Two-Step, One-Flask Synthesis of a Meso-Substituted Phlorin

Article abstract of DOI:10.1021/acs.joc.6b00571

A two-step, one-flask reaction of pyrrole with pentafluorobenzaldehyde and acetone was investigated to determine the potential for a streamlined synthesis of a phlorin and/or 5-isocorrole as an alternative to stepwise, dipyrromethanecarbinol routes. Analytical-scale reactions were performed examining the effect of reactant concentration, reactant ratio, acid catalyst (TFA or BF₃·OEt₂), concentration of acid catalyst, oxidant quantity, and reaction time on the distribution of phlorin and 5-isocorrole as well as three additional porphyrinoids (porphodimethene, porphyrin, and corrole). Phlorin was observed ubiquitously in yields up to 20-26%, whereas 5-isocorrole was not detected. Promising reaction conditions for the one-flask synthesis of the phlorin were performed on a preparative scale. The best reaction condition afforded the phlorin in an isolated yield of 20-21% (249-268 mg). Preliminary attempts to extend the methodology to the preparation of phlorins derived from other ketones resulted in a low yield of phlorin from acetophenone (5%) and no detectable phlorin from benzophenone. The discovery of reaction conditions for the two-step, one-flask synthesis of a phlorin provides easier access to this interesting compound, and provides encouragement for the further study of reactions of pyrrole with an aldehyde and a ketone.

Full text of DOI:10.1021/acs.joc.6b00571

Article
pubs.acs.org/joc
Two-Step, One-Flask Synthesis of a Meso-Substituted Phlorin
Dongjoon Kim, Hao-Jung Chun, Christopher C. Donnelly, and G. Richard Geier, III*
Colgate University, Department of Chemistry, 13 Oak Drive, Hamilton, New York 13346, United States
S
* Supporting Information
ABSTRACT: A two-step, one-flask reaction of pyrrole with pentafluorobenzaldehyde and acetone was investigated to determine
the potential for a streamlined synthesis of a phlorin and/or 5-isocorrole as an alternative to stepwise, dipyrromethanecarbinol
routes. Analytical-scale reactions were performed examining the effect of reactant concentration, reactant ratio, acid catalyst (TFA
or BF3·OEt2), concentration of acid catalyst, oxidant quantity, and reaction time on the distribution of phlorin and 5-isocorrole as
well as three additional porphyrinoids (porphodimethene, porphyrin, and corrole). Phlorin was observed ubiquitously in yields
up to 20−26%, whereas 5-isocorrole was not detected. Promising reaction conditions for the one-flask synthesis of the phlorin
were performed on a preparative scale. The best reaction condition afforded the phlorin in an isolated yield of 20−21% (249−
268 mg). Preliminary attempts to extend the methodology to the preparation of phlorins derived from other ketones resulted in a
low yield of phlorin from acetophenone (5%) and no detectable phlorin from benzophenone. The discovery of reaction
conditions for the two-step, one-flask synthesis of a phlorin provides easier access to this interesting compound, and provides
encouragement for the further study of reactions of pyrrole with an aldehyde and a ketone.
INTRODUCTION
■
The calix[4]phyrin family of porphyrinoids, of which phlorin is
a member, have structures intermediate to porphyrin and
calix[4]pyrrole (porphyrinogen).1,2 Calix[4]phyrins contain a
combination of sp2- and sp3-hybridized carbon atoms at the
four meso-positions, whereas porphyrins have entirely sp2-
hybridized meso-carbon atoms and calix[4]pyrroles have only
sp3-hybridized meso-carbon atoms. Many calixphyrins have
been prepared and studied, including species with expanded,
contracted, N-confused, and/or heteroatom modified core
structures.1,2
Two members of the broader calixphyrin family have been of
particular interest to usphlorin3−5 and 5-isocorrole.6 We
were initially interested in phlorin and 5-isocorrole due to their
intriguing structural relationship to the better known porphyrin
and corrole macrocycles. Phlorin possesses four bridging meso-
carbons (like porphyrin), but it has three internal N−H groups
(like corrole). In a complementary fashion, 5-isocorrole has
three bridging meso-carbon atoms and a direct bipyrrole
linkage (like corrole), but it has two internal N−H groups (like
porphyrin). Phlorin and 5-isocorrole also have the structural
distinction of a single sp3-hybridized meso-carbon atom,
adapting dipyrromethanecarbinol routes initially developed for
the rational synthesis of porphyrins bearing different meso-
interrupting the macrocycle conjugation found in porphyrin
and corrole.
In the course of our prior studies of phlorin and 5-isocorrole,
Received: March 16, 2016
we developed stepwise syntheses for both compounds4,6 by
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.6b00571
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
substituents in defined positions.7−9 Our published syntheses of
a phlorin and a 5-isocorrole are adequate for the preparation of
100−200 mg quantitiessufficient for spectroscopic character-
ization, X-ray structure determination, and studies of stability
toward light and air (many phlorins have poor stability toward
light and air3−5,10). Additionally, in collaboration with Ziegler
and co-workers, a small number of metal complexes of a 5-
isocorrole have been prepared and characterized.11 We have
also applied our stepwise synthesis of phlorin to the preparation
of phlorins bearing different substituents at the sp3-hybridized
meso-position.5 The stepwise methodology has been recently
utilized by Rosenthal and co-workers in their independent
studies of phlorin spectroscopy, structure, electrochemistry, and
anion binding.12−15
ward, one-flask syntheses. Examples include meso-substituted
porphyrin,16−18 N-confused porphyrin,19−21 and corrole.22,23
Unfortunately, calixphyrins such as phlorin and 5-isocorrole
are not obvious candidates for a one-flask approach. The
presence of both sp2- and sp3-hybridized meso-positions lowers
the symmetry of calixphyrins relative to the porphyrins
commonly prepared from the one-flask reaction of pyrrole
with an aldehyde. The synthesis of calixphyrins is more
analogous to the preparation of porphyrins bearing two
different meso-substituents. While such porphyrins have been
prepared by the one-flask reaction of pyrrole with two different
aldehydes, the reaction inevitably produces a mixture of up to
six porphyrins bearing different combinations of meso-
substituents.24 The product mixture lowers yield and
complicates purification. A one-flask synthesis of phlorin and/
or 5-isocorrole via the reaction of pyrrole with an aldehyde and
a ketone would be expected to produce a complicated mixture
of porphyrinoids and linear oligomers (Scheme 2). Note that
only a subset of the possible products are shown in the scheme.
Despite the utility of stepwise syntheses of phlorin and 5-
isocorrole, these routes have important limitations as illustrated
by the stepwise synthesis of a phlorin (Scheme 1).4 While the
Scheme 1. Stepwise Synthesis of a Meso-Substituted Phlorin
Scheme 2. Reaction of Pyrrole, Pentafluorobenzaldehyde,
and Acetone Leading to a Number of Possible Porphyrinoids
and Linear Oligomers
Given the complicated reaction mixture expected from the
reaction of pyrrole with an aldehyde and a ketone, it is not
surprising that there are few reported examples. Furuta, Osuka,
Ishikawa, and co-workers isolated an N-confused calix[4]phyrin
species in a yield of 3% from the reaction of pyrrole with p-
tolualdehyde and acetone.25 Hu and co-workers have
investigated the reaction of pyrrole with arylaldehydes and
acenapthenequinone (a 1,2-diketone) with the stated goal of
obtaining a phlorin.26 Instead, depending on the reaction
synthesis is reasonably concise, the preparation of dipyrro-
methane and acyldipyrromethane building block molecules is
required, which adds time to the synthesis, places limits on the
scale of the synthesis, and increases the degree of required
synthetic expertise. A one-flask approach could address these
limitations, and potentially increase the relevance of phlorin
and 5-isocorrole. It is not surprising that many of the most
widely studied porphyrinoids can be obtained by straightfor-
B
DOI: 10.1021/acs.joc.6b00571
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
ab
,
Table 1. Comparison of Porphyrinoid Yields from Preliminary Analytical-Scale Experiments
c
c
c
entry [pyrrole], mM [aldehyde], mM [acetone], mM
acid
TFA
[acid], mM % yield of porphyrin
% yield of corrole
% yield of phlorin
d
1a
2a
3a
4a
5a
1b
2b
3b
4b
5b
10
10
10
10
10
10
10
10
10
10
10
−
−
2.5
2.5
5.0
−
215
215
215
215
215
10
20
22
19
12
11
32
50
43
25
21
2
7
6
9
7
0
1
1
4
3
−
d
7.5
7.5
5.0
5.0
10
TFA
−
TFA
5
TFA
5
TFA
12
d
BF3·OET2
BF3·OET2
BF3·OET2
BF3·OET2
BF3·OET2
−
d
7.5
7.5
5.0
5.0
−
10
−
2.5
2.5
5.0
10
3
3
4
10
10
a
The reactions were performed in CH2Cl2 with the indicated reactants on a 10 mL scale at room temperature. The reactions were monitored at 15
b
c
min, 1 h, and 4 h upon oxidation of an aliquot (1.2 mL) with DDQ (2.0 mg). 5-Isocorrole 2 and porphodimethene 3 were not detected. The
highest yield (HPLC) of any of the three time points is reported. Phlorin 1 could not be obtained from this reaction as acetone was not present.
d
condition, they isolated 5-(8-ethoxycabonyl-1-naphthyl)-
10,15,20-triarylporphyrins in yields of 1−11%,26,27 a β,β′-linked
porphyrin-chlorin heterodimer and a π-extended porphyrin in
yields of 3−4% and 7−8% respectively,28 or 5,10- and 5,15-
porphodimethenes in yields of 4−7%.29 Overall, these reports
show that it is possible to obtain lower symmetry porphyrinoids
from mixed reactions of pyrrole with an aldehyde and a ketone,
albeit in generally modest yield.
Herein, we report an investigation of a two-step (acid
catalyzed condensation followed by oxidation), one-flask
reaction of pyrrole with pentafluorobenzaldehyde and acetone
(Scheme 2). The objectives of the present work were to explore
the interplay of reaction conditions on the yields of a variety of
potential porphyrinoid products, and to determine whether a
one-flask reaction could afford phlorin 1, 5-isocorrole 2, and/or
other calixphyrins (e.g., porphodimethene 3) in a meaningful
yield with straightforward purification. Pentafluorobenz-
aldehyde was employed as the aldehyde as the electron-
withdrawing pentafluorophenyl substituent is known to
stabilize phlorin4 and corrole30 toward degradation upon
exposure to light and air. Acetone was employed as the ketone
for consistency with the geminal dimethyl groups present in our
prior stepwise syntheses of phlorin4 and 5-isocorrole.6
Key to this work was the development of an HPLC method
that allowed monitoring crude, reaction mixtures for the yield
of phlorin 1, 5-isocorrole 2, porphodimethene 3, porphyrin 4,
and corrole 5. Other potential porphyrinoid products (e.g., 10-
isocorrole) were not included due to the absence of authentic
compounds required for HPLC method development and
calibration of detector response. A small number of initial
analytical-scale reactions were carried out to determine the
scope of reaction conditions to explore. Further analytical-scale
reactions were then performed to systematically investigate the
effect of reactant concentration, reactant ratio, aldehyde
concentration, acid catalyst, acid catalyst concentration, oxidant
quantity, and reaction time. Promising reaction conditions were
repeated on a preparative scale. Refined reaction conditions
were applied to reactions with two other ketones (aceto-
phenone and benzophenone).
isocorrole 2, porphodimethene 3, porphyrin 4, and corrole 5
from which an HPLC method could be developed. The five
compounds are also a representative sampling of possible
products of the reaction. The HPLC method was adapted from
prior studies in which we monitored crude reaction mixtures for
yields of phlorin 1,4 5-isocorrole 2 and porphodimethene 3,6 or
5-isocorrole 2 and porphyrin 46 (see the Supporting
Information for an expanded discussion of HPLC method
development). Reactions were also monitored by TLC to
detect other colored compounds in addition to the five
porphyrinoids assessed by HPLC.
Preliminary Analytical-Scale Experiments. Ten reaction
conditions were selected as the starting point for the study
(Table 1). The reaction conditions were modeled after refined
two-step, one-flask conditions for the preparation of porphyrin
4,31 with varying ratios of pyrrole, pentafluorobenzaldehyde,
and acetone consistent with the stoichiometric requirements for
porphyrinoids 1−5. Entries 1a,b and 2a,b afforded good levels
of porphyrin 4, a low yield of corrole 5, and no detectable
phlorin 1, 5-isocorrole 2, or porphodimethene 3, consistent
with the absence of acetone in these reactions. The yield of
corrole 5 was higher for entries 2a,b given the more favorable
ratio of pyrrole to aldehyde for corrole synthesis. Reactions
carried out in the presence of acetone (entries 3−5) afforded
lower yields of porphyrin 4, similar yields of corrole 5, and low
to modest yields of phlorin 1. 5-Isocorrole 2 and
porphodimethene 3 were not detected. The consistent
presence of phlorin 1 in all reactions containing acetone was
encouraging, especially the observation of a yield of phlorin 1 of
12% (entry 5a).
Survey of Acetone Concentration. Encouraged by the
results of the preliminary experiments, a systematic survey of
reaction conditions was performed targeting a one-flask
synthesis of phlorin 1. As calixphyrins, such as phlorin 1,
possess defined ratios of sp2- and sp3-hybridized meso-
positions, the concentration of acetone relative to pyrrole and
pentafluorobenzaldehyde was expected to be an important
parameter. Thus, a survey of acetone concentration was carried
out for reactions utilizing pyrrole (10 mM) and pentafluoro-
benzaldehyde (2.5, 5.0, or 7.5 mM) with catalysis by TFA (215
mM) or BF3·OEt2 (10 mM).
A representative plot of the yield of phlorin 1 as a function of
acetone concentration for reactions with pyrrole, pentafluoro-
benzaldehyde (5.0 mM), and TFA is provided in Figure 1,
panel A1. Panel A2 provides a summary of the highest yield of
phlorin 1, porphyrin 4, and corrole 5 observed at any of the
RESULTS AND DISCUSSION
■
Reaction Monitoring. The reaction of pyrrole with
pentafluorobenzaldehyde and acetone could produce a number
of possible products (Scheme 2). We sought to monitor
analytical-scale reactions for as many of these products as
possible. From prior studies,4,6,30 we had authentic phlorin 1, 5-
C
DOI: 10.1021/acs.joc.6b00571
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 1. Yields of porphyrinoids from the reactions of pyrrole (10 mM), pentafluorobenzaldehyde (5.0 mM), and acetone as a function of acetone
concentration. (A) Reactions with TFA (215 mM). Panel A1 provides the yield of phlorin 1 at condensation reaction times of 15 min, 1 h, and 4 h,
and panel A2 provides the highest yield of phlorin 1, porphyrin 4, and corrole 5 observed from the three time points. (B) Reactions with BF3·OEt2
(10 mM). Panel B1 provides the yield of phlorin 1 at condensation reaction times of 15 min, 1 h, and 4 h, and panel B2 provides the highest yield of
phlorin 1, porphyrin 4, and corrole 5 observed from the three time points. The reactions were performed in CH2Cl2 at room temperature, and were
monitored by HPLC. Porphodimethene 3 was observed in low yields (0−2%), and 5-isocorrole 2 was not detected.
three time points. (Note: 5-isocorrole 2 and porphodimethene
3 are not included in this or other plots as 5-isocorrole was not
detected and the yield of porphodimethene was generally very
low.) The highest yield of phlorin 1 (26%) was obtained at
acetone concentrations of 40−80 mM. However, detectable
phlorin 1 was observed from all acetone concentrations
surveyed (2.5−320 mM). The yields of porphyrin 4 and
corrole 5 declined with increasing acetone concentration as
would be expected from increased incorporation of ketone
units into oligomer precursors. Porphodimethene 3 was
observed in low yield (<2%) in reactions with ≥20 mM
acetone. TLC analyses of crude, oxidized aliquots revealed a
number of colored compounds of varying polarity in addition
to the five porphyrinoids assessed by HPLC. However, none of
the additional pigments were of sufficient prevalence or
intensity to merit further investigation. On the basis of the
results of HPLC monitoring, an acetone concentration of 80
mM was found to provide the best balance of the yield of
phlorin 1 (26%) while minimizing the yield of other
porphyrinoids, in particular porphyrin 4 (4%). Analogous
reactions utilizing BF3·OEt2 (10 mM) provided similar trends,
albeit with lower maximum yields of phlorin 1 (16%) and
higher yields of porphyrin 4 (Figure 1, panels B1 and B2).
Porphodimethene 3 was observed in low yield (<1%) in
reactions with ≥20 mM acetone. An acetone concentration of
80 mM was again judged to be the best condition.
(BF3·OEt2). With 7.5 mM aldehyde, the highest yields of
phlorin 1 were 17% (TFA) and 14% (BF3·OEt2). Phlorin 1 was
a fairly ubiquitous product, with at least low levels detected
from almost every reaction. In the reactions with 2.5 mM
aldehyde, porphodimethene 3 was observed in low yield (<3%)
in reactions with ≥20 mM acetone. In the reactions with 7.5
mM aldehyde, porphodimethene 3 was observed only at the
highest acetone concentration in a yield of ∼3%. For reactions
utilizing 2.5 mM or 7.5 mM pentafluorobenzaldehyde with
either TFA or BF3·OEt2 catalysis, an acetone concentration of
80 mM was again found to strike a good balance between the
yield of phlorin and the yields of other porphyrinoids.
Survey of Acid Catalysis Conditions. The concentration
of acid catalyst is known to impact the trajectory and maximum
yield of two-step, one-flask syntheses of porphyrinoids.24,32
Thus, a survey of TFA and BF3·OEt2 concentration was carried
out for reactions utilizing pyrrole (10 mM), pentafluoro-
benzaldehyde (2.5, 5.0, or 7.5 mM), and acetone (80 mM).
A representative plot of the yield of phlorin 1 as a function of
TFA concentration for reactions with pyrrole, pentafluoro-
benzaldehyde (5.0 mM), and acetone is provided in Figure 2,
panel A1. Panel A2 provides a summary of the highest yield of
phlorin 1, porphyrin 4, and corrole 5 observed at any of the
three time points. The highest yield of phlorin 1 (24%) was
obtained at TFA concentrations of 215−290 mM. The range of
TFA concentrations providing the best yield of phlorin 1 was in
agreement with the initial choice of 215 mM. Detectable
phlorin 1 was observed with TFA concentrations of 55−500
mMa nearly 10-fold range. Porphodimethene 3 was observed
in low yield (<2%) from reactions with ≥160 mM TFA.
Analogous reactions surveying the concentration of BF3·OEt2
Experiments utilizing 2.5 mM or 7.5 mM pentafluoro-
benzaldehyde provided similar trends, though differing
maximum yields (see the Supporting Information for a
complete set of plots). The highest yields of phlorin 1 for
reactions with 2.5 mM aldehyde were 20% (TFA) and 10%
D
DOI: 10.1021/acs.joc.6b00571
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 2. Yields of porphyrinoids from the reactions of pyrrole (10 mM), pentafluorobenzaldehyde (5.0 mM), and acetone (80 mM) as a function of
acid concentration. (A) Reactions with TFA. Panel A1 provides the yield of phlorin 1 at condensation reaction times of 15 min, 1 h, and 4 h, and
panel A2 provides the highest yield of phlorin 1, porphyrin 4, and corrole 5 observed from the three time points. (B) Reactions with BF3·OEt2. Panel
B1 provides the yield of phlorin 1 at condensation reaction times of 15 min, 1 h, and 4 h, and panel B2 provides the highest yield of phlorin 1,
porphyrin 4, and corrole 5 observed from the three time points. The reactions were performed in CH2Cl2 at room temperature, and were monitored
by HPLC. Porphodimethene 3 was generally observed in low yields (0−2%) with the exception of the lowest concentrations of BF3·OEt2 (0.5−1.0
mM) which afforded 3 in yields of 8−9%. 5-Isocorrole 2 was not detected.
provided generally similar trends, albeit with lower maximum
yields of phlorin 1 (15%) and higher yields of porphyrin 4
(Figure 2, panels B1 and B2). The highest yields of phlorin 1
were obtained with BF3·OEt2 concentrations of 7.0−13 mM.
The range of BF3·OEt2 concentrations providing the best yield
of phlorin was again in agreement with the initial choice of 10
mM.
Experiments utilizing 2.5 mM or 7.5 mM pentafluoro-
benzaldehyde afforded similar trends, though differing max-
imum yields (see the Supporting Information for a complete set
of plots). TFA was generally a better catalyst than BF3·OEt2 for
the synthesis of phlorin 1. The highest yields of phlorin 1 from
reactions with 2.5 mM aldehyde were 19% (TFA, 215 mM)
and 14% (BF3·OEt2, 10 mM). At 7.5 mM aldehyde, the highest
yields of phlorin 1 were 20% (TFA, 160−290 mM) and 18%
(BF3·OEt2, 10−26 mM). In the reactions with 2.5 mM
aldehyde, porphodimethene 3 was observed in modest yield
(4%) in reactions with the lowest concentrations of BF3·OEt2
(0.5−1.0 mM). For reactions utilizing 2.5 mM or 7.5 mM
aldehyde, the initial selections of 215 mM TFA and 10 mM
BF3·OEt2 were found to be appropriate.
Investigation of Oxidation Conditions. In two-step, one-
flask syntheses of porphyrins, it is common to employ a
stoichiometric quantity of oxidant (e.g., DDQ) in the oxidation
step (assuming quantitative formation of the porphyrinogen
precursor). However, we have found that syntheses of
corrole,30 phlorin,4 and 5-isocorrole6 from stepwise dipyrro-
methanecarbinol routes are sensitive to the quantity of oxidant,
and that the optimal amount of oxidant is not always in
accordance with the calculated stoichiometric quantity. Thus,
we examined oxidant quantity at the beginning of the study,
and again after the surveys of acetone and acid catalyst
concentrations.
At the beginning of the study, oxidation of aliquots (1.2 mL)
from a handful of trial reactions was done with both 1.0 and 2.0
mg of DDQ. These quantities of DDQ were selected based on
amounts we had previously found to be effective in stepwise
syntheses of phlorin 1,4 5-isocorrole 2,6 and corrole 530 carried
out at similar reactant concentrations. From comparison of the
two DDQ quantities, we found that higher yields of all
porphyrinoids detected were generally obtained from 2.0 mg of
DDQ. Thus, the preliminary analytical-scale experiments
summarized in Table 1 utilized 2.0 mg of DDQ. Upon
completion of these experiments, DDQ quantity (0.25−16 mg)
was investigated further before proceeding with the systematic
survey of reaction conditions. The examination of DDQ
quantity supported continued oxidation of reaction aliquots
with 2.0 mg of DDQ.
After conducting systematic surveys of the concentration of
acetone and acid catalyst, DDQ quantity was again investigated.
Representative plots of the yield of phlorin 1, porphyrin 4, and
corrole 5 as a function DDQ quantity used in the oxidation of
aliquots (1.2 mL) are provided in Figure 3. Under catalysis by
TFA, the maximum yield of phlorin 1 (22%) was obtained with
2.0 mg of DDQ. At lower and higher DDQ quantities, the yield
of phlorin 1 declined sharply. The highest yields of porphyrin 4
and corrole 5 were obtained with 3.0 mg of DDQ, and there
was no decline in yield with higher quantities of DDQ. The
analogous reaction mediated by BF3·OEt2 afforded similar
results, with the maximum yield of phlorin (18%) obtained
from 1.0 to 2.0 mg of DDQ. Experiments utilizing 2.5 mM or
7.5 mM pentafluorobenzaldehyde provided similar results (see
the Supporting Information for a complete set of plots).
E
DOI: 10.1021/acs.joc.6b00571
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
pentafluorobenzaldehyde (7.5 mM), and acetone (160 mM)
mediated by TFA (215 mM) in an attempt to further suppress
porphyrin 4 formation.
A representative plot of the yields of phlorin 1, porphyrin 4,
and corrole 5 as a function of time for the reaction with pyrrole
(10 mM), pentafluorobenzaldehyde (5.0 mM), and TFA (215
mM) is provided in Figure 4A. A near maximum yield of
Figure 3. Yields of porphyrinoids from the reaction of pyrrole (10
mM), pentafluorobenzaldehyde (5.0 mM), and acetone (80 mM) at a
reaction time of 1 h as a function of DDQ quantity used to oxidize
reaction aliquots (1.2 mL). (A) TFA (215 mM). (B) BF3·OEt2 (10
mM). The reaction was performed in CH2Cl2 at room temperature,
and the oxidized aliquots were assessed by HPLC. A consistent, low
yield porphodimethene 3 (1−2%) was observed with DDQ quantities
≥2.0 mg. 5-isocorrole 2 was not detected.
Figure 4. Yields of porphyrinoids from the reaction of pyrrole (10
mM), pentafluorobenzaldehyde (5.0 mM), and acetone (80 mM) as a
function of condensation reaction time (1 min to 24 h). (A) TFA (215
mM). (B) BF3·OEt2 (10 mM). The reactions were performed in
CH2Cl2 at room temperature, and monitored by HPLC. Porphodi-
methene 3 was observed in low yields (0−2%), and 5-isocorrole 2 was
not detected. Note the logarithmic scale for time.
Overall, oxidation of aliquots of reaction mixtures required a
fairly exacting quantity of DDQ, with 2.0 mg providing maximal
or near maximal yields for all porphyrinoids detected from each
of the reaction conditions investigated.
It is interesting that a 3-fold difference in the concentration
of pentafluorobenzaldehyde had a negligible effect on the
optimal quantity of DDQ. For each porphyrinoid detected, the
aldehyde was generally the limiting reagent. Thus, the
stoichiometric quantity of DDQ required for the preparation
of each porphyrinoid from their reduced precursor declines
with decreasing aldehyde concentration. For example, the
stoichiometric quantities of DDQ required for the preparation
of phlorin 1 (assuming quantitative formation of its reduced
precursor during the condensation step) are 1.4, 0.91, and 0.45
mg for aliquots (1.2 mL) from reactions performed with
aldehyde concentrations of 7.5, 5.0, and 2.5 mM, respectively.
Nonetheless, the composition of the overall, complex reaction
mixture appears to be such that a consistent amount of DDQ is
required over the range of aldehyde concentration investigated.
In practical terms, the quantity of oxidant did not need be
adjusted for analytical-scale reactions performed with different
concentrations of the limiting reagent.
phlorin 1 (21%) was obtained by a reaction time of 15 min, and
remained fairly constant from 4 to 8 h. The observation of a
consistent yield of phlorin 1 over an extended period of time is
of practical significance as the precise timing of the oxidation
should not impact reproducibility of preparative-scale syn-
theses. Porphyrin 4 (2−3% yield) was observed from 4 min to
24 h, and a very low level of corrole 5 (<1%) was detected from
1 to 2 min. Porphodimethene 3 (<2% yield) was observed from
15 min to 24 h. The analogous reaction with BF3·OEt2 afforded
similar results (Figure 4B). The maximum yield of phlorin 1
(15%) was obtained from 15 min to 1 h, with only modest
decline in yield by 4 h. The yield of porphyrin 4 was fairly
constant (5−6%) from 4 min to 24 h. Porphodimethene 3
(<1%) was observed from 30 min to 8 h.
Reaction Time-Course Experiments. The survey of
reaction parameters provided promising conditions in terms
of the yield of phlorin 1 and level of porphyrinoid byproducts.
Seven conditions were selected for reaction monitoring from 1
min to 24 h. Six reactions utilized pyrrole (10 mM) and
acetone (80 mM), along with pentafluorobenzaldehyde (2.5,
5.0, or 7.5 mM) and TFA (215 mM) or BF3·OEt2 (10 mM).
The seventh reaction was performed with pyrrole (10 mM),
The yield trajectories from the other reactions were similar
(see the Supporting Information for the complete set of plots).
Reactions utilizing TFA generally provided higher yields of
phlorin 1 (20−25%) than reactions with BF3·OEt2 (13−15%).
Reactions mediated by BF3·OEt2 also provided higher levels of
porphyrin 4 relative to analogous reactions with TFA.
Reactions performed with a higher concentration of penta-
fluorobenzaldehyde resulted in higher levels of porphyrin 4.
F
DOI: 10.1021/acs.joc.6b00571
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Preparative-Scale Phlorin Syntheses. To confirm results
of analytical-scale experiments, three reaction conditions were
performed on a preparative scale: (1) pyrrole (10 mM),
pentafluorobenzaldehyde (5.0 mM), acetone (80 mM), and
TFA (215 mM); (2) pyrrole (10 mM), pentafluorobenz-
aldehyde (7.5 mM), acetone (160 mM), and TFA (215 mM);
and (3) pyrrole (10 mM), pentafluorobenzaldehyde (2.5 mM),
acetone (80 mM), and TFA (215 mM). The reaction
conditions were selected based on the yield of phlorin 1 and
other porphyrinoids observed in the analytical-scale reactions.
Each condition afforded similarly good yields of phlorin 1, but
differing quantities of porphyrin 4. We sought to determine the
extent to which the porphyrin (along with the very low levels of
porphodimethene 3 and corrole 5) would complicate
purification of phlorin 1. We found in our analytical-scale
experiments that the level of porphyrin 4 can be suppressed
without a significant decline in the yield of phlorin 1 by
decreasing the concentration of aldehyde. However, doing so
increases the volume of solvent required to prepare a given
quantity of phlorin 1. Thus, it was important to determine the
level of porphyrin 4 and other porphyrinoids that can be
tolerated.
The reaction utilizing a pentafluorobenzaldehyde concen-
tration of 5.0 mM was performed first. On the basis of
observations from the analytical-scale reactions, we suspected
that this condition would provide the best balance between the
yield of phlorin 1, suppression of other porphyrinoid products,
and concentration of the limiting reagent thereby minimizing
solvent volume. Preoxidation monitoring of the reaction by
HPLC provided the following yield estimates: phlorin 1 (21%),
isocorrole 2 (not detected), porphodimethene 3 (1%),
porphyrin 4 (4%), and corrole 5 (0.4%). These values were
similar to yields obtained from HPLC monitoring of the
analogous analytical-scale reaction: phlorin 1 (22−26%),
isocorrole 2 (not detected), porphodimethene 3 (1%),
porphyrin 4 (3−4%), and corrole 5 (trace levels). Monitoring
of the reaction mixture by HPLC after bulk oxidation with
DDQ afforded similar estimates of the yield of phlorin 1 (22%),
isocorrole 2 (not detected), porphodimethene 3 (1%),
porphyrin 4 (4%), and corrole 5 (0.4%). The reaction mixture
was purified by filtration through a pad of silica gel, followed by
silica gel column chromatography. Isolated quantities of
porphodimethene 3, porphyrin 4, and corrole 5 corresponded
to yields of 1%, 4%, and 0.5%, respectively. Fractions containing
phlorin 1 were further purified by neutral alumina chromatog-
raphy affording phlorin 1 in good purity (297 mg, 24% yield).
Crystallization afforded 266 mg of phlorin (21% yield), in
agreement with HPLC monitoring and with results on the
analytical-scale. This condition was performed two additional
times, with similar results. Isolated yields for each porphyrinoid
from the two additional trials were: phlorin 1 (postalumina
column, 24%, 22%), phlorin 1 (postcrystallization, 21%, 20%)
porphodimethene 3 (1%, 1%), porphyrin 4 (5%, 5%) and
corrole 5 (0.6%, 0.5%).
the analogous analytical-scale reaction: phlorin 1 (21%)
porphodimethene 3 (1%), porphyrin 4 (6%), and corrole 5
(not detected).
The reaction performed at lower pentafluorobenzaldehyde
concentration (2.5 mM) proceeded well overall, but the larger
solvent volume was inconvenient and allowed a higher level of
impurity to pass through the initial silica pad. Nonetheless,
further purification of the phlorin was not overly impacted.
Isolated yields for each porphyrinoid were phlorin 1
(postalumina column, 22%), phlorin 1 (postcrystallization,
18%) porphodimethene 3 (2%), and porphyrin 4 (1%). 5-
Isocorrole 2 and corrole 5 were not detected. These yields were
in good agreement with yields obtained from HPLC
monitoring of the analogous analytical-scale reaction: phlorin
1 (19−22%) porphodimethene 3 (2%), porphyrin 4 (2%), and
corrole 5 (not detected). Given that the yield of phlorin was
similar for the three reaction conditions and that the presence
of porphodimethene 3, porphyrin 4 and corrole 5 were found
to not adversely complicate purification of phlorin under any of
the conditions investigated, it is not necessary to perform the
reaction at the lowest concentration of pentafluorobenz-
aldehyde.
Attempted Reactions with Benzophenone and Aceto-
phenone. Although the reaction of pyrrole with pentafluor-
obenzaldehyde and acetone was the focus of the present work,
we were interested in the extension of the findings to other
ketones. Previously, we found that the presence of different
substituents at the sp3-hybridized meso-position of phlorin can
impact structure, spectroscopic properties, and stability.5 While
a comprehensive examination of additional ketones was beyond
the scope of the present work, we attempted reactions using
acetophenone and benzophenone (Scheme 3). These ketones
Scheme 3. Attempted Two-Step, One-Flask Syntheses of
Phlorins Bearing Different Substituents at the sp3-
Hybridized Meso-Position
The reaction performed at higher pentafluorobenzaldehyde
concentration (7.5 mM) proceeded similarly well with the
exception of a low level of residual impurity present in the
phlorin isolated from neutral alumina chromatography which
was removed by crystallization. Isolated yields for each
porphyrinoid were phlorin 1 (postcrystallization, 17%)
porphodimethene 3 (0.7%), and porphyrin 4 (6%). 5-
Isocorrole 2 and corrole 5 were not detected. The isolated
yields are similar to yields obtained from HPLC monitoring of
were selected due to their different electronic and steric aspects
relative to acetone, and we had previously prepared phlorins 6
and 7 via stepwise, dipyrromethanecarbinol routes.5 Each
ketone was subjected to a reaction condition found to work
well for the preparation of phlorin 1: pyrrole (10 mM),
pentafluorobenzaldehyde (5.0 mM), ketone (80 mM), and
G
DOI: 10.1021/acs.joc.6b00571
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
TFA (215 mM) in CH2Cl2 at room temperature for 1 h
followed by oxidation with DDQ.
EXPERIMENTAL SECTION
■
General Experimental Methods. 1H NMR (400 MHz), 13C
NMR (100 MHz), and absorption spectra were collected routinely.
Column chromatography was performed on silica (Sorbent Tech-
nologies, standard grade, 230−400 mesh, 60 Å) or neutral alumina
(Fisher, 80−200 mesh). CH2Cl2 used in analytical-scale reactions was
distilled from K2CO3 and stored over 4-Å Linde molecular sieves.
CH2Cl2 used in sample preparation for HPLC determination of yield
of porphyrinoids 1−5 was passed through a pad of basic alumina prior
to use. Pyrrole was distilled from CaH2 and stored at −15 °C. The
distilled pyrrole was used prior to the appearance of any discoloration.
All other chemicals are reagent grade and were used as received.
Authentic phlorin 1,4 5-isocorrole 2,6 porphodimethene 3,6,33,34 and
corrole 535 used as standards for HPLC method development and
detector response calibration were prepared as described in the
literature. Porphyrin 4 was available commercially.
Both reactions were challenging. The level of phlorin in the
crude reaction mixtures appeared to be low (acetophenone
reaction) or absent (benzophenone reaction) as qualitatively
judged by TLC analysis. In addition, residual, excess ketone was
more difficult to remove from the crude reaction mixtures due
to the lower volatility of acetophenone and especially
benzophenone which could not be removed by evaporation.
Purification of phlorin 6 (derived from acetophenone) was
hampered by a more complicated reaction mixture, by the more
similar chromatographic retention of phlorin 6 to corrole 5, and
by the lower yield of phlorin 6. After chromatography on silica
gel, neutral alumina, and silica gel again, phlorin 6 of modest
purity was isolated (64 mg, 5% yield). The reaction also
afforded porphyrin 4 (6% yield) and corrole 5 (4% yield).
Although phlorin 7 did not appear to be present in the reaction
involving benzophenone, the reaction mixture was subjected to
silica gel chromatography from which porphyrin 4 and corrole
5 were isolated in yields of 8% and 7%, respectively. No phlorin
7 was detected. The chromatographic purification of porphyrin
4 and corrole 5 was complicated by the presence of the excess
benzophenone. The excess benzophenone streaked down the
column negatively impacting the separation of porphyrin 4 and
corrole 5, and contaminating fractions with a low level of
benzophenone. A second silica column was required to separate
the porphyrin 4 and corrole 5 from each other and from the
remaining benzophenone.
HPLC Determination of Yields of Porphyrinoids 1−5.
Analytical-scale reactions of pyrrole with pentafluorobenzaldehyde
and acetone were monitored for the yield of phlorin 1, 5-isocorrole 2,
porphodimethene 3, porphyrin 4, and corrole 5 by adaptation of a
literature method for the analysis of phlorin 14 and 5-isocorrole 26 in
crude reaction mixtures. An aliquot (1.2 mL) of a condensation
reaction mixture was transferred by adjustable pipet to a 1.8 mL
microcentrifuge tube containing DDQ (2.00 mg, 0.00881 mmol), the
mixture was vortex mixed for ∼5 s, and centrifuged for ∼5 s.
[Microfuge tubes containing DDQ were prepared by dispensing 220
μL of a DDQ solution (40 mM in toluene) in each microfuge tube,
and evaporating the solvent under a vacuum at room temperature.]
Triethylamine (2 equiv relative to acid) was added, the mixture was
vortex mixed for ∼5 s, and centrifuged for ∼5 s. A portion of the
oxidized reaction mixture (0.70 mL) was transferred via adjustable
pipet to a Pasteur pipet filled two-thirds full with silica gel (∼1.5 g).
The sample was eluted with three 1 mL portions of CH2Cl2 dispensed
from a solvent pump, and solvent was driven off the silica pad with a
hand-held pipet tool. The eluent was transferred to an autosampler
vial. HPLC analysis was performed with an injection volume of 1 μL, a
normal phase silica column (Alltech, Altima, 5 μ, 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 portions
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 = 7−14 min, 2 mL/min; T = 14−15 min,
linear decrease to 1 mL/min. The phlorin 1, 5-isocorrole 2,
porphodimethene 3, porphyrin 4, and corrole 5 eluted at 8.3, 7.3,
7.5, 8.0, and 8.7 min, respectively. Detection was performed at 434 nm
(phlorin 1), 418 nm (5-isocorrole 2), 420 nm (porphodimethene 3),
410 nm (porphyrin 4), and 406 nm (corrole 5). The yields of
porphyrinoids 1−5 were determined from the peak area by application
of a detector response factor determined from the calibration of the
detector response for each compound. Representative chromatograms
and further details on the HPLC method development, control and
reproducibility experiments, detector response calibration, and the
calculation of porphyrinoid yields from HPLC peak areas may be
found in the Supporting Information.
It is not surprising that conditions identified for the
preparation of phlorin 1 from acetone did not provide similarly
good yields of phlorins 6 and 7 given sharp difference in
ketones. Two-step, one-flask syntheses of porphyrins can
require substantial refinement of reaction conditions based on
the nature of the aldehyde.31 Nonetheless, it is encouraging that
the reaction with acetophenone afforded some phlorin 6.
CONCLUSION
■
The two-step, one-flask reaction of pyrrole with pentafluoro-
benzaldehyde and acetone was investigated with the goal of
achieving a streamlined synthesis of calixphyrins such as
phlorin. Phlorin 1 was found to be a fairly ubiquitous product
of the reaction, with yields reaching 20−26% in analytical-scale
reactions monitored by HPLC. An isolated yield of phlorin 1 of
20−21% (249−268 mg) was obtained from three trials of the
reaction of pyrrole (10 mM), pentafluorobenzaldehyde (5.0
mM), and acetone (80 mM) mediated by TFA (215 mM). The
preparation of a phlorin in good yield from a two-step, one-
flask approach shows that it is possible to obtain meaningful
quantities of a calixphyrin in a streamlined fashion. This
observation provides encouragement for further studies of two-
step, one-flask reactions of pyrrole with an aldehyde and a
ketone. The scope of phlorins that can be directly prepared
from the methodology, the extent to which conditions can be
further refined to afford phlorins that are not prepared in good
yield (or at all) from the present conditions, and the potential
to obtain other calixphyrins (e.g., 5-isocorrole) in good yield
from a wider survey of reaction conditions could be further
investigated. In the meantime, improved access to phlorin 1 will
facilitate studies of phlorin properties, reactivity, and metal
chemistry.
General Procedure for Analytical-Scale Reactions, Exempli-
fied by the Reaction of Pyrrole (10 mM), Pentafluorobenz-
aldehyde (5.0 mM), and Acetone (80 mM) with Catalysis by
TFA (215 mM). The reaction was performed at room temperature in
a tightly capped 20 mL vial, and stirred with a micro stir bar. CH2Cl2
(10 mL) was dispensed into the vial, followed by pyrrole (50 μL, 0.10
mmol, 2.0 M solution in CH2Cl2), pentafluorobenzaldehyde (25 μL,
0.050 mmol, 2.0 M solution in CH2Cl2), and acetone (59 μL, 0.80
mmol). After stirring briefly, TFA (165 μL, 2.15 mmol) was added.
The reactions were monitored by HPLC at 0.25, 1, and 4 h as
described above. TLC was performed on the crude, oxidized mixtures
[silica, CH2Cl2/hexanes (1:1)].
Survey of Acetone Concentration. Reactions were performed as
described in the general procedure, with an acetone concentration of
2.5, 5.0, 10, 20, 40, 80, 160, or 320 mM. Reactions with a low
H
DOI: 10.1021/acs.joc.6b00571
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
concentration of acetone (≤20 mM) utilized a 1.0 M solution of
acetone in CH2Cl2.
NMR spectrum). Crystallization from CH2Cl2/hexanes with gradual
evaporation of the CH2Cl2 (50−60 °C) gave fine, needle crystals that
were stored overnight at −15 °C. The crystals were collected by
vacuum filtration aided by rinsing with pentane, affording dark purple
Survey of Acid Catalysis Conditions. Reactions were performed
as described in the general procedure, with a TFA concentration of 10,
17, 30, 55, 90, 160, 215, 290, or 500 mM; or a BF3·OEt2 concentration
of 0.5, 1.0, 1.9, 3.6, 7.0, 10, 13, 26, or 50 mM. Reactions with a low
concentration of TFA (≤17 mM) utilized a 2.0 M solution of TFA in
CH2Cl2. Reactions with a low concentration of BF3·OEt2 (≤1.9 mM
and 3.6−13 mM) utilized a 0.25 or 1.0 M solution of BF3·OEt2 in
CH2Cl2, respectively.
Survey of Oxidation Conditions Exemplified by the Reaction
of Pyrrole (10 mM), Pentafluorobenzaldehyde (5.0 mM), and
Acetone (80 mM) with Catalysis by TFA (215 mM). The reaction
was performed at room temperature in tightly capped 20 mL vial, and
stirred with a micro stir bar. CH2Cl2 (15 mL) was dispensed into the
vial, followed by pyrrole (75 μL, 0.15 mmol, 2.0 M solution in
CH2Cl2), pentafluorobenzaldehyde (38 μL, 0.075 mmol, 2.0 M
solution in CH2Cl2), and acetone (88 μL, 1.2 mmol). After stirring
briefly, TFA (247 μL, 3.23 mmol) was added. At 1 h, aliquots (1.2
mL) of the reaction were transferred to microfuge tubes containing
0.25, 0.5, 1.0, 2.0, 3.0, 4.0, 8.0, or 16 mg of DDQ. The oxidized
mixtures were prepared for HPLC analysis and analyzed as described
above. TLC was performed on the crude, oxidized mixtures [silica,
CH2Cl2/hexanes (1:1)].
1
crystals of phlorin 1 (249−268 mg, 20−21%). H NMR (CDCl3 and
DMSO-d6), UV−vis (CH2Cl2), and LD-MS analyses were consistent
with published values.4
5,5-Dimethyl-10,15,20-tris(pentafluorophenyl)phlorin 1
[Reaction of Pyrrole (10 mM), Pentafluorobenzaldehyde (7.5
mM), and Acetone (160 mM) with Catalysis by TFA (215 mM)].
To a 2-L round-bottom flask containing a stir bar were added CH2Cl2
(600 mL), pyrrole (0.416 mL, 6.00 mmol), pentafluorobenzaldehyde
(0.556 mL, 4.50 mmol), and acetone (7.05 mL, 96.0 mmol). After
stirring briefly, the reaction was initiated by the addition of TFA (9.88
mL, 129 mmol). The flask was tightly capped, and the reaction was
stirred at room temperature. At a reaction time of 30 min the reaction
was monitored by HPLC as described above, and the yield of phlorin 1
was found to be satisfactory. At a reaction time of 30 min, the reaction
mixture was oxidized by the addition of DDQ (1.00 g, 4.41 mmol) at
room temperature. After ∼1 min, triethylamine (19.8 mL, 142 mmol,
1.1 equiv relative to the acid) was added and the mixture was stirred at
room temperature. At oxidation reaction times of 30 min and 1 h, an
aliquot (1.2 mL) of the reaction mixture was removed for HPLC
analysis. At 1 h, the reaction mixture was filtered through a pad of silica
gel and eluted with CH2Cl2 (∼400 mL) until the eluant was no longer
green. Further purification of the impure product was carried out by
silica gel chromatography as described above. Fractions containing a
mixture of porphodimethene 3 (12 mg, 0.7%) and porphyrin 4 (65
mg, 5.9%) were collected and analyzed as described above. The
fractions containing phlorin (355 mg) were purified further by
chromatography on neutral alumina as described above, affording
phlorin (285 mg) in good purity (see the Supporting Information for
the HPLC chromatogram, UV−vis spectrum, and 1H NMR spectrum).
Crystallization from CH2Cl2/hexanes gave dark purple crystals of
Reaction Time-Course Experiments. Reaction monitoring as a
function of time was performed as described above for the general
procedure with the exception of using a 20 mL reaction volume in a
tightly capped, 50 mL pear shaped flask. The reactions were monitored
by HPLC as described above at 1 min, 2 min, 4 min, 8 min, 15 min, 30
min, 1, 2, 4, 8, and 24 h. TLC was performed on the crude, oxidized
mixtures [silica, CH2Cl2/hexanes (1:1)].
5,5-Dimethyl-10,15,20-tris(pentafluorophenyl)phlorin 1,
[Reaction of Pyrrole (10 mM), Pentafluorobenzaldehyde (5.0
mM), and Acetone (80 mM) with Catalysis by TFA (215 mM)].
To a 2-L round-bottom flask containing a stir bar were added CH2Cl2
(900 mL), pyrrole (0.624 mL, 9.00 mmol), pentafluorobenzaldehyde
(0.556 mL, 4.50 mmol), and acetone (5.29 mL, 72.0 mmol). After
stirring briefly, the reaction was initiated by the addition of TFA (14.8
mL, 194 mmol). The flask was tightly capped, and the reaction was
stirred at room temperature. At reaction times of 30 min and 1 h, the
reaction was monitored by HPLC as described above, and the yield of
phlorin 1 was found to be satisfactory. At a reaction time of 1 h, the
reaction mixture was oxidized by the addition of DDQ (1.50 g, 6.61
mmol) at room temperature. After ∼1 min, triethylamine (29.7 mL,
213 mmol, 1.1 equiv relative to the acid) was added and the mixture
was stirred at room temperature. At oxidation reaction times of 30 min
and 1 h an aliquot (1.2 mL) of the reaction mixture was removed for
HPLC analysis. At 1 h, the reaction mixture was filtered through a pad
of silica gel and eluted with CH2Cl2 (∼400 mL) until the eluant was
no longer green. The filtrate was concentrated to a dark, viscous liquid.
The impure phlorin was diluted with CH2Cl2 (50 mL), adsorbed onto
silica gel (40 g), evaporated to dryness, and subjected to
chromatography [silica, CH2Cl2/hexanes (1:5), (1:3), (1:2), (1:1)].
With a solvent composition of CH2Cl2/hexanes (1:3), fractions
containing a mixture of porphodimethene 3 and porphyrin 4 eluted,
followed closely by fractions containing a mixture of largely porphyrin
4 and corrole 5. The quantity of porphodimethene 3 (20 mg, 1.3%),
porphyrin 4 (48 mg, 4.4%), and corrole 5 (7 mg, 0.6%) were estimated
based on the dry mass of the collected fractions and the integrated
peak areas of the 1H NMR spectra of the fractions (see the Supporting
Information). Upon increasing the polarity of the solvent to CH2Cl2/
hexanes (1:2), fractions containing phlorin 1 were collected, and
evaporated to dryness affording 354−377 mg of impure phlorin. The
impure phlorin was redissolved in CH2Cl2 (25 mL), adsorbed onto
neutral alumina (15 g), evaporated to dryness, and subjected to
chromatography [neutral alumina, CH2Cl2/hexanes (1:4)]. After
elution of a bright red band of unknown structure, a dark green
band containing phlorin 1 was collected, and evaporated to dryness
affording phlorin (274−297 mg) of good purity (see the Supporting
1
phlorin 1 (212 mg, 17%). H NMR (CDCl3 and DMSO-d6), UV−vis
(CH2Cl2), and LD-MS analyses were consistent with published
values.4
5,5-Dimethyl-10,15,20-tris(pentafluorophenyl)phlorin 1
[Reaction of Pyrrole (10 mM), Pentafluorobenzaldehyde (2.5
mM), and Acetone (80 mM) with Catalysis by TFA (215 mM)].
To a 2-L round-bottom flask containing a stir bar were added CH2Cl2
(1.8 L), pyrrole (1.25 mL, 18.0 mmol), pentafluorobenzaldehyde
(0.556 mL, 4.50 mmol), and acetone (10.6 mL, 144 mmol). After
stirring briefly, the reaction was initiated by the addition of TFA (29.6
mL, 387 mmol). The flask was tightly capped, and the reaction was
stirred at room temperature. At reaction times of 30 min and 1 h the
reaction was monitored by HPLC as described above, and the yield of
phlorin 1 was found to be satisfactory. At a reaction time of 1 h, the
reaction mixture was oxidized by the addition of DDQ (3.00 g, 13.2
mmol) at room temperature. After ∼1 min, triethylamine (59.4 mL,
426 mmol, 1.1 equiv relative to the acid) was added and the mixture
was stirred at room temperature. At oxidation reaction times of 30 min
and 1 h an aliquot (1.2 mL) of the reaction mixture was removed for
HPLC analysis. At 1 h, the reaction mixture was filtered through a pad
of silica gel and eluted with CH2Cl2 (∼400 mL) until the eluant was
no longer green. Further purification of the impure product was
carried out by silica gel chromatography as described above. Fractions
containing a mixture of porphodimethene 3 (34 mg, 2.2%) and
porphyrin 4 (14 mg, 1.3%) were collected and analyzed as described
above. The fractions containing phlorin (624 mg) were purified further
by chromatography on neutral alumina as described above, affording
phlorin (276 mg) in good purity (see the Supporting Information for
the HPLC chromatogram, UV−vis spectrum, and 1H NMR spectrum).
Crystallization from CH2Cl2/hexanes gave dark purple crystals of
1
phlorin 1 (230 mg, 18%). H NMR (CDCl3 and DMSO-d6), UV−vis
(CH2Cl2), and LD-MS analyses were consistent with published
values.4
5-Methyl-5-phenyl-10,15,20-tris(pentafluorophenyl)phlorin
6. To a 2-L round-bottom flask containing a stir bar were added
CH2Cl2 (900 mL), pyrrole (0.624 mL, 9.00 mmol), pentafluoro-
1
Information for the HPLC chromatogram, UV−vis spectrum, and H
I
DOI: 10.1021/acs.joc.6b00571
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
benzaldehyde (0.556 mL, 4.50 mmol), and acetophenone (8.40 mL,
72.0 mmol). After stirring briefly, the reaction was initiated by the
addition of TFA (14.8 mL, 194 mmol). The flask was tightly capped,
and the reaction was stirred at room temperature. At a reaction time of
1 h, the reaction mixture was oxidized by the addition of DDQ (1.50 g,
6.61 mmol) at room temperature. After ∼1 min, triethylamine (29.7
mL, 213 mmol, 1.1 equiv relative to the acid) was added and the
mixture was stirred at room temperature. At an oxidation reaction time
of 1 h, the reaction mixture was filtered through a pad of silica gel and
eluted with CH2Cl2 (∼450 mL) until the eluant was no longer green.
The filtrate was concentrated to a dark, viscous liquid. The sample was
placed under a vacuum at 60 °C to remove as much of the excess
acetophenone as possible. The impure phlorin was diluted with
CH2Cl2 (50 mL), adsorbed onto silica gel (40 g), evaporated to
dryness, and subjected to chromatography [silica, CH2Cl2/hexanes
(1:5), (1:3), (1:2), (1:1)]. With a solvent mixture of CH2Cl2/hexanes
(1:3), fractions containing porphyrin 4 (64 mg, 5.8%) eluted, followed
closely by fractions containing an impure mixture of corrole 5 and
phlorin 6. The impure mixture of corrole and phlorin was redissolved
in CH2Cl2 (25 mL), adsorbed onto neutral alumina (15 g), evaporated
to dryness, and subjected to chromatography [neutral alumina,
CH2Cl2/hexanes (1:5), (1:4), CH2Cl2, and CH2Cl2/MeOH (20:1)].
A dark green band containing phlorin 6 eluted rapidly. The band was
collected and evaporated to dryness affording phlorin (85 mg) of
modest purity. Corrole 5 was retained strongly by the neutral alumina,
requiring elution with CH2Cl2 and CH2Cl2/MeOH (20:1). Fractions
containing corrole were collected and evaporated to dryness affording
corrole (47 mg, 4.0%) with good purity (see the Supporting
Information for UV−vis and 1H NMR spectra). Due to the low
quantity of phlorin, crystallization from CH2Cl2/hexanes was not
successful. Instead, the phlorin was further purified by silica gel
chromatography. The impure phlorin was dissolved with a minimal
volume of CH2Cl2/hexanes (1:5), and passed through a silica gel
column, eluting with CH2Cl2/hexanes (1:3). Upon evaporation of
solvent, phlorin 6 (64 mg, 4.7%) contaminated with a low level of
impurities was isolated (see the Supporting Information for UV−vis
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.6b00571.
■
S
Discussion of HPLC method development; control and
reproducibility experiments, and detector response
calibration; calculation of porphyrinoid yield from
HPLC peak area; plots of porphyrinoid yields from
analytical-scale reactions; summary of conditions afford-
ing porphodimethene 3 in yields exceeding 2%; HPLC
chromatograms, UV−vis spectra, and 1H NMR spectra of
phlorin 1 isolated from preparative-scale reactions;
estimation of the yield of porphodimethene 3, porphyrin
4, and corrole 5 from preparative-scale reactions; and
1
UV−vis spectra and H NMR spectra of 5-methyl-5-
phenylphlorin 6 and corrole 5 from the reaction of
pyrrole, pentafluorobenzaldehyde, and acetophenone
mediated by TFA. (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: ggeier@colgate.edu.
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank David B. Long for the preparation of corrole 5 used
as a standard for HPLC method development and detector
response calibration. Summer research fellowships for DK were
provided by the Justus ’43 and Jayne Schlichting Student
Research Fund administered by Colgate University.
1
1
and H NMR spectra). H NMR (CDCl3), UV−vis (CH2Cl2), and
REFERENCES
LD-MS analyses were consistent with published values.5
■
(1) Sessler, J. L.; Zimmerman, R. S.; Bucher, C.; Kral, V.; Andrioletti,
B. Pure Appl. Chem. 2001, 73, 1041−1057.
Attempted Synthesis of 5,5-Diphenyl-10,15,20-tris-
(pentafluorophenyl)phlorin 7. To a 2-L round-bottom flask
containing a stir bar were added CH2Cl2 (900 mL), pyrrole (0.624
mL, 9.00 mmol), pentafluorobenzaldehyde (0.556 mL, 4.50 mmol),
and benzophenone (13.1 g, 72.0 mmol). After stirring briefly, the
reaction was initiated by the addition of TFA (14.8 mL, 194 mmol).
The flask was tightly capped, and the reaction was stirred at room
temperature. At a reaction time of 1 h, the reaction mixture was
oxidized by the addition of DDQ (1.50 g, 6.61 mmol) at room
temperature. After ∼1 min, triethylamine (29.7 mL, 213 mmol, 1.1
equiv relative to the acid) was added and the mixture was stirred at
room temperature. TLC analysis [silica, CH2Cl2/hexanes (1:1)]
revealed the presence of porphyrin 4 and corrole 5, but not phlorin
7. At an oxidation reaction time of 1 h, the reaction mixture was
filtered through a pad of silica gel and eluted with CH2Cl2 (∼500 mL).
The filtrate was concentrated to a dark, viscous liquid. The sample was
diluted with CH2Cl2 (50 mL), adsorbed onto silica gel (40 g),
evaporated to dryness, and subjected to chromatography [silica,
CH2Cl2/hexanes (1:5), (1:3), (1:2)]. With a solvent mixture of
CH2Cl2/hexanes (1:3), fractions containing a mixture of porphyrin 4
and corrole 5 contaminated with benzophenone eluted quickly. No
phlorin 7 was observed upon further elution of the column. To
determine the yield of porphyrin 4 and corrole 5, the mixture of
porphyrin 4 and corrole 5 contaminated with benzophenone was
subjected to a second silica gel column, again eluting with CH2Cl2/
hexanes (1:5), (1:3), (1:2). With a lower initial level of impurity, the
porphyrin (87 mg, 8.0%) cleanly eluted first followed by the corrole
(2) Dehaen, W. Top. Heterocycl. Chem. 2010, 24, 75−102.
(3) LeSaulnier, T. D.; Graham, B. W.; Geier, G. R., III Tetrahedron
Lett. 2005, 46, 5633−5637.
(4) O’Brien, A. Y.; McGann, J. P.; Geier, G. R., III J. Org. Chem. 2007,
72, 4084−4092.
(5) Bruce, A. M.; Weyburne, E. S.; Engle, J. T.; Ziegler, C. J.; Geier,
G. R., III J. Org. Chem. 2014, 79, 5664−5672.
(6) Flint, D. L.; Fowler, R. L.; LeSaulnier, T. D.; Long, A. C.;
O’Brien, A. Y.; Geier, G. R., III J. Org. Chem. 2010, 75, 553−563.
(7) Lee, C. H.; Li, F.; Iwamoto, K.; Dadok, J.; Bothner-By, A. A.;
Lindsey, J. S. Tetrahedron 1995, 51, 11645−11672.
(8) Rao, P. D.; Dhanalekshmi, S.; Littler, B. J.; Lindsey, J. S. J. Org.
Chem. 2000, 65, 7323−7344.
(9) Geier, G. R., III; Callinan, J. B.; Rao, D. P.; Lindsey, J. S. J.
Porphyrins Phthalocyanines 2001, 5, 810−823.
(10) Jeandon, C.; Krattinger, B.; Ruppert, R.; Callot, H. J. Inorg.
Chem. 2001, 40, 3149−3153.
(11) Costa, R.; Geier, G. R., III; Ziegler, C. J. Dalton Trans. 2011, 40,
4384−4386.
(12) Pistner, A. J.; Yap, G. P. A.; Rosenthal, J. J. Phys. Chem. C 2012,
116, 16918−16924.
(13) Pistner, A. J.; Lutterman, D. A.; Ghidiu, M. J.; Ma, Y.-Z.;
Rosenthal, J. J. Am. Chem. Soc. 2013, 135, 6601−6607.
(14) Pistner, A. J.; Lutterman, D. A.; Ghidiu, M. J.; Walker, E.; Yap,
G. P. A.; Rosenthal, J. J. Phys. Chem. C 2014, 118, 14124−14132.
(15) Nieto-Pescador, J.; Abraham, B.; Pistner, A. J.; Rosenthal, J.;
Gundlach, L. Phys. Chem. Chem. Phys. 2015, 17, 7914−7923.
(16) Rothemund, P.; Mennotti, A. R. J. Am. Chem. Soc. 1941, 63,
267−270.
1
(88 mg, 7.3%) with no contamination from benzophenone. H NMR
(CDCl3), UV−vis (CH2Cl2), and LD-MS analyses were consistent
with published values.22
J
DOI: 10.1021/acs.joc.6b00571
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(17) Adler, A. D.; Longo, F. R.; Finarelli, J. D.; Goldmacher, J.;
Assour, J.; Korsakoff, L. J. Org. Chem. 1967, 32, 476.
(18) Lindsey, J. S.; Schreiman, I. C.; Hsu, H. C.; Kearney, P. C.;
Marguerettaz, A. M. J. Org. Chem. 1987, 52, 827−836.
(19) Furuta, H.; Asano, T.; Ogawa, T. J. Am. Chem. Soc. 1994, 116,
767−768.
(20) Chmielewski, P. J.; Latos-Grazynski, L.; Rachlewicz, K.; Glowiac,
T. Angew. Chem., Int. Ed. Engl. 1994, 33, 779−781.
(21) Geier, G. R., III; Haynes, D. M.; Lindsey, J. S. Org. Lett. 1999, 1,
1455−1458.
(22) Gross, Z.; Galili, N.; Saltsman, I. Angew. Chem., Int. Ed. 1999, 38,
1427−1429.
(23) Paolesse, R.; Jaquinod, L.; Nurco, D. J.; Mini, S.; Sagone, F.;
Boschi, T.; Smith, K. M. Chem. Commun. 1999, 1307−1308.
(24) Lindsey, J. S. In The Porphyrin Handbook; Kadish, K. M., Smith,
K., Guilard, R., Eds.; Academic Press: San Diego, CA, 2000; Vol. 1, pp
45−118.
(25) Furuta, H.; Ishizuka, T.; Osuka, A.; Uwatoko, Y.; Ishikawa, Y.
Angew. Chem., Int. Ed. 2001, 40, 2323−2325.
(26) Yang, J.; Jiang, J.; Fang, W.; Kai, X.; Hu, C.; Yang, Y. J.
Porphyrins Phthalocyanines 2011, 15, 197−201.
(27) Xu, G.; Jiang, J.; Hu, C. J. Porphyrins Phthalocyanines 2013, 17,
392−398.
(28) Kai, X.; Jiang, J.; Hu, C. Chem. Commun. 2012, 48, 4302−4304.
(29) Feng, Z.; Kai, X.; Jiang, J.; Hu, C. Chin. J. Chem. 2012, 30,
1715−1721.
(30) Geier, G. R., III; Chick, J. F. B.; Callinan, J. B.; Reid, C. G.;
Auguscinski, W. P. J. Org. Chem. 2004, 69, 4159−4169.
(31) Geier, G. R., III; Lindsey, J. S. Tetrahedron 2004, 60, 11435−
11444.
(32) Geier, G. R., III; Lindsey, J. S. J. Chem. Soc., Perkin Trans. 2
2001, 677−686.
(33) Bernatkova, M.; Andrioletti, B.; Kral, V.; Rose, E.; Vaissermann,
J. J. Org. Chem. 2004, 69, 8140−8143.
(34) Dolensky, B.; Kroulik, J.; Kral, V.; Sessler, J. L.; Dvorakova, H.;
Bour, P.; Bernatkova, M.; Bucher, C.; Lynch, V. J. Am. Chem. Soc.
2004, 126, 13714−13722.
(35) Wasbotten, I. H.; Wondimagegn, T.; Ghosh, A. J. Am. Chem.
Soc. 2002, 124, 8104−8116.
K
DOI: 10.1021/acs.joc.6b00571
J. Org. Chem. XXXX, XXX, XXX−XXX