Phlorins bearing different substituents at the sp3-hybridized meso-position

Article abstract of DOI:10.1021/jo5008256

Phlorins bearing different substituents at the sp³-hybridized meso-position were investigated. The extent to which different substituents at this unique position can influence phlorin spectroscopic properties, structure, and stability is of int

Full text of DOI:10.1021/jo5008256

Article
pubs.acs.org/joc
Phlorins Bearing Different Substituents at the sp3-Hybridized Meso-
Position
Alexandra M. Bruce,† Emily S. Weyburne,† James T. Engle,‡ Christopher J. Ziegler,*,‡
and G. Richard Geier, III*,†
†Colgate University, Department of Chemistry, 13 Oak Drive, Hamilton, New York 13346, United States
‡Department of Chemistry, University of Akron, Akron, Ohio 44325, United States
S
* Supporting Information
ABSTRACT: Phlorins bearing different substituents at the sp3-hybridized meso-position were investigated. The extent to which
different substituents at this unique position can influence phlorin spectroscopic properties, structure, and stability is of interest
given that such substituents are not in direct conjugation with the phlorin macrocycle. While the effect of various substituents at
the sp2-hybridized positions has been the subject of prior investigations, the impact of different substituents at the saturated
carbon atom has not been systematically examined. In this study, phlorins with different combinations of geminal methyl and
phenyl substituents were prepared in yields of 24−49% via dipyrromethane + dipyrromethanedicarbinol routes, and their NMR
spectra, UV−vis spectra, X-ray crystal structures, and stability toward light and air were compared. The nature of the substituents
at the sp3-hybridized position was found to impact spectroscopic properties, structure, and stability to varying degrees. Thus, the
choice of substituents at the sp3-hybridized meso-position provides a further option for altering phlorin properties.
anions.4−6 Phlorins bearing a hydrogen atom at the sp3-
INTRODUCTION
■
hybridized position are tautomeric to chlorins.7
Phlorin is a structurally interesting member of the porphyrinoid
First identified by Woodward in the 1960s,8 phlorins have
family. In addition to the typical sp2-hybridized meso-carbon
since been reported in a wide range of contexts.1,2 Yet despite
atoms found in porphyrin, phlorin also possesses an sp3-
past and present interest in phlorins, many phlorins suffer from
hybridized meso-position. Thus, phlorin is an example of a
poor stability toward light and air, rendering their preparation,
calixphyrina hybrid family of molecules with structures
purification, and study difficult. Phlorins bearing a hydrogen
intermediate to porphyrin and calixpyrrole (porphyrinogen).1,2
atom at the sp3-hybridized meso-position can readily undergo
Additionally, the structure of phlorin is complementary to the
oxidation to afford an aromatic porphyrin. Phlorins geminally
well-known corrole system.3 While phlorin has a similar core
size to porphyrin (i.e., all four meso-positions are present), the
central core environment is reminiscent of corrole (i.e., the
presence of three N−H groups). Unlike either porphyrin or
corrole, the tetrahedral geometry of the sp3-hybridized carbon
atom affords a nonplanar macrocycle structure.
disubstituted at the sp3-hybridized position are susceptible to
rapid oxidation leading to ring-opened products.9 Phlorin
stability can be enhanced by the incorporation of substituents
on one or more of the core nitrogen atoms.10−12 However,
substitution in this way radically alters the structure of the
phlorin and its coordination environment.
Several years ago, we began investigating the potential to
enhance phlorin stability by the judicious incorporation of
substituents at the oxidatively prone sp2-hybridized meso-
positions. The preparation and study of phlorins bearing
Phlorins also display interesting properties. The presence of
the sp3-hybridized carbon atom interrupts the main macrocycle
conjugation pathway. As a result, the electronic properties of
phlorin differ from porphyrin. Phlorins display a longer
wavelength absorption than is typical of porphyrins. Phlorins
can be oxidized by up to three electrons at modest potentials.4
Upon protonation, phlorins have been shown to bind
Received: April 12, 2014
Published: June 11, 2014
© 2014 American Chemical Society
5664
dx.doi.org/10.1021/jo5008256 | J. Org. Chem. 2014, 79, 5664−5672

The Journal of Organic Chemistry
Article
Scheme 1. Preparation of Dipyrromethanes 1a−c and
Diacyldipyrromethanes 2a−c
sterically bulky mesityl13 and electron-withdrawing pentafluoro-
phenyl14 substituents revealed that phlorin stability could be
increased and practical stability could be achieved. 5,5-
Dimethyl-10,15,20-tris(pentafluorophenyl)phlorin 3a prepared
in the course of these efforts is one of the most stable free-base
phlorins reported.14 Rosenthal and co-workers have utilized this
phlorin in their independent studies of phlorin spectroscopy,
structure, electrochemistry, and anion binding.4 Recently,
Rosenthal and co-workers prepared and studied a series of
five phlorins based on phlorin 3a but bearing different aryl
substituents at the 15-position.15 The nature of the substituent
at this sp2-hybridized position was found to impact phlorin
spectroscopic and electrochemical properties. However, a
comparison of the stability of the various phlorins was not
reported. Taken together, it is clear that the nature of the
substituents at the sp2-hybridized meso-positions impacts
phlorin properties and provides a means for tuning phlorin
properties.
Since establishing that phlorin stability can be enhanced by
the choice of substituents at the peripheral sp2-hybridized
meso-positions, we have turned our attention to substituents at
the sp3-hybridized position (the 5-position). The potential
impact of different substituents at this location is less obvious as
these substituents are not in direct conjugation with the
macrocycle. To the best of our knowledge, a comparative study
of phlorins differently substituted at only the 5-position has not
been reported. Thus, we sought to determine whether the
introduction of one or two aryl substituents at the sp3-
hybridized meso-position (altering steric size and electronic
environment) would impact phlorin spectroscopic properties,
structure, and stability. Herein, we report the preparation,
spectroscopic characterization (NMR and UV−vis), X-ray
structural characterization, and assessment of stability of three
different phlorins. Each phlorin has pentafluorophenyl groups
at the sp2-hybridized meso-positions and bears either two
methyl substituents (3a), a methyl and a phenyl substituent
(3b), or two phenyl substituents (3c) at the sp3-hybridized
meso-position.
prepared from the reaction of pyrrole with an appropriate
ketone according to the literature (Scheme 1). 5,5-Dimethyldi-
pyrromethane 1a was prepared using a method reported by
Lindsey and co-workers.16 Dipyrromethanes 1b17 and 1c18
were prepared following intriguing methods from Gambarotta
and co-workers that utilized ethanol as solvent and
methanesulfonic acid as the acid catalyst. We found the
preparation of 1c to be particularly convenient as analytically
pure dipyrromethane (5.1 g, 24%) crystallized directly from the
crude reaction mixture as reported in the literature.18 5-
Pentafluorophenyldipyrromethane 1d, required for the prep-
aration of all three phlorins, was obtained following a procedure
reported by Lindsey and co-workers.19
Diacyldipyrromethanes 2a−c were prepared from dipyrro-
methanes 1a−c in accordance with a general procedure
reported for the preparation of other diacyldipyrromethanes
(Scheme 1).20 The purification of 2a was refined from our
earlier publication.14 We found that the final crystallization step
takes place with better reproducibility when the crude reaction
mixture is first partially purified via a short silica gel column.
With this modification, we obtained 0.59 g (21% yield)
compared to a variable 0.37−0.83 g (13−29% yield) from our
prior method. The purification of 2b was hampered by an
inability to identify suitable conditions for crystallization.
Instead, sufficient purity was obtained via two silica gel
columns using different solvent systems. Evaporation of solvent
RESULTS AND DISCUSSION
Preparation of Dipyrromethanes and Diacyldipyrro-
methanes. 5,5-Disubstituted dipyrromethanes 1a−c were
■
5665
dx.doi.org/10.1021/jo5008256 | J. Org. Chem. 2014, 79, 5664−5672

The Journal of Organic Chemistry
Article
Scheme 2. Preparation of Phlorins 3a−c
afforded a yellow foam that could be ground to an amorphous
yellow powder (1.33 g, 35%). Modification of the general
diacylation procedure was required for the preparation of 2c
due to poor solubility of 1c in toluene. The reaction proceeded
with better yield if 1c was first dissolved in the THF solution of
ethylmagnesium bromide, followed by dilution in toluene
immediately prior to the addition of pentafluorobenzoyl
chloride. After purification by chromatography and crystal-
lization, 2c was obtained in a yield similar to that of the other
diacyldipyrromethanes (0.91 g, 22%).
could be determined, but not which resonance corresponds to
which proton. 13C NMR spectra were also recorded for each
phlorin in CDCl3 (see the Supporting Information). Poor
solubility of phlorin 3c in DMSO-d6 prevented comparison of
13C NMR spectra in that solvent.
Comparison of 1H NMR spectra recorded in CDCl3 is
shown in Figure 1. The presence of different substituents at the
sp3-hybridized meso-position impacted the chemical shifts of
the β-pyrrole protons of the phlorin ring. As expected, the
equivalent protons on the adjacent 3 and 7 positions were the
most affected. These protons produced a signal at 6.78 ppm in
3a. Replacement of a methyl group with a phenyl substituent in
3b caused an upfield shift of 0.5 ppm (6.28 ppm). Replacement
of both methyl groups with phenyl substituents in 3c gave rise
to a signal at an intermediate position of 6.48 ppm. Protons at
positions 2 and 8 were less altered, appearing over a narrower
range of 6.78 to 6.95 ppm. Interestingly, the positions of the
signals due to the β-protons on the other pyrrole rings were
also affected, progressively shifting upfield from 3a to 3b to 3c
(7.36 and 7.12 ppm; 7.29 and 7.03 ppm; and 7.06 and 6.79
ppm, respectively). Thus, different substituents at the sp3-
hybridized meso-position can impact the environment of
protons at some distance. Assuming similar assignments for
the 1H NMR spectra reported for four phlorins bearing
different substituents at the 15-position,15 altering substituents
at the 5-position had a greater effect on the NMR spectra.
Across the family of phlorins bearing electron-withdrawing,
electron-donating, and sterically bulky aryl substituents at the
15-position, in direct conjugation with the phlorin ring, H3/7
ranged from 6.77 to 6.84 ppm (0.07 ppm), H2/8 from 6.83 to
6.99 ppm (0.16 ppm), and H12/13/18/17 from 7.01 to 7.27 ppm
(0.26 ppm) and 7.16 to 7.44 ppm (0.28 ppm). All of these
ranges are smaller than observed across the series of phlorins
3a−c.
Preparation of Phlorins. Phlorins 3a−c were prepared in
accordance with our published methodology for the prepara-
tion of phlorin 3a (Scheme 2).14 Dipyrromethanedicarbinol
species 2-OH were prepared immediately before use by NaBH4
reduction of 2a−c and used without purification as reported in
the literature.20 Phlorins were prepared from the reaction of 2-
OH with (5-pentafluorophenyl)dipyrromethane 1d on a 0.500
mmol scale at room temperature in CH2Cl2 with TFA catalysis
(100 mM) followed by oxidation with DDQ. Each phlorin was
isolated from polar and insoluble byproducts by passage of the
crude reaction mixture through a silica pad, washing with
CH2Cl2. Satisfactory purification of each phlorin could be
achieved by passage through a neutral alumina column
[CH2Cl2/hexanes (1:4)] followed by crystallization from
CH2Cl2/hexaneswithout need for additional silica gel
chromatography as employed in our prior work.14 Each phlorin
showed similar behavior during chromatography and crystal-
lization procedures. Methodology originally developed for the
preparation and purification of 3a was not as efficient when
applied to 3b and 3c (Scheme 2), showing that subtle changes
in the nature of substituents can affect product yield in difficult
to predict ways.20,21 Nonetheless, the isolated yields of phlorins
2b and 2c are typical of syntheses of porphyrinoids via
dipyrromethanedicarbinol routes.20,22 Each phlorin was readily
obtained in greater than 100 mg quantity.
1H NMR spectra were also recorded in DMSO-d6 as we had
earlier observed sharper, distinct signals from the core N−H
protons in this solvent.14 Different substituents at the sp3-
hybridized meso-position again impacted the NMR spectra in
interesting ways (Figure 2). Protons at positions 3 and 7 shifted
progressively upfield from 3a to 3b to 3c (6.73, 6.27, and 6.13
ppm, respectively), a range of 0.60 ppm. Once again, H2/8 were
less impacted ranging from 7.09 to 6.96 ppm, whereas
H12/13/18/17 were shifted upfield in 3c relative to 3a,b by
1
NMR Analyses. H NMR spectra were recorded for each
phlorin. The spectra were partially assigned with the aid of
1H−1H COSY and NOESY spectra (see the Supporting
Information for a discussion of the peak assignments).
Confident assignments could be made for most peaks except
for those arising from the two nonequivalent pairs of protons
on the pyrrole rings furthest from the 5-position (i.e., H12/18
and H13/17). The two signals that arise from these protons
5666
dx.doi.org/10.1021/jo5008256 | J. Org. Chem. 2014, 79, 5664−5672

The Journal of Organic Chemistry
Article
Figure 2. 1H NMR spectra of phlorins 3a−c showing an expansion of
the aromatic region (DMSO-d6). The numbering system is shown in
Figure 1. The peaks indicated with an asterisk (*) correspond to
protons H12/H18 and H13/H17. However, the precise assignment could
not be made for these two peaks.
Figure 1. 1H NMR spectra of phlorins 3a−c showing an expansion of
the aromatic region (CDCl3). The numbering system is illustrated for
phlorin 3b. The peaks indicated with an asterisk (*) correspond to
protons H12/H18 and H13/H17. However, the precise assignment could
not be made for these two peaks.
∼0.25 ppm for both signals. The impact of different
substituents at the 5-position on the chemical shift of the N−
H peaks was particularly pronounced. Proton NH23 was found
at 6.10 and 6.11 pm for 3a and 3b respectively, but appeared
much farther downfield at 7.69 ppm for 3c. A similar downfield
shift was observed for NH21/22. The signal appeared at 6.89 and
6.98 for 3a and 3b, respectively, but at 8.38 ppm for 3c. Again,
different substituents at the sp3-hybridized meso-position
impact the environment of protons on the phlorin ring, even
those at greater distance from the 5-position.
Figure 3. UV−vis spectra of dimethylphlorin 3a, methylphenylphlorin
3b, and diphenylphlorin 3c in CHCl3.
UV−vis Analyses. Absorption spectra for phlorins 3a−c
were measured in CHCl3 (Figure 3) and toluene (see the
Supporting Information). The spectra for phlorins 3a and 3b
were nearly identical. The primary point of difference was that
the longer wavelength band for 3b was slightly blue-shifted
relative to that of 3a (647 and 653 nm, respectively, in CHCl3;
651 and 654, respectively, in toluene). Spectra for phlorin 3c
differed more sharply. The longer wavelength band was red-
shifted by ∼20 nm to 667 and 674 nm in CHCl3 and toluene,
respectively. The molar absorptivity was also diminished by
∼20% (∼21000 M−1 L−1 in both solvents) relative to 3a,b
(∼27000 M−1 L−1 for both compounds in both solvents). The
effect on the absorption spectra of different substituents at the
sp3-hybridized meso-position is similar to that reported for the
family of phlorins bearing different substituents at the 15-
position.15 In that report, the position of the longer wavelength
band varied from 655 to 670 nm, a range of 15 nm. Thus,
altering substituents at the 5-position provides another option
for tuning absorption properties.
Crystallographic Analyses. Phlorins 3a−c were charac-
terized by X-ray diffraction (Figure 4). X-ray data collection and
structure parameters are listed in the Supporting Information.
Phlorin 3a was recently elucidated in the monoclinic space
group P21.4 In our study, we isolated a second crystal form in
the orthorhombic space group P212121. The loss of conjugation
caused by the presence of the sp3-hybridized carbon atom in
the phlorin macrocycle results in nonplanar structures for
phlorins 3a−c, in contrast to most aromatic porphyrinoids. In
phlorins 3a−c, the two pyrrole units opposite the sp3-
hybridized carbon atom are largely coplanar. However, the
pyrrole rings immediately adjacent to the 5-position are
puckered out of the macrocycle plane, rising above the plane
5667
dx.doi.org/10.1021/jo5008256 | J. Org. Chem. 2014, 79, 5664−5672

The Journal of Organic Chemistry
Article
Figure 4. Top: structures of phlorins 3a−c with 35% thermal ellipsoids. Nonionizable hydrogen atoms have been omitted for clarity. Bottom: side
views of phlorins 3a−c with 35% thermal ellipsoids. Perfluorophenyl groups and nonionizable hydrogen atoms have been omitted for clarity.
defined by the 11 atoms of the opposite dipyrrole unit. The
substituents on the sp3-hybridized carbon atom do affect the
degree of nonplanarity, but some variability is also observed, as
will be discussed below.
single in character. The pyrrole rings on either side of the sp3-
hybridized carbon atom show electron delocalization in their
carbon−carbon bonding, all indicative of localized aromaticity
in these pyrrole rings. In contrast, the other rings show
alternating double and single bonds with the lengths varying by
more than 0.2−0.3 Å. The lone carbon−nitrogen double bond
of the macrocycle (located on the deprotonated pyrrole ring,
C14−N24) can be identified as it is the shortest carbon−
nitrogen bond within the structure at ∼1.34−1.36 Å. This
factor, plus the electron density on the difference map, led to
assignment of the three ionizable protons to the two pyrroles
immediately adjacent to the sp3-hybridized carbon atom and to
one of the opposite pyrrole units.
The deviation from planarity in phlorins 3a−c is variable and
partially dependent on the identity of the substituents on the
sp3-hybridized carbon atom. As can be seen in the side-on views
in Figure 4, the 5,5-diphenylphlorin 3c shows the greatest
degree of planarity, 5-methyl-5-phenylphlorin 3b is the most
nonplanar, and the new crystal form of 5,5-dimethylphlorin 3a
is intermediate. Phlorin 3b exhibits the largest distances,
∼1.016 and ∼1.017 Å, in spite of the increased steric bulk of
the nearby phenyl ring. This contrasts with the diphenyl variant
3c which has the smallest distances of ∼0.811 and ∼0.812 Å.
However, a comparison of the previously elucidated structure
of 3a to the orthorhombic form indicates that packing forces
play a role in the degree of planarity of the phlorins. In the
orthorhombic form of 3a, the distances measure ∼0.902 and
∼0.923 Å, considerably longer than that seen for the
monoclinic form (∼0.843 and ∼0.801 Å). In spite of some
variability of planarity due to packing forces, the increased steric
bulk of the phenyl substituents in 3c affects the geometry at the
sp3-hybridized carbon atom. The phenyl−carbon−phenyl bond
angle is larger (110.7°) than for an ideal tetrahedron and larger
than the corresponding angles in 3a and 3b (108.5° and 108.4°
respectively). The angle between the adjacent pyrrole rings is
also greater in 3c compared to 3a and 3b (109.4°, 106.7°, and
107.4°, respectively). Taken together, the nature of substituents
at the sp3-hybridized meso-position can impact the structure of
the phlorin.
There are only three structurally elucidated examples of
phlorins in the literature, excluding core modified,23 expanded
core modified,24 expanded,25 or confused26 phlorin systems.
The first example published by Krattinger and Callot,11 bears a
phenyl N-substituent on one of the pyrrole nitrogen atoms
adjacent to the sp3-hybridized meso-position. The bulky N-
substituent causes the pyrrole to invert, such that this phlorin is
not a valid structural comparison to 3a−c. In 2005, Gryko and
co-workers presented the structure of a phlorin serendipitously
isolated from the reaction of (5-pentafluorophenyl)-
dipyrromethane with 4-nitrobenzaldehyde.27 The resultant
phlorin has a dipyrrin and p-nitrophenyl group attached to
the sp3-hybridized meso-position. The macrocycle deviates
from planarity as observed from phlorins 3a−c. The two
pyrrolic nitrogen atoms (N21 and N22) deviate from the plane
of the dipyrrole across from the sp3-hybridized carbon atom,
and the three ionizable hydrogen atoms are assigned to the
pyrrole nitrogen positions immediately adjacent to the sp3-
hybridized carbon position (N21 and N22) and to one of the
opposite pyrroles (N23). Notably, the deviation from planarity
is less than observed from 3a−c as will be discussed below. The
deviation from planarity in Gryko’s phlorin structure can be
assessed by measurement of the distance of the nitrogen atoms
adjacent to the sp3-hybridized carbon atom (N21 and N22) from
the plane defined by the opposite dipyrromethene unit. The
phlorin bearing the dipyrrin unit exhibits deviations of ∼0.850
and ∼0.636 Å for the two out of plane pyrrolic nitrogen atoms.
The most recent phlorin X-ray crystal structure presented in
the literature is for the dimethyl substituted species 3a reported
by Rosenthal and co-workers.4 This phlorin has the same
protonation state and bonding features as seen in the Gryko
structure. This macrocycle is also nonplanar, with nitrogen
deviations from the dipyrromethene unit of ∼0.843 and ∼0.801
Å.
In the structures of 3a−c determined in this study, the sp3
character of the disubstituted meso-carbon atom is readily
observed in the bond lengths to adjacent atoms, all of which are
Phlorin Stability. The stability toward light and air of
phlorins 3a−c in dilute solution was investigated. Similar to our
5668
dx.doi.org/10.1021/jo5008256 | J. Org. Chem. 2014, 79, 5664−5672

The Journal of Organic Chemistry
Article
previous studies of the stability of meso-substituted corroles,21
octaphyrins,28 phlorins bearing mesityl substituents,13 and
phlorin 3a;14 solutions of each phlorin were prepared in a
variety of solvents, and changes in UV−vis spectra were
followed as a function of time of exposure to ambient room
light (i.e., light from conventional overhead fluorescent
lighting). The solutions were monitored closely for 8 h of
exposure, followed by monitoring once a day for a period of 2
weeks. Additionally, solutions of phlorins 3a−c were monitored
by HPLC for a period of 2 weeks of continuous exposure to
light. For each solvent condition, solutions of 3a−c were made
at the same time so that changes in UV−vis spectra or HPLC
peak area were monitored under identical conditions for the
three phlorins.
In our initial experiments, solutions of phlorins 3a−c were
prepared in hexanes, toluene, CH2Cl2, THF, ethyl acetate,
acetone, acetonitrile, and methanol such that the absorbance of
the long wavelength peak (∼650 nm) was initially ∼0.7. As the
molar absorptivity of 3c at that wavelength is ∼20% lower than
that of 3a,b, we also carried out analyses in a subset of solvents
(toluene, CH2Cl2, THF, and methanol) on phlorin solutions of
approximately uniform concentration (∼0.025 mM). The
additional experiments were done to be certain that any
differences in phlorin stability were not due to differences in
initial conditions.
Plots of overlaid UV−vis spectra for phlorins 3a−c in
CH2Cl2 recorded from solutions of uniform initial concen-
tration over a period of 14 days of exposure to light are shown
in Figure 5 (see the Supporting Information for the complete
set of overlaid spectra for solutions of uniform initial
absorbance and concentration for 8 h and 14 d of exposure
to light). The overlaid spectra reveal gradual changes in the
UV−vis spectra for each phlorin, with an apparent faster change
for 3a followed by 3b, with 3c appearing to be the most stable.
Fairly clean isosbestic points are observed for each phlorin.
These qualitative trends are evident across the solvents and
initial conditions (similar absorbance or similar concentration)
investigated (see the Supporting Information). Under all
conditions, phlorin 3a consistently showed the fastest
spectroscopic changes. Phlorin 3c generally showed the slowest
rate of change, with the exception of THF solutions where 3b
showed a slightly more gradual change.
Comparison of changes in the UV−vis spectra as a function
of exposure to light was also made by plotting absorbance as a
function of time of exposure. In our earlier studies,13,14 we
plotted change in absorbance of the peak at ∼650 nm.
However, in this study we observed in some solvents (most
significantly acetone with 3c) that an isosbestic point was
present very close to that peak causing a likely under-reporting
of absorbance change. Thus, in this study we also plotted the
change in absorbance of the primary peak found at ∼430 nm
(see the Supporting Information for additional discussion).
Figure 6 provides illustrative plots following changes at both
peaks for the solutions of phlorins 3a−c that provided the
spectra shown in Figure 5 (see the Supporting Information for
the complete set of plots). The apparent magnitude of the
difference in phlorin stability does depend on which peak is
monitored, with greater differences between 3a−c generally
observed when following the peak at shorter wavelength.
However, the overall detection of differences in stability, and
the general order of stability do not depend on the wavelength
monitored. The initial conditions (similar absorbance or similar
concentration) also did not have a significant impact on the
Figure 5. UV−vis spectra of CH2Cl2 solutions of phlorins 3a−c
recorded after exposure to light and air for 0−14 days. The arrows
indicate the direction of the change.
detection of differences in stability and the general order of
stability. Based on graphical representation of changes in the
UV−vis spectra, phlorin 3c is generally the most stable and 3a
is the least stable of the three phlorins.
To confirm the conclusions drawn from UV−vis analyses,
analogous experiments were carried out with HPLC monitor-
ing. Solutions of phlorins 3a−c were prepared in toluene,
CH2Cl2, and THF, and the area of the phlorin peak was
followed over 14 days of exposure to light. An illustrative result
is shown in the bottom panel of Figure 6. While a slower rate of
decomposition for all three phlorins was observed in the HPLC
experiments, the overall trends are in good agreement with the
UV−vis findings. Phlorin 3c shows the slowest and smallest
decline in peak area over time, whereas 3a shows the greatest
changes. The overall slower rate of change for all phlorins could
stem from the ∼10-fold higher phlorin concentration required
for HPLC monitoring. The solutions used in the HPLC
experiments were noticeably darker in color which likely
impacted the depth of light penetration. Comparison of
methods for assessing phlorin stability and representing the
data as shown in Figures 5 and 6 highlight the importance of
applying multiple approaches to distinguish general trends from
potential anomalies. Both UV−vis and HPLC assessment of
5669
dx.doi.org/10.1021/jo5008256 | J. Org. Chem. 2014, 79, 5664−5672

The Journal of Organic Chemistry
Article
finding provides encouragement toward the preparation and
study of additional phlorins bearing a wider array of
substituents at the sp3-hybridized meso-position.
EXPERIMENTAL SECTION
■
General Experimental Methods. 1H NMR (400 MHz), 13C
NMR (100 MHz), and absorption spectra were collected routinely.
Melting points are uncorrected. Column chromatography was
performed on silica (Merck, 230−400 mesh, 60 Å) or neutral alumina
(Fisher, 80−200 mesh). THF used in the reduction of diacyl
dipyrromethanes was stored over 4-Å Linde molecular sieves. Toluene
used in the synthesis of diacyldipyrromethanes was obtained from a
solvent purification system. All other chemicals are reagent grade and
were used as obtained. Dipyrromethanes 1a,16 1b,17 1c,18 and 1d19
were prepared as described in the literature.
5,5-Dimethyl-1,9-bis(pentafluorobenzoyl)dipyrromethane
(2a). The reaction of 5,5-dimethyldipyrromethane 1a (0.871 g, 5.00
mmol) with ethylmagnesium bromide (25.0 mL, 250 mmol, 1 M in
THF) in dry toluene (100 mL) under argon, followed by addition of a
solution of pentafluorobenzoyl chloride (1.80 mL, 12.5 mmol) in dry
toluene (12.5 mL) was carried out as described in the literature.20
Purification of 2a from the crude reaction mixture was refined from
our earlier publication,14 and was carried out as follows. The crude
reaction mixture was subjected to chromatography [silica gel, CH2Cl2
followed by CH2Cl2/EtOAc (25:1)]). The diacyl dipyrromethane
eluted quickly after the increase in solvent polarity, and afforded a light
yellow solid (774 mg) upon evaporation of the solvent. The diacyl
dipyrromethane was further purified by crystallization from CH2Cl2/
hexanes with gradual evaporation of the CH2Cl2 over ∼1 h at ∼45 °C
yielding a white crystalline solid (21%, 586 mg). Melting point and 1H
NMR analyses were consistent with published values.14 Anal. Calcd for
C25H12F10N2O2: C, 53.40; H, 2.15; N, 4.98. Found: C, 53.45; H, 2.14;
N 5.08.
5-Methyl-5-phenyl-1,9-bis(pentafluorobenzoyl)-
dipyrromethane (2b). Following a general diacylation procedure
from the literature,20 ethylmagnesium bromide (30.0 mL, 300 mmol, 1
M in THF) was added dropwise over 15 min to a solution of 5-methyl-
5-phenyldipyrromethane 1b (1.42 g, 6.00 mmol) in dry toluene (120
mL) under argon. The reaction mixture was stirred for 30 min at room
temperature, and a solution of pentafluorobenzoyl chloride (2.16 mL,
15.0 mmol) in dry toluene (15 mL) was added dropwise over 10 min.
The reaction mixture was stirred for 30 min at room temperature. The
reaction was quenched by addition of a saturated aqueous NH4Cl (90
mL). The organic layer was separated and washed with water (60 mL)
and then brine (60 mL) followed by drying over Na2SO4. The crude
product mixture was concentrated yielding a golden foam. The crude
reaction mixture was subjected to chromatography [silica gel, CH2Cl2/
hexanes (2:1) followed by CH2Cl2]. The diacyldipyrromethane eluted
quickly after the increase in solvent polarity and afforded a light green
foam (1.98 g) upon evaporation of the solvent. Attempts to further
purify 2b by crystallization or precipitation were not successful. Thus,
the sample was subjected to further chromatography [silica gel,
hexanes followed by hexanes/EtOAc (20:1)]. Evaporation of solvent
and drying under vacuum yielded a yellow foam which was ground to
Figure 6. Plots of absorbance (top and middle panels) or HPLC peak
area (bottom panel) as a function of time of exposure to light for
CH2Cl2 solutions of phlorins 3a−c. The absorbance wavelengths
monitored for the top panel were 655 nm (3a), 650 nm (3b), and 674
nm (3c). The absorbance wavelengths monitored for the middle panel
were 439 nm (3a), 436 nm (3b), and 431 nm (3c). The initial
concentration of each phlorin sample for UV−vis monitoring was
approximately the same (∼0.025 mM), and note that the same
solutions of 3a−c were utilized for the two plots. HPLC monitoring
(bottom panel) was performed at 434 nm (3a), 434 nm (3b), and 430
nm (3c). The initial concentration of each phlorin sample for HPLC
monitoring was approximately the same (∼0.30 mM).
phlorin stability revealed differences between the three
phlorins. Thus, substituents at the sp3-hybridized meso-position
can impact phlorin stability, despite not being in direct
conjugation with the macrocycle.
1
an amorphous yellow powder (35%, 1.33 g): H NMR (CDCl3, 400
MHz) δ 9.33 (s, 2H), 7.36 (m, 3H), 7.13 (m, 2H), 6.68 (m, 2H), 6.16
(dd, J = 2.7, 4.0 Hz, 2H), 2.17 (s, 3H); 13C{1H} NMR (CDCl3, 100
MHz) δ 28.2, 45.6, 111.3, 113.7 (t, J = 20 Hz), 121.8, 127.1, 128.1,
129.1, 131.2, 137.6 (d, J = 256 Hz), 142.3 (d, J = 264 Hz), 143.2, 144.0
(d, J = 250 Hz), 146.4, 172.1; HRMS (ESI-TOF) m/z [M + H] calcd
for C30H14F10N2O2H 625.0974, found 625.0975; IR νmax (thin film)/
cm−1 1615.
5,5-Diphenyl-1,9-bis(pentafluorobenzoyl)dipyrromethane
(2c). Ethylmagnesium bromide (30.0 mL, 300 mmol, 1 M in THF)
was added dropwise over 15 min to 5,5-diphenyldipyrromethane 1c
(1.79 g, 6.00 mmol) under argon. [Note: it is important for this
specific reaction that the ethylmagnesium bromide solution is added to
the dipyrromethane prior to adding the toluene solvent as the
dipyrromethane dissolves poorly in toluene leading to a very low level
CONCLUSION
■
Phlorins bearing different substituents at the sp3-hybridized
meso-position were prepared and their NMR spectra, UV−vis
spectra, X-ray crystal structures, and stability toward light and
air were compared. The nature of the substituents at the sp3-
hybridized position were found to impact spectroscopic
properties, structures, and stability of the phlorins. Although
not in direct conjugation with the phlorin macrocycle, the
selection of substituents at the sp3-hybridized position provides
an additional approach for altering phlorin properties. This
5670
dx.doi.org/10.1021/jo5008256 | J. Org. Chem. 2014, 79, 5664−5672

The Journal of Organic Chemistry
Article
of acylation.] The reaction mixture was stirred for 30 min at room
temperature. Dry toluene (120 mL) was added, immediately followed
by the dropwise addition of a solution of pentafluorobenzoyl chloride
(2.16 mL, 15.0 mmol) in dry toluene (15 mL) over 10 min. The
reaction mixture was stirred for 30 min at room temperature. The
reaction was quenched by addition of a saturated aqueous NH4Cl (90
mL). The organic layer was separated and washed with water (60 mL)
and then brine (60 mL) followed by drying over Na2SO4. The crude
product mixture was concentrated, yielding an orange-brown film. The
crude reaction mixture was subjected to chromatography [silica gel,
CH2Cl2/hexanes (1:1) followed by CH2Cl2/hexanes (2:1)]. The
diacyldipyrromethane eluted quickly after the increase in solvent
polarity and afforded a light yellow-green film (1.88 g) upon
evaporation of the solvent. The diacyl dipyrromethane was further
purified by crystallizing twice from CH2Cl2/hexanes with gradual
evaporation of the CH2Cl2 over ∼1 h at ∼50 °C yielding a light yellow
4.7 Hz, 2H); (DMSO-d6, 400 MHz) δ 2.61 (s, 3H), 6.11 (s, 1H), 6.27
(d, J = 7.2 Hz, 2H), 6.93 (t, J = 7.3 Hz, 1H), 6.98 (s, 2H), 7.02 (d, J =
7.8 Hz, 2H), 7.04 (m, 2H), 7.22 (m, 2H), 7.32 (d, J = 5.0 Hz, 2H),
7.61 (d, J = 5.0 Hz, 2H); 13C {1H} NMR (CDCl3, 100 MHz) δ 29.7,
45.3, 91.3, 106.9, 111.0, 121.5, 125.7, 126.6, 126.9, 127.7, 128.7, 131.4,
132.0, 144.3, 144.8, 155.2 (additional weak signals were present at
∼114 ppm and in the region of 135−147 ppm due to carbon atoms of
the C6F5 substituents); HRMS (ESI-TOF) m/z [M + H] calcd for
C45H19F15N4H 901.1443, found 901.1439.
5,5-Diphenyl-10,15,20-tris(pentafluorophenyl)phlorin (3c).
Phlorin 3c was prepared from 2c (343 mg, 0.500 mmol) and purified
as described above for the preparation of 3b. Column chromatography
[alumina, CH2Cl2/hexanes (1:4)] afforded 3c (460 mg) that was
subsequently crystallized from CH2Cl2/hexanes providing dark
purple/green crystals (28%, 135 mg): λabs (toluene, ε × 103) 354
(25.3), 413 (48.1), 431 (49.4), 674 (21.0); λabs (CHCl3, ε × 103) 354
1
1
crystalline solid (22%, 0.911 g): mp 218−219 °C; H NMR (CDCl3,
(25.9), 412 (47.0), 430 (47.8), 667 (20.9); H NMR (CDCl3, 400
400 MHz) δ 9.12 (s, 2H), 7.40 (m, 6H), 7.10 (m, 4H), 6.71 (m, 2H),
6.22 (dd, J = 2.7, 4.0 Hz, 2H); 13C{1H} NMR (CDCl3, 100 MHz) δ
56.7, 113.6 (t, J = 21 Hz), 113.8, 121.3, 128.4, 129.0, 129.1, 131.0,
137.6 (d, J = 256 Hz), 141.9, 142.4 (d, J = 255 Hz), 143.9 (d, J = 252
Hz), 144.4, 172.2; MS (LDI-TOF) m/z [M + H] calcd for
C35H16F10N2O2H 687.11, found 687.07; IR νmax (thin film)/cm−1
1626. Anal. Calcd for C35H16F10N2O2: C, 61.23; H, 2.35; N, 4.08.
Found: C, 61.52; H, 2.38; N 4.12.
MHz) δ 6.48 (d, J = 4.0 Hz, 2H), 6.78 d, J = 4.0 Hz, 2H), 6.79 (d, J =
5.1 Hz, 2H), 6.98 (m, 4H), 7.06 (d, J = 5.0 Hz, 2H), 7.27 (m, 6H);
(DMSO-d6, 400 MHz) δ 6.13 (dd, J = 1.9, 3.8 Hz, 2H), 6.85 (d, J = 6.0
Hz, 4H), 6.96 (dd, J = 1.4, 3.7 Hz, 2H), 7.07 (d, J = 5.1 Hz, 2H),
7.28−7.34 (m, 6H), 7.37 (d, J = 5.0 Hz, 2H), 7.69 (s, 1H), 8.38 (s,
2H); 13C {1H} NMR (CDCl3, 100 MHz) δ 56.4, 91.6, 107.3, 116.0,
120.9, 127.5, 127.8, 128.0, 128.8, 129.7, 131.8, 132.3, 142.8, 144.8,
156.3 (additional weak signals were present at ∼114 ppm and in the
region of 135−147 ppm due to carbon atoms of the C6F5
substituents); HRMS (ESI-TOF) m/z [M + H] calcd for
C50H21F15N4H 963.1599, found 963.1597.
Crystallographic Data. X-ray quality crystals of phlorins 3a−c
were obtained from diffusion of hexane into pyridine solution. Single
crystals of 3a−c were coated in Fomblin oil, mounted on pins, and
placed on a goniometer head under a stream of nitrogen cooled to 100
K. For all samples, the frames were integrated with the Bruker SAINT
software package using a narrow-frame algorithm. For all three
structures, disordered solvent molecules were removed from the
structure via the Platon SQUEEZE program. For 3a, the data were
collected on a Bruker SMART APEX CCD-based X-ray diffractometer
system equipped with a Mo-target X-ray tube (λ = 0.71073 Å)
operated at 2000 W power. The detector was placed at a distance of
5.009 cm from the crystal. Absorption corrections were carried out
using the SADABS program. For 3b and 3c, the data were collected on
an APEX2 CCD diffractometer with Cu Kα radiation (λ = 1.54178).
Data were corrected for absorption effects using the multiscan method
and the structure was solved and refined using the Bruker SHELXTL
Software Package29 until the final anisotropic full-matrix, least-squares
refinement of F2 converged.
5,5-Dimethyl-10,15,20-tris(pentafluorophenyl)phlorin (3a).
The reduction of 2a (281 mg, 0.500 mmol) with NaBH4 (946 mg,
25.0 mmol) in THF/methanol (40 mL, 3:1) yielding 2a-OH,
immediately followed by condensation with 1d (156 mg, 0.500
mmol) in the presence of TFA (1.54 mL, 20.0 mmol) in CH2Cl2 (200
mL), oxidation with DDQ (166 mg, 0.730 mmol), and filtration of the
crude reaction mixture through a pad of silica gel were carried out as
described in the literature.14,20 Purification of 3a from the partially
purified mixture obtained from the silica pad was refined from our
earlier publication,14 and was carried out as follows. The sample was
adsorbed onto neutral alumina (15 g) and purified by chromatography
[alumina, CH2Cl2/hexanes (1:4)] affording 3a (288 mg) that was
subsequently crystallized from CH2Cl2/hexanes with gradual evapo-
ration of the CH2Cl2 over ∼1 h at ∼50 °C providing dark purple/
1
green crystals (49%, 206 mg): H NMR (CDCl3 and DMSO-d6) and
LD-MS analyses were consistent with published values:14 λabs (toluene,
ε × 103) 358 (28.4), 419 (39.8), 438 (45.4), 654 (26.7); λabs (CHCl3),
ε × 103) 358 (29.2), 416 (41.9), 437 (48.4), 653 (27.7); 13C {1H}
NMR (CDCl3, 100 MHz) δ 25.1, 34.8, 91.5, 107.1, 108.3, 121.7, 126.9,
127.6, 131.2, 131.4, 144.2, 154.9 (additional weak signals were present
at ∼114 ppm and in the region of 135−147 ppm due to carbon atoms
of the C6F5 substituents).
Phlorin Stability Experiments (UV−vis, Uniform Absorb-
ance).14 In a darkened laboratory, solutions of phlorins 3a−c were
prepared in hexanes, toluene, CH2Cl2, THF, ethyl acetate, acetone,
acetonitrile, and methanol. The concentration of each solution was
adjusted so that the maximum absorbance of the long wavelength peak
(638−679 nm) was ∼0.7. UV−vis spectra were recorded in the dark,
and then the solutions were exposed to room lights (conventional
overhead fluorescent lighting). Spectra were recorded at 15 min, 30
min, 1 h, 1.5 h, 2 h, 4 h, 6 h, 8 h, 1 d, 1.5 d, 2 d, 3 d, 4 d, 5 d, 6 d, 7 d, 8
d, 9 d, 10 d, 11 d, 12 d, 13 d, and 14 d of continuous exposure to room
lights. Decomposition of phlorins 3a−c was inferred from changes in
the intensity of the peak at 638−679 nm (longer wavelength peak)
and from changes in the intensity of the peak at 426−439 nm (shorter
wavelength peak). Data were normalized with respect to the maximum
absorbance of the monitored peak prior to exposure to light. Control
experiments carried out previously established that phlorin solutions
are stable in the absence of light.13,14
5-Methyl-5-phenyl-10,15,20-tris(pentafluorophenyl)phlorin
(3b). Following a literature procedure,14,20 the reduction of 2b (312
mg, 0.500 mmol) with NaBH4 (946 mg, 25.0 mmol) in THF/
methanol (40 mL, 3:1) afforded the corresponding dicarbinol 2b-OH,
which was used without purification. The reduction reaction was
monitored by TLC [alumina, EtOAc/hexanes (1:3)]. The dicarbinol
was dried under vacuum for 30 min and then immediately subjected to
condensation with 1d (156 mg, 0.500 mmol) in the presence of TFA
(1.54 mL, 20.0 mmol) in CH2Cl2 (200 mL) for 15 min at room
temperature. Oxidation of the reaction mixture was carried out by the
addition of DDQ (166 mg, 0.730 mmol) at room temperature. After 5
min, triethylamine (14 mL, 100 mmol) was added, and the mixture
was stirred at room temperature for a further 30 min. The reaction
mixture was filtered through a pad of silica gel and eluted with CH2Cl2
until the eluent was no longer green. The filtrate was concentrated and
adsorbed onto neutral alumina (15 g) and purified by chromatography
[alumina, CH2Cl2/hexanes (1:4)] affording 3b (223 mg) that was
subsequently crystallized from CH2Cl2/hexanes with gradual evapo-
ration of the CH2Cl2 over ∼1 h at ∼50 °C providing dark purple/
green crystals (24%, 107 mg): λabs (toluene, ε × 103) 357 (27.1), 415
(41.2), 435 (45.0), 651 (25.9); λabs (CHCl3, ε × 103) 357 (27.7), 413
Phlorin Stability Experiments (UV−vis, Uniform Concen-
tration).14 In a darkened laboratory, solutions of phlorins 3a−c were
prepared in toluene, CH2Cl2, THF, and methanol. The concentration
of each solution was ∼0.025 mM. UV−vis spectra were recorded in
the dark, and then the solutions were exposed to room lights, and
monitored as described above.
1
(42.0), 434 (46.3), 647 (26.2); H NMR (CDCl3, 400 MHz) δ 2.67
(s, 3H), 6.28 (m, 2H), 6.95 (m, 3H), 7.03−7.05 (m, 6H), 7.29 (d, J =
5671
dx.doi.org/10.1021/jo5008256 | J. Org. Chem. 2014, 79, 5664−5672

The Journal of Organic Chemistry
Article
Phlorin Stability Experiments (HPLC, Uniform Concentra-
tion).14 In a darkened lab, solutions of phlorins 3a−c were prepared
in toluene, CH2Cl2, and THF. The concentration of each solution was
∼0.30 mM. The samples were analyzed in the dark by HPLC, and
then the solutions were exposed to light (conventional overhead
fluorescent lighting). Chromatograms were recorded at 2 h, 4 h, 8 h, 1
d, 1.5 d, 2 d, 3 d, 4 d, 5 d, 6 d, 7 d, 8 d, 9 d, 12 d, 13 d, and 14 d of
continuous exposure to room lights. Decomposition of phlorins 3a−c
was inferred by a decline in the peak area. HPLC analysis was
performed with an injection volume of 1 μL, a normal-phase silica
column (Alltech, Altima, 4.6 mm × 250 mm), using an isocratic
solvent mixture of 95.5% hexanes and 4.5% acetone. The hexanes
solvent was 50% water saturated by mixing equal quantities of hexanes
and hexanes stored over water. The solvent flow rate was controlled as
follows: T = 0−6 min, 1 mL/min; T = 6−7 min, linear increase to 2
mL/min; T = 7−13 min, 2 mL/min; T = 13−14 min, linear decrease
to 1 mL/min Phlorins 3a−c eluted at 8.5, 7.5, and 7.2 min,
respectively. Detection was performed at 434 nm (3a−b) and 430 nm
(3c).
(12) Krattinger, B.; Callot, H. J. Eur. J. Chem. 1999, 1857−1867.
(13) LeSaulnier, T. D.; Graham, B. W.; Geier, G. R., III. Tetrahedron
Lett. 2005, 46, 5633−5637.
(14) O’Brien, A. Y.; McGann, J. P.; Geier, G. R., III. J. Org. Chem.
2007, 72, 4084−4092.
(15) Pistner, A. J.; Lutterman, D. A.; Ghidiu, M. J.; Ma, Y.-.Z.;
Rosenthal, J. J. Am. Chem. Soc. 2013, 135, 6601−6607.
(16) Littler, B. J.; Miller, M. A.; Hung, C.-H.; Wagner, R. W.; O’Shea,
D. F.; Boyle, P. D.; Lindsey, J. S. J. Org. Chem. 1999, 64, 1391−1396.
(17) Berube, C. D.; Yazdanbakhsh, M.; Gambarotta, S.; Yap, G. P. A
Organometallics 2003, 22, 3742−3747.
(18) Freckmann, D. M. M.; Dube, T.; Berube, C. D.; Gambarotta, S.;
Yap, G. P. A. Organometallics 2002, 21, 1240−1246.
(19) Laha, J. K.; Dhanalekshmi, S.; Taniguchi, M.; Ambroise, A.;
Lindsey, J. S. Org. Process Res. Dev. 2003, 7, 799−812.
(20) Rao, P. D.; Dhanalekshmi, S.; Littler, B. J.; Lindsey, J. S. J. Org.
Chem. 2000, 65, 7323−7344.
(21) Geier, G. R., III; Chick, J. F. B.; Callinan, J. B.; Reid, C. G.;
Auguscinski, W. P. J. Org. Chem. 2004, 69, 4159−4169.
(22) Geier, G. R., III; Callinan, J. B.; Rao, D. P.; Lindsey, J. S. J.
Porphyrins Phthalocyanines 2001, 5, 810−823.
ASSOCIATED CONTENT
* Supporting Information
■
(23) Gupta, I.; Frohlich, R.; Ravikanth, M. Chem. Commun. 2006,
̈
S
3726−3728.
NMR spectra of diacyldipyrromethanes 2a−c, NMR spectra of
phlorins 3a−c, discussion of the partial assignment of 1H NMR
spectra for phlorins 3a−c, UV−vis spectra of phlorins 3a−c,
overlaid UV−vis spectra of phlorins 3a−c exposed to light,
plots of the absorbance of phlorins 3a−c as a function of time
of exposure to light, plots of HPLC peak area of phlorins 3a−c
as a function of time of exposure to light, and crystallographic
data for phlorins 3a−c. This material is available free of charge
via the Internet at http://pubs.acs.org.
(24) Gokulnath, S.; Chandrashekar, T. K. Org. Lett. 2008, 10, 637−
640.
(25) Higashino, T.; Osuka, A. Chem.Asian. J. 2013, 8, 1994−2002.
(26) Liu, B.; Li, X.; Xu, X.; Stepien, M.; Chmielewski, P. J. J. Org.
Chem. 2013, 78, 1354−1364.
(27) Gryko, D. T.; Koszarna, B. Eur. J. Org. Chem. 2005, 3314−3318.
(28) Geier, G. R., III; Grindrod, S. C. J. Org. Chem. 2004, 69, 6404−
6412.
(29) Sheldrick, G. M. SHELXTL, Crystallographic Software Package,
Version 6.10, Bruker-AXS, Madison, WI, 2000.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: ziegler@uakron.edu.
*E-mail: ggeier@colgate.edu.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Support for this work was provided by the National Science
Foundation under Grant No. 0517882 (GRG). High-resolution
mass spectra were obtained at the North Carolina State
University Department of Chemistry Mass Spectrometry
Facility. Funding was obtained from the North Carolina
Biotechnology Center and the North Carolina State University
Department of Chemistry.
REFERENCES
■
(1) Sessler, J. L.; Zimmerman, R. S.; Bucher, C.; Kral, V.; Andrioletti,
B. Pure Appl. Chem. 2001, 73, 1041−1057.
(2) Dehaen, W. Top. Heterocycl. Chem. 2010, 24, 75−102.
(3) Aviv-Harel, I.; Gross, Z. Chem.Eur. J. 2009, 15, 8382−8394.
(4) Pistner, A. J.; Yap, G. P. A.; Rosenthal, J. J. Phys. Chem. C 2012,
116, 16918−16924.
(5) Ka, J. W.; Lee, C. H. Tetrahedron Lett. 2001, 42, 4527−4529.
(6) Hong, S. J.; Ka, J. W.; Won, D. H.; Lee, C. H. Bull. Korean Chem.
Soc. 2003, 24, 661−663.
(7) Whitlock, H. W.; Oester, M. Y. J. Am. Chem. Soc. 1973, 95, 5738−
5741.
(8) Woodward, R. B. Ind. Chim Belg. 1962, 27, 1293−1308.
(9) Jeandon, C.; Krattinger, B.; Ruppert, R.; Callot, H. J. Inorg. Chem.
2001, 40, 3149−3153.
(10) Setsune, J. −i.; Ikeda, M.; Iida, T.; Kitao, T. J. Am. Chem. Soc.
1988, 110, 6572−6574.
(11) Krattinger, B.; Callot, H. J. Chem. Commun. 1996, 1341−1342.
5672
dx.doi.org/10.1021/jo5008256 | J. Org. Chem. 2014, 79, 5664−5672