Steric/π-Electronic Insulation of the carbo-Benzene Ring: Dramatic Effects of tert-Butyl versus Phenyl Crowns on Geometric, Chromophoric, Redox, and Magnetic Properties

Article abstract of DOI:10.1002/chem.201800835

Hexa-tert-butyl-carbo-benzene (C₁₈tBu₆) and three phenylated counterparts (C₁₈tBu_mPh_6?m; m=4, 2) have been synthesized. The peralkylated version (m=6) provides experimental access to intrinsic features of the insulated C₁₈ core independently from the influence of π-conjugated substituent. Over the series, structural, spectroscopic, and electrochemical properties are compared with those of the hexaphenylated reference (m=0). Anchoring tBu substituents at the C₁₈ macrocycle is shown to enhance stability and solubility, and to dramatically modify UV/Vis absorption and redox properties. Whereas all carbo-benzenes reported previously were obtained as dark-reddish/greenish solids, crystals and solutions of C₁₈tBu₆ happen to be yellow (λ_max=379 vs. 472 nm for C₁₈Ph₆). In comparison to C₁₈Ph₆, the reduction of C₁₈tBu₆ remains reversible, but occurs at twice as high an absolute potential (E_1/2=?1.36 vs. ?0.72 V). Systematic XRD analyses and DFT calculations show that the C₁₈ ring symmetry is the nearest to D_6h for m=6, which indicates a maximum geometric aromaticity. According to calculated nucleus-independent chemical shifts (NICS), the macrocyclic magnetic aromaticity is also maximum for C₁₈tBu₆: NICS(0)=?17.2 ppm versus (?18.0±0.1) ppm for the theoretical references C₁₈H₆ and C₁₈F₆, and ?13.5 ppm for C₁₈Ph₆. Accurate correlations of NICS(0) with experimentally recorded or calculated maximum UV/Vis absorption wavelengths, λ_max, and chemical hardness, η=E_LUMO?E_HOMO, are evidenced.

Full text of DOI:10.1002/chem.201800835

A Journal of
Accepted Article
Title: Steric/π-electronic insulation of the carbo-benzene ring:
dramatic effects of tert-butyl vs phenyl crowns on geometric,
chromophoric, redox and magnetic properties
Authors: Dymytrii Listunov, Carine Duhayon, Albert Poater, Serge
Mazères, Alix Saquet, Valérie Maraval, and Remi Chauvin
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201800835
Link to VoR: http://dx.doi.org/10.1002/chem.201800835
Supported by

10.1002/chem.201800835
Chemistry - A European Journal
FULL PAPER
Steric/π-electronic insulation of the carbo-benzene ring: dramatic
effects of tert-butyl vs phenyl crowns on geometric,
chromophoric, redox and magnetic properties
Dymytrii Listunov,[a,b] Carine Duhayon, Albert Poater,[c] Serge Mazères,[d] Alix Saquet,[a,b] Valérie
Maraval,*[a,b] Remi Chauvin*[a,b]
Abstract: Hexa-tert-butyl-carbo-benzene C18tBu6 and three
phenylated counterparts C18tBumPh6–m (m
4, 2) have been
Introduction
=
synthesized. The peralkylated version (m = 6) provides experimental
access to intrinsic features of the insulated C18 core independently
from the influence of π-conjugated substituent. Over the series,
structural, spectroscopical and electrochemical properties are
compared with those of the hexaphenylated reference (m = 0).
Anchoring tBu substituents at the C18 macrocycle is shown to enhance
stability and solubility, and to dramatically modify UV-vis absorption
and redox properties. Whereas all the carbo-benzenes reported
hitherto were obtained as dark-reddish/greenish solids, crystals and
solutions of C18tBu6 happen to be yellow (lmax = 379 nm vs 472 nm for
C18Ph6). By comparison to C18Ph6, reduction of C18tBu6 remains
reversible but occurs at a twice higher absolute potential (E1/2 = –1.36
V vs –0.72 V). Systematic X-ray diffraction analyses and DFT
calculations show that the C18 ring symmetry is the nearest to D6h for
m = 6, indicating a maximum geometrical aromaticity. According to
calculated nucleus independent chemical shifts, the macrocyclic
magnetic aromaticity is also found to be maximum for C18tBu6:
NICS(0) = –17.2 ppm, vs –18.0 ± 0.1 ppm for the theoretical
references C18H6 and C18F6, and –13.5 ppm for C18Ph6. Accurate
correlations of NICS(0) with experimental or calculated maximum UV-
Within the family of cumulene-containing conjugated cyclic
hydrocarbons,[1a] carbo-benzenes[1b,c] display particular optical,[2]
electrical,[3] magnetic,[4] or mesogenic[5] properties, that can be a
priori attributed to their common C18 core, combining unique
geometrical and π-electronic features. In spite of a 4n+2 πz-
electron count, however, the moderate energetic aromaticity of
the C18 ring (ca one third of the C6 ring of benzene)[6] enables
conjugation with π-unsaturated substituents. All the carbo-
benzene cores reported to date are actually substituted by either
six alkynyl groups (as in 1)[7] or at least three aryl groups (as in
2),[8] most of them with four phenyl groups (as in 3),[1b] and up to
six in 4 (Fig. 1).[8,9] The outer π-conjugation is evidenced by low
dihedral angles ꢀ between the C18 ring mean plane and the mean
planes of the aryl substituents (0 ≤ ꢀ < 20°). No more than two or
three of the remaining substituents can be H or alkyl groups (e.g.
in 2a-b[8] or 3a-c),[10,11] and since the first unsuccessful attempt at
synthesis of the unsubstituted carbo-benzene C18H6,[12] no
experimental example of a π-isolated C18 core has been at
disposal for appraising its intrinsic electronic properties. With the
additional requirement of preventing π-π stacking of C18 rings, the
challenge is here addressed by decorating the macrocycle with a
crown of six tert-butyl substituents. Beyond the C18tBu6 target 5,
partially phenylated congeners C18tBumPh6-m (6, 7, 8 in Scheme
1) are also envisaged. Comparison with the hexaphenylated
reference 4 (m = 0) is ultimately envisaged with the view to
appraising the incremental effect of outer conjugation with phenyl
groups, e. g. by UV-vis absorption spectroscopy. All the known
carbo-benzenes are indeed highly chromophoric, with a UV-vis
profile reminiscent of the UV-vis profile of porphyrins:[2,13] one
main absorption band and smaller bands at higher wavelengths
can be compared with the Soret and Q bands of porphyrins,
respectively. Furthermore, the UV-vis spectra of carbo-benzenes
is also interpretable through the Gouterman four-orbital model
devised for porphyrins.[14] The maximum absorption wavelengths
vis absorption wavelength lmax and chemical hardness h = ELUMO
EHOMO are evidenced.
–
[a]
Dr. D. Listunov, Dr. C. Duhayon, Dr. A. Saquet, Dr. V. Maraval, Prof.
R. Chauvin
CNRS, LCC (Laboratoire de Chimie de Coordination), 205 route de
Narbonne, BP 44099, 31077 Toulouse Cedex 4, France.
E-mail: vmaraval@lcc-toulouse.fr; chauvin@lcc-toulouse.fr
Dr. D. Listunov, Dr. C. Duhayon, Dr. A. Saquet, Dr. V. Maraval, Prof.
R. Chauvin
Université de Toulouse, UPS, ICT-FR 2599, 118 route de Narbonne,
31062 Toulouse Cedex 9, France.
Dr. A. Poater
[b]
[c]
[d]
Institut de Química Computacional i Catàlisi and Departament de
Química, Universitat de Girona, Campus Montilivi, 17071 Girona,
Catalonia, Spain
l
max of all the known carbo-benzenes with a single C18 ring are in
the range 472 ± 48.5 nm, the mean value corresponding to the
max value of C18Ph6, 4 (lmax = 472 nm).[8,9] This range of variation
Dr. S. Mazères
UMR CNRS 5089, IPBS (Institut de Pharmacologie et de Biologie
Structurale), 205 route de Narbonne, 31077 Toulouse cedex,
France.
l
is twice larger than the one reported for porphyrins (420 ± 25 nm),
whatever the nature and position of the substituents on the
tetrapyrolic C20 macrocycle.[15] This difference indicates that the
C18 carbo-benzene core is more prone to external π-
delocalization than the C20N4 porphine core, thus revealing a
lower aromatic character of the former. Remarkably, the mean
value at 420 nm corresponds to the lmax value of the meso-
Supporting information for this article is given via a link at the end of
the document. It includes experimental synthesis procedures and
characterization data for all new compounds, crystallographic tables
and figures, voltammograms, NMR spectra, and detailed
computational data based on DFT-optimized Cartesian coordinates
of main compounds discussed in the main text.
1
This article is protected by copyright. All rights reserved.

10.1002/chem.201800835
Chemistry - A European Journal
FULL PAPER
tetraphenyl derivative P1, to be compared with the hexaphenyl
derivative 4 in the carbo-benzene series.
The synthesis of 5, 6, 7 and 8 has been challenged through an
adapted [8+10] macrocyclization route to their respective
[6]pericyclynediol precursors 9, 10, 11 and 12 (Scheme 1).[17] The
common di-tert-butylated synthon 13 was envisaged by a method
previously described for the preparation of the di-phenylated
version 14.[18]
Et3Si
SiEt3
Ar
R
OMe R2
tBu
tBu
Et3Si
SiEt3
R
Ar
OMe
tBu
R2
OMe
H
tBu
OMe
R1
C8 dinucleophile
HO
R1
R2
OH
=
tBu
13: R2 = tBu
14: R2 = Ph
H
R1
R1
Ar
R
2a: R = H, Ar = 4-tBu-C6H4
2b: R = tBu, Ar = Ph
1
+
R2
OMe
O
O
Et3Si
SiEt3
R1
R1
R2
5: R1 = R2
6: R1 = Ph, R2
7: R1 tBu, R2 = Ph
8: R1 = R2 = Ph
R2
C10 dielectrophile
OMe R2
2c: R = Ph, Ar = 4-tBu-C6H4
=
tBu
9: R1 = R2 tBu
=
=
tBu
10: R1 = Ph, R2 = tBu
11: R1 = tBu, R2 = Ph
12: R1 = R2 = Ph
Ph
Ph
Ph
Ph
R2
OMe
=
OMe R2
Scheme 1. Retrosynthetic [8+10] route to the fully and partially tert-butyl-
substituted carbo-benzenes from the common precursor 13.
R
R
Ph
Ph
Ph
Ph
Ph
Ph
4
3
Results and Discussion
R = H, aryl (≠ Ph), alkynyl, i-Pr or n-alkyl
3a: R = H
3b: R = n-C14H29
3c: R = i-Pr
1. Synthesis of the di-tert-butyl-triyne C8 precursor 13
3d: R = 4-NMe2-C6H4
The preparation of 13 was initiated from the ketone 15, obtained
in quantitative yield by AlCl3-promoted acylation of bis-
trimethylsilylacetylene with pivaloyl chloride (Scheme 2).[19,20]
Reaction of 15 with ethynylmagnesium bromide gave the diynol
16 in 80 % yield. Selective O-methylation of 16 by treatment with
one equivalent of n-BuLi and MeI led to 17 with 92 % yield.
Deprotonation of 17 with EtMgBr followed by addition to 15 gave
the triynol 18, which was readily O-methylated to the diether 19.
Subsequent proto-desilylation with K2CO3/MeOH afforded the
triyne 13, with 85 % yield from 19, and 42 % overall yield over six
steps from pivaloyl chloride.
A more straightforward method consisted in the direct addition of
the terminal diyne 16 to the ketone 15 in the presence of an
excess of LiHMDS, giving the triynediol 20 with 87 % yield.
Subsequent double O-methylation and C-desilylation allowed
access to 13 with 58 % yield over four steps only. This shorter
procedure was also applied in the preparation of the phenylated
triyne 14, widely used for the synthesis of carbo-benzenes of type
3 (Figure 1).[1b,10,11,13,18] The triynes 18-20 and 13 were obtained
as quasi-statistical mixtures of stereoisomers, and used as such
in the next steps. Structures were confirmed by X-ray diffraction
(XRD) analysis of single crystals of rac-16, rac-erythro (R*,R*)-18
and meso-13 spontaneously resolved by deposition from
DCM/pentane solutions (Supporting Information Figure S.3.1).[21]
R
HN
N
N
R
R
NH
R
P1: R = Ph
P2: R = tBu
5
Figure 1. Examples of know carbo-benzene derivatives and related porphyrins,
and the target 5.
The lowest known maximum absorption wavelengths amongst
carbo-benzenes with a single C18 ring were measured for the
triaryl representatives 2a and 2b (lmax = 424 and 427 nm),[8b] while
the highest value was recorded for the π-donor-substituted
hexaaryl carbo-benzene 3d (lmax = 521 nm).[3,16] A first open issue
for the target 5 is thus whether the π-insulating effect of the six
tBu groups will overcome their s-donating inductive effect that
could parallel the π-donating mesomeric effect of the two 4-
NMe2C6H4 groups of 3d. Furthermore, the target series
C18tBumPh6-m lends itself to systematic investigations of
fluorescence, structural and electrochemical properties. The
variation of electronic and magnetic properties vs m is also
addressed by DFT calculations below.
2
This article is protected by copyright. All rights reserved.

10.1002/chem.201800835
Chemistry - A European Journal
FULL PAPER
OMe tBu
19, 98 %
OMe tBu
18, 68 %
OMe tBu
H
1) EtMgBr
THF, 0 °C
1) H
THF, 0 °C
MgBr
tBu
1) n-BuLi
RO
O
tBu
OMe
tBu
OH
tBu
OMe
H
K2CO3
MeOH
2) MeI, DMSO
tBu
2) H+
TMS
2) 15
THF, - 78 °C
TMS
3) H+
TMS
TMS
TMS
TMS
H
13, 85 %
15
16, R = H, 80 %
1) n-BuLi
2) MeI, DMSO
1) n-BuLi
17, R = Me, 92 %
OH tBu
2) MeI, DMSO
THF, - 78 °C
tBu
OH
1) LiHMDS
16
2) 15
THF, - 78 °C
3) H+
TMS
TMS
20, 87 %
Scheme 2. Synthesis of the triyne 13, common precursor of the target carbo-benzenes 5-8 (Scheme 1).
effect of the replacement of phenyl groups by tert-butyl groups is
imputable to their mutual hindrance in 13 and 21.
2. Synthesis of dialdehyde and diketone C10 precursors
[8+10] Macrocyclization from the C8 triyne 13 was first envisaged
with the tert-butylated C10 dialdehyde 21 (R2 = H in Scheme 1).
Following a procedure used in the phenylated series from the
triyne 14,[17c,18] 13 was first deprotonated with two equivalents of
n-BuLi before addition to para-formaldehyde, and the resulting
diol 22 was oxidized with IBX to give 21 with 68 % yield over two
steps (Scheme 3).
OMe tBu
OMe tBu
tBu
OMe
tBu
OMe
IBX
1,2-DCE, ∆
H
H
HOH2C
CH2OH
22, 68 %
21, quant.
OMe tBu
O
O
As an alternative dielectrophile, the diketone 23 was
prepared in one step through "pseudo-Sonogashira" coupling of
tBu
OMe
1) n-BuLi
PhC(O)Cl
Pd(PPh3)2Cl2
2) (CH2O)n
3) H+
13 with benzoyl chloride,[22] using
a procedure previously
Ph
Ph
implemented with an arylated triyne (R2 = 4-nPent-C6H4 in
Scheme 1),[23] giving 23 with 54 % yield (Scheme 3). The structure
of meso-23 was confirmed by XRD analysis of single crystals,
spontaneously resolved from a solution of the diastereoisomeric
mixture (Figure S.3.1 in Supporting Information).[21,24]
CuI, Et3N, THF
OMe tBu
O
O
23, 54 %
OMe tBu
tBu
OMe
tBu
OMe
tBuC(O)Cl
H
H
13
Pd(PPh3)2Cl2
The Pd-catalytic method did not allow access to the di-tert-
butyl diketone 24 from 13 and pivaloyl chloride. Instead, addition
of the dilithium salt of 13 to two equivalents of pivalaldehyde gave
the diol 25, which, after oxidation with MnO2, led to 24 with 45 %
yield from 13 over two steps.
1) n-BuLi
tBu
tBu
CuI, Et3N, THF
2) tBuCHO
3) H+
24, 65 %
O
O
OMe tBu
tBu
OMe
MnO2
DCM
From the triyne 14 and pivaloyl chloride, however, the pseudo-
Sonogashira procedure allowed access to the di-tert-butyl
diketone 26, required for the synthesis of 7. The failure of the Pd-
catalytic method for the preparation of 24 can thus be ascribed to
the mutual hindrance of the tert-butyl groups in both 13 and
pivaloyl chloride (Scheme 3).
tBu
tBu
OH
HO
25, 69 %
OMe Ph
OMe Ph
Ph
OMe
O
Ph
OMe
H
tBuC(O)Cl
Pd(PPh3)2Cl2
CuI, Et3N, THF
3. Synthesis of the pericyclynediols 9-12 and carbo-benzenes 5-8
tBu
tBu
H
14
O
26, 52 %
The [6]pericyclynediol 27 (Scheme 4) was first envisaged as a
precursor of the [6]pericyclynedione 28, tetra-tert-butyl analogue
of 29, widely used for the synthesis of tetraphenyl-carbo-
benzenes of type 3 (Figure 1).[1b,10,11,13,18] However, treatment of
21 with the bromomagnesium salt of 13 did not give any trace of
27, letting the triyne unreacted while giving the diol 22, likely
resulting from Mg-mediated reduction of 21. Deprotonation of 13
with n-BuLi was also attempted, but only trace amounts of 27 and
monoadduct 30 could be detected. At last, trans-metallation of the
dilithium salt of 13 with CeCl3 prior to the addition of 21 allowed
isolation of 27 with a very low 1 % yield. Here again, the dramatic
Scheme 3. Synthesis of C10 dialdehyde and diketones from the triyne 13 or 14.
The synthesis of the [6]pericyclynediols 9-12 was then envisaged
by [8+10] macrocyclization of the diketones 23, 24 and 26 with
either the triynes 13 or 14 (Schemes 1, 3, 5). The target products
were thus obtained in 14-36 % yield by mixing the reactants in the
presence of LiHMDS.
3
This article is protected by copyright. All rights reserved.

10.1002/chem.201800835
Chemistry - A European Journal
FULL PAPER
tBu
OMe
OMe
tBu
OMe
OMe tBu
tBu
tBu
tBu
OMe
tBu
OMe
OMe tBu
HO
H
H
HO
H
O
tBu
OMe
H
1) [B]
O
+
+
2)
H
O
OH
OMe
H
OH
HO
H
H
H
13
tBu
OMe tBu
OMe
OMe
tBu
tBu
OMe
22,[B] = EtMgBr, nd
[B] = n-BuLi, 0 %
30, [B] = EtMgBr, 0 %
[B] = n-BuLi, traces
27, [B] = EtMgBr, 0 %
[B] = n-BuLi, traces
OMe tBu
21
[B] = n-BuLi, CeCl3, 0 %
[B] = n-BuLi, CeCl3, 3 %
[B] = n-BuLi, CeCl3, 1 %
?
MnO2
OMe
tBu
OMe Ph
tBu
tBu
OMe
OMe
Ph
OMe
OMe
O
O
O
O
Ph
OMe tBu
OMe Ph
29
28
Scheme 4. Outcome of reactions of the C8 triyne 13 with the C10 dialdehyde 21, while targeting the [6]pericyclynediol 27 (nd = yield not determined)
OMe
tBu
tBu
tBu
Reduction of 9-12 with SnCl2/HCl in DCM at room temperature
(r.t.) afforded the corresponding carbo-benzenes 5-8, decorated
with two to six tert-butyl groups (Scheme 5). Each reaction was
monitored by thin layer chromatography (TLC), showing both the
disappearance of the starting pericyclynediol spot and the
appearance of a strongly coloured spot. While the formation of 6-
8 was found to be completed after ca 15 minutes, significant
production of hexa-tert-butyl-carbo-benzene 5 required stirring
overnight at r. t. The yield was also smaller for 5 (30 %) than for 6
(54 %), 8 (56 %) and 7 (quantitative). While 6, 7 and 8 were
isolated as dark red or violet-greenish solids, the hexa-tert-butyl
tBu
tBu
OMe
tBu
1) SnCl2
HCl
HO
1) LiHMDS
13
tBu
tBu
2)
24
2) NaOH
tBu
OH
OMe
OMe tBu
9, 29 %
tBu
tBu
5, 30 %
OMe
tBu
tBu
tBu
tBu
OMe
1) SnCl2
HCl
HO
Ph
1) LiHMDS
Ph
Ph
t
2) NaOH
Ph
OH
2) 23
target 5 was found to be a yellow solid. Bu groups having no
intrinsic chromophoric properties, this unique colour within the
series can thus be directly imputed to the C18 ring itself (see
section 5). The tert-butyl substituent also have a dramatic effect
on both the solubility (s) and stability. As a qualitative comparison,
at least 30 mg of 5-8 could be dissolved in 1 mL of DCM or
chloroform, while less than 5 mg of 3 or 4 could be (e g. s = 0.3
mg/mL for 3b in chloroform).[9]
tBu
OMe
OMe tBu
10, 36 %
tBu
tBu
6, 54 %
OMe
tBu
tBu
tBu
tBu
OMe
tBu
1) SnCl2
HCl
HO
1) LiHMDS
tBu
tBu
2)
26
2) NaOH
tBu
OH
OMe
Di-tert-butyl-tetraphenyl-carbo-benzene
8
was found to
decompose at 280 °C,[9] so at a significantly higher temperature
Ph
OMe Ph
Ph
tBu
Ph
tBu
than other tetraphenyl congeners, 3 or 4. The counterparts 5-7
11, 24 %
7, quant.
t
bearing four or six Bu groups, were found to be stable beyond
OMe
tBu
tBu
OMe
Ph
OH
OMe
300 °C.
1) SnCl2
HCl
HO
Ph
1) LiHMDS
14
Ph
Ph
2) 23
2) NaOH
Ph
OMe Ph
Ph
Ph
8, 56 %
12, 14 %
Scheme 5. Synthesis of the carbo-benzenes 5-8 by [8+10] macrocyclization of
the triynes 13 or 14 with the diketones 23, 24 and 26, followed by reductive
aromatization of the resulting [6]pericyclynediols 9-12.
4
This article is protected by copyright. All rights reserved.

10.1002/chem.201800835
Chemistry - A European Journal
FULL PAPER
The 1H NMR spectrum of 6-8 in CDCl3 solution exhibits classical
features of other phenyl-substituted carbo-benzenes, namely a
strong deshielding of the ortho-1H nuclei at 9.5-9.6 ppm due to the
t
diatropic ring current over the C18 macrocycle.[1b,4,8,13,25] The Bu
1H nuclei of 5-8 also resonate at relatively low field (ca 2.5 ppm;
Figure 2),[11] with chemical shifts deshielded by 1 ppm as
compared to 1,2,3,4- and 2,3,4-5-tetra-tert-butylbenzene (1.27-
1.49 ppm).[26] Notably, the hexa-tert-butylbenzene, parent of 5,
has been studied at the theoretical level only.[27] The magnetic
aromaticity of 5-8 has been scrutinized in more detail by NICS
calculation (see section 7).
Figure 2. 1H NMR spectra of the tert-butylated carbo-benzenes 5-8 (CDCl3).
4. X-Ray crystallography of 5-8
The tert-butylated carbo-benzenes 5-8 were characterized by
XRD analysis of single crystals deposited from DCM or chloroform
solutions (Figure 3, Table 1).[21,28] A weak diffracting power of
crystals of 8 prevented high-quality resolution (R = 10 %), but was
sufficient for structural confirmation. The previously reported
crystal structure of 4 is here considered as a reference.[9] In the
crystals of 4-7, the asymmetric unit contains a single molecule
(two molecules in the crystal of 8), with a quite regular hexagonal
C18 core, and classical bond lengths and angles (spC-spC ≈ 1.22-
1.23 Å, and sp2C-spC ≈ 1.37-1.38 Å). The maximum deviation
from C18 planarity, ∆1, is found minimal for 5 (0.037 Å): in spite of
the steric crowding due to the six tBu groups in 5, these
substituents induce three times less distortion than the six phenyl
groups in 4 (∆1 = 0.11 Å). The same trend is found after DFT-
optimization of the geometry of 4 and 5 in the gas phase (∆1 =
0.007 vs 0.060 Å; see section 7 and Supporting Information).
Figure 3. Molecular views of the X-ray crystal structures of 5-8. Numbering of
the non-equivalent C atoms indicative of the crystal symmetry. For clarity,
hydrogen atoms, disordered atoms, and solvent molecules are omitted.
5
This article is protected by copyright. All rights reserved.

10.1002/chem.201800835
Chemistry - A European Journal
FULL PAPER
The macrocycle of 6 is also almost planar with ∆1= 0.077(1) Å in
the crystal (even strictly vanishing in the theoretical gas phase),
while those of the less symmetrical counterparts 7 and 8 (ideally
C2v, instead of D2h for 6, C6h or C6v or D3d for 5) are at least twice
more distorted with ∆1 = 0.152(2) Å for 7, 0.282(5) Å for 8. As far
as structural aromaticity is inversely related to geometrical
distortion, complementary to ∆1, another indicator thereof is the
maximum difference of diagonal distances (ca 8.0 ± 0.2 Å in
generic carbo-benzenes), ∆2, which is found three times larger for
4 than for 5, both in the crystal state (0.38 vs 0.12 Å) and in the
DFT-theoretical gas phase (0.163 vs 0.054 Å, see section 7: albeit
limited, discrepancies in absolute or relative values between
experimental and computational data can be ascribed to crystal
packing effects). Over the C18tBumPh6-m series, the mean index
(∆1 + ∆2)/2 may formally support a variation of structural
aromaticity as 4 < 8 < 7, 6 < 5, i. e. with increasing number m of
tBu substituents, and decreasing number of Ph substituents. The
Table 1. Selected crystallographic data for 5-8 and 4 (see Figure 4).
Species
Space group
4•CHCl3
P212121
2 X 2
18.24°
65.88°
68.80°
0. 11
5
P21/c
2
6
P21/c
2
7
C2/c
2
8•CHCl3
P21/n
2
no of mean plane
orientations and
angles a between
them
47.09(2)° 69.58(4)° 49.70(1)
°
7.76°
∆1 = deviation from
0.037(2) 0.077(1) 0.152(2) 0.244(5)[b]
[0.007]
0.12
C18 planarity (Å)[a]
∆2 = largest diagonal
difference (Å)[a]
[0.060]
0.38
[0.163]
[0.000]
0.11
[0.120]
[0.030]
0.13
[0.036]
[0.101]
0.17[b]
[0.054]
[0.117]
(n. a.)
Distance d1
between fully
eclipsed[c] C18 ring
2.55
4.29
3.52
3.52
mean planes (Å)
…
Shortest C C
3.55
5.65
3.58
3.45
3.60
-
distance d2 between
C18 rings (Å)
t
structural effect of Bu groups on the carbo-benzene ring is thus
opposite to that exerted at the meso positions on the porphyrin
core, P2 being reported to be highly distorted.[29]
F = columnar
82.0°
66.5°
55.2°
59.7°
verticality deviation
Shortest distance d3
between a tBu H
atom and a C18
mean plane
The angles qi between the mean plane of the C18 ring and those
of the phenyl substituents were not found to increase upon
increasing the number of tBu substituents, the largest value being
30.8° for the three mixed tBu/Ph representatives 6-8, close to the
29.9° measured for hexaphenyl-carbo-benzene 4. qi values of the
same order of magnitude are obtained from DFT optimization in
the gas phase (see below). The π-conjugation of the central C18
ring with phenyl substituents is thus not significantly affected in
the partially t-butylated carbo-benzenes 6-8.
[d]
1.396
(1.574)
1.536
(1.739)
1.237
(1.458)
0.887
(1.202)
(centroid) (Å)
Angles qi between 2.9-29.9°
the C18 and Ph mean [0.1-22.5°]
planes [calcd.][a]
-
[-]
12.8°
5.5-21.2° 12.9- 30.8°
[0.0°] [16.2-17.0°] [0.8-24.1°]
[a] In square brackets, value calculated in the theoretical gas phase (see
section 7). [b] Mean differences for the two non-equivalent molecules of 8:
(0.205(4) + 0.282(5))/2 = 0.244(5) Å; (0.18 + 0.15)/2 = 0.16 Å. [c] Eclipse
along the b direction for 4-7, not applicable for 8. [d] A distance of 1.31 Å is
measured between a Cl atom of the chloroform solvate molecule and both
the closest C18 ring mean plane and centroid thereof.
In the crystals of 4-7, the C18 rings pack in a fully eclipsed manner
(without intercalation of any other molecular fragment) in parallel
columns along the b direction, with quasi-regular hexagonal C18
sections slanted vs the column axis in four ways for 4 (of relative
angles rsa = 18.2°, 65.9, 68.8°) or in two ways for 5 (rsa = 47.1°),
6 (rsa = 69.6°) and 7 (rsa = 49.6°). The two molecules in the
asymmetric unit of 8, belong to parallel columns (along the c
direction), the C18 sections of which are just slightly slanted vs
each other (rsa = 7.8°).
Except in the crystal of 5, the shortest C-C distance between
C18 rings, in the range 3.5 ± 0.1 Å, corresponds to a C-C van der
Waals contact: looser than the previously observed attractive π-
π stacking in the crystal of 3b (3.23 Å, Figure 1),[11b] but quite short
t
considering the Bu bulkiness. Truly columnar arrangements are
found in crystals of 6 and 7, where the projections of two closest
parallel C18 hexagons along their common normal line (taken as
"vertical") on one of their mean planes intersect over ca one third
of the C18 hexagon surface area, with a columnar deviation from
verticality (vs the b axis) F = 55.2° and 59.7°, respectively, close
to the one reported for 3b (F = 55.0°).[11b] Notably, in all the
Figure 4. Illustration of the structural descriptors d1, d2, d3, ∆1, ∆2, a and F
defined in Table 1 in the case of the carbo-benzene 6.
t
crystals where applicable, a H atom of a Bu substituent points
towards the center of a nearest C18 ring, either within a column
(as for 6) (Figure 4) or between neighbouring columns (as for 7).
5. Absorption and emission spectroscopy of 5-8
The carbo-benzenes C18tBumPh6–m (4-8) are chromophoric
compounds, their solid state color varying with m. While 4 and 8
(m = 0, 2) are dark-red, the diphenyl versions 6 and 7 (m = 4) are
orange-red, and the hexa-tert-butyl version 5 (m = 6, without any
phenyl substituent) is yellow. The yellow color of 5,
6
This article is protected by copyright. All rights reserved.

10.1002/chem.201800835
Chemistry - A European Journal
FULL PAPER
unprecedented within the carbo-benzene family, is attributed to
the sole p-delocalization around the C18 ring in the absence of
external π-delocalization. Solutions of 5 in organic solvents are
also yellow, the maximum absorption wavelength lmax varying
from 374 nm in methanol to 379 nm in chloroform (see Figure 5
and Supporting Information). In contrast, all the carbo-benzenes
reported to date, bearing at least three aryl or alkynyl substituents,
give red or violet solutions.[1b,2,8-11,13,16]
very high e values, any emitted photon being instantly reabsorbed
in moleculo.[32]
Table 2. Emission properties of the carbo-benzenes 5-8 in chloroform solutions.
[c]
labsmax
[calcd][a]
(nm)
472
[509, 511]
10–3
e
lem
[calcd][a]
(nm)
F
(%)
No
Life-
time
(ns)
-
(L.mol-1.cm-1)
[calcd f][b]
3638b/2349
[2.3, 2.1]
[calcd f][b]
48,9
5
-
-
Replacement of two tBu groups of 5 by two Ph groups induces a
bathochromic shift of 30-35 nm in 6 and 7 (lmax = 408 and 415
[570, 571]
[2.6, 2.5]
379
[387, 386]
266
[2.0, 2.0]
446, 485, 505, 530
[420, 421]
0.03
1.9
t
nm). Further Bu → Ph replacement induces a similar shift of ca
[2.4, 2.4]
[d]
[d]
30-35 nm in 8 (lmax = 447 nm), consistent with the values reported
for the three regio-isomers of the tetraphenyl-carbo-benzene
C18Ph4H2 (lmax = 438-444 nm),[10] and p-iPr2-C18Ph4 (lmax = 446
nm).[11a] Ultimate tBu → Ph replacement is accompanied by
another 25 nm red-shift in 4 (lmax = 472 nm).[8,9]
6
408
411
530, 580
-
-
[422, 425]
[1.8, 2.0]
[428, 431]
[1.8, 2.0]
7
415
[431, 435]
407
[1.5, 2.2]
520, 580
[470, 483]
0.1
[1.7, 2.6]
0.1
0.4
8
447
347
520, 550
0.1
In the porphyrin series (Figure 1), the distortion induced by
replacement of the Ph substituents of P1 by tBu substituents in P2
is accompanied by a 20 nm bathochromic shift of the Soret band:
from 420 nm for P1[30] to 446 nm for P2.[15a] The puckered
conformation adopted by hindered meso-substituted porphyrins is
indeed known to generally induce such a shift.[15,31] The optical
behavior vs Ph → tBu replacement is thus opposite in the carbo-
benzene series, with a 100 nm hypsochromic shift from 4 to 5: this
is explained by the absence of steric crowding around the sp2-C
vertices of the C18 macrocycle.
[476, 478]
[1.7, 2.2]
[530, 534]
[2.0, 2.5]
[a]See section 7 and Supporting Information. [b] Oscillator strength; see section
7 and Supporting Information. [c]Excitation at lexc = labsmax. [d]The fluorescence
of 6 was too low to allow accurate determination of the quantum yield and life
time.
6. Electrochemistry of 5-8
Electrochemical properties of 5-8 were investigated by square-
wave (SW) and cyclic voltammetry (CV) (Table 3). In all cases,
three reduction and three oxidation waves were evidenced by
SWV. The first reduction of 5-8 was found to be reversible, at a
potential increasing with the number of phenyl substituents,
consistently with the overall π-conjugation extent: –1.36 V for
hexa-tert-butyl-carbo-benzene 5, –1.04 ± 0.01 V for the diphenyl
counterparts 6 and 7, –0.88 V for the tetraphenyl congener 8, and
–0.72 V for the hexaphenyl reference 4.[9] The second reduction
process was also found to be reversible for 6-7 (–1.50 ± 0.07 V)
and 8 (–1.31 V), but was not for the homoleptic references 5 (–
1.75 V) and 4 (–1.15 V).[9] In oxidation regime, the first process,
occurring invariably at 1.13 ± 0.01 V, was found reversible for 5-8
(at high scan rate only for 8). In contrast, the oxidized product 4+
was observed to polymerize at the anode (Ep = 1.14 V).[9] The
ox
oxidation reversibility is thus found to increase with the number of
tBu substituents at the C18 ring, their s-donor effect being indeed
prone to stabilize the oxidized species.
Figure 5. Normalized absorption spectra of the carbo-benzenes 4-8 in
The electrochemical behavior of meso-substituted porphyrins
follows the same trend: alkyl substituents make the compounds
more difficult to reduce and easier to oxidize, at least in a
reversible manner.[33]
chloroform solutions.
The emission properties of 5-8 in the UV-vis region were
measured to be very weak, with quantum yields below 1 %,
lifetime below 2 ns, and Stoke shifts above 70 nm (Table 2,
calculated data in Supporting Information). This is in line with
trends reported for most of the carbo-benzenes, even those
bearing fluorophoric substituents:[2] to date, the sole example of
significant fluorescence has been reported for a bisindolyl
derivative and explained by the exceptionally low molar extinction
coefficient of the latter.[32] The high extinction coefficient of 5-8 (e
=
260 000-400 000 L.mol-1.cm-1) and their extremely low
fluorescence are in accordance therewith. It was indeed proposed
that the very weak fluorescence of carbo-benzenes is due to their
7
This article is protected by copyright. All rights reserved.

10.1002/chem.201800835
Chemistry - A European Journal
FULL PAPER
concentrated over the C18 core, for which the tBu groups thus act
as π-insulators (Supporting Information).
As also expected from the strong s-donating character of Bu
Table 3. SWV and CV data for the carbo-benzenes 4-8.
No
Reductions
Oxidations
t
[c]
red [d]
[c]
ox [d]
E1/2[a] (DEP)[b]
E1/2[a](DEP)[b]
RIP
EP
RIP
EP
substituents, 5 exhibits a maximal chemical potential (µ = –0.139
Ha), i.e. a minimum electronegativity in the Mulliken-Pearson
scale,[35e] and a minimal electrophilicity[35b] (w = µ/(2h) = 0.094 Ha)
within the series: 5 is thus predicted more stable than 4 (w = 0.135
Ha) towards all kinds of nucleophiles (in addition to the steric
49
5
–0.72 (0.44)
-1.36 (0.74)
0.58
0.96
–1.15
-
-
1.14
1.13 (0.68) 0.95
1.12 (0.63) 1.02
1.14 (0.69) 1.04
-1.75
-2.00
1.84
2.13
t
6
7
8
-1.03 (0.73)
-1.43 (0.68)
0.89
1.03
protection brought by the crown of Bu groups). For comparison,
1.68[e]
2.27
consistently with the high electronegativity of fluorine atoms,
C18F6 exhibits a minimal chemical potential (µ = –0.181 Ha) and
a maximal electrophilicity (w = 0.153 Ha).
-2.03
-1.05 (0.68)
0.87
-1.57[e]
-1.84
1.75
2.03
TD-DFT calculations of excited states allowed simulation of the
absorption spectra. As previously shown for other carbo-
benzenes, [1c,2,13] the lowest energy excitations |S0>→|S1> and
|S0> → S2> take place in the blue-green region with a weak
oscillator strength, each of them involving near-frontier one-
electron transitions, HàL & H-1àL+1 and HàL+1 & H-1àL (see
section 6.4 in the Supporting Information). In accordance with the
four-orbital Gouterman model, the same one-electron transitions
are involved in the two next excitations |S0>→|S3> and |S0>→|S4>
underlying the most intense absorption band, shifted from the blue
region for 4 to the violet region for 5-8. These two excitations are
close both in energy and oscillator strength and can thus be
averaged to calculate lmax values, which happen to fit perfectly
with the experimental values (Figure S.6.4., eqn. 1).
-0.88 (0.73)
-1.31 (0.73)
0.99
0.85
-
-
1.18[e]
1.72
-1.86
2.00
Measurements performed at r. t. in DCM; supporting electrolyte: [n-Bu4N][PF6]
(0.1 M); working electrode: Pt; reference electrode: saturated calomel electrode
(SCE, 0.242 V vs the hydrogen electrode); scan rate: 0.2 V.s-1 unless otherwise
red
noted. [a] Half-wave potential E1/2 = (EP + EPox)/2, in V/SCE. [b] Separation
between the two peak potentials: DEP = ǀEPred - EPoxǀ, in V. [c] Peak current ratio
RIP = ǀIPox/IPredǀ. [d] EP values measured from CV in V/SCE. [e] Reversibility
observed at high scan rate only (5 V.s-1).
7. DFT computational characterization of 5-8
The geometry of the isolated carbo-benzenes 4-8, C18H6 and
C18F6 was optimized at the B3PW91/6-31G(d,p) level without
symmetry constraint using the Gaussian09 program (Supporting
Information).[34] The calculated bond lengths reproduce
crystallographic data within a 0.01 Å difference limit, without
significant variation with the substitution pattern (Csp–Csp ≈ 1.24
Å, Csp-Csp2 ≈ 1.38 Å , C–tBu≈ 1.54 Å, C–Ph ≈ 1.48 Å). A quasi-
D6h symmetry is obtained for the C18 core of 5, with a negligible
deviation from planarity: ∆1 = 0.007 Å for 5 vs 0.060 Å for 4,
confirming the trend observed in the crystal state (Table 1). As
stated in section 4, crystallography and DFT calculations
converge for a relative structural aromaticity in the order: 4 < 8 <
7, 6 < 5. Regarding the qi angles between the means planes of
the C18 core and Ph substituents, differences with experimental
values reveal crystal packing effects (Table 2). In the absence of
such effects, it is remarkable that qi = 0° for 6 whose two Ph
substituents are adjacent to two o-tBu substituents: the latter thus
induce less strain than the former allowing perfect π-conjugation
of the C18 core with such "isolated" Ph substituents.
The near-frontier molecular orbitals (i.e. non-strictly degenerated
HOMO and HOMO-1 on one hand, LUMO and LUMO+1 on the
other hand), though remaining mainly localized on the C18 core
operating as a nodal surface (thus of πz nature), are more
sensitive to the substituents in terms of energy (see Supporting
Information). As expected, the HOMO-LUMO gap h, i.e. twice the
original definition of chemical hardness by Pearson,[35] is minimal
for 4, exhibiting the largest π-conjugation extent (h = 0.084 Ha;
see Table 4). The gap is consistently much larger for 5 (h = 0.103
Ha), albeit smaller than for C18H6 (h = 0.108 Ha) and C18F6 (h =
0.106 Ha). While the HOMOs and LUMOs of phenylated or
fluorinated carbo-benzenes are partly delocalized over the
substituents, those of C18H6 and 5 are almost exclusively
l
max(exptl.) ≈ 0.745 lmax(calcd.) + 92.2, R > 0.999 (eqn. 1).
Table 4. DFT calculated electronic properties of 4-8, 3c, C18H6 and C18F6
(B3PW91/6-31G(d,p) or 6-31+G(d,p) level).
a)
a)
EHOMO ELUMO
µb)
h c)
w d)
NICS(0)/
NICS(1)e)
EHOMO–1 ELUMO+1
C18H6 -0.214 -0.106
-0.214 -0.106
-0.160
-0.151
-0.148
-0.139
-0.144
-0.144
-0.148
-0.181
0.108
0.085
0.084
0.103
0.088
0.093
0.088
0.106
0.119
0.135
0.119
0.094
0.118
0.112
0.124
0.153
-18.1/-16.5
-14.7/-13.4
-13.5/-12.4
-17.2/-15.7
-15.7/-14.3
-15.9/-14.6
-14.6/-13.4
-17.9/-18.9
3c[11a] -0.191 -0.106
-0.197 -0.098
4
5
6
7
8
-0.193 -0.109
-0.195 -0.108
-0.190 -0.087
-0.192 -0.085
-0.189 -0.100
-0.196 -0.090
-0.190 -0.098
-0.194 -0.092
-0.192 -0.104
-0.194 -0.101
C18F6 -0.234 -0.127
-0.234 -0.127
[a] HOMO and LUMO energies, in Ha.; due to a near D6h symmetry, HOMO and
LUMO are quasi-degenerated with HOMO–1 and LUMO+1, respectively (see
Supporting Information). [b] Chemical potential, µ = (EHOMO + ELUMO)/2, in
Ha/e; [c] chemical hardness, h = ELUMO – EHOMO, in Ha/e2. [d] Electrophilicity, w
= µ2/2h, in Ha. [e] Calculated at the B3PW91/6-31+G(d,p)//6-31G(d,p) level
within the GIAO formalism, in ppm.
The variation of local aromaticity of the C18 ring has also been
appraised for the magnetic criterion. Although accurate data
8
This article is protected by copyright. All rights reserved.

10.1002/chem.201800835
Chemistry - A European Journal
FULL PAPER
would rely on current density maps, relevant approximations can
be gained from a single numerical index: the nucleus independent
chemical shift (NICS), calculated at the B3PW91/6-31+G(d,p)
level,[36] at either the ring center (NICS(0)) or 1 Å above (NICS(1)).
A tight correlation between NICS(0) and NICS(1) is found over the
series 4-8, C18H6 and p-iPr-C18Ph4 (3c):[11a] NICS(1) = 0.90
NICS(0) – 0.3 (Table 4 and Supporting Information). Taking the
theoretical unsubstituted parent C18H6 as a reference, the NICS(0)
value has been previously observed to decrease whichever
substitution takes place.[11a,13,37] This is confirmed and refined in
the present series, showing that the magnetic aromaticity
decreasing effect is stronger for aryl substituents than for alkyl
counterparts: for homoleptic carbo-benzenes C18R6, the
NICS(0)/NICS(1) values indeed increase from –18.1/–16.5 ppm
for R = H, to –17.2/–15.7 ppm for R = tBu (5) and –13.5/–12.4 ppm
for R = Ph (4). The effect is thus ca five time stronger for Ph
groups than for tBu group.
the LUMO level appears to be quantitatively accurate (Figure
S.6.7 in the Supporting Information).
Notably, NICS values of the phenyl substituents are found below
–5.4 ppm, vs –8.3 ppm for the benzene molecule (see Supporting
Information),[11a,36,37] showing that their magnetic aromatic
character is also severely reduced by attachment to the C18 core:
the C6-C18 π-conjugation (qi < 30°, Table 2) indeed decreases
their local chemical hardness and thus their magnetic aromaticity
(see discussion below).
In the tetraphenyl series, o- or p-R2-C18Ph4, the NICS(0)/NICS(1)
values of the C18 core remain almost constant, at –14.6/–13.4
ppm for R = o-tBu (8) and –14.7/–13.4 ppm for R = p-iPr (3c).[11a]
The effects of substitution by tBu and F are comparable: NICS(0)
= –17.2 and –17.9 ppm, respectively.
NICS(0) is found to decrease vs chemical hardness or HOMO-
LUMO gap h, and even more regularly with NICS-effective
counterpart heff (as defined in section 6.6 of the Supporting
Information).[38] More remarkably still, NICS(0) increases with
calculated lmax values (or experimental values through the above
eqn. 1) through an accurate linear correlation for 3c,[11a] 4-8, C18H6
and C18F6 (Figure 6a and Supporting Information). This
correlation, albeit over a limited set of compounds, spans a bridge
between the energetic and magnetic criteria of aromaticity:[39a]
from a heuristic viewpoint, hard (i.e. poorly polarizable) electronic
systems, primarily characterized by large HOMO-LUMO gaps and
more generally by low lmax values, should indeed be more prone
to resist deviation of a ring current induced by an external
magnetic field.[6] The correlation is empirical in nature, but the
practical consideration of lmax instead of l1, the first excitation
wavelength with vanishing oscillator strength (see section 6.6 in
the Supporting Information), is also intuitively justified by the
common dynamical nature of the phenomena underlying NICS
and lmax (a large oscillator strength or transition dipole moment
features a high absorption probability). Nevertheless, as far as the
spectral pattern remains the same, l1 also correlates with lmax
(Figure S.4.2 in the Supporting Information).
Figure 6. a (top) Variation of NICS(0) with calculated values of UV-vis
absorption wavelengths of highest oscillator strengths (actually arithmetic mean
over two very close excited states: see Table 4 and Supporting Information). b
red
(bottom) Variation of the first reduction potential E1/2 (Table 3) vs NICS(0)
following Haley's scheme.[40]
Conclusions
The above-described preparation and characterization of tert-
t
butylated/ phenylated carbo-benzenes show that Bu groups in
ortho position do not induce distortion of the C18 ring, and that the
presence of at least three aryl substituents is not a requirement
for experimental stability. The complete series C18tBumPh6–m
exhibits regular variations of key properties vs m = 0, 2, 4, 6:
increasing solubility, stability and chemical hardness (h) on one
hand, decreasing maximum UV-vis absorption wavelength (lmax),
reduction potential and NICS on the other hand. In the considered
Further, following a scheme proposed by Haley et al. for a series
of antiaromatic indenofluorenes,[40] a regular correlation is also
found between the first reduction potential E1/2red (Table 3) and
NICS(0) (Figure 6b). As very recently disclosed in the carbo-OPE
series,[1c] the expected qualitative correlation between E1/2red and
9
This article is protected by copyright. All rights reserved.

10.1002/chem.201800835
Chemistry - A European Journal
FULL PAPER
copper Kα wavelength equipped with a Dectris Pilatus 200K hybrid detector.
The structures were solved using SUPERFLIP, and refined by means of
least-squares procedures on F using the programs of the PC version of
CRYSTALS.[41] Atomic scattering factors were taken from the international
tables for X-ray crystallography.[42] The asymmetric unit consists in a whole
carbo-benzene series, the correlations of NICS vs lmax, h and
E1/2red illustrate that aromatic systems are hard chemical systems.
More practically, as the poor solubility of arylated carbo-benzenes
hampers their use for applications (e. g. high-quality thin film
deposition for organic photovoltaics),[9] the solubilizing effect of
tert-butyl substituents opens promising prospects. In particular,
the implemented methodology for the synthesis of 6 could be
applied for the preparation of soluble versions p-R2-C18tBu4 of
poorly tractable p-R2-C18Ph4 leads, while deserving deeper
studies of their photo-physical properties.[2,13,16]
molecule (with or without solvent), except for
8 (2 molecules per
asymmetric unit) and for 6, 13 and 23 (half a molecule per asymmetric unit).
All non-hydrogen atoms were refined anisotropically. Hydrogen atoms
were refined using a riding model. Absorption corrections were introduced
using the program MULTISCAN. For 5, 7, 8 and 18, several disorders were
modelled with respect to the electronic densities. Moreover, crystals of 5
were found to be inherently twinned.
Computational methods
Experimental Section
All the DFT static calculations have been performed at the GGA level with
the Gaussian09 set of programs,[34] using the B3PW91 functional of Becke,
Perdew and Wang.[43] The electronic configuration of the molecular
systems was described with the 6-31G(d,p) basis set for H, C, and O (6-
31G** keyword in Gaussian) for the geometry optimizations.[44] The
geometry optimizations were carried out without symmetry constraints,
and the characterization of the stationary points was performed by
analytical frequency calculations. Using the equilibrium geometries,
calculations of electronic spectra (absorption and emission) and NICS
were performed at the TDDFT level with the 6-31+G(d,p) basis set, within
the framework of the GIAO formalism.[45] Data were treated with ROTAX
which gave the twin law between the two components (0.94: 0.06).
Refinement of the final models led to imperfect but reasonable solutions.
The same data were obtained after running single point energy
calculations with the program Firefly v. 8.1.1 from the geometries
optimized using Gaussian09.[46] Molecular orbitals were edited using the
interface program MacMolPlt, v. 7.7.[47]
General
THF, diethyl ether (Et2O), pentane and dichloromethane (DCM) were dried
with a PureSolv-MD-5 Innovative Technology system for the purification of
solvents. All other reagents were used as commercially available. In
particular, solutions of n-BuLi were 1.6 M or 2.5 M in hexane, solutions of
EtMgBr were 3 M in THF, solutions of HCl were 2 M in diethyl ether,
solutions of ethynylmagnesium bromide were 0.5 M in THF. All reactions
were carried out under argon atmosphere using Schlenk and vacuum line
techniques. Column chromatography was carried out on silica gel (60 Å,
C.C 70-200 μm). Silica gel thin layer chromatography plates (60F254, 0.25
mm) were revealed under UV-light and/or by treatment with an ethanolic
solution of phosphomolybdic acid (20%). The following analytical
instruments were used, 1H and 13C NMR: Avance 400 and Avance 400 HD
spectrometers; mass spectroscopy: TSQ 7000 Thermo Electron and
Voyager DE-ST Perseptive Biosystems spectrometers; UV-Visible:
Perkin-Elmer UV-Vis Win-Lab Lambda 35; IR: Perkin Elmer 1725. NMR
chemical shifts are given in ppm with positive values to high frequency
relative to the tetramethylsilane reference. Coupling constants J are in
Hertz. UV-Visible extinction molar coefficient ε is in L.mol-1.cm-1 and
wavelengths λ in nm.
Acknowledgements
The authors thank the Toulouse IDEX Emergence program 2014
(Carbo-device project) for the post-doctoral grant of D. L. and for
funding. R. C. thanks the Centre National de la Recherche
Scientifique (CNRS) for half a teaching sabbatical in 2015-2016.
In the framework of the CNRS ЯÉCIPROCS network, this work
has benefited from the X-ray facility of the Biophysical and
Structural Chemistry platform at IECB, CNRS UMS 3033,
INSERM US 001, Bordeaux University. Dr. Brice Kauffmann is
particularly acknowledged for the DRX analysis of compound 5.
Voltammetry
Voltammetric measurements were carried out with a potentiostat Autolab
PGSTAT100 controled by GPES 4.09 software. Experiments were
performed at room temperature in a home-made airtight three-electrode
cell connected to a vacuum/agon line. The reference electrode consisted
of a saturated calomel electrode (SCE) separated from the solution by a
bridge compartment. The counter electrode was a platinum wire of ca 1
cm2 apparent surface. The working electrode was a Pt microdisk (0.5 mm
diameter). The supporting electrolyte [n-Bu4N][PF6] was used as received
(Fluka, 99 % electrochemical grade) and simply degassed under argon.
The solutions used in the electrochemical study was 10-3 M in carbo-
benzene and 0.1 M in supporting electrolyte. Before each measurement,
the solutions are degassed by bubbling argon, and the working electrode
was polished with a polishing machine (Presi P230). Typical instrumental
parameters for recording square-wave voltammograms were: SW
frequency f = 20 Hz, SW amplitude Esw = 20 mV, scan increment dE = 0.5
mV.
Keywords: aromaticity • carbo-benzene • electronic insulation •
substituent effect • tert-butyl group
[1]
a) For a recent review, see: J. L. Marshall, D. Lehnherr, B. D. Linder, R.
R. Tykwinski, ChemPlusChem 2017, 82, 967-1001; b) K. Cocq, C.
Lepetit, V. Maraval, R. Chauvin, Chem. Soc. Rev. 2015, 44, 6535-6559,
and references therein; c) most recent article on carbo-benzenes: C. Zhu,
A. Poater, C. Duhayon, B. Kauffmann, A. Saquet, R. Chauvin, Angew.
Chem. Int. Ed. 2018, in press. DOI: 10.1002/anie.201713411.
See for example: I. Baglai, M. de Anda-Villa, R. M. Barba-Barba, C.
Poidevin, G. Ramos-Ortíz, V. Maraval, C. Lepetit, N. Saffon-Merceron,
J.-L. Maldonado, R. Chauvin, Chem. Eur. J. 2015, 21, 14186-14195.
Z. Li, M. Smeu, A. Rives, V. Maraval, R. Chauvin, M. A. Ratner, E.
Borguet, Nat. Commun. 2015, 6:6321.
Crystallography
[2]
[3]
Excepted for the hexa-tertbutyl-carbo-benzene 5, intensity data were
collected at low temperature on an Apex2 Bruker diffractometer equipped
with a 30W air-cooled microfocus source (λ = 0.71073 Å) or on a Gemini
Oxford Diffraction diffractometer (λ = 0.71073 Å). Data collection for 5 was
performed on a high flux microfocus Rigaku FRX rotating anode at the
10
This article is protected by copyright. All rights reserved.

10.1002/chem.201800835
Chemistry - A European Journal
FULL PAPER
[4]
See for example: C. Poidevin, J.-P. Malrieu, G. Trinquier, C. Lepetit, F.
Allouti, M. E. Alikhani, R. Chauvin, Chem. Eur. J. 2016, 22, 1-15.
C. Zhu, T.-H. Wang, C.-J. Su, S.-L. Lee, A. Rives, C. Duhayon, B.
Kauffmann, V. Maraval, C.-h. Chen, H.-F. Hsu, R. Chauvin, Chem.
Commun. 2017, 53, 5902-5905.
[22] R. J. Cox, D. J. Ritson, T. A. Dane, J. Berge, J. P. H. Charmant, A.
Kantacha, Chem. Commun. 2005, 1037-1039.
[5]
[23] K. Cocq, N. Saffon-Merceron, Y. Coppel, C. Poidevin, V. Maraval, R.
Chauvin, Angew. Chem. Int. Ed. 2016, 55, 15133-15136.
[24] Crystallographic data for 23: C32H34O4, M = 482.62, monoclinic, space
group P21/c, a = 6.1720(3) Å, b = 19.9189(9) Å, c = 10.8454(5) Å, a =
90°, b = 92.902(3)°, g = 90°, V = 1331.62(6) Å3, Z = 2, T = 100 K, 18919
collected reflexions, 3318 unique reflexions (Rint= 0.064), R[I>3σ(I)] =
0.101 for 2575 contributing reflexions.
[6]
[7]
[8]
R. Chauvin, C. Lepetit, V. Maraval, L. Leroyer, Pure Appl. Chem. 2010,
82, 769-800.
C. Zou, C. Duhayon, V. Maraval, R. Chauvin, Angew. Chem. Int. Ed.
2007, 46, 4337-4341.
a) Y. Kuwatani, N. Watanabe, I. Ueda, Tetrahedron Lett. 1995, 46, 119-
122; b) R. Suzuki, H. Tsukuda, N. Watanabe, Y. Kuwatani, I. Ueda,
Tetrahedron 1998, 54, 2477-2496.
[25] A. Soncini, P. W. Fowler, C. Lepetit, R. Chauvin, Phys. Chem. Chem.
Phys. 2008, 10, 957-964.
[26] A. Krebs, E. Franken, S. Müller, Tetrahedron Lett. 1981, 22, 1675-1678.
[27] a) J. Zhang, D. M. Ho, R. A. Pascal Jr, J. Am. Chem. Soc. 2001, 123,
10919-10926; b) D. Nori-Shargh, D. Tahmassebi, M. Poukalhor, M. M.
Amini, S. Jameh-Bozorghi, F. Deyhimi, M. Khanizadeh, M.
Malekhosseini, Phosphorus, Sulfur, and Silicon 2006, 181, 2419-2434.
[28] a) Crystallographic data for 5: C42H54, M = 598.87, monoclinic, space
group P21/c, a = 11.9018(3) Å, b = 10.7319(3) Å, c = 29.1858(7) Å , a =
90°, b = 101.639(2)°, g = 90°, V = 3654.23(9) Å3, Z = 4, T = 120 K, 24805
collected reflexions, 7164 unique reflexions (Rint = 0.020), R[I>3σ(I)] =
0.077 for 6193 contributing reflexions; b) crystallographic data for 6:
C46H46, M = 598.87, monoclinic, space group P21/c, a = 14.8944(12) Å,
b = 6.1633(5) Å, c = 19.8087(16) Å, a = 90°, b = 98.675(3)°, g = 90°, V =
1797.61(15) Å3, Z = 2, T = 100 K, 43406 collected reflexions, 5873 unique
[9]
C. Zhu, C. Duhayon, D. Romero-Borja, J.-L. Maldonado,G. Ramos-Ortíz,
A. Saquet, V. Maraval, R. Chauvin, New. J. Chem. 2017, 41, 3908-3914,
and references therein.
[10] K. Cocq, V. Maraval, N. Saffon-Merceron, R. Chauvin, Chem. Rec. 2015,
15, 347-361.
[11] a) K. Cocq, N. Saffon-Merceron, A. Poater, V. Maraval, R. Chauvin,
Synlett 2016, 27, 2105-2112; b) C. Zhu, A. Rives, C. Duhayon, V.
Maraval, R. Chauvin, J. Org. Chem. 2017, 82, 925-935.
[12] R. Chauvin, Tetrahedron Lett. 1995, 36, 401-404.
[13] L. Leroyer, C. Lepetit, A. Rives, V. Maraval, N. Saffon-Merceron, D.
Kandaskalov, D. Kieffer, R. Chauvin, Chem. Eur. J. 2012, 18, 3226-3240.
[14] a) M. Gouterman, J. Chem. Phys. 1959, 30, 1139-1161; b) M. Gouterman,
J. Mol. Spectrosc. 1961, 6, 138-163; c) P. J. Spellane, M. Gouterman, A.
Antipas, S. Kim, Y. C. Liu, Inorg. Chem. 1980, 19, 386-391.
[15] a) M. O. Senge, I. Bischoff, N. Y. Nelson, K. M. Smith, J. Porphyrins
Phthalocyanines 1999, 3, 99-116; b) A. B. Ormond, H. S. Freeman, Dyes
and Pigments 2013, 96, 440-448.
reflexions (Rint
= 0.059), R[I>3σ(I)] = 0.043 for 3782 contributing
reflexions; c) crystallographic data for 7: C46H46, M = 598.87, monoclinic,
space group C2/c, a = 32.232(2) Å, b = 10.6973(7) Å, c = 23.3273(14) Å,
a = 90°, b = 114.6906(16)°, g = 90°, V = 7307.8(4) Å3, Z = 8, T = 100 K,
95255 collected reflexions, 12797 unique reflexions (Rint
= 0.033),
[16] A. Rives, I. Baglai, V. Malytskyi, V. Maraval, N. Saffon-Merceron, Z.
Voitenko, R. Chauvin, Chem. Commun. 2012, 48, 8763-8765.
[17] For early references on pericyclynes see: a) L. T. Scott, G. J. DeCicco,
J. L. Hyun, G. Reinhardt, J. Am. Chem. Soc. 1983, 105, 7760-7761; b) L.
T. Scott, G. J. DeCicco, J. L. Hyun, G. Reinhardt, J. Am. Chem. Soc.
1985, 107, 6546-6555. For an early reference on hexaoxy[6]pericyclynes,
see: c) C. Saccavini, C. Tedeschi, L. Maurette, C. Sui-Seng, C. Zou, M.
Soleilhavoup, L. Vendier, R. Chauvin, Chem. Eur. J. 2007, 13, 4895-
4913.
R[I>3σ(I)] = 0.059 for 10529 contributing reflexions; d) crystallographic
data for 8: C50.50H38.50Cl1.50, M = 698.54, monoclinic, space group P21/n,
a = 9.6382(5) Å, b = 16.4828(9) Å, c = 48.783(2) Å, a = 90°, b =
93.921(3)°, g = 90°, V = 7731.8(4) Å3, Z = 8, T = 100 K, 51106 collected
reflexions, 12872 unique reflexions (Rint= 0.056), R[I>2.7σ(I)] = 0.101 for
9535 contributing reflexions.
[29] For the synthesis of the 5,10,15,20-tetra-tert-butyl-porphyrin see: a) T.
Ema, M. O. Senge, N. Y. Nelson, H. Ogoshi, K. M. Smith, Angew. Chem.
Int. Ed. 1994, 33, 1879-1881. For the XRD structure see: b) M. S. Somma,
C. J. Medforth, N. Y. Nelson, M. M. Olmstead, R. G. Khoury, K. M. Smith,
Chem. Commun. 1999, 121-1222.
[18] L. Leroyer, C. Zou, V. Maraval, R. Chauvin, C. R. Chimie 2009, 12, 412-
429.
[19] For the synthesis of 15 by AlCl3-promoted acylation, see: O. Klein, H.
Hopf, J. Grunenberg, Eur. J. Org. Chem. 2009, 2141-2148.
[20] For the characterization of 15, see: a) L. Birkofer, K. Richtzenhain, Chem.
Ber. 1979, 112, 2829-2836; b) S. Eagon, C. Delieto, W. J. McDonald, D.
Haddenham, J. Saavedra, B. Singaram, J. Org. Chem. 2010, 75, 7717-
7725.
[30] K. M. Smith, Porphyrins and Metalloporphyrins, Elsevier, Amsterdam,
1975.
[31] C. J. Medforth, M. O. Senge, K. M. Smith, L. D. Sparks, J. A. Shelnutt, J.
Am. Chem. Soc. 1992, 114, 9859-9869.
[32] I. Baglai, V. Maraval, Z. Voitenko, Y. Volovenko, R. Chauvin, French-
Ukrainian J. Chem. 2013, 1, 48-53.
[21] CCDC 1523150-1523156 (18, 13, 7, 5, 6, 8, 16) and CCDC 1525999 (23)
contain the supplementary crystallographic data for this paper. The data
can be obtained free of charge from the Cambridge Crystallographic
[33] C. Paliteiro, A Sobral, Electrochemica Acta 2005, 50, 2445-2541.
[34] Gaussian 09, Revision A.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P.
Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr.,
J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin,
V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A.
Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J.
Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R.
Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G.
Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A.
D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J.
Fox,. Gaussian, Inc., Wallingford CT, 2009.
Data
Centre
via
http://www.ccdc.cam.ac.uk/getstructures.
a)
Crystallographic data for 16: C12H20OSi, M = 208.38, monoclinic, space
group P21/n, a = 14.2363(3) Å, b = 6.51895(12) Å, c = 14.4834(3) Å, a =
90°, b = 95.7989(18)°, g = 90°, V = 1337.27(3) Å3, Z = 4, T = 120 K, 39847
collected reflexions, 4560 unique reflexions (Rint= 0.029), R[I>3σ(I)] =
0.030 for 3852 contributing reflexions; b) crystallographic data for 18:
C23H40O2Si2, M = 404.74, triclinic, space group P-1, a = 11.1519(4) Å, b
= 11.1745(5) Å, c = 11.2463(4) Å, a = 92.786(2)°, b = 110.218(2)°, g =
96.628(2)°, V = 1300.44(5) Å3, Z = 2, T = 100 K, 70763 collected
reflexions, 7830 unique reflexions (Rint= 0.036), R[I>3σ(I)] = 0.057 for
5605 contributing reflexions; c) crystallographic data for 13: C18H26O2, M
= 274.40, monoclinic, space group P21/c, a = 7.4651(8) Å, b = 6.2827(6)
Å, c = 18.2015(17) Å, a = 90°, b = 99.828(3)°, g = 90°, V = 841.14(8) Å3,
Z = 2, T = 100 K, 16601 collected reflexions, 2349 unique reflexions (Rint
0.036), R[I>3σ(I)] = 0.101 for 1867 contributing reflexions.
=
[35] a) R. G. Pearson, Proc. Natl. Acad. Sci. USA 1986, 83, 8440-8441; b) P.
K. Chattaraj, S. Giri, S. Duley, J. Chem. Phys. Lett. 2010, 1, 1064-1067;
c) R. G. Pearson, Chemical Hardness: Applications from Molecules to
11
This article is protected by copyright. All rights reserved.

10.1002/chem.201800835
Chemistry - A European Journal
FULL PAPER
[40] C. K. Frederickson, L. N. Zakharov, M. M. Haley, J. Am. Chem. Soc.
2016,138, 16827-16838.
[41] P. W. Betteridge, J. R. Carruthers, R. I. Cooper, K. Prout, D. J. Watkin,
J. Appl. Cryst. 2003, 36, 1487.
Solids; Wiley-VCH: New York, 1997; d) A. Poater, A. Gallegos, R. Carbó-
Dorca, J. Poater, M. Solà, L. Cavallo, A. P. Worth, J. Comput. Chem.
2009, 30, 275-284; e) R. G. Pearson, Inorg. Chem. 1988, 27, 734-740.
[36] P. v. R. Schleyer, C. Maerker, A. Dransfeld, H. J. Jiao, J. R. V. Hommes,
J. Am. Chem. Soc. 1996, 118, 6317-6318.
[42] International Tables for X-ray Crystallography, vol. IV, Kynoch Press,
Birmingham, England, 1974.
[37] C. Godard, C. Lepetit, R. Chauvin, Chem. Commun. 2000, 1833-1834.
[38] In the absence of strict D6h symmetry, the active frontier MOs are not
strictly pairwise degenerated (HOMO ≠ HOMO-1, LUMO ≠ LUMO+1):
near to degeneracy, a relevant Pearson's hardness should thus involve
all of them four. Effective HOMO and LUMO energies can thus be
[43] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648-5652; b) J. P. Perdew, Y.
Wang, Phys. Rev. B 1992, 45, 13244-13249.
[44] a) R. Ditchfield, W. J. Hehre, J. A. Pople, J. Chem. Phys. 1971, 54, 724;
b) T. Clark, J. Chandrasekhar, G. W. Spitznagel, P. v. R. Schleyer, J.
Comp. Chem. 1983, 4, 294-301.
defined as EHOMOeff = w EHOMO-1 + (1-w) EHOMO and ELUMOeff = w' ELUMO+1
+
[45] P. v. R. Schleyer, C. Maerker, A. Dransfeld, H. J. Jiao, J. R. V. Hommes,
J. Am. Chem. Soc. 1996, 118, 6317-6318.
(1-w') ELUMO, where the weighting coefficients w, w' must be equal to unity
for perfect degeneracy, and vanish as drawing farther from it. Taking the
simplest form w = k/(k + EHOMO – EHOMO-1) and w' = k/(k + ELUMO+1 – ELUMO),
and defining the effective hardness as heff = ELUMOeff – EHOMOeff, a perfect
linear correlation is obtained for k = 0.0072: NICS(0) = 5.0 – 214 heff with
a reliability coefficient R > 0.9999 over a set of seven compounds: 4, 6,
[46] a)
A.
A.
Granovsky,
Firefly
version
8,
http://classic.chem.msu.su:gran:firefly:index.html) which is partially
based on the GAMESS (US) source code; b) M. W. Schmidt, K. K.
Baldridge, J. A. Boatz, S. T. Elbert, M. S. Gordon, J. H. Jensen, S.
Koseki, N. Matsunaga, K. A. Nguyen, S. Su, T. L. Windus, M. Dupuis, J.
A. Montgomery, J. Comput. Chem. 1993, 14, 1347-1363.
i
7, 8, p- Pr2C18Ph4, C18H6 and C18F6. The NICS(0) value predicted for 5
by this equation (with heff = 0.106 Ha) is –17.7 ppm, instead of –17.2 ppm
calculated (a perfect fit would be obtained for the value heff = 0.104 Ha).
[47] B. M. Bode, M. S. J. Gordon, J. Mol. Graph. Model. 1998, 16, 133-138.
See Supporting Information.
[39] For a seminal paper about the aromaticity issue, see: M. K. Cyranski, T.
Krygowski, A. R. Katritzky, P. v. R. Schleyer, J. Org. Chem. 2002, 67,
1333-1338, and citations therein.
12
This article is protected by copyright. All rights reserved.

10.1002/chem.201800835
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
FULL PAPER
λabs = 379 nm
hν
λem = 446 nm
hν
A crown of six tert-butyl
groups acts as a steric
bumper and p-electronic
insulator of the carbo-
benzene ring, unveiling
intrinsic features of the C18
aromatic core.
D. Listunov, C. Duhayon, A.
Poater, S. Mazères, A. Saquet,
V. Maraval,* R. Chauvin*
Page No. – Page No.
Steric/π-electronic insulation
of the carbo-benzene ring:
dramatic effects of tert-butyl
vs phenyl crowns on
geometric, chromophoric,
redox and magnetic
NICS(0) =
–17.2 ppm
e–
properties
e–
Epox = +1.13 V
E1/2red = –1.36 V
13
This article is protected by copyright. All rights reserved.