity

NMR 1H and 13C chemical shifts were computed to help identify the structure of any

[12]circulenes that might be prepared and to aid in assessing their aromatic character. The first step was

to assess the performance of the computational methodology. Following guidance provided by Tantillo

at the CHESHIRE web site,24 chemical shifts were computed at B3LYP/6-31+G(d,p) with solvent

(chloroform) modeled using SMD for comparison with the experimental values of R-17 and I-2.11

Excellent linear correlations are seen for both the proton and carbon chemical shifts for I-2 and for the

proton shifts for R-1 (see Supporting Materials). The slopes for these correlations are very close to one.

The computed proton chemical shifts of R-11, 8.39 and 7.53 ppm, are in excellent agreement with

experiment, 8.27 and 7.52 ppm. 4a Thus, the computed chemical shifts of the [12]circulenes can be used

without any further corrections.

The computed chemical shifts for all of the [12]circulenes are reported in the Supporting

Materials. The chemical shifts for the three lowest energy ring, infinitene, and Möbius infinitene isomers

are displayed in Figure 9. Some characteristic signal similarities are apparent for each set of isomers. For

the low energy ring isomers, which have a pseudo-planar structure, the chemical shifts of the interior

protons are far downfield, from 9.7 to 11.8 ppm. For the infinitenes, the chemical shifts of the protons in

the crossing region of the figure-eight are shifted slightly upfield, between 6 and 7 ppm. The most

distinctive proton signals for M-1 and M-2 arise from the protons positioned above the crossing and

pointing down towards the face of the phenyl rings below; their signals are shifted downfield to 10.8

and 9.4 ppm in M-1 and 9.2 ppm in R-2.

26
 8.65 8.25 7.90
8.18 7.82 6.86 7.60
8.25 8.28
A B 7.66
C B 7.41
A B
8.34
8.17
10.91 A C
D
9.40 8.00
10.75 D
7.90
G F E
9.23 7.91
7.65 7.72
I-1 M-1 7.46
R-1
8.25 8.91 8.12
7.62 7.33 7.68
6.43 8.12
8.31 8.18
C D E A 8.33
8.58 7.05 B 8.24
8.48 B F B 8.10
A C
8.65 8.23 9.19 7.89
9.40 A G 8.22 C
9.01
10.90 8.72 7.89
F D
7.50 E 7.97
7.76 7.89
R-2 I-2 M-2
8.45 8.12 7.58 7.36
8.57 6.96 8.06
A 8.49 7.08 7.80 8.19
6.21 A
B 9.44 A 6.73 8.04 B
7.46 9.37 8.14 6.17 8.16
LG 7.31
C 8.42 B H C
8.22
8.71 D 8.06 C 8.14
8.90 F K I
8.18 D
11.12
E 7.34 8.24
D 7.73 J 9.00
F 8.31 8.07 E 8.01 E F
7.34 8.22
G 7.64
8.47 7.90 7.88 8.09 7.82
8.03 7.52 7.24 7.86
R-3 I-3 M-3

Figure 9. Computed 1H chemical shifts (ppm) of select low-energy [12]circulenes

Chemical shifts are often indicative of aromatic character. Almost all of the [12]circulenes

examined here exhibit chemical shifts in the typical aromatic region (7-9 ppm). There are two notable

exceptions: the two highest energy isomers R-18 and R-19. R-19 is a cyclic acene with only A-type (para

fusion) connectivity and R-18 has 10 out of the 12 rings with A connectivity. This connectivity results in a

ribbon-like structure necessitating significant non-planarity of the phenyl rings, which might imply a

lessening of their aromatic character. In fact, the proton chemical shifts of R-19 are at 2.7 ppm, while

most of the chemical shifts of R-18 are below 6 ppm, with two as low as 1.6 ppm. Despite the fact that

27
all of the examined isomers (except R-1) are non-planar and require some amount of distortion of the

phenyl rings from planarity, only when that distortion becomes extreme do non-aromatic-like proton

chemical shifts result.

Nucleus Independent Chemical Shifts (NICS) is a widely utilized method to assess aromaticity

based on magnetic effects.25 NICS values around -10 ppm are indicative of aromatic character, while

values near zero indicate non-aromatic character. NICS is typically evaluated either at the center of the

ring, NICS(0), or at a point 1 Å above or below the ring, NICS(1). NICS(1)26 is often preferred as it

minimizes the role of the σ-electrons. However, the geometries of the infinitenes and Möbius infinitenes

often involve stacked rings or close approaches of atoms and a ring face, making NICS(1) problematic.

Therefore, NICS(0) values were computed and reported in the Supporting Materials. NICS(0) values at

the center of each phenyl ring for the three lowest energy isomers of the three classes are listed in Table

6.

Another commonly used method for assessing aromaticity is the Harmonic Oscillator Model

of Aromaticity (HOMA), based on bond distances about the ring.27 HOMA values range from 1,

representing the aromatic character of benzene, to 0, having alternating bonds about the ring. The

HOMA values for each phenyl ring of the three lowest energy isomers are also listed in Table 6.

28
 Table 6. NICS and HOMA values of select low-energy [12]circulenes
ring HOMA NICS HOMA NICS HOMA NICS
R-1 R-2 R-3
A 0.36 -3.28 0.55 -5.09 0.66 -6.03
B 0.92 -9.21 0.76 -6.83 0.71 -6.68
C 0.49 -4.93 0.59 -5.45
D 0.90 -9.50 0.73 -5.90
E 0.35 -2.62 0.50 -5.34
F 0.92 -9.19 0.89 -9.68
G 0.39 -3.69 0.33 -1.55
I-1 I-2 I-3
A 0.87 -11.42 0.69 -7.01 0.54 -5.20
B 0.48 -5.53 0.64 -6.84 0.89 -9.76
C 0.66 -5.75 0.48 -4.93 0.52 -3.46
D 0.46 -5.75 0.70 -6.48
E 0.43 -3.56
F 0.90 -11.04
G 0.47 -4.39
H 0.78 -5.45
I 0.69 -4.55
J 0.72 -4.87
K 0.68 -5.10
L 0.68 -6.82
M-1 M-2 M-3
A 0.93 -9.50 0.48 -4.89 0.34 -0.62
B 0.38 -2.54 0.85 -9.80 0.86 -10.1
C 0.91 -9.66 0.36 -3.93 0.33 -5.39
D 0.44 -4.01 0.73 -5.51 0.50 -6.09
E 0.79 -5.83 0.69 -4.97 0.67 -6.02
F 0.26 -2.33 0.67 -5.48 0.52 -4.01
G 0.88 -8.53 0.87 -9.62

Inspection of the values of NICS and HOMA in Table 6 indicates a correlation between the two,

though the linear fit is poor for the infinitenes and Möbius infinitenes. In general though, both measures

agree on whether an individual phenyl ring has strong or weak aromatic character. For example, the

phenyl rings of R-1 alternate between having little aromatic character (NICS of -3.3ppm, HOMA of 0.4)

or strong aromatic character (NICS of -9.2 ppm, HOMA of 0.9). This result is consistent with the recent

29
AFM study8 that shows alternating aromatic and non-aromatic rings in R-1. Similar alternation of

aromatic character is seen in all three Möbius infinitenes M-1, M-2, and M-3 and in I-3. In contrast, I-1

has every fourth ring with large aromaticity (NICS of -11.4 ppm) with the other rings showing modest

aromaticity (NICS of -5.5 to -5.8 ppm). The rings of I-2 show even less variation, with NICS values ranging

from -5 to -7 ppm.

NICS, particularly when applied to polycyclic compounds, is a local index and may not reflect a

more global assessment of a molecule’s aromaticity.28 Magnetically induced current density may provide

a more global metric of such molecules. For example, kekulene R-1 exhibits a diatropic current along the

outer periphery and a paratropic current around the inner ring.29 Computation30 of the magnetically

induced current density of I-2 shows two non-intersecting diatropic current-density pathways, indicating

aromatic character.

These data are not providing a consistent message regarding the aromatic character of the

infinitenes. The root cause is the poor definition of aromaticity,31 one that is just a collection of

properties, including stabilization energy, structural properties like planarity and bond length, and

magnetic properties including chemical shifts, NICS, and induced current density.

In examining this new class of twisted polycyclic aromatic compounds, our interest resides in

their possible preparation. That centers our focus on their energetics and possible aromatic stabilization

energy. Examination of acyclic linear and bent acenes provide some guidance here. Aromatic

stabilization generally decreases with increasing size of the acene.32 This can be seen in the decreasing

stability of the linear acenes relative to anthracene (Table 7).

30
 Table 7. Stabilization energya (kcal mol-1) of the linear acenes
Stabilization
Length energy
5 3.9
6 6.5
7 9.4
8 12.4
9 15.4
10 18.5
11 21.5
12 24.6
a
Stabilization energy defined by reaction analogous to Reaction 3, see Supporting Materials for details.

Clar’s sextet theory33 argues that the dominant resonance structure of an aromatic compound is

the one that maximizes the number of 6-member rings with 3 alternating double bonds. For the linear

acenes, every resonance structure can have only one such “aromatic” 6-member ring (Figure 10).

Therefore, fusing another ring adds two more double bonds that can aid to delocalize the π-electrons

but does nothing to create additional aromaticity.

Figure 10. Dominant resonance structure and its Clar representation for anthracene and phenanthrene.

On the other hand, bent acenes, like phenanthrene, have a resonance structure with more than

one Clar-type 6-member ring (see Figure 10). An interesting consequence of this is that acenes with

multiple kinks, which allows for more Clar-type rings, are more stable than the reference compounds

31
anthracene and phenanthrene. This can be seen for isomers of nonacene (Table 8) and other examples

are provided in the Supporting Materials. (Isomers with adjacent meta fusion were omitted to examine

only planar structures.)

Table 8. Stabilization energya (kcal mol-1) and NICS (PPM) of some non-linear nonacenes.

-2.1 -9.6
1.5
-8.0 -10.4
-12.1 -12.5
-12.3
-7.7 -8.4

AAACAAA AACACAA
-4.4
-4.0 -10.6
-10.4
-3.1
-10.9

-11.1 -5.6

-9.2 -9.8
ACACACA CACACAC

Nonacene SE
AAACAAA -0.1
AACACAA -7.0
ACACACA -9.1
CACACAC -7.7
a
Stabilization energy defined by reaction analogous to Reaction 3, see Supporting Materials for details
and additional examples.

The NICS values of the bent nonacenes correspond to the Clar sextets. In particular, every ring

that occupies a kink position has a decidedly reduced, i.e., less negative, NICS value, consistent with

their diminished aromatic character.

32
 Despite the indication from the NICS values, which might suggest reduced aromaticity of these

bent nonacenes, they are all more stable than the reference compounds anthracene and phenanthrese.

Since these two molecules are commonly considered to be aromatic, it is logical to conclude that the

bent nonacenes, are likewise aromatic.

It is therefore consistent to conclude that kekulene R-1 is also aromatic, as it too is more stable

than the reference compounds (six anthracene and six phenanthrene molecules). Despite the fact that

the NICS values of the phenyl rings alternate -9.2 and -3.3 ppm, suggesting that every other ring is

strongly aromatic and weakly aromatic, the energetic criterion decidedly argues for the aromaticity of

kekulene.

None of the other ring circulenes, nor any of infinitene or Möbius infinitenes, are more stable

than their reference aromatic compounds. All of these molecules display NICS values at the kinked

phenyl positions that are near zero, but that alone cannot discount the possibility of overall aromaticity.

In fact, as discussed above, the kinked positions allow for adjacent rings to adopt the Clar sextet. So, all

six of the anthracenyl phenyl rings of kekulene R-1 adopt the Clar sextet, alternating with the

phenanthrenyl phenyl rings that do not fulfil the sextet. Similarly, the phenyl rings in I-1, I-2, and M-1

(among others) also have alternating rings that fulfil the Clar sextet. This too suggests that these

molecules are aromatic.

The case for aromaticity of M-1 is further buttressed by its energy partitioning. The overall strain

energy of M-1 is 28.0 kcal mol-1. This strain can be portioned into a distortion energy and a stabilization

energy. The distortion energy, as defined in Scheme 1, is 57.7 kcal mol-1. That implies a stabilization

energy of nearly 30 kcal mol-1. That stabilization energy can be associated with delocalization and/or

aromaticity. Taken together with the alternating Clar sextet and NICS value among the phenyl rings, the

far downfield chemical shifts, it seems apparent that M-1 is aromatic.

33
Highly twisted circulenes

With larger circulenes, ones with more phenyl rings, other topologies are possible. In particular,

highly twisted geometries can be obtained. To highlight these topologies, the geometries of four highly

twisted circulenes (HTC), see Figure 11, were computed. The intent here is to demonstrate the range of

structural possibilities and not to identify the lowest energy possibilities.

HTC-1 is a [20]circulene and has Lk = 3. This is a Möbius topology with one and one-half twists.

Its isomer HTC-2 has Lk = 4, making two full twists. HTC-2 is 111 kcal mol-1 lower in energy than HTC-1.

HTC-3 is a [28]circulene with Lk = 5, making two and one-half twists. Its isomer HTC-4 has Lk=6, making

three full twists. HTC-4 lies 85 kcal mol-1 below HTC-3. The structures are displayed in Figure 12, though

their three-dimensional shape is probably better viewed through a movie (see the Supporting Material)

or with interactive molecular visualization software.

34
 HTC-1 CCCCCAAADAACADAACAAA

HTC-2 CCCCCADADACCCCCADADA

HTC-3 CCCCCADADACACAACADAADADACACA

HTC-4 CCCCCADADACACADDDDDADADADADA

Figure 11. Examples of highly twisted circulenes (HTC)

35
 HTC-1

HTC-2

HTC-3

HTC-4

Figure 12. Structures of HTC-1 – HTC-4

36
 An interesting structural feature of these highly twisted circulenes are their π-stacking regions.

HTC-1 has one crossover where the phenyl rings stack. The centers of these two phenyl rings are

separated by 3.25 Å, and the ring centers are offset very slightly. On the other hand, the two crossover

regions in HTC-2 position the phenyl rings significantly offset. This can be seen in the closeup truncated

view of the crossover region of HTC-2 in Figure 13. The center of each phenyl ring sits above a carbon

atom of the phenyl ring in the other phenyl of the stacked pair, with a separation of 3.186 Å. This so-

called parallel displaced arrangement of the benzene dimer is more energetically favorable than the

arrangements where the two rings are directly above one another.34 This parallel displaced arrangement

is also seen in HTC-3 and HTC-4, where the separation is 3.231 and 3.205 Å, respectively. The distance

between the two phenyl rings in the central crossover of HTC-4 is 3.536 Å, a long distance, suggesting

little favorable stacking energy resulting from this crossover.

Figure 13. Crossover region of HTC-2 showing one of the helicene regions. The rest of the molecule has
been deleted and the hydrogens omitted. The purple ball defines the center of phenyl ring, and the
labeled distance (Å) is the closest approach between the center of one phenyl ring and a carbon atom of
the other phenyl of the stacked pair.

37
 The computed strain energies of the two Möbius structures HTC-1 and HTC-3 are quite large,

133 and 85 kcal mol-1, respectively. Further exploration of the possible isomers might identify more

amenable targets examples. However, the strain energies of the two Hückel structures HTC-2 and HTC-4

reveal them to be viable synthetic targets. The strain energy of HTC-4 is only 9 kcal mol-1, and HTC-2 is

estimated to be more stable than its reference compounds by about 10 kcal mol-1.

Conclusions

Fused polyphenyl systems can express a range of interesting topologies. [12]circulenes

exhibiting a linking number of zero (ring7 and ribbon4a) or two (infinitene11 I-2) have been synthesized.

The computations presented here identified eight isomers of I-2 with Lk = 2, including one isomer I-1

that is predicted to be energetically more stable than the original infinitene I-2. We have also identified

fifteen [12]circulene isomers that have a Möbius topology, with Lk = 1. The most stable Möbius

infinitene M-1 is lower in energy than any of the infinitene isomers.

The aromatic character of the circulenes is a matter of debate, resting largely upon what criteria

one chooses to utilize. The 1H chemical shifts of almost all of the [12]circulenes examined here fall in the

range typically associated with aromatic compounds (6-8ppm). The exception are the ribbon structures

with a preponderance of para fusion, causing significant distortion of many of the phenyl rings away

from planarity. NICS and HOMA values indicate an alternating strong aromaticity and weak aromaticity

for the phenyl rings in many of the low energy ring structures, infinitenes, and Möbius infinitenes.

However, this behavior reflects a maximization of Clar structures. Lastly, comparison of the energies of

the [12]circulenes with reference compounds (phenanthrene and anthracene) along with estimation of

distortion energies suggests that many of the [12]circulenes exhibit stabilization energies comparable to

those of linear or bent acenes. It appears that these compounds deserve to be categorized as aromatic.

38
 These computations suggest that a number of infinitenes and Möbius infinitenes should be

viable synthetic targets. Additionally, the low strain energies of more highly twisted circulene suggest

that they too may be good candidates for synthesis. The circulenes appear to be ground for fruitful

exploration, elaborating molecules with unusual topologies that may lead to interesting properties and

utilities.

Supporting Information

Part 1: Complete citation for Ref. 15, InChIs and InChIKeys of the [12]circulenes, Tables S1-S6, Figures

S1-S5. Part 2: Coordinates of the [12]circulenes and HTC1-4, computed 1H and 13C NMR chemical shifts

and NICS(0) values of the [12]circulenes. Movies depicting rotation of HTC-1 – HTC-4.

Notes

The author declares no competing financial interest.

Acknowledgments

The author thanks Monmouth University for the resources where the computations were performed.

References

(1) Majewski, M. A.; Stępień, M. Bowls, Hoops, and Saddles: Synthetic Approaches to Curved Aromatic

Molecules. Angew. Chem. Int. Ed. 2019, 58, 86-116.

(2) (a) Allen, M. J.; Tung, V. C.; Kaner, R. B. Honeycomb Carbon: A Review of Graphene. Chem. Rev. 2010,

110, 132-145. (b) Ghuge, D. A.; Shirode, R. A.; Kadam, J. V. Graphene: A Comprehensive Review. Curr.

Drug Targets 2017, 18, 724-733. (c) Madurani, K. A.; Suprapto, S.; Machrita, N. I.; Bahar, S. L.; Illiya, W.;

39
Kurniawan, F. Progress in Graphene Synthesis and its Application: History, Challenge and the Future

Outlook for Research and Industry. ECS J. Solid State Sci. Tech. 2020, 9, 093013.

(3) (a) Wu, Y.-T.; Siegel, J. S. Aromatic Molecular-Bowl Hydrocarbons: Synthetic Derivatives, Their

Structures, and Physical Properties. Chem. Rev. 2006, 106, 4843-4867. (b) Wu, Y.-T.; Siegel, J. S.

Synthesis, Structures, and Physical Properties of Aromatic Molecular-Bowl Hydrocarbons. Top. Curr.

Chem. 2014, 349, 63-120. (c) Stuparu, M. C. Corannulene: A Curved Polyarene Building Block for the

Construction of Functional Materials. Acc. Chem. Res. 2021, 54, 2858-2870.

(4) (a) Povie, G.; Segawa, Y.; Nishihara, T.; Miyauchi, Y.; Itami, K. Synthesis of a carbon nanobelt. Science

2017, 356, 172-175. (b) Cheung, K. Y.; Gui, S.; Deng, C.; Liang, H.; Xia, Z.; Liu, Z.; Chi, L.; Miao, Q.

Synthesis of Armchair and Chiral Carbon Nanobelts. Chem 2019, 5, 838-847.

(5) (a) Yamago, S.; Kayahara, E.; Iwamoto, T. Organoplatinum-Mediated Synthesis of Cyclic π-Conjugated

Molecules: Towards a New Era of Three-Dimensional Aromatic Compounds. Chem. Rec. 2014, 14, 84-

100. (b) Takaba, H.; Omachi, H.; Yamamoto, Y.; Bouffard, J.; Itami, K. Selective Synthesis of

[12]Cycloparaphenylene. Angew. Chem. Int. Ed. 2009, 48, 6112-6116. (c) Golder, M. R.; Jasti, R.

Syntheses of the Smallest Carbon Nanohoops and the Emergence of Unique Physical Phenomena. Acc.

Chem. Res. 2015, 48, 557-566. (d) Lewis, S. E. Cycloparaphenylenes and related nanohoops. Chem. Soc.

Rev. 2015, 44, 2221-2304. (e) Hermann, M.; Wassy, D.; Esser, B. Conjugated Nanohoops Incorporating

Donor, Acceptor, Hetero- or Polycyclic Aromatics. Angew. Chem. Int. Ed. 2021, 60, 15743-15766.

(6) Kumar, R.; Aggarwal, H.; Srivastava, A. Of Twists and Curves: Electronics, Photophysics, and

Upcoming Applications of Non-Planar Conjugated Organic Molecules. Chem. Eur. J. 2020, 26, 10653-

10675.

(7) Diederich, F.; Staab, H. A. Benzenoid versus Annulenoid Aromaticity: Synthesis and Properties of

Kekulene. Angew. Chem. Int. Ed. Engl. 1978, 17, 372-374.

40
(8) Pozo, I.; Majzik, Z.; Pavliček, N.; Melle-Franco, M.; Guitián, E.; Peña, D.; Gross, L.; Pérez, D. Revisiting

Kekulene: Synthesis and Single-Molecule Imaging. J. Am. Chem. Soc. 2019, 141, 15488-15493.

(9) Christoph, H.; Grunenberg, J.; Hopf, H.; Dix, I.; Jones, P. G.; Scholtissek, M.; Maier, G. MP2 and DFT

calculations on circulenes and an attempt to prepare the second lowest benzolog, [4]circulene. Chem. -

Eur. J. 2008, 14, 5604-5616.

(10) Heilbronner, E. Molecular Orbitals in homologen Reihen mehrkerniger aromatischer

Kohlenwasserstoffe: I. Die Eigenwerte yon LCAO-MO's in homologen Reihen. Helv. Chim. Acta 1954, 37,

921-935.

(11) Krzeszewski, M.; Ito, H.; Itami, K. Infinitene: A Helically Twisted Figure-Eight [12]Circulene

Topoisomer. J. Am. Chem. Soc. 2022, 144, 862-871.

(12) Rappaport, S. M.; Rzepa, H. S. Intrinsically Chiral Aromaticity. Rules Incorporating Linking Number,

Twist, and Writhe for Higher-Twist Möbius Annulenes. J. Am. Chem. Soc. 2008, 130, 7613-7619.

(13) (a) Walba, D. M.; Richards, R. M.; Haltiwanger, R. C. Total synthesis of the first molecular Moebius

strip. J. Am. Chem. Soc. 1982, 104, 3219-3221. (b) Walba, D. M.; Homan, T. C.; Richards, R. M.;

Haltiwanger, R. C. Topological Stereochemistry. 9. Synthesis and Cutting in Half of a Molecular Mobius

Strip. New J. Chem. 1993, 17, 661-681. (c) Martin-Santamaria, S.; Lavan, B.; Rzepa, H. S. Mobius

Aromatics Arising from a C=C=C Ring Component. Chem. Commun. 2000, 1089-1090. (d) Ajami, D.;

Oeckler, O.; Simon, A.; Herges, R. Synthesis of a Möbius aromatic hydrocarbon. Nature 2003, 426, 819-

821. (e) Rzepa, H. S. A Double-Twist Möbius-Aromatic Conformation of [14]Annulene. Org. Lett. 2005, 7,

4637-4639. (f) Castro, C.; Chen, Z.; Wannere, C. S.; Jiao, H.; Karney, W. L.; Mauksch, M.; Puchta, R.;

Hommes, N. J. R. v. E.; Schleyer, P. v. R. Investigation of a Putative Möbius Aromatic Hydrocarbon. The

Effect of Benzannelation on Möbius [4n]Annulene Aromaticity. J. Am. Chem. Soc. 2005, 127, 2425-2432.

(g) Castro, C.; Karney, W. L.; Valencia, M. A.; Vu, C. M. H.; Pemberton, R. P. Möbius Aromaticity in

[12]Annulene: Cis-Trans Isomerization via Twist-Coupled Bond Shifting. J. Am. Chem. Soc. 2005, 127,

41
9704-9705. (h) Castro, C.; Karney, W. L.; Vu, C. M. H.; Burkhardt, S. E.; Valencia, M. A. Computational

Evaluation of the Evidence for Tri-trans-[12]Annulene. J. Org. Chem. 2005, 70, 3602-3609. (i) Ajami, D.;

Hess, K.; Köhler, F.; Näther, C.; Oeckler, O.; Simon, A.; Yamamoto, C.; Okamoto, Y.; Herges, R. Synthesis

and Properties of the First Möbius Annulenes. Chem. Eur. J. 2006, 12, 5434-5445. (j) Herges, R. Topology

in Chemistry:  Designing Möbius Molecules. Chem. Rev. 2006, 106, 4820-4842. (k) Tahara, K.; Tobe, Y.

Molecular Loops and Belts. Chem. Rev. 2006, 106, 5274-5290. (l) Allan, C. S. M.; Rzepa, H. S. Chiral

Aromaticities. A Topological Exploration of Möbius Homoaromaticity. J. Chem. Theory Comput. 2008, 4,

1841-1848. (m) Rzepa, H. S. Lemniscular Hexaphyrins as Examples of Aromatic and Antiaromatic Double-

Twist Möbius Molecules. Org. Lett. 2008, 10, 949–952. (n) Stępień, M.; Sprutta, N.; Latos-Grażyński, L.

Figure Eights, Möbius Bands, and More: Conformation and Aromaticity of Porphyrinoids. Angew. Chem.

Int. Ed. 2011, 50, 4288-4340. (o) Schaller, G. R.; Herges, R. Moebius molecules with twists and writhes.

Chem. Commun. 2013, 49, 1254-1260. (p) Schaller, G. R.; Topić, F.; Rissanen, K.; Okamoto, Y.; Shen, J.;

Herges, R. Design and synthesis of the first triply twisted Möbius annulene. Nature Chem. 2014, 6, 608-

613. (q) Tanaka, T.; Osuka, A. Chemistry of meso-Aryl-Substituted Expanded Porphyrins: Aromaticity and

Molecular Twist. Chem. Rev. 2017, 117, 2584-2640. (r) Fan, Y.-Y.; Chen, D.; Huang, Z.-A.; Zhu, J.; Tung, C.-

H.; Wu, L.-Z.; Cong, H. An isolable catenane consisting of two Möbius conjugated nanohoops. Nature

Commun. 2018, 9, 3037. (s) Naulet, G.; Sturm, L.; Robert, A.; Dechambenoit, P.; Röhricht, F.; Herges, R.;

Bock, H.; Durola, F. Cyclic tris-[5]helicenes with single and triple twisted Möbius topologies and Möbius

aromaticity. Chem. Sci. 2018, 9, 8930-8936. (t) Nishigaki, S.; Shibata, Y.; Nakajima, A.; Okajima, H.;

Masumoto, Y.; Osawa, T.; Muranaka, A.; Sugiyama, H.; Horikawa, A.; Uekusa, H.; Koshino, H.; Uchiyama,

M.; Sakamoto, A.; Tanaka, K. Synthesis of Belt- and Möbius-Shaped Cycloparaphenylenes by Rhodium-

Catalyzed Alkyne Cyclotrimerization. J. Am. Chem. Soc. 2019, 141, 14955-14960. (u) Dash, S.; Ghosh, A.;

Srinivasan, A.; Suresh, C. H.; Chandrashekar, T. K. Protonation-Triggered Hückel and Möbius Aromatic

Transformations in Nonaromatic Core-Modified [30]Hexaphyrin(2.1.1.2.1.1) and Annulated

42
[28]Hexaphyrin(2.1.1.0.1.1). Org. Lett. 2019, 21, 9637-9641. (v) Senthilkumar, K.; Kondratowicz, M.; Lis,

T.; Chmielewski, P. J.; Cybińska, J.; Zafra, J. L.; Casado, J.; Vives, T.; Crassous, J.; Favereau, L.; Stępień, M.

Lemniscular [16]Cycloparaphenylene: A Radially Conjugated Figure-Eight Aromatic Molecule. J. Am.

Chem. Soc. 2019, 141, 7421-7427. (w) Jiang, X.; Laffoon, S. D.; Chen, D.; Pérez-Estrada, S.; Danis, A. S.;

Rodríguez-López, J.; Garcia-Garibay, M. A.; Zhu, J.; Moore, J. S. Kinetic Control in the Synthesis of a

Möbius Tris((ethynyl)[5]helicene) Macrocycle Using Alkyne Metathesis. J. Am. Chem. Soc. 2020, 142,

6493-6498. (x) Schaub, T. A.; Prantl, E. A.; Kohn, J.; Bursch, M.; Marshall, C. R.; Leonhardt, E. J.; Lovell, T.

C.; Zakharov, L. N.; Brozek, C. K.; Waldvogel, S. R.; Grimme, S.; Jasti, R. Exploration of the Solid-State

Sorption Properties of Shape-Persistent Macrocyclic Nanocarbons as Bulk Materials and Small

Aggregates. J. Am. Chem. Soc. 2020, 142, 8763-8775. (y) Bachrach, S. M.; Rzepa, H. S.

Cycloparaphenylene Möbius trefoils. Chem. Commun. 2020, 56, 13567-13570.

(14) Chai, J.-D.; Head-Gordon, M. Long-range corrected hybrid density functionals with damped atom-

atom dispersion corrections. Phys. Chem. Chem. Phys. 2008, 10, 6615-6620.

(15) Frisch, M. J.; et al. Gaussian-16, rev. B.01; Gaussian, Inc.: Wallingford CT, 2016.

(16) Rzepa, H. S.; Rappaport, S. LinkTW and LinkTWM Programs for computing Linking numbers, Twist

and Writhe for cyclic annulenes. 2014.

https://figshare.com/articles/dataset/LinkTW_and_LinkTWM_Programs_for_computing_Linking_numb

ers_Twist_and_Writhe_for_cyclic_annulenes_/1207997 (accessed 12/1/2022).

(17) OEIS Foundation. Number of n-bead necklaces with 3 colors. In The On-line Encyclopedia of Integer

Sequences; 2022; https://oeis.org/A001867.

(18) Heller, S. R.; McNaught, A.; Pletnev, I.; Stein, S.; Tchekhovskoi, D. InChI, the IUPAC International

Chemical Identifier. J. Cheminformatics 2015, 7, 23.

(19) Bachrach, S. M. On the Resonance Energy of Methylenecyclopropene and Cyclopropenone. J. Org.

Chem. 1990, 55, 4961-4963.

43
(20) Bachrach, S. M. The Group Equivalent Reaction: An Improved Method for Determining Ring Strain

Energy. J. Chem. Ed. 1990, 67, 907-908.

(21) Wheeler, S. E.; Houk, K. N.; Schleyer, P. v. R.; Allen, W. D. A Hierarchy of Homodesmotic Reactions

for Thermochemistry. J. Am. Chem. Soc. 2009, 131, 2547-2560.

(22) (a) Cioslowski, J.; O'Connor, P. B.; Fleischmann, E. D. Is Superbenzene Superaromatic? J. Am. Chem.

Soc. 1991, 113, 1086-1089. (b) Aihara, J. Is superaromaticity a fact or an artifact? The kekulene problem.

J. Am. Chem. Soc. 1992, 114, 865-868. (c) Jiao, H.; Schleyer, P. v. R. Is Kekulene Really Superaromatic?

Angew. Chem. Int. Ed. 1996, 35, 2383-2386. (d) Zhou, Z. Are kekulene, coronene, and corannulene

tetraanion superaromatic? Theoretical examination using hardness indices. J. Phys. Org. Chem. 1995, 8,

103-107. (e) Haags, A.; Reichmann, A.; Fan, Q.; Egger, L.; Kirschner, H.; Naumann, T.; Werner, S.;

Vollgraff, T.; Sundermeyer, J.; Eschmann, L.; Yang, X.; Brandstetter, D.; Bocquet, F. C.; Koller, G.;

Gottwald, A.; Richter, M.; Ramsey, M. G.; Rohlfing, M.; Puschnig, P.; Gottfried, J. M.; Soubatch, S.; Tautz,

F. S. Kekulene: On-Surface Synthesis, Orbital Structure, and Aromatic Stabilization. ACS Nano 2020, 14,

15766-15775. (f) Fan, W.; Han, Y.; Wang, X.; Hou, X.; Wu, J. Expanded Kekulenes. J. Am. Chem. Soc. 2021,

143, 13908-13916. (g) Aihara, J.-i.; Makino, M.; Ishida, T.; Dias, J. R. Analytical Study of Superaromaticity

in Cycloarenes and Related Coronoid Hydrocarbons. J. Phys. Chem. A 2013, 117, 4688-4697.

(23) Ehrlich, S.; Bettinger, H. F.; Grimme, S. Dispersion-Driven Conformational Isomerism in σ-Bonded

Dimers of Larger Acenes. Angew. Chem. Int. Ed. 2013, 41, 10892–10895.

(24) Lodewyk, M. W.; Siebert, M. R.; Tantillo, D. J. Chemical Shift Repository (Cheshire). 2019.

http://cheshirenmr.info/ (accessed 12/1/2022).

(25) Schleyer, P. v. R.; Maerker, C.; Dransfeld, A.; Jiao, H.; Hommes, N. J. R. v. E. Nucleus-Independent

Chemical Shifts: A Simple and Efficient Aromaticity Probe. J. Am. Chem. Soc. 1996, 118, 6317-6318.

44
(26) Schleyer, P. v. R.; Jiao, H.; van Eikema Hommes, N. J. R.; Malkin, V. G.; Malkina, O. L. An Evaluation

of the Aromaticity of Inorganic Rings: Refined Evidence from Magnetic Properties. J. Am. Chem. Soc.

1997, 119, 12669-12670.

(27) (a) Kruszewski, J.; Krygowski, T. M. Definition of aromaticity basing on the harmonic oscillator

model. Tetrahedron Letters 1972, 13, 3839-3842. (b) Krygowski, T. M.; Szatylowicz, H.; Stasyuk, O. A.;

Dominikowska, J.; Palusiak, M. Aromaticity from the Viewpoint of Molecular Geometry: Application to

Planar Systems. Chem. Rev. 2014, 114, 6383-6422, lode. (c) Dobrowolski, J. C. Three Queries about the

HOMA Index. ACS Omega 2019, 4, 18699-18710.

(28) (a) Orozco-Ic, M.; Barroso, J.; Charistos, N. D.; Muñoz-Castro, A.; Merino, G. Consequences of

Curvature on Induced Magnetic Field: The Case of Helicenes. Chem. - Eur. J. 2020, 26, 326-330. (b)

Monaco, G.; Zanasi, R. Analysis of the Nucleus-Independent Chemical Shifts of [10]Cyclophenacene: Is It

an Aromatic or Antiaromatic Molecule? J. Phys. Chem. Lett. 2017, 8, 4673-4678.

(29) (a) Steiner, E.; Fowler, P. W.; Jenneskens, L. W.; Acocella, A. Visualisation of counter-rotating ring

currents in kekulene. Chem. Commun. 2001, 659-660. (b) Orozco-Ic, M.; Dimitrova, M.; Barroso, J.;

Sundholm, D.; Merino, G. Magnetically Induced Ring-Current Strengths of Planar and Nonplanar

Molecules: New Insights from the Pseudo-π Model. J. Phys. Chem A 2021, 125, 5753-5764.

(30) (a) Orozco-Ic, M.; Valiev, R. R.; Sundholm, D. Non-intersecting ring currents in [12]infinitene. Phys.

Chem. Chem. Phys. 2022, 24, 6404-6409. (b) Monaco, G.; Zanasi, R.; Summa, F. F. Magnetic

Characterization of the Infinitene Molecule. J. Phys. Chem A 2022, 126, 3717-3723.

(31) Bachrach, S. M. Computational Organic Chemistry; Wiley, 2014.

(32) (a) Solà, M. Forty years of Clar's aromatic π-sextet rule. Frontiers Chem. 2013, 1, 1-8, Review. (b)

Sakai, S.; Kita, Y. Structure prediction and aromaticity of acenes and phenylenes series. A combination

method between asymmetric Kekulé structures. J. Phys. Org. Chem. 2012, 25, 840-849.

(33) Clar, E. The Aromatic Sextet; J. Wiley, 1972.

45
(34) (a) Karton, A.; Martin, J. M. L. Prototypical π–π dimers re-examined by means of high-level

CCSDT(Q) composite ab initio methods. J. Chem. Phys. 2021, 154, 124117. (b) Sinnokrot, M. O.; Sherrill,

C. D. Highly Accurate Coupled Cluster Potential Energy Curves for the Benzene Dimer:  Sandwich, T-

Shaped, and Parallel-Displaced Configurations. J. Phys. Chem. A 2004, 108, 10200-10207.

46