Conformational and electronic consequences in crafting extended, π-conjugated, light-harvesting macrocycles

Article abstract of DOI:10.1002/chem.201102488

The synthesis of a new series of free-base, Ni^II and Zn ^II 2,3,12,13-tetra(ethynyl)-5,10,15,20-tetraphenyl porphyrins is described. Upon heating, two of the four ethynyl moieties undergo Bergman cyclization to afford the monocyclized 2,3-diethynyl-5,20-diphenylpiceno[10,11, 12,13,14,15-jklmn]porphyrin in 30%, 10%, and trace yields, respectively. The structures of all products were investigated by using quantum chemical calculations and the free-base analogue was isolated and crystallized; all compounds show significant deviation from the idealized planar structure. No fully-cyclized bispiceno[20,1,2,3,4,5,10,11,12,13,14,15-fghij]porphyrin was isolated from the reaction mixture. To understand why only two of the four enthynyl groups undergo Bergman cyclization, the reaction coordinates were examined by using DFT at the PWPW91/cc-pVTZ(-f) level coupled to a continuum solvation model. The barrier to cyclization of the second pair of ethynyl groups was found to be 5.5 kcal mol^-1 higher than the first, suggesting a negative cooperative effect and significantly slower rate for the second cyclization. Cyclization reactions for model porphyrin-enediynes with ethene- and H-functionality substitutions at the meso-phenyl rings were also examined, and found to have a similar barrier to diradical formation for the second cyclization event as for the first in these highly planar molecules. By enforcing an artificial 30° cant in two of the pyrrole rings of the porphyrin, the second barrier was increased by 2 kcal mol^-1 in the ethene model system; this suggests that the disruption of the π conjugation of the extended porphyrin structure is the cause of the increased barrier to the second cyclization event.

Full text of DOI:10.1002/chem.201102488

FULL PAPER
DOI: 10.1002/chem.201102488
Conformational and Electronic Consequences in Crafting Extended,
p-Conjugated, Light-Harvesting Macrocycles
Leigh J. K. Boerner, Shivnath Mazumder, Maren Pink, Mu-Hyun Baik,* and
Jeffrey M. Zaleski*[a]
Abstract: The synthesis of a new series
phyrin was isolated from the reaction
mixture. To understand why only two
of the four enthynyl groups undergo
Bergman cyclization, the reaction coor-
dinates were examined by using DFT
at the PWPW91/cc-pVTZ(-f) level cou-
pled to a continuum solvation model.
The barrier to cyclization of the second
pair of ethynyl groups was found to be
5.5 kcalmolꢀ1 higher than the first, sug-
gesting a negative cooperative effect
and significantly slower rate for the
second cyclization. Cyclization reac-
tions for model porphyrin–enediynes
with ethene- and H-functionality sub-
stitutions at the meso-phenyl rings
were also examined, and found to have
a similar barrier to diradical formation
for the second cyclization event as for
the first in these highly planar mole-
cules. By enforcing an artificial 308
cant in two of the pyrrole rings of the
porphyrin, the second barrier was in-
creased by 2 kcalmolꢀ1 in the ethene
model system; this suggests that the
disruption of the p conjugation of the
extended porphyrin structure is the
cause of the increased barrier to the
second cyclization event.
of free-base, NiII and ZnII 2,3,12,13-
tetraACHTUNGTRENNUNG(ethynyl)-5,10,15,20-tetraphenyl
porphyrins is described. Upon heating,
two of the four ethynyl moieties under-
go Bergman cyclization to afford the
monocyclized 2,3-diethynyl-5,20-diphe-
nylpiceno[10,11,12,13,14,15-jklmn]por-
phyrin in 30%, 10%, and trace yields,
respectively. The structures of all prod-
ucts were investigated by using quan-
tum chemical calculations and the free-
base analogue was isolated and crystal-
lized; all compounds show significant
deviation from the idealized planar
structure. No fully-cyclized bispiceno-
Keywords: aromaticity · computa-
tional chemistry
· cyclization ·
fused-ring systems · porphyrinoids
ACHTUNGTRENNUNG[20,1,2,3,4,5,10,11,12,13,14,15-fghij]por-
Introduction
turn translates into perturbations in the electronic system.[5]
Thus, effective construction of aromatic molecules with ex-
tended conjugation requires a subtle balance of appropriate
geometric conformation fused to electronic stability derived
from increased delocalization. Proper harnessing of these
properties can result in large, rigid, exocyclic, aromatic ar-
chitectures that possess long-wavelength absorption, high
transmission efficiencies, and large emission quantum
yields.[4c,6]
Extended aromatic conjugation in macrocyclic structures
plays a prominent role in light-harvesting events in photo-
synthesis, optical generation of singlet O2 in photodynamic
therapy,[1] molecular wires,[2] and photoinduced energy or
charge transport in dye-sensitized, solar-energy materials.[3]
In addition to their optical significance, ground-state elec-
tronic properties, such as redox potentials, are critical to
metalloenzyme function and the stabilization of high oxida-
tion states in key intermediates (e.g., cytochrome P450). The
consequences of extended conjugation can also dictate
chemical transformations at the macrocycle periphery due
to the modulation of net electro- or nucleophilicity at fringe
functionalities.[4] Such transformations can also be affected
by structural conformations caused by steric strain, which in
The most straightforward and widely employed strategy
for preparation of these p-extended, light-harvesting motifs
involves fusion of a rigid, aromatic, ring system to a conju-
gated porphyrinoid backbone. Incorporation of substruc-
tures, such as anthracene, at the meso-position,[4c] and nap-
thyl[4d] or azulene[4b] to the porphyrin core have resulted in
molecules with long-wavelength absorption up to 855 nm
and large two-photon cross sections for near-infrared excita-
tion.[4b] The advanced optical properties of these motifs are
realized in spite of the fact that none of these constructs are
rigidly planar and fully conjugated across the entirety of the
porphyrinoid backbone. Thus, development and understand-
ing of unusual synthetic methodologies toward fusion of pe-
ripheral, aromatic substituents is important for further en-
hancement of the optical properties of these versatile por-
phyrinoid chromophores.
[a] Dr. L. J. K. Boerner, S. Mazumder, Dr. M. Pink, Prof. M.-H. Baik,
Prof. J. M. Zaleski
Department of Chemistry and Molecular Structure Center
Indiana University, Bloomington, Indiana 47405 (USA)
E-mail: mbaik@indiana.edu
zaleski@indiana.edu
Supporting information for this article, including details about experi-
mental methods, estimation of barrier heights for 4-TS and 9-TS as
well as X-ray structure data collection, structure solution, and refine-
ment for Ni2b, Zn2c, H3a, and H8a, including CIF files, is available
on the WWW under http://dx.doi.org/10.1002/chem.
In a conceptually unrelated work, Smith et al.[7] showed
that alkynes fused to the b,b’-positions of one of the four
Chem. Eur. J. 2011, 17, 14539 – 14551
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
14539

backbone pyrroles could be thermally activated at high tem-
peratures (1908C) to undergo Bergman cyclization, forming
a six-membered, 1,4-phenyl diradical capable of addition to
the adjacent phenyl rings at the macrocycle meso-posi-
porphyrins have the potential to cyclize and systematically
aromatize the two halves of the macrocycle backbone; this
creates the highly conjugated bispicenoporphyrin skeleton
(Scheme 1). If structural flexibility permits the formation of
tions.[8] This thermal rearrangement affords three new C C
six additional C C bonds, the bispicenoporphyrin product
ꢀ
ꢀ
bonds per dialkyne, and upon
rearomatization through H2
elimination, extends the por-
phyrin conjugation, redshifting
the p–p* electronic transitions.
The resulting piceno unit is a
remarkable example of chemi-
cal stitching of adjacent aromat-
ic segments that could find utili-
ty in chemical synthesis. Subse-
Scheme 1. Bergman cyclization of tetraalkynylporphyrins.
quent work on dialkynylpor-
phyrins with hydrogen or halo-
gen substitution at the termini positions[9] demonstrated that
the high thermal barrier to this process could be reduced by
a steric or an inductive effect; this latter observation has
been computationally purported to reduce electron repul-
sion in the transition state.[10] The decreased barrier to the
primary cyclization event also permits the photocyclization
process to occur at 108C, further promoting formation of
the extended, aromatized, piceno product under accessible
conditions. Attachment of a 1,2-diamino arenediyne at the
b,b’-pyrrole positions can also generate a cyclized product,
albeit without creation of the extended piceno-unit.[11]
The tools to extend this conceptual framework to penta-,
hexa-, and octaalkynyl porphyrins to create larger aromatic
constructs by a multistep, ring-closure event is established[12]
and several of these constructs have been crystallographical-
ly characterized.[13] Interestingly, across this series, a redshift
in the electronic spectrum of B and Q bands of 40 nm is ob-
served, or about 13 nm per alkyne unit. As electronically at-
tractive as this series is, formation of a single chemical prod-
uct by Bergman cyclization/meso-phenyl-ring substitution is
unfortunately hindered with addition of each successive di-
should be a highly aromatized, thermodynamically stable
structure. Because polymerization always accompanies
single-enediyne-unit cyclization steps,[14] a precise assess-
ment of reaction progress and mechanism is difficult. DFT
calculations are therefore used to computationally probe the
single and double cyclization/rearomatization events along
the reaction profile to the bispicenoporphyrin product. As
an additional probe of molecular strain, computational eval-
uation of the meso-phenyl-substituted base structure, togeth-
er with hydrogen and ethenyl derivatives, was carried out to
examine the contribution of steric crowding to the pyrrole-
ring distortion across the structure, and the effect of this dis-
tortion on the electronic energy of the system and corre-
sponding reaction barriers along the reaction profile were
evaluated.
Results and Discussion
Syntheses: A series of new compounds were prepared by
using 5,10,15,20-tetraphenylporphyrin (TPP) as a starting
material.[15] The synthesis involves selective bromination at
the antipodal porphyrin positions by refluxing five equiva-
lents of N-bromosuccinimide (NBS) with TPP in CHCl3 for
24 h.[16] Bromination occurs quite readily and affords
2,3,12,13-tetrabromo-5,10,15,20-tetraphenylporphyrin (1a)
in high yield (80%).
ACHTUNGTRENNUNGalkyne pair, to the limit at which no isolable aromatized
structure is detected upon either heating or photochemical
activation. This suggests that the reaction profile is either
dominated by side-product (i.e., polymer) formation, or that
a considerable energetic penalty exists to more than one
cyclization event—likely caused by molecular strain commu-
nicated throughout the macrocycle backbone. This hypothe-
sis is supported by the solid-state thermal cyclization tem-
peratures of the octaalkynylporphyrins, which increase as
the deviation of the macrocycle from planarity increases,[13]
suggesting that the strain of the molecule can override the
thermodynamic gains of formation of a large extended p
system, which should be the thermodynamically favored
structure.
Nickel is inserted into the macrocycle by refluxing 1a in a
solution of NiACTHNUTRGNE(UNG OAc)2·4H2O in acetic acid and chloroform,
and the product is isolated as the red–purple crystalline
(2,3,12,13-tetrabromo-5,10,15,20-tetraphenylporphyrinato)-
nickel(II) (1b) in 89% yield (see Scheme 2). The brominat-
ed 2,3,12,13-positions were ethynylated by treating 1 with
six equivalents of trimethyl(trimethylstannanylethynyl)silane
and a Pd0 catalyst in an inert atmosphere. Addition of the
trimethylsilylethynyl units is a facile reaction and affords the
2,3,12,13-tetrakis(trimethylsilylethynyl)-5,10,15,20-tetraphen-
ylporphyrins (2a,b) in high yields (ꢁ75%).
As a mean to determine which of these issues are respon-
sible for the observed reaction profile and specifically define
the origin and magnitude of molecular strain in these ex-
tended p structures, four alkyne units were installed at the
antipodal b,b’-pyrrole positions. The resulting tetraalkynyl-
14540
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14539 – 14551

Extended, p-Conjugated, Light-Harvesting Macrocycles
FULL PAPER
Bergman cyclization: Unlike
the dialkynylporphyrins,[9b] pho-
tochemical activation of the tet-
raalkynylporphyrins does not
result in isolable cyclized prod-
ucts. However, thermal Berg-
man cyclization of 3 affords the
half-cyclized 2,3-diethynyl-5,20-
diphenylpiceno[10,11,12,13,14,
15-fghij]porphyrin (8) as well as
aggregated and polymerized
side products. The fully cyclized
bispicenoCAHUTGTNREN[UNG 20,1,2,3,CAHTUNGTRENNNUG
AHCTUNGTERG1NNUN 3,14,15-fghij]porphyrin
cannot be isolated from the re-
action mixture. Yields of 8 are
highly dependent on the central
metal. The free-base analogue
3a undergoes cyclization most
readily producing 8a in up to
30% yield. The NiII product 8b
can be isolated in yields up to
10%, and the ZnII product 8c
only forms in trace amounts. In
Scheme 2. Synthesis of tetraalkynylporphyrins; TMS=trimethylsilane.
addition, attempts to convert
isolated 8a into fully cyclized
Zinc is inserted into 2,3,12,13-tetrakis(trimethylsilylethyn-
yl)-5,10,15,20-tetraphenylporphyrin (2a) by stirring with Zn-
ACHTUNGTRENNUNG(OAc)2·4H2O for one hour in CHCl3 and methanol to yield
13a were unsuccessful with conversion to side products
(mixture of aromatic and polymeric material) in near quan-
titative yields at elevated temperatures. Selected conditions
for these reactions are shown in Table 1.
(2,3,12,13-tetrakis(trimethylsilylethynyl)-5,10,15,20-tetraphe-
nylporphyrinato)zinc(II) (2c). Free-base and ZnII 2,3,12,13-
tetrakis(trimethylsilylethynyl)-5,10,15,20-tetraphenylpor-
phyrin were deprotected at room temperature with five
equivalents of tetrabutylammonium fluoride (TBAF) in
THF to afford the corresponding desilylated products 3a
and c in high yields. The free-base derivative was successful-
ly crystallized from chlorobenzene/methanol by slow diffu-
sion and the structure was determined by X-ray crystallogra-
phy. To deprotect the NiII compound (2b) was added to
THF, methanol, and water. Potassium carbonate was added,
and the reaction mixture was stirred at 258C for 24 h. This
gave (2,3,12,13-tetraethynyl-5,10,15,20-tetraphenylporphyri-
nato)nickel(II) (3b) in good yield (80%).
Table 1. Reaction conditions for Bergman cyclization of 3 and 8.
H donor Equiv T [8C]
t
3 [%][a] 8 [%] Side
product [%][b]
3a toluene solvent 115
3a toluene solvent 155
2 h
10 min 30
0
30
15
20
10
10
0
70
65
80
90
40
50
77
30
85
100
100
100
10
3a CHD[c]
3a CHD
3b CHD
100
1000
350
80
80
105
8 h
12 h
1.5 h
1.5 h
2 h
1 h
2 h
1 h
5 h
0
0
50
50
20
70
15
0
–
–
–
3b toluene solvent 115
3b toluene solvent 130
3c toluene solvent 115
3c toluene solvent 110
3
trace
trace
0
0
0
90
3c CHD
8a CHD
100
500
120
200
8a toluene solvent 120
8a toluene solvent 65
7 h
24 h
Solid-state thermal reactivity: Differential scanning calorim-
etry (DSC) has been shown to be a reliable tool for measur-
ing Bergman cyclization temperatures of metalloenediyne
structures in the solid state (absence of an H donor), and ef-
fectively correlates the temperature with the distance be-
tween the alkyne termini.[7,9b,10b,13,17,18] Solid-state cyclization
temperatures for 3a–c all exhibit a major exothermic peak
between 149 and 1508C, which reveals that activation of all
three tetraalkynyl derivatives occurs at a comparable tem-
perature. These solid-state cyclization temperatures are
lower than the dialkynyl counterparts by around 108C;[9b]
this suggests that the tetraalkynyl compounds may be slight-
ly more reactive towards Bergman cyclization.
[a] Recovered starting material. [b] Mixture of aromatic and polymeric
materials. [c] CHD=1,4-cyclohexadiene.
The thermal Bergman cyclization reaction of dialkynyl
porphyrins was originally shown by Smith et al.[7] to proceed
in high yields (ca. 70%), albeit at elevated temperatures.
Subsequent work demonstrated that alkyne termini substitu-
tion[9a] and photochemical activation[9b] are both viable
routes to lower the thermal reaction barrier of the cycliza-
tion reaction. In the case of the tetraalkyne scaffold, the
Chem. Eur. J. 2011, 17, 14539 – 14551
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14541

J. M. Zaleski, M.-H. Baik et al.
ability to isolate 8 from solution suggests that sequential
cyclization of each half of the molecule is the plausible path
of reaction. Within this theme, the proposed mechanism
(Scheme 3), adapted from Smith et al.,[7] involves thermal
intermediates, or more likely, the result of higher reaction
barriers associated with the second cyclization event.
Structures of tetraalkynylporphyrins: Compound 3a crystal-
lizes as brown blocks in
monoclinic space group by slow
evaporation from a chloroben-
zene/methanol cosolvent
a
AHCTUNGTRENNUNG
system. A variety of methods
and solvents were employed to
grow crystals of compounds 3b
and 3c, but crystals obtained
for both proved too small for
structure determination. To es-
tablish geometric parameters
for these molecules, DFT-com-
puted structures for 3b–c were
obtained (Table 2). The crystal
structure of 3a was also com-
pared to the computational
counterpart to confidently dem-
onstrate the ability of the quan-
tum calculations to reliably re-
produce/predict the structures
of the metallated porphyrins
(Figure 1).
Scheme 3. Proposed mechanism for the stepwise Bergman cyclization of tetraalkynylporphyrins.
activation of one dialkynyl unit to generate the bis(1,4-di-
ACHTUNGTRENNUNGradical) species 4. The radicals are then transferred to the
adjacent meso-phenyl rings by electrophilic aromatic substi-
tution to afford species 6, which is quenched by two succes-
Table 2. Structural parameters for tetraalkynylporphyrins 3a–c.
ꢀ
Alkyne
termini[a]
[ꢁ]
Ca
Alkyne Alkyne
[b]
CSx
[ꢁ]
CSy
[ꢁ]
Cmeso SB[e]
[ꢁ] [ꢁ]
TB[f]
[ꢁ]
C
sive hydrogen transfers from the H donor to give the tetra-
hydro species 7. The reduced aromatic structure is then oxi-
dized upon workup, which results in the monocylized, por-
phyrinic enediyne 8. Due to the radical nature of the reac-
tion and abundance of reactive alkynes, polymeric and
oligomeric products are readily observed from this type of
reaction[19] and serve as an obstacle to achieve high yields.
Initial attempts to generate 13 saw a high degree of poly-
merization, discernible as a layer of a sticky black substance
on the sides and bottom of the reaction flask. CHD concen- N N distance was measured and divided by two. The x axis is defined as the
trations were then iteratively varied to achieve a better ratio
of hydrogen donor to reactant, and the qualitative amounts
of polymerized material were minimized. However, the
presence of the fully cyclized bispiceno[20,1,2,3,4,5,10,11,-
3a
crystal
3a
model
3b
model
3c
4.11
4.13
4.17
4.13
2.08 2.07 106.3[c]
2.12 2.06 106.5[c]
1.95 1.91 106.9[c]
2.12 2.06 106.5[c]
126.2[d]
126.1[d]
120.8[d]
126.1[d]
1.40 1.42
1.42 1.41
1.40 1.41
1.41 1.41
1.19
1.22
1.22
1.22
model
ꢀ
ꢀ
[a] Terminal C C distance. [b] Core size or M N distance; in 2H porphyrins,
ꢀ
horizontal axis, and y as the vertical axis. [c] Angle measured for alkyne-sub-
stituted pyrroles, in degrees. [d] Angle reported for meso-position adjacent to
measured pyrrole ring, in degrees. [e] Single bond. [f] Triple bond.
ACHTUNGTRENNUNG12,13,14,15-fghij]porphyrin product (13) has only been de-
tected in trace amounts by mass spectrometry. When tolu-
ene was substituted for chlorobenzene/benzene and CHD,
which acts as both solvent and hydrogen donor, the amounts
of side product were further diminished, but polymerization
was still the dominant reaction pathway. It is unknown at
which step in the mechanism the oligomerization/polymeri-
zation occurs, but diradical 6 is a likely candidate, because
bimolecular radical coupling or addition is competitive with
hydrogen abstraction by 6 to form 7. In fact, both dimers
and trimers of the porphyrin product have been detected by
mass spectrometry. This leads to the question of whether the
inability to isolate 13 is simply due to poor quenching of the
The X-ray and calculated structures of 3a are in excellent
ꢀ
ꢀ
agreement; the Cb alkyne single bond, Ca Cmeso distance,
and core size in the y direction are all within 0.01 ꢁ. Only
very minor differences in the alkyne termini distance, triple-
bond length, and core size in the x direction exist; this indi-
cates a high degree of accuracy in the calculated structures.
The nature of this agreement allows the use of calculated
structures instead of X-ray crystal structures for 3b–c. Nota-
ble parameters for these include core size, a significant
structural parameter in discussing planarity of porphyrins,[20]
which is 0.2 ꢁ smaller for 3b than for 3a or c. This feature
is common for NiII porphyrins, and leads to distinctively ruf-
14542
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14539 – 14551

Extended, p-Conjugated, Light-Harvesting Macrocycles
FULL PAPER
one side of the molecule, the remaining antipodal alkynes
are displaced farther out of the plane of the porphyrin, and
away from the remaining meso-phenyl rings; this in turn
also forces the two remaining alkynes apart from each
other. In 8a, the alkyne termini distance increases from 4.11
to 4.27 ꢁ, and from 4.13 to 4.23 ꢁ for 8c. Because the
alkyne termini distance is a significant factor in the cycliza-
tion reactivity, this intuitively understandable increase of the
alkyne distance provides one explanation for the inability to
obtain fully cyclized structures from the reaction mixture.
The cyclization of the first enediyne unit results in a nearly
threefold increase in deviation from the plane of 8a (0.102
to 0.288 ꢁ), and a nearly fivefold increase in deviation from
the plane of 8c (0.055 to 0.269 ꢁ; Figure 3, Table 3). In addi-
Figure 1. X-ray crystal structure of 3a and computed structures for tet-
raalkynylporphyrins 3b–c.
fled structures for substituted NiII porphyrin deriva-
ACHTUNGTRENNUNG
tives.[5,20a,21]
The degree of the ruffling can be seen in the calculated
deviation from a planar geometry for the tetraalkynylpor-
phyrins, illustrated in Table 3 and Figure 3. Compound 3b
exhibits the classic ruffled conformation, showing a planar
deviation of 0.322 ꢁ. As a result, the alkynes also distort
away from each other and cause a large alkyne termini dis-
tance (4.17 ꢁ). For 3a and c, the main porphyrin core is rel-
atively flat, with deviations from the plane of 0.102 and
0.055 ꢁ, respectively, from the idealized porphyrin frame-
work. The ZnII compound 3c is slightly more planar than
the free-base analogue, as is common for ZnIITPP deriva-
tives. The dialkyne units in both structures point slightly
above and below the main porphyrin plane, ꢂ0.88 ꢁ for the
alkyne termini carbon atoms, due to the steric encumbrance
of the meso-phenyl groups. However, the alkyne pairs lie
planar to each other and have respective alkyne termini dis-
tances of 4.11 and 4.13 ꢁ.
Partial thermal Bergman cyclization of 3a–c produces
new structures 8a–c that are highly distorted (Figure 2) due
ꢀ
to formation of three new C C bonds: one from the devel-
Figure 3. Structural overlays of deviation from plane for 3 (solid lines)
and the subsequent cyclized structures 8 (dotted lines).
Figure 2. Structures of monocyclized porphyrinic enediynes.
tion to these changes, the main core of the porphyrin under-
goes significant transformations as well. The core of all com-
pounds becomes antisymmetric upon cyclization (Table 2
and Table 4). For the free-base tetraalkyne 3a, CSx and CSy
are nearly equivalent at 2.08 and 2.07 ꢁ, respectively. Upon
cyclization, however, the core is forced to increase in size
along the y direction (CSy =2.17 ꢁ, in the direction along
the length of the piceno unit) and compress along x (CSx =
1.95 ꢁ) to accommodate the expanded p-piceno unit.
Whereas the flexible free-base porphyrin core can accom-
modate these structural changes, the more rigid ZnII struc-
ture is less likely to do so. Similar to 3a, the ZnII porphyrin
oping benzyne ring, and two more from the radical addition
at the 1,4-positions of the newly formed ring adjacent to the
meso-phenyl substituents. The resulting aromatic system
flattens and forces the rest of the molecule into a distorted
geometry. For 8a and c, this structural distortion subse-
quently increases the steric interaction on the opposite side
of the molecule between the remaining dialkyne unit and
the adjacent meso-phenyl rings. To minimize this unfavora-
ble interaction, the entire porphyrin distorts to a pseudo-
saddled conformation. Thus, as a result of the cyclization of
Chem. Eur. J. 2011, 17, 14539 – 14551
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14543

J. M. Zaleski, M.-H. Baik et al.
Table 3. Calculated deviation from plane[a] for porphyrinic enediynes 3
and 8 [ꢁ].
Electronic spectroscopy: Representative electronic absorp-
tion spectra for 3a and 8a are shown in Figure 4. The tet-
raalkynyl 3a exhibits a B band at l=441 nm, with a molar
[c]
[d]
[e]
Mean[b]
NApyrrole
NBpyrrole
Calkyne
3a crystal
8a crystal
3b model
8b model
3c model
8c model
0.102
0.288
0.322
0.336
0.055
0.269
ꢀ0.192
ꢀ0.302
0.034
0.010
0.091
0.192
0.036
0.056
ꢀ0.886
ꢀ1.584
ꢀ0.674
0.760
0.080
ꢀ0.115
0.016
ꢀ0.091
0.279
0.014
[a]Deviation of the indicated atom from the idealized planar structure.
[b]The mean deviation of the 24 core atoms of the porphyrin macrocycle
from the idealized planar structure. [c]Nitrogen of cyclized/enediyne pyr-
role ring. [d]Nitrogen of uncyclized pyrrole ring. [e]Alkyne termini C for
uncyclized enediyne.
Table 4. Structural parameters for porphyrinic enediynes 8.
Alkyne
termini[a]
[ꢁ]
New
[b]
ꢀ
CSx
[ꢁ]
CSy
[ꢁ]
Ca Cmeso
[ꢁ]
ꢀ
C C
[ꢁ]
Figure 4. Absorption spectra for 3a (solid line) and cyclized 8a (dashed
line).
8a
crystal
8a
model
8b
model
8c
4.27
4.19
4.13
4.23
1.95
1.97
1.89
2.03
2.17 104.8[c]
2.17 105.0[c]
1.95 106.2[c]
2.11 107.5[c]
120.3[d]
120.7[d]
116.1[d]
120.5[d]
1.39
1.41
1.44
1.46
1.41
1.41
absorptivity of 3.41ꢂ105 mꢀ1 cmꢀ1. The four Q bands (535,
574, 627, and 686 nm) are a distinctive feature for free-base
porphyrins, and indicate D2h symmetry for the p-electron-
conjugated ring system. The absorption spectrum for cy-
clized product 8a is bathochromatically shifted due to the
increased aromaticity, with the B band at l=472 nm (e=
1.95ꢂ105 mꢀ1 cmꢀ1). The three discernable Q bands fall at
636, 658, 688, and 741 nm with extinction coefficients rang-
ing from 1.84ꢂ105 to 4.13ꢂ105 mꢀ1 cmꢀ1. It is known that
structural perturbations, such as the addition of fused ben-
zene rings, break the accidental degeneracy of the 1a1u and
1a2u HOMO set.[23] This diminishes the ideally forbidden
and allowed characters of the Q and B transitions, respec-
tively, and causes the Q bands to gain intensity, at the ex-
pense of the B band.[5h,23,24]
1.39
1.41
model
ꢀ
ꢀ
ꢀ
[a] Terminal C C distance. [b] M N distance; in 2H porphyrins, N N dis-
tance was measured, then divided by two. The x axis is defined as the
horizontal axis and y as the vertical axis given in Figure 2. [c] The angle
measured is the pyrrole on the new benzene ring in 8, measured in de-
grees. [d] Angle measured was from meso-position adjacent to measured
pyrrole ring, in degrees.
3c has a relatively symmetrical core size, with CSx and CSy
distances of 2.12 and 2.06 ꢁ, respectively. In cyclized 8c,
these distances shift to CSx =2.03 and CSy =2.11 ꢁ; this
highlights that the presence of ZnII in the cavity restricts the
divergence of the porphyrin from a symmetric core and
leads to diminished change relative to the free-base ana-
logue. Nonplanar ZnII porphyrins are known, but they are
exceedingly rare and are only observed when ZnII has a co-
ordinating axial ligand.[22] Because only a very small amount
(trace yield) of monocyclized ZnII porphyrin can be isolated
from solution, it appears that the half-cyclized ZnII structure
is highly susceptible to both further cyclization reactions,
and competing side reactions that occur in the high-temper-
ature solution. Distorted ZnII porphyrins have been shown
to be more susceptible to oxidation,[22b] and strained, exocy-
clic ZnII porphyrins in particular can react with ambient
oxygen to form decomposition products.[3c]
The compound 8b displays fewer structural distortions
upon cyclization than 8a or 8c. The deviation from the
plane differs very little from the uncyclized tetraalkyne, that
is, <0.02 ꢁ. Unlike the free-base and ZnII derivatives, the
core size changes only very slightly; CSx decreases from 1.95
to 1.89 ꢁ, whereas CSy increases from 1.91 to 1.95 ꢁ. In ad-
dition, the alkyne-termini separation actually decreases
slightly from 4.17 to 4.13 ꢁ.
Computed reaction profiles: The mass-balanced[25] reaction
profile for Bergman cyclization of free-base tetraalkynylpor-
phyrin 3a can be seen in Figure 5 and Figure 6. As expected,
the overall process is highly exothermic as a result of forma-
ꢀ
tion of p-extended porphyrin 13 in which six new C C s
bonds have been formed in place of four p bonds in 3a. Ar-
omatization of two equivalents of CHD into benzene and
release of four equivalents of H2 molecules also favor the
overall reaction equilibrium entropically. The fully cyclized
and rearomatized product 13 lies 245.1 kcalmolꢀ1 below
starting material 3. The overall reaction profile has several
steps and a total of six calculated reaction barriers. The first
step affords the stable, intermediate, 1,4-diradical from the
tetraalkynylporphyrin 3 by traversing the transition state 3-
TS, which has a barrier of 17.4 kcalmolꢀ1.[26] The product of
this reaction, diradical 4, reacts stepwise with each adjacent
meso-phenyl moiety, first to form the separated diradical 5.
This transient intermediate then adds to the second meso-
phenyl ring to form diradical 6. The addition of two equiva-
C
lents of H is expected to be a fast reaction, thus we assume
14544
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14539 – 14551

Extended, p-Conjugated, Light-Harvesting Macrocycles
FULL PAPER
C
diradical intermediate 11. Two equivalents of H add to this
species to give the quenched intermediate 12, which elimi-
nates two H2 molecules to finally form the fully cyclized 13,
which lies 120.8 kcalmolꢀ1 below the half-cyclized 8 and an
overall of 245.1 kcalmolꢀ1 below the starting porphyrinic
enediyne 3.
The diradical species 4–6 and 9–11 have been computed
as open-shell singlets within a broken symmetry MO frame-
work, to properly model the 1,4-diradicals resulting from
Bergman cyclization as diradicals with the unpaired elec-
trons adopting opposite spins.[10a,12c,18a,28] Addition of meso-
phenyl rings to these diradicals is facile with activation bar-
riers of approximately 3 kcalmolꢀ1 (see the Supporting In-
formation for details). Spin densities from Mulliken popula-
tion analysis are enumerated in Table 5 and show conver-
ꢀ
Table 5. Energies, C C distance, and spin density for diradicals and asso-
ciated transition states.
ꢀ
DE
(gas)
DG
(sol)
C C distance
Spin
Figure 5. Reaction profile for the first cyclization of the Bergman cycliza-
tion of tetraalkynylporphyrin.
[kcalmolꢀ1
]
[kcalmolꢀ1
]
[ꢁ][a]
density[b]
3-TS
4
17.34
17.42
2.13
1.45
2.16
1.45
C59
0.01
ꢀ0.01
0.75
C62
C59
C62
C79
C82
C79
C82
ꢀ1.50
22.00
4.24
2.00
22.97
8.73
ꢀ0.75
8-TS
9
0.00
0.00
0.79
ꢀ0.78
ꢀ
[a] Length of new or forming C C bonds. [b] Atomic spin densities from
Mulliken analysis, as portion of an electron at 1,4-carbon atoms. Specific
carbon atom positons are defined in Figure 5 and 6.
gence to the open-shell singlet states for species 4 and 9. For
diradical 4, spin density at C59 and C62 (for numbering, see
Figure 5) is approximately ꢂ0.75, which shows strong diradi-
cal character at the 1,4-carbon atoms, and the radical orbital
clearly shows unpaired alpha spin density on C59, and un-
paired beta spin on C62 (Figure 7). Diradical 9 shows a mar-
ginally higher spin density of ꢂ0.79 at C79 and C82 (for
numbering, see Figure 6). The two diradicals also differ in
relative energies with 9 being 6.7 kcalmolꢀ1 higher in energy
Figure 6. Reaction profile for the second cyclization of the Bergman cyc-
lization of tetraalkynylporphyrin.
ꢀ
than 4 relative to the respective starting enediynes. The C
C distances for the new delocalized bonds are both 1.45 ꢁ,
that both equivalents add simultaneously to give the
quenched species 7, which lies 80.7 kcalmolꢀ1 below the
starting enediyne. The transition state between 6 and 7
cannot be calculated easily, due to difficulty to calculate the
proper reaction trajectory for this intermolecular reaction,
but we expect it to be relatively low in energy. Likewise, the
release of two molecules of H2 to form the rearomatized,
half-cyclized 8 is also expected to be associated with a low
barrier, which cannot be modeled reliably.[27]
The cyclization mechanism of the second enediyne, shown
in Figure 6, is analogous to the first half: Compound 8 tra-
verses 8-TS with a barrier height of 23.0 kcalmolꢀ1 to form
diradical species 9, which is 8.7 kcalmolꢀ1 higher in energy
than 8. Subsequently, one radical adds to the adjacent
phenyl ring to form 10, and the second adds to afford the
ꢀ
which is usual for a resonance-stabilized C C double bond.
The transition states for the diradical formation (Table 5)
show no significant diradical character; this is consistent
with previous studies.[18a,29] Structurally, the transition state
ꢀ
is late, with C C distances at 2.13 to 2.16 ꢁ. The alkyne
triple bonds are slightly elongated, from 1.22 ꢁ in 3 and 8 to
1.26–1.27 ꢁ in 3-TS and 8-TS, respectively; this shows that
the bonds still retain significant triple bond character. Thus,
the transition state can be characterized as structurally late,
but electronically early. The most notable characteristic of
3-TS and 8-TS, however, is the difference in relative energy:
8-TS lies 5.5 kcalmolꢀ1 higher than 3-TS. Inspection of the
differences in energy terms, summarized in Table 5, reveals
that practically the same energy difference can be found in
Chem. Eur. J. 2011, 17, 14539 – 14551
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14545

J. M. Zaleski, M.-H. Baik et al.
Figure 8. Structures of model-porphyrinic-enediyne systems with C2H3
and H at the meso-positions.
Figure 7. Isosurface plots (0.05 a.u.) of the MOs for 4 and 9 that show di-
ACHTUNGTRENNUNGradical character.
ꢀ
Table 6. Energies, C C distance, and spin density for diradicals and asso-
ciated transition states for model systems.
the electronic component. A more detailed analysis suggests
that the structural difference of how the enediynes are ori-
ented in relation to the rest of the porphyrin system during
the first and second cyclization is responsible for the ener-
getic difference. As discussed above, 3 displays a planar por-
phyrin core, with the alkyne pairs pointing slightly above
and below the plane, as shown in Figure 1. The meso-phenyl
rings, which in 3 are almost orthogonal to the porphyrin p
system, need to rotate into the plane to couple with the 1,4-
ꢀ
DE
G
DG
G
C C distances
Spin
[kcalmolꢀ1
]
[kcalmolꢀ1
]
[ꢁ][a]
density[b]
14-TS
15
19.35
19.97
2.16
1.45
2.17
1.45
2.10
1.45
2.15
1.45
C44
0.00
0.00
C47
C44
C47
C48
C51
C48
C51
C39
C42
C39
C42
C41
C44
C41
C44
1.22
19.02
1.41
5.29
19.79
5.43
0.81
ꢀ0.81
16-TS
17
0.00
0.00
ꢀ0.81
0.81
0.00
19-TS
20
25.25
7.82
26.19
12.10
26.00
12.38
ꢀ
diradical and to form the new C C bonds. To accommodate
0.00
this new piceno unit, the porphyrin core must elongate in
the y direction; this in turn distorts the angles at the meso
carbon atoms, and pushes the adjacent pyrrole moieties out
of the plane. The overall effect is a loss of porphyrin planari-
ty, which destabilizes the p system. This translates into a
higher barrier for the second cyclization than the first and
increases the likelihood that 8 will follow alternative, unpro-
ductive reaction pathways, such as polymerization, radical–
radical coupling, or Stork acylation.[30]
ꢀ0.83
0.83
0.00
21-TS
22
25.09
8.07
0.00
ꢀ0.83
0.83
[a] Length of new or formed C C bond. [b] Atomic spin densities from
Mulliken analysis, as portion of an electron at 1,4-carbon atoms. Specific
carbon atom positions are defined in Figure 8.
To probe how this strain impacts the transition states, two
other porphyrinic enediyne models were considered, each
with a smaller functional group at the porphyrin meso-posi-
tion. Compound 19 has hydrogen at this position, and 14
ethenyl units (C2H3). Both systems are completely planar
system. For H, both the first and second cyclizations are
higher by 8.8 and 3.0 kcalmolꢀ1, respectively, compared to
the Ph-substituted analogue. These model systems suggest
that stabilization from an additional conjugated p system at
the meso-position aids in making the cyclization feasible.
The H-substituted derivative lacks this feature, resulting in a
higher barrier of about 26 kcalmolꢀ1. Whereas this factor
favors the first cyclization event for Ph relative to C2H3, the
bulkiness of the meso-phenyl group perturbs the planarity of
the porphyrin core in the second cyclization event; this
makes it energetically more costly than C2H3.
Diradical character as assessed by Mulliken spin density
(MSD) is reported in Table 6 for these model systems. For
C2H3, diradicals of 15 and 17 both have MSDs of ꢂ0.81,
compared to the H system, in which diradicals 20 and 22
have spins of ꢂ0.83. These values emphasize further that
ꢀ
(Figure 8). The C C distances and the diradical character in
the respective transition states closely resemble those of the
Ph system (Table 6). Activation barriers for the formation of
1,4-diradicals of these model systems were found to be simi-
lar for the first and second cyclization event. For C2H3, the
activation energies are 20.0 and 19.8 kcalmolꢀ1, for first and
second cyclization events, respectively, whereas they are
26.2 and 26.0 kcalmolꢀ1 for H (Table 6). These model sys-
tems differ from the phenyl (Ph)-substituted system in that
for C2H3 the first cyclization requires 2.6 kcalmolꢀ1 higher
activation energy, whereas the second cyclization is lower in
energy by 3.2 kcalmolꢀ1 compared with the Ph-substituted
14546
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14539 – 14551

Extended, p-Conjugated, Light-Harvesting Macrocycles
FULL PAPER
the model systems reliably mimic the electronics of the real
systems.
With the insight from the small model system at hand, we
attempted to reproduce the electronic impact of the struc-
tural distortion, discussed above, in another computational
experiment. An artificial 308 cant is introduced in the pyr-
role rings relative to the plane of the porphyrin in 16 and
this angle is fixed when the transition state of the second
Bergman cyclization is calculated. Optimized geometries of
16 and 16’ are shown in Figure 9. Interestingly, the second
Figure 9. Comparison of optimized geometries of 16, 16’ and 21’. In 16’
and 21’ pyrroles are fixed at 308 cant.
cyclization barrier increases from 19.8 to 21.8 kcalmolꢀ1, as
illustrated in Figure 10. Inspection of 16’ clearly shows sig-
nificant bending of the enediyne moiety from the porphyrin
plane as a result of canting of the pyrrole. This is reflected
in an activation-barrier increase by 2.0 kcalmolꢀ1 from 16 to
16’. This distortion perturbs the resonance gained from the
conjugation of the meso-position into the porphyrin system,
and raises the barrier for the second cyclization. However,
for H, similar canting of pyrrole rings showed no change in
the activation barrier (21-TS and 21’-TS in Figure 10). Thus
the distortion that the first cyclization event causes to the
porphyrin ring in the Ph-substituted system is the cause of
the increased barrier to the second cyclization event, due to
the decreased stabilization from reduced p overlap.
Figure 10. Diagram that illustrates the relative energy changes of the cyc-
lization transition states upon substitution at the meso-position in por-
phyrinic enediynes. Bonds in bold indicate molecules frozen in the dis-
torted geometry (pyrroles are fixed at 308 cant).
Conclusion
higher than to the first; this suggests a negative cooperative
effect and a significantly slower rate for the second cycliza-
tion. Cyclization reactions for model porphyrinenediynes
with ethene and H functionality at the meso-positions were
also found to have the same barrier to diradical formation
for the second cyclization event as for the first in these
highly planar molecules. Enforcing an artificial 308 cant in
two of the pyrrole rings of the porphyrin increases the
second barrier by 2 kcalmolꢀ1 in the ethene model, indicat-
ing that disruption of the p conjugation in the extended por-
phyrin structure is the cause of the increased barrier to the
second cyclization event. Combined, this suggests that main-
taining a planar aromatic structure is paramount to promote
combined or tandem cyclization reactions in the pursuit of
extended p structures.
Heating of a new series of free-base, NiII, and ZnII 2,3,12,13-
tetraACHTUNGTRENNUNG(ethynyl)-5,10,15,20-tetraphenyl porphyrins promotes
two of the four ethynyl moieties to undergo Bergman cycli-
zation to afford the monocyclized 2,3-diethynyl-5,20-diphe-
nylpiceno[10,11,12,13,14,15-jklmn]porphyrin in 30%, 10%,
and trace yields, respectively. No fully cyclized bis-
(piceno[20,1,2,3,4,5,10,11,12,13,14,15-fghij])porphyrin
was
isolated from the reaction mixture. The X-ray and computa-
tional structures of picenoporphyrin products show signifi-
cant deviation from the idealized planar structure as well as
from that of the starting material. Thus the distortion on the
porphyrin backbone imparted by the newly formed piceno
unit is clearly a complicating factor in the second cyclization
event. Computation of the reaction coordinates with DFT at
the PWPW91/cc-pVTZ(-f) level coupled to a continuum sol-
vation model reveal that the barrier to cyclization of the
second pair of ethynyl groups was found to be 5.5 kcalmolꢀ1
Chem. Eur. J. 2011, 17, 14539 – 14551
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14547

J. M. Zaleski, M.-H. Baik et al.
(1.70), 615.0 nm (3.50ꢂ10ꢀ4 mꢀ1 cmꢀ1); MALDI-TOF-MS: m/z: 1054 [M]+
; HRMS (EI): m/z: calcd for C64H60N4NiSi4: 1055.2234; found: 1055.3373.
Experimental Section
Compound 2c: To a solution of 2a (500 mg, 0.501 mmol) in CHCl3
(200 mL), 1.2 equivalents of ZnACHTNUGTRNEUNG(OAc)2·H2O (130 mg, 0.601 mmol) in
Materials and general procedures: All chemicals and solvents used were
of the highest purity available from Aldrich and Strem. Air-sensitive re-
actions were carried out under nitrogen by using Schlenk techniques and
air-sensitive compounds were handled in an inert-atmosphere dry box.
Compounds were purified by flash chromatography with activated neu-
tral aluminum oxide or silica gel. All NMR (1H and 13C) spectra were re-
corded on a VXR 400, i400, or Gem 300 NMR spectrometer with the re-
sidual proton resonance of the solvent as an internal reference. Infrared
spectra (KBr) were measured with a Nicolet 510P FTIR spectrophotome-
ter. MALDI-TOF data were obtained with a Bruker Biflex III Maldi-
TOF mass spectrometer. ESI and EI-HRMS spectra were recorded on
PE-Sciex API III Triple Quadrupole and Thermo Finnigan MAT 95 XP
high-mass-resolution spectrometer, respectively. Elemental analyses were
obtained from Robertson Microlit Laboratories. Electronic absorption
spectra were acquired on a Perkin–Elmer Lambda 19 UV/Vis/near-IR
spectrometer. All DSC traces were measured on a TA Instruments Q10
MeOH (50 mL) were added. The reaction mixture was stirred at room
temperature for 1 h to afford a green solution. The solvent was removed
under reduced pressure and the resulting solid was purified by silica-gel
column chromatography with CH2Cl2/hexane (1:1) (yield: 95%). Crystals
for X-ray diffraction were obtained by slow diffusion from dichlorome-
thane/methanol. 1H NMR (CDCl3): d=0.23 (s, 36H, 4
ACTHUNGTRNEUNG(CH3)3Si), 7.63–
7.68 (m, 8H, meso-ArH), 7.75–7.77 (m, 4H, meso-ArH), 8.06 (d, J=
7.2 Hz, 8H, meso-ArH), 8.59 ppm (s, 4H, b-pyrrolic H); 13C NMR
(CDCl3): d=ꢀ1.37, 99.49, 110.65, 121.60, 127.14, 128.90, 133.16, 133.55,
134.94, 142.50, 147.34, 152.30 ppm. IR (KBr): n˜ =630, 659, 698, 759, 795,
866, 895, 1002, 1024, 1072, 1121, 1173, 1245, 1321, 1352, 1485, 1598, 2136,
2955, 3052 cmꢀ1; HRMS (ESI): m/z: calcd for C64H60N4Si4Zn: 1061.3181;
found 1061.3212.
Compound 3a: To
a solution of 2a (200 mg, 0.200 mmol) in THF
(40 mL), 1m solution of TBAF in THF (1 mL) was added. The reaction
mixture was stirred at room temperature for 2 h. After completion of the
reaction, the solvent was evaporated and the resulting residue was dis-
solved in CH2Cl2. This solution was washed with water, dried over
Na2SO4 and evaporated under reduced pressure. The crude compound
was purified by activated, neural, aluminum-oxide, column chromatogra-
phy with CH2Cl2/hexane (1:1) as solvent (yield: 90%). Crystals for X-ray
diffraction were obtained by slow diffusion from chlorobenzene/metha-
nol. 1H NMR (CDCl3): d=3.57 (s, 4H, alkynyl H), 7.65–7.70 (m, 8H,
meso-ArH), 7.78–7.83 (m, 4H, meso-ArH), 8.08 (d, J=8.1 Hz, 8H, meso-
ArH), 8.83 ppm (s, 4H, b-pyrrolic H); 13C NMR (CDCl3): d=90.58,
120.47, 124.10, 128.62, 129.52, 134.19, 134.67, 140.04, 141.61 ppm; IR
(KBr): n˜ =608, 687, 699, 735, 768, 753, 801, 960, 1002, 1031, 1074, 1089,
1157, 1137, 1175, 1251, 1281, 1342, 1376, 1442, 1476, 1507, 1551, 1598,
1713, 1828, 1892, 2105, 3047, 3290 cmꢀ1; UV/Vis (CH2Cl2): lmax (e)=442.0
(34.14), 535.5 (2.19), 574.0 (1.41), 627.0 (0.76), 686.0 nm (1.11ꢂ
10ꢀ4 mꢀ1 cmꢀ1); elemental analysis calcd (%) for C52H30N4·H2O: C 85.68,
H 4.43, N 7.69; found: C 85.44, H 4.14, N 7.25.
DSC at a heating rate of 108C minꢀ1
.
Porphyrin precursors syntheses: 5,10,15,20-Tetraphenylporphyrin was
synthesized by using the Adler method (yield: 20%).[15] 1H NMR
(CDCl3): d=7.76–7.80 (m, 12H, meso-ArH), 8.25 (dd, J=1.6, 1.6 Hz,
8H, meso-ArH), 8.88 ppm (s, 8H, b-pyrrolic H).
Compound 1a: Compound 1a was prepared as described by Crossley
(yield: 80%).[16] 1H NMR (CDCl3): d=7.76–7.81 (m, 12H, meso-ArH),
8.18 (dd, J=1.5, 2.1 Hz, 8H, meso-ArH), 8.70 ppm (s, 4H, b-pyrrolic H).
Compound 1b: To a solution of 1a (1.0 g, 1.1 mmol) in CHCl3 (500 mL),
a solution of NiACHTUNGTRENNUNG(OAc)2·4H2O (0.60 g, 2.4 mmol) in acetic acid (20 mL)
was added. The reaction mixture was stirred at reflux temperature for
8 h. After completion of the reaction, the reaction mixture was cooled to
room temperature and washed with water (4ꢂ500 mL). The organic
layer was dried over Na2SO4 and concentrated under reduced pressure.
The crude product was triturated with MeOH and filtered to afford red–
purple crystals (yield: 89%). 1H NMR (CDCl3): d=7.72–7.62 (m, 12H,
meso-ArH), 7.90–7.84 (m, 8H, meso-ArH), 8.54 ppm (s, 4H, b-pyrrolic
H); MALDI-TOF-MS m/z: 986 [M]+.
Compound 3b: To a solution of 2b (0.80 g, 0.76 mmol) in THF (150 mL),
MeOH (75 mL) was added followed by the addition of water (7.5 mL)
and K2CO3 (1.3 g, 9.40 mmol). The reaction mixture was stirred at 258C
for 24 h. After completion of the reaction, the solvent was evaporated to
dryness. The purple solid was washed with water and purified on activat-
ed, neutral, aluminum-oxide, column by using CH2Cl2/hexane (1:4) as
eluent (yield: 80%). 1H NMR (CDCl3): d=3.49 (s, 4H, alkynyl H), 7.57
(dd, J=7.6, 7.2 Hz, 8H, meso-ArH), 7.68 (dd, J=7.6, 7.2 Hz, 4H, meso-
ArH), 7.84 (d, J=7.2 Hz, 8H, meso-ArH), 8.70 ppm (s, 4H, b-pyrrolic
H); 13C NMR (CDCl3): d=90.49, 119.10, 127.06, 128.52, 131.87, 133.43,
133.71, 139.78, 140.04, 144.08 ppm; IR (KBr): n˜ =633, 697, 742, 793, 833,
889, 1022, 1071, 1116, 1198, 1340, 1442, 1514, 1599, 1807, 2096, 3047,
Compound 2a: To (Ph3P)4Pd, a solution of 1a (500 mg, 0.538 mmol) in
dry THF (30 mL) was added at room temperature. Trimethyl(trimethyl-
stannanylethynyl)silane (842 mg, 3.23 mmol) in dry THF (15 mL) was
then added to the reaction mixture and heated to reflux (70–808C) for
6 h. The solvent was removed under reduced pressure and the resulting
solid was purified by activated, neutral, aluminum-oxide, column chroma-
tography with 40% CH2Cl2 in hexane (yield: 90%). 1H NMR (CDCl3):
d=0.18 (m, 36H, 4ACHTUNGTRENNUNG(CH3)3Si), 7.69–7.73 (m, 8H, meso-ArH), 7.78–7.81
(m, 4H, meso-ArH), 8.17 (d, J=7.2 Hz, 8H, meso-ArH), 8.63 ppm (s,
4H, b-pyrrolic H); 13C NMR (CDCl3): d=1.01, 99.35, 109.67, 120.55,
127.40, 128.88, 129.42, 134.23, 135.76, 140.49, 141.60, 151.75 ppm; IR
(KBr): n˜ =629, 657, 702, 757, 800, 853, 898, 1001, 1031, 1100, 1140, 1243,
3289 cmꢀ1
; UV/Vis (CH2Cl2): lmax (e)=440.0 (41.46), 557.0 (1.87),
609.0 nm (3.64ꢂ10ꢀ4 mꢀ1 cmꢀ1); MALDI-TOF-MS: m/z: 766 [M]+; ele-
mental analysis calcd (%) for C52H28N4·H2O: C 79.31, H 4.10, N 7.11;
found: C 79.47, H 3.79, N 7.01.
1470, 1599, 2131, 2926, 3365 cmꢀ1
C64H62N4Si4: 999.4124; found: 999.4152.
; HRMS (EI): m/z: calcd for
Compound 2b: To a suspension of 1b (0.95 g, 0.96 mmol) and (Ph3P)4Pd
(0.15 g, 0.13 mmol) in dry THF (80 mL), a solution of trimethyl(trime-
thylsilylethynyl)tin (1.6 g, 6.10 mmol) in dry THF (40 mL) was added.
The reaction mixture was stirred at 70–758C for 6 h. After completion of
the reaction, the mixture was cooled to 258C and the solvent was evapo-
rated under reduced pressure. The crude solid was purified on activated,
neutral, aluminum-oxide column by using 15% CH2Cl2 in hexane as
eluent (yield: 87%). Crystals for X-ray diffraction were obtained by slow
diffusion from dichloromethane/hexanes. 1H NMR (CDCl3): d=0.14 (s,
Compound 3c: To
a solution of 2c (500 mg, 0.50 mmol) in THF
(150 mL), 1m solution of TBAF in THF (1 mL) was added. The reaction
mixture was stirred at room temperature for 2 h. After completion of the
reaction, the solvent was evaporated and the resulting residue was dis-
solved in CH2Cl2. This solution was washed with water, dried over
Na2SO4, filtered, and evaporated under reduced pressure. The crude
compound was purified by silica-gel column chromatography (yield:
55%). 1H NMR (CDCl3): d=3.61 (s, 4H, alkynyl H), 7.61–7.67 (m, 8H,
meso-ArH), 7.76–7.78 (m, 4H, meso-ArH), 8.03 (d, J=6.9 Hz, 8H, meso-
ArH), 8.82 ppm (s, 4H, b-pyrrolic H); 13C NMR (CDCl3): d=90.61,
121.26, 126.61, 128.22, 131.97, 133.22, 133.96, 142.13, 147.26, 151.71 ppm;
IR (KBr): n˜ =619, 673, 699, 746, 757, 766,794, 831, 845, 891, 912, 976,
1009, 1033, 1072, 1105, 1166, 1177, 1194, 1245, 1321, 1345, 1441, 1490,
1575, 1598, 1702, 1817, 2091, 2919, 3305 cmꢀ1; UV/Vis (CH2Cl2): lmax
(e)=442.0 (53.09), 571.0 (1.65), 617.5 nm (3.47ꢂ10ꢀ4 mꢀ1 cmꢀ1); HRMS
(ESI): m/z: calcd for C52H28N4Zn [M+H]+ 773.1678; found: 773.1703.
36H, 4ACHTUNGTRENNUNG(CH3)3Si), 7.59 (dd, J=8.0, 6.8 Hz, 8H, meso-ArH), 7.68 (dd, J=
7.6, 7.2 Hz, 4H, meso-ArH), 7.89 (d, J=7.2 Hz, 8H, meso-ArH),
8.44 ppm (s, 4H, b-pyrrolic H); 13C NMR (CDCl3): d=1.03, 98.57, 110.34,
119.53, 127.39, 132.59, 133.348, 134.53, 140.09, 140.79, 145.27 ppm; IR:
n˜ =631, 644, 662, 697, 720, 745, 757, 792, 833, 894, 995, 1005, 1020, 1073,
1131, 1156, 1176, 1200, 1247, 1338, 1372, 1442, 1493, 1520, 1599, 2139,
2954 cmꢀ1; UV/Vis (CH2Cl2): lmax (e)=447.0 (35.35), 512.0 (0.83), 551.5
14548
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14539 – 14551

Extended, p-Conjugated, Light-Harvesting Macrocycles
FULL PAPER
Thermal cyclization of 3a: A solution of 3a (20 mg, 0.028 mmol) in tolu-
ene (15 mL) was heated in a pressure tube at 1158C for 2 h. The solvent
was evaporated under reduced pressure and the resulting solid was puri-
fied by activated, neutral, aluminum-oxide, column chromatography by
using 2% ethyl acetate in CH2Cl2 to produce 8a and 13a in 30% and
trace yields, respectively. Characterization data for 13a: MALDI-TOF
MS: m/z: 707 [MH]+. Characterization data for 8a: Crystals for X-ray
diffraction were obtained by slow diffusion from dichloromethane/metha-
necessary to converge to a plausible open-shell state and this was moni-
tored by carefully analyzing the computed MSDs and the MOs.
The energy components were calculated following the standard protocol.
The change in free energy in solution DGACTHNUGTRNEUNG(sol) was calculated as follows:
DGðsolÞ ¼ DGðgasÞ þ DDGsolv
DGðgasÞ ¼ DHðgasÞꢀTDSðgasÞ
DHðgasÞ ¼ DEðSCFÞ þ DZPE
ð1Þ
ð2Þ
ð3Þ
1
nol. H NMR (CDCl3): d=3.62 (s, 2H, alkynyl H), 7.75 (dd, 2H, piceno-
H), 7.75 (m, 6H, meso-ArH), 7.78 (q, 4H, meso-ArH), 8.16 (d, J=
7.2 Hz, 4H, piceno-H), 8.49 (s, 2H, piceno-H), 8.59 (d, J=7.6 Hz, 2H,
piceno-H), 8.74 (d, J=6.4 Hz 2H, piceno-H), 9.36 ppm (m, 4H, b-pyrrol-
ic H); IR (KBr): n˜ =664, 698, 754, 795, 822, 850, 866, 965, 1001, 1081,
1442, 1552, 1710 cmꢀ1; UV/Vis (CH2Cl2): lmax (e)=473.5 (20.33), 658.0
(3.56), 688.0 (3.78), 741.5 nm (4.14ꢂ10ꢀ4 mꢀ1 cmꢀ1); MALDI-TOF MS:
m/z: 708 [M]+; HRMS (internally calibrated MALDI-TOF-MS): calcd
for C52H28N4 [M]+: 708.2314; found: 708.2416.
DGACHTNURGTNE(UGN gas) is gas-phase Gibbs free-energy change, DDGsolv is free energy of
solvation difference as calculated with the continuum solvation model.
The entropy of the solvent is implicitly included in the dielectric continu-
um model as the cavitation energy. DS
of the solute. DH(gas) is gas-phase enthalpy change, DE
tronic energy change as computed from the SCF procedure, DZPE is
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
zero-point energy correction difference and
(298.15 K).
T is the temperature
Thermal cyclization 3b: To a solution of 3b (35 mg, 0.046 mmol) in
chloroACHTUNGTRENNUNGbenzene (15 mL), CHD (1.5 mL) was added. The mixture was
stirred at 1058C in a pressure tube for 1.5 h. The reaction mixture was
cooled to room temperature, loaded on a silica-gel column, and eluted
with hexane (500 mL) to remove the chlorobenzene. The column was fur-
ther eluted with 50% CH2Cl2 in hexane to afford 8b in 10% yield.
1H NMR (CDCl3): d=3.47 (s, 2H, alkynyl H), 7.47 (dd, J=7.6, 7.2 Hz,
2H, piceno-H), 7.63 (dd, J=8.0, 7.2 Hz 4H, meso-ArH), 7.75–7.69 (m,
4H, meso-ArH), 7.89 (d, J=6.8 Hz 4H, piceno-H), 8.45 (s, 2H, piceno-
H), 8.50 (d, J=8.0 Hz, 2H, piceno-H), 8.79 (d, J=4.8 Hz, 2H, b-pyrrolic
H), 9.03 (d, J=8.0 Hz, 2H, piceno-H), 9.32 ppm (d, J=4.8 Hz, 2H, b-pyr-
rolic H); 13C NMR (CDCl3): d=88.63, 96.12, 106.92, 121.31, 123.12,
124.24, 125.08, 125.63, 126.93, 128.36, 129.23, 129.42, 129.85, 132.70,
133.38, 136.60, 138.52, 138.60, 140.04, 145.47 ppm; IR (KBr): n˜ =574, 627,
662, 698, 762, 790, 820, 840, 879, 985, 1000, 1018, 1077, 1135, 1249, 1295,
1378, 1434, 1505, 2110, 3306 cmꢀ1; UV/Vis (CH2Cl2): lmax (e)=472.0
(8.08), 587.0 (0.44), 635.0 nm (7.76ꢂ10ꢀ4 mꢀ1 cmꢀ1); MALDI-TOF-MS:
m/z: 765 [M+H]+, 764 [M]+; HRMS (internally calibrated MALDI-
TOF-MS): calcd for C52H26N4Ni [M]+: 764.1511; found: 764.1479.
Acknowledgements
The generous support of the National Science Foundation (CHE-
0956447, 0645381, 1001589), the Research Corporation (Cottrell and
Scialog Awards to MHB) is gratefully acknowledged. LJKB thanks the
Graduate Assistance in Areas of National Need program, Indiana Uni-
versity College of Arts and Sciences and Department of Chemistry fel-
lowship programs, and Richard L. Lord for many helpful discussions.
[1] a) R. Bonnett, B. D. Djelal, A. Nguyen, J. Porphyrins Phthalocya-
nines 2001, 5, 652; b) D. E. J. G. J. Dolmans, D. Fukumura, R. K.
Jain, Nat. Rev. Cancer 2003, 3, 380; c) R. Bonnett, G. Martinez, Tet-
rahedron 2001, 57, 9513.
[2] a) S. G. DiMagno, V. S. Y. Lin, M. J. Therien, J. Org. Chem. 1993, 58,
5983; b) H. T. Uyeda, Y. Zhao, K. Wostyn, I. Asselberghs, K. Clays,
A. Persoons, M. J. Therien, J. Am. Chem. Soc. 2002, 124, 13806;
c) Y.-P. An, Z. Yang, M. A. Ratner, J. Chem. Phys. 2011, 135,
044706/1; d) N. Bennett, G. Xu, L. J. Esdaile, H. L. Anderson, J. E.
MacDonald, M. Elliott, Small 2010, 6, 2604; e) R. Charvet, K.
Ariga, J. P. Hill, Q. Ji, A. H. Khan, S. Acharya, Chem. Commun.
2011, 47, 6825; f) V. Gandhi, M. L. Thompson, T. D. Lash, Tetrahe-
dron 2010, 66, 1787; g) M. Koepf, J. Conradt, J. Szmytkowski, J. A.
Wytko, L. Allouche, H. Kalt, T. S. Balaban, J. Weiss, Inorg. Chem.
2011, 50, 6073; h) F. Malvolti, P. Le Maux, L. Toupet, M. E. Smith,
W. Y. Man, P. J. Low, E. Galardon, G. Simonneaux, F. Paul, Inorg.
Chem. 2010, 49, 9101; i) G. Sedghi, V. M. Garcia-Suarez, L. J. Es-
daile, H. L. Anderson, C. J. Lambert, S. Martin, D. Bethell, S. J. Hig-
gins, M. Elliott, N. Bennett, J. E. Macdonald, R. J. Nichols, Nature
Nanotechnology 2011, 6, 517.
[3] a) A. Osuka, H. Shimidzu, Angew. Chem. 1997, 109, 93; Angew.
Chem. Int. Ed. Engl. 1997, 36, 135; b) J. D. Spence, T. D. Lash, J.
Org. Chem. 2000, 65, 1530; c) A. K. Sahoo, S. Mori, H. Shinokubo,
A. Osuka, Angew. Chem. 2006, 118, 8140; Angew. Chem. Int. Ed.
2006, 45, 7972.
[4] a) A. N. Cammidge, P. J. Scaife, G. Berber, D. L. Hughes, Org. Lett.
2005, 7, 3413; b) K. Kurotobi, K. S. Kim, S. B. Noh, D. Kim, A.
Osuka, Angew. Chem. 2006, 118, 4048; Angew. Chem. Int. Ed. 2006,
45, 3944; c) N. K. S. Davis, M. Pawlicki, H. L. Anderson, Org. Lett.
2008, 10, 3945; d) S. Hayashi, M. Tanaka, H. Hayashi, S. Eu, T.
Umeyama, Y. Matano, Y. Araki, H. Imahori, J. Phys. Chem. C 2008,
112, 15576.
Thermal cyclization of 3c: A solution of 3c (20 mg, 0.046 mmol) in tolu-
ene (15 mL), was stirred at 1108C in a pressure tube for 2 h. Upon com-
pletion of the reaction, the solvent was removed under reduced pressure.
The crude compound was purified by silica-gel column chromatography
to afford 8c in trace yields. MALDI-TOF-MS: m/z: 773 [M+H]+.
CCDC-825981 (H8a), 825982 (Zn2c), 825983 (H3a), and 825984 (Ni2b)
contain the supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge Crystallograph-
ic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Computational details: All calculations were carried out with DFT[31] as
implemented in the Jaguar 7.0 suite[32] of quantum chemistry programs.
The spin-unrestricted (UDFT) approach was used in all cases. Geometry
optimizations were performed with the PWPW91[33] and the 6-31G**
basis sets with Ni and Zn represented by using the Los Alamos LACVP
basis that includes relativistic, effective, core potential.[34] The energies
were reevaluated by additional single-point calculations at each opti-
mized geometry with Dunningꢃs[35] correlation consistent triple-z basis set
cc-pVTZ(-f) with the standard double set of polarization functions. For
metals, we used a modified version of LACVP, designed as LACV3P, in
which the exponents were decontracted to match the effective core po-
tential with triple-z quality. Vibrational-frequency calculations were done
at the PWPW91/6-31G** level of theory to confirm proper convergence
to minima and saddle points for equilibrium and transition-state geome-
tries, respectively, and to derive the zero-point-energy (ZPE) and entropy
corrections at room temperature. Solvation energies were evaluated by a
self-consistent reaction field (SCRF)[36] approach based on accurate nu-
merical solutions of the Poisson–Boltzmann equation.[37] Solvation calcu-
lations were carried out at the gas-phase geometry by using 6-31G**/
PWPW91 theory and employing a dielectric constant of e=2.13 for tolu-
ene. To obtain open-shell, broken-symmetry, singlet solutions is difficult
and manual adjustments to the initial-guess wave function were often
[5] a) Y. Song, R. E. Haddad, S.-L. Jia, S. Hok, M. M. Olmstead, D. J.
Nurco, N. E. Schore, J. Zhang, J.-G. Ma, K. M. Smith, S. Gazeau, J.
Pecaut, J.-C. Marchon, C. J. Medforth, J. A. Shelnutt, J. Am. Chem.
Soc. 2005, 127, 1179; b) C. J. Medforth, M. O. Senge, K. M. Smith,
L. D. Sparks, J. A. Shelnutt, J. Am. Chem. Soc. 1992, 114, 9859;
Chem. Eur. J. 2011, 17, 14539 – 14551
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14549

J. M. Zaleski, M.-H. Baik et al.
c) J. A. Shelnutt, X.-Z. Song, J.-G. Ma, S.-L. Jia, W. Jentzen, C. J.
Medforth, Chem. Soc. Rev. 1998, 27, 31; d) R. E. Haddad, S.
Gazeau, J. Pecaut, J.-C. Marchon, C. J. Medforth, J. A. Shelnutt, J.
Am. Chem. Soc. 2003, 125, 1253; e) S. G. DiMagno, A. K. Wertsch-
ing, C. R. Ross II, J. Am. Chem. Soc. 1995, 117, 8279; f) L.-M. Jin,
L. Chen, J.-J. Yin, J.-M. Zhou, C.-C. Guo, Q.-Y. Chen, J. Org. Chem.
2006, 71, 527; g) M.-S. Liao, T. Kar, S. M. Gorun, S. Scheiner, Inorg.
Chem. 2004, 43, 7151; h) J. Mack, Y. Asano, N. Kobayashi, M. J.
Stillman, J. Am. Chem. Soc. 2005, 127, 17697; i) K. E. Thomas, I. H.
Wasbotten, A. Ghosh, Inorg. Chem. 2008, 47, 10469.
45, 4136; c) M. O. Senge, T. Ema, K. M. Smith, J. Chem. Soc. Chem.
Commun. 1995, 733.
[23] M. Gouterman in The Porphyrins, Vol. 3 (Ed.: D. Dolphin), Aca-
demic Press, New York, 1978.
[24] a) M. Gouterman, G. Wagniere, L. C. Snyder, J. Mol. Spectrosc.
1963, 11, 108; b) E. J. Baerends, G. Ricciardi, A. Rosa, S. J. A. van
Gisbergen, Coord. Chem. Rev. 2002, 230, 5.
[25] Porphyrins that underwent cyclization required corrections for mass
balance, due to the fact that the cyclization of each enediyne results
in a net loss of two H atoms. Also taken under consideration was
the source of the H atoms and the overall entropy gain of the reac-
tions therein. Corrections were made as suggested by Cramer (C. J.
Cramer, J. Am. Chem. Soc. 1998, 120, 6261), and were performed as
follows. To energies of the starting enediyne 3 were added the calcu-
lated energies of two equivalents of CHD, a common H donor in
Bergman cyclization reactions. This was repeated for compounds 3-
TS through 6. Compound 7 was corrected with one equivalent of
CHD and one equivalent of benzene, and compound 8 was correct-
ed with one equivalent each of CHD, benzene, and H2 molecules.
Compounds 8-TS through 11 were also corrected this way. Com-
pound 12 was corrected with two equivalents of benzene and one
equivalent H2, whereas 13 was corrected with two equivalents each
of benzene and H2 molecules. This assured that the relative energies
of all compounds could be directly compared to each other, because
all had the same number of net atoms.
[26] Although solid-state DSC temperatures (ca. 1508C) and cyclization
reaction conditions in solution (80–1558C) seem incongruent with
the initial Bergman cyclization barrier of 17.4 kcalmolꢀ1, reaction of
3a at 408C for 120 h generates 50% conversion and 10% isolated
yield of product 8a; this indicates that the calculated reaction barri-
er is indeed markedly lower than these optimal reaction tempera-
tures imply.
[27] Experimental thermal Bergman cyclization of 3 to 8 is typically per-
formed anaerobically, with isolation and purification conducted aer-
obically. Under these conditions, 8 is presumed to be formed via tet-
rahydro species 7, which subsequently converts to 8 either through
loss of H2 (anerobically) or oxidation by O2 (aerobically). Thermo-
dynamically, loss of 2H2 is downhill by 45 kcalmolꢀ1, and control re-
actions with anaerobic analyses illustrate distinct formation of a pi-
cenoporphyrin product. Additionally, computation of the reaction of
7+O2!8+2H2O is downhill by about 145 kcalmolꢀ1. Whereas oxi-
dation-assisted formation of the picenoporphyrin product is clearly
favored and has been demonstrated at ambient temperature by
using external addition of 2,3-dichloro-5,6-dicyano-1,4-benzoquinone
(DDQ; M. Nath, J. C. Huffman, J. M. Zaleski, Chem. Commun.
2003, 858a), the H2 evolution pathway is the most fundamental of
viable reaction trajectories.
[6] a) H. Imahori, T. Umeyama, S. Ito, Acc. Chem. Res. 2009, 42, 1809;
b) N. K. S. Davis, A. L. Thompson, H. L. Anderson, Org. Lett. 2010,
12, 2124; c) N. K. S. Davis, A. L. Thompson, H. L. Anderson, J. Am.
Chem. Soc. 2011, 133, 30.
[7] H. Aihara, L. Jaquinod, D. J. Nurco, K. M. Smith, Angew. Chem.
2001, 113, 3547; Angew. Chem. Int. Ed. 2001, 40, 3439.
[8] a) R. R. Jones, R. G. Bergman, J. Am. Chem. Soc. 1972, 94, 660;
b) R. G. Bergman, Acc. Chem. Res. 1973, 6, 25; c) B. H. Long, J.
Golik, S. Forenza, B. Ward, R. Rehfuss, J. C. Dabrowiak, J. J. Catino,
S. T. Musial, K. W. Brookshire, T. W. Doyle, Proc. Natl. Acad. Sci.
USA 1989, 86, 2.
[9] a) M. Nath, J. C. Huffman, J. M. Zaleski, J. Am. Chem. Soc. 2003,
125, 11484; b) M. Nath, M. Pink, J. M. Zaleski, J. Am. Chem. Soc.
2005, 127, 478.
[10] a) M. Prall, A. Wittkopp, A. A. Fokin, P. R. Schreiner, J. Comput.
Chem. 2001, 22, 1605; b) P. R. Schreiner, J. Am. Chem. Soc. 1998,
120, 4184.
[11] J. D. Spence, E. D. Cline, D. M. Llagostera, P. S. OꢃToole, Chem.
Commun. 2004, 180.
[12] a) H.-K. Chang, S. Datta, A. Das, A. Odedra, R.-S. Liu, Angew.
Chem. 2007, 119, 4828; Angew. Chem. Int. Ed. 2007, 46, 4744; b) X.
Han, Y. Zhang, K. K. Wang, J. Org. Chem. 2005, 70, 2406; c) P. R.
Schreiner, M. Prall, V. Lutz, Angew. Chem. 2003, 115, 5935; Angew.
Chem. Int. Ed. 2003, 42, 5757; d) Y. Yamamoto, A. Nagata, H.
Nagata, Y. Ando, Y. Arikawa, K. Tatsumi, K. Itoh, Chem. Eur. J.
2003, 9, 2469; e) I. V. Alabugin, K. Gilmore, S. Patil, M. Manoharan,
S. V. Kovalenko, R. J. Clark, I. Ghiviriga, J. Am. Chem. Soc. 2008,
130, 11535.
[13] T. Chandra, B. J. Kraft, J. C. Huffman, J. M. Zaleski, Inorg. Chem.
2003, 42, 5158.
[14] a) D. W. Smith, Jr., H. V. Shah, K. P. U. Perera, M. W. Perpall, D. A.
Babb, S. J. Martin, Adv. Funct. Mater. 2007, 17, 1237; b) X. Cheng, J.
Ma, J. Zhi, X. Yang, A. Hu, Macromolecules 2010, 43, 909.
[15] A. D. Adler, F. R. Longo, J. D. Finarelli, J. Goldmacher, J. Assour,
L. Korsakoff, J. Org. Chem. 1967, 32, 476.
[16] M. J. Crossley, P. L. Burn, S. S. Chew, F. B. Cuttance, I. A. Newsom,
J. Chem. Soc. Chem. Commun. 1991, 1564.
[17] a) B. P. Warner, S. P. Millar, R. D. Broene, S. L. Buchwald, Science
1995, 269, 814; b) K. C. Nicolaou, Y. P. Hong, W. M. Dai, Z. J. Zeng,
W. Wrasidlo, J. Chem. Soc. Chem. Commun. 1992, 1542.
[18] a) M. Prall, A. Wittkopp, P. R. Schreiner, J. Phys. Chem. A 2001,
105, 9265; b) M. Nath, J. C. Huffman, J. M. Zaleski, Chem.
Commun. 2003, 858.
[28] C. J. Cramer, J. Am. Chem. Soc. 1998, 120, 6261.
[29] I. Alabugin, B. Breiner, M. Manoharan, Adv. Phys. Org. Chem.
2007, 42, 1.
[30] G. Stork, A. Brizzolara, H. Landesman, J. Szmuszkovicz, R. Terrell,
J. Am. Chem. Soc. 1963, 85, 207.
[31] a) R. G. Parr, W. Yang, Density-Functional Theory of Atoms and
Molecules, Oxford University Press, New York, 1989; b) T. Ziegler,
Chem. Rev. 1991, 91, 651.
[32] Jaguar 7.0, Schrçdinger, LLC, New York, NY, 2007.
[33] a) J. C. Slater, Quantum Theory of Molecules and Solids, Vol. 4: The
Self-Consistent Field for Molecules and Solids, McGraw-Hill, New
York, 1974; b) J. P. Perdew, Electronic Structure of Solids, Akademie
Verlag, Berlin, 1991; c) J. P. Perdew, Y. Wang, Phys. Rev. B 1992, 45,
13244.
[19] J. D. Rule, S. R. Wilson, J. S. Moore, J. Am. Chem. Soc. 2003, 125,
12992.
[20] a) L. D. Sparks, C. J. Medforth, M. S. Park, J. R. Chamberlain, M. R.
Ondrias, M. O. Senge, K. M. Smith, J. A. Shelnutt, J. Am. Chem.
Soc. 1993, 115, 581; b) M. O. Senge, Chem. Commun. 2006, 243.
[21] a) K. M. Barkigia, M. W. Renner, L. R. Furenlid, C. J. Medforth,
K. M. Smith, J. Fajer, J. Am. Chem. Soc. 1993, 115, 3627; b) W. W.
Kalisch, M. O. Senge, Angew. Chem. 1998, 110, 1156; Angew. Chem.
Int. Ed. 1998, 37, 1107; c) D. J. Nurco, C. J. Medforth, T. P. Forsyth,
M. M. Olmstead, K. M. Smith, J. Am. Chem. Soc. 1996, 118, 10918;
d) M. O. Senge, I. Bischoff, Eur. J. Org. Chem. 2001, 1735; e) M. O.
Senge, W. W. Kalisch, Inorg. Chem. 1997, 36, 6103; f) M. O. Senge,
M. W. Renner, W. W. Kalisch, J. Fajer, J. Chem. Soc., Dalton Trans.
2000, 381.
[34] P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299.
[35] T. H. Dunning, J. Chem. Phys. 1989, 90, 1007.
[36] a) J. Tomasi, M. Persico, Chem. Rev. 1994, 94, 2027; b) C. J. Cramer,
D. G. Truhlar, Structure and Reactivity in Aqueous Solution, ACS
Symposium Series 568, American Chemical Society, Washington,
D. C. , 1994; c) C. J. Cramer, D. G. Truhlar, Chem. Rev. 1999, 99,
2161.
[22] a) P. Bhyrappa, C. Arunkumar, B. Varghese, Inorg. Chem. 2009, 48,
3954; b) P. Bhyrappa, M. Sankar, B. Varghese, Inorg. Chem. 2006,
14550
www.chemeurj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14539 – 14551

Extended, p-Conjugated, Light-Harvesting Macrocycles
FULL PAPER
[37] a) D. J. Tannor, B. Marten, R. Murphy, R. A. Friesner, D. Sitkoff, A.
Nicholls, B. Honig, M. Ringnalda, W. A. Goddard, III, J. Am.
Chem. Soc. 1994, 116, 11875; b) B. Marten, K. Kim, C. Cortis, R. A.
Friesner, R. B. Murphy, M. N. Ringnalda, D. Sitkoff, B. Honig, J.
Phys. Chem. 1996, 100, 11775.
Received: August 10, 2011
Published online: November 14, 2011
Chem. Eur. J. 2011, 17, 14539 – 14551
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14551