N-confused tetraphenylporphyrin and tetraphenylsapphyrin formation in one-flask syntheses of tetraphenylporphyrin

Article abstract of DOI:10.1021/jo982068r

The yields of tetraphenylporphyrin (TPP), N-confused tetraphenylporphyrin (NC-TPP), and tetraphenylsapphyrin (TPS) were examined under diverse conditions in the one-flask room-temperature reaction of pyrrole and benzaldehyde. The conditions examined included TFA and BF₃- etherate catalysis, acid concentrations, pyrrole and benzaldehyde concentrations, pyrrole: benzaldehyde ratios, the addition of salt, and reaction times. The quantities of TPP, NC-TPP, and TPS were determined by HPLC. The yield of NC-TPP mirrored that of TPP during systematic variation of reaction parameters but was consistently lower (≤ 7.5%). The rate of NC-TPP formation also resembled that of TPP. Excess pyrrole was not required to form NC-TPP. About 2-fold more NC-TPP was obtained with BF₃-etherate compared with TFA catalysis. TPS was observed in only a few of the reactions examined. The yield of TPS peaked early and declined well before the maximum yields of TPP and NC-TPP were reached. The highest yield of TPS observed was 1.2%. TPS was produced in higher yields by TFA compared with BF₃-etherate catalysis. Additional pigments with chromatographic properties spanning the range of TPP and NC-TPP were observed in very low yield but were not characterized. N- confused tetramesitylporphyrin was isolated from the reaction of mesitaldehyde and pyrrole under BF₃-ethanol cocatalysis. Examination of the Adler reaction of benzaldehyde and pyrrole revealed the presence of NC-TPP and TPS in the filtrate but not in the crystalline TPP material. The substantial chromatographic differences between the porphyrins and N-confused porphyrins make for straightforward isolation of the purified porphyrin devoid of contaminating N-confused porphyrin. These studies indicate that the scope of conditions yielding NC-TPP is extremely broad.

Full text of DOI:10.1021/jo982068r

1596
J . Org. Chem. 1999, 64, 1596-1603
N-Con fu sed Tetr a p h en ylp or p h yr in a n d Tetr a p h en ylsa p p h yr in
F or m a tion in On e-F la sk Syn th eses of Tetr a p h en ylp or p h yr in
G. Richard Geier, III and J onathan S. Lindsey*
Department of Chemistry, North Carolina State University, Raleigh, North Carolina 27695-8204
Received October 14, 1998
The yields of tetraphenylporphyrin (TPP), N-confused tetraphenylporphyrin (NC-TPP), and
tetraphenylsapphyrin (TPS) were examined under diverse conditions in the one-flask room-
temperature reaction of pyrrole and benzaldehyde. The conditions examined included TFA and
BF3-etherate catalysis, acid concentrations, pyrrole and benzaldehyde concentrations, pyrrole:
benzaldehyde ratios, the addition of salt, and reaction times. The quantities of TPP, NC-TPP, and
TPS were determined by HPLC. The yield of NC-TPP mirrored that of TPP during systematic
variation of reaction parameters but was consistently lower (e7.5%). The rate of NC-TPP formation
also resembled that of TPP. Excess pyrrole was not required to form NC-TPP. About 2-fold more
NC-TPP was obtained with BF3-etherate compared with TFA catalysis. TPS was observed in only
a few of the reactions examined. The yield of TPS peaked early and declined well before the
maximum yields of TPP and NC-TPP were reached. The highest yield of TPS observed was 1.2%.
TPS was produced in higher yields by TFA compared with BF3-etherate catalysis. Additional
pigments with chromatographic properties spanning the range of TPP and NC-TPP were observed
in very low yield but were not characterized. N-confused tetramesitylporphyrin was isolated from
the reaction of mesitaldehyde and pyrrole under BF3-ethanol cocatalysis. Examination of the Adler
reaction of benzaldehyde and pyrrole revealed the presence of NC-TPP and TPS in the filtrate but
not in the crystalline TPP material. The substantial chromatographic differences between the
porphyrins and N-confused porphyrins make for straightforward isolation of the purified porphyrin
devoid of contaminating N-confused porphyrin. These studies indicate that the scope of conditions
yielding NC-TPP is extremely broad.
In tr od u ction
conditions of equimolar pyrrole and benzaldehyde in
either refluxing propionic acid4 or CH2Cl2 at room tem-
perature with TFA or BF3-etherate catalysis.5
In 1994, the groups of Latos-Grazynski1 and Furuta2
independently described the isolation of a porphyrin
isomer in about 5% yield from the acid-catalyzed con-
densation of pyrrole and an aryl aldehyde followed by
oxidation. The isomer, a 2-aza-21-carba-5,10,15,20-tet-
raarylporphyrin (also called carbaporphyrin or N-con-
fused or N-inverted porphyrin), possesses seven R-link-
ages and one â-pyrrole linkage rather than all R-linkages
as in the parent porphyrin. In both cases, the condensa-
tion conditions were somewhat unusual. Latos-Grazynski
et al. used 70 mM pyrrole and 40 mM p-tolualdehyde in
CH2Cl2 at room temperature with 8 mM BF3-etherate
catalysis,1 whereas Furuta et al. used equimolar pyrrole
and benzaldehyde in the solvent tert-butyl alcohol/CH2-
Cl2 (1:1) at room temperature with HBr catalysis.2 In
1995, Latos-Grazynski’s group used 24 mM pyrrole and
8 mM benzaldehyde with 3 mM BF3-etherate catalysis
and isolated tetraphenylsapphyrin in 1.1% yield.3 The
reaction of benzaldehyde and pyrrole yielding tetraphe-
nylporphyrin (TPP), N-confused TPP (NC-TPP), and
tetraphenylsapphyrin (TPS) is shown in Scheme 1. No
reports have appeared in which the yields of all three
macrocycles (TPP, NC-TPP, and TPS) have been quan-
titated under conditions using excess pyrrole1,3 or HBr
catalysis in protic solvents2 or under the more typical
The isolation of NC-TPP and TPS from the reaction of
benzaldehyde and pyrrole raised a number of important
questionsbeyondgeneralinterest in thenewmacrocycles.6-8
(1) How widespread are NC-TPP and TPS in one-flask
porphyrin syntheses? (2) Are ostensibly pure porphyrin
preparations contaminated with N-confused porphyrin
and sapphyrin? (3) Are other porphyrinic macrocycles
readily isolable? (4) How does the presence of NC-TPP
and TPS affect our understanding of the reaction mech-
anism?
In this paper, we describe our studies aimed at ad-
dressing these questions. We have investigated the effects
of numerous reaction parameters on the yields of TPP,
NC-TPP, and TPS obtained in the one-flask room-
temperature porphyrin synthesis. The parameters stud-
ied include TFA and BF3-etherate catalysis, acid concen-
tration,reactant (pyrrole and benzaldehyde)concentrations,
(4) Adler, A. D.; Longo, F. R.; Finarelli, J . D.; Goldmacher, J .; Assour,
J .; Korsakoff, L. J . Org. Chem. 1967, 32, 476.
(5) Lindsey, J . S.; Schreiman, I. C.; Hsu, H. C.; Kearney, P. C.;
Marguerettaz, A. M. J . Org. Chem. 1987, 52, 827-836.
(6) (a) Latos-Grazynski, L.; Chmielewski, P. J . New J . Chem. 1997,
27, 691-700. (b) Chmielewski, P. J .; Latos-Grazynski, L. Inorg. Chem.
1997, 36, 840-845. (c) Chmielewski, P. J .; Latos-Grazynski, L.;
Glowiak, T. J . Am. Chem. Soc. 1996, 118, 5690-5701. (d) Chmielewski,
P. J .; Latos-Grazynski, L. J . Chem. Soc., Perkin Trans. 2 1995, 503-
509.
(7) (a) Ishikawa, Y.; Yoshida, I.; Akaiwa, K.; Koguchi, E.; Sasaki,
T.; Furuta, H. Chem. Lett. 1997, 453-454. (b) Ariga, K.; Kunitake, T.;
Furuta, H. J . Chem. Soc., Perkin Trans. 2 1996, 667-672.
(8) Ghosh, A. Angew. Chem., Int. Ed. Engl. 1995, 34, 1028-1030.
(1) Chmielewski, P. J .; Latos-Grazynski, L.; Rachlewicz, K.; Glowiak,
T. Angew. Chem., Int. Ed. Engl. 1994, 33, 779-781.
(2) Furuta, H.; Asano, T.; Ogawa, T. J . Am. Chem. Soc. 1994, 116,
767-768.
(3) Chmielewski, P. J .; Latos-Grazynski, L.; Rachlewicz, K. Chem.
Eur. J . 1995, 1, 68-73.
10.1021/jo982068r CCC: $18.00 © 1999 American Chemical Society
Published on Web 02/17/1999

TPS and NC-TPP Formation in Syntheses of TPP
J . Org. Chem., Vol. 64, No. 5, 1999 1597
chromatography was done using silica gel (Baker, 40 µm) or
Brockman basic alumina (Fisher, 60-325 mesh, activity I)
deactivated to activity II. Activity II alumina was obtained by
adding 3 mL H2O to 100 g alumina with intermittent shaking.
All other materials and solvents were used as received.
Isola tion of TP P , NC-TP P , a n d TP S for HP LC Ca li-
br a tion . The one-flask method of Latos-Grazynski reported
for the synthesis of TPS3 was used with slight modifications.
Samples of benzaldehyde (6.1 mmol) and pyrrole (18.2 mmol)
were added to CH2Cl2 (750 mL), then BF3-etherate was added
(2.3 mmol), and the solution was stirred for 1 h at room
temperature. p-Chloranil (6.1 mmol) was then added, and the
mixture was refluxed for 1 h. The mixture was concentrated
and then purified by column chromatography. Column chro-
matography on basic alumina (activity grade II) with CH2Cl2
gave a mixture of TPP, TPS, NC-TPP, and a variety of other
pigments. (Note that higher activity basic alumina produced
strong retention of TPS and NC-TPP, whereas lower activity
basic alumina resulted in poor retention of undesired, more
polar materials.) A second basic alumina column chromatog-
raphy was performed with hexanes and CH2Cl2 (beginning
with 25% CH2Cl2 and ending with 100% CH2Cl2, increasing
at about 25% intervals). TPP eluted first with 25% CH2Cl2,
TPS eluted with about 50% CH2Cl2, and NC-TPP eluted with
about 75% CH2Cl2. TPP and NC-TPP eluted as dark bands,
whereas TPS did not produce an obvious band. A variety of
other pigments also eluted between TPP and NC-TPP. Frac-
tions containing TPP and NC-TPP were identified by UV-vis
spectroscopy and comparison to published spectra.2 Detection
of TPS-containing fractions by UV-vis spectroscopy was
difficult as a result of the low TPS concentration; small
shoulders at 491 and 517 nm on the main absorption band
(centered at 420 nm) were the only indication.3 TPP was
further purified by silica gel chromatography with CH2Cl2, and
NC-TPP was further purified by silica gel chromatography
with CH2Cl2 containing 1% triethylamine (TEA). TPP and NC-
TPP were each recrystallized from methanol and CH2Cl2. The
recrystallization of NC-TPP resulted in poor recovery. TPS was
further purified by silica gel chromatography with CH2Cl2
containing TEA. This purification required close attention to
the concentration of TEA, with 0.5-2% being typical. The
chromatography was monitored by UV-vis spectroscopy. TPS
was not recrystallized owing to the small amount of isolated
material. TPP was isolated in 10% yield (90 mg), NC-TPP in
3% yield (30 mg), and TPS was isolated in about 0.5% yield (4
mg). The yield of TPS was slightly lower than the literature
report of 1.1%.3 Analytical data for NC-TPP and TPS (UV-
vis, 1H and 13C NMR, and LD-MS) were identical to published
data.1-3 The NC-TPP was identical to an authentic sample
provided to us by Latos-Grazynski’s group.
Gen er a l Rea ction Con d ition s. CH2Cl2, pyrrole, and ben-
zaldehyde were dispensed into 20-mL vials containing a micro
stir bar. The volume of CH2Cl2 was selected so that the final
volume of the reaction mixture would be 10 mL, assuming
additivity of volumes. Benzaldehyde and pyrrole were dis-
pensed as neat reagent or as 1 or 0.1 M stock solutions in
CH2Cl2, depending on the desired final concentration. For
experiments using salt, finely ground, dried NaCl was added.
The reactions were initiated by addition of TFA or BF3-
etherate (neat acid or 1 or 0.1 M stock solutions in CH2Cl2).
The reaction mixtures were stirred, tightly capped, and
shielded from light for the desired length of time. The reactions
were terminated by addition of DDQ. For 10 mM reactions,
DDQ (35 mg, 0.15 mmol) was used. For reactions at higher
concentration, 0.85 equiv of DDQ (based on limiting pyrrole
or benzaldehyde; 200 mg for 100 mM reactions) was used. The
workup for analysis began 1 min after DDQ addition. One-
hour reactions were done in parallel at 12 min intervals.
P r ea n a lysis Rea ction Wor k u p . One minute after DDQ
oxidation, the samples were basified by addition of triethy-
lamine (8-fold molar excess relative to acid). Addition of
triethylamine was crucial for accurate detection of NC-TPP,
otherwise NC-TPP was removed during the pre-HPLC sample
cleanup (see Supporting Information). Strongly adsorbing
black materials were removed prior to HPLC sample analysis
Sch em e 1
reactant ratios, the addition of salt, and reaction times.
These studies revealed that NC-TPP is formed under
essentially all conditions leading to TPP, whereas TPS
was found to a more limited extent. Accordingly, we have
delineated the chromatographic properties of NC-TPP to
avoid contamination of TPP with NC-TPP. To extend our
studies concerning the formation and isolation of N-
confused porphyrins, we have also identified and isolated
N-confused tetramesitylporphyrin (NC-TMP) formed from
the BF3-ethanol cocatalyzed reaction of mesitaldehyde
and pyrrole (1:1) leading to tetramesitylporphyrin (TMP).
Additionally, the Adler synthesis4 of TPP in refluxing
propionic acid was examined for NC-TPP and TPS
formation. Taken together, these studies establish the
scope of the formation of NC-TPP and TPS under diverse
reaction conditions.
Exp er im en ta l Section
Ma ter ia ls. Pyrrole (Acros) was distilled from calcium
hydride, stored at -15 °C, and used prior to any discoloration.
CH2Cl2 (Fisher, ACS grade) used as reaction solvent was
distilled from potassium carbonate and stored over 4 Å Linde
molecular sieves. Benzaldehyde (Aldrich, 99.5%), mesitalde-
hyde (Acros, 98%), TFA (Aldrich), BF3-etherate (Aldrich,
redistilled grade), and DDQ (Aldrich) were used as received.
Analytical glass-backed silica gel 60 plates (EM Science) and
alumina oxide 60 plates (E. Merck) were used for TLC. Column

1598 J . Org. Chem., Vol. 64, No. 5, 1999
Ta ble 1. Com p a r ison of Yield s Obta in ed u n d er Selected Rea ction Con d ition sa
Geier and Lindsey
entrya
acid
[PhCHO], mM
[pyrrole], mM
NaCl, equivb
%TPP (SD)c
%NC-TPP (SD)c
% TPS
1
2
3
4
5
6
7
8
9
BF3-etherate, 8 mM
BF3-etherate, 3 mM
BF3-etherate, 1 mM
BF3-etherate, 1 mM
BF3-etherate, 10 mM
BF3-etherate, 10 mM
TFA, 20 mM
TFA, 20 mM
TFA, 64 mM
TFA, 64 mM
40
8
10
70
24
10
17d (3.7)
9.2 (2.0)
27 (4.3)
49 (5.1)
24 (1.2)
42 (2.5)
39 (1.8)
39 (1.7)
28 (1.4)
28 (1.0)
6.1d(1.3)
2.9 (0.7)
5.1 (0.7)
5.4 (0.6)
5.6 (0.3)
5.8 (1.0)
2.2 (0.2)
2.0 (0.2)
4.0 (0.3)
3.9 (0.2)
0
0.6e
0
0
0
0
10
10
25
1
100
100
10
10
100
100
100
100
10
10
100
100
<0.2
0
<0.2
0
25
1
10
a
All reactions were performed in CH2Cl2 at room temperature for 1 h, followed by DDQ oxidation. The yields were determined by
HPLC. For reaction conditions employed in entries 3-10, see refs 5, 12, and 13. Equivalents relative to pyrrole. c The standard deviation
b
d
from the mean (generally 7-8 experiments). The reaction of p-tolualdehyde and pyrrole afforded the porphyrin in 16% yield and the
N-confused porphyrin in 4% yield (ref 1). e TPS has been isolated in 1.1% yield (ref 3).
by passing a carefully measured volume of crude reaction
mixture (1.5 mL for 10 mM reactions or 0.5 mL for reactions
at higher concentration) through a Pasteur pipet column
packed with 1.4 g of activity II basic alumina. The sample was
eluted with three 1-mL portions of CH2Cl2. Solvent was driven
off the column by application of mild pressure using a
handheld pipet tool. The eluant was transferred immediately
to an autosampler vial and capped. Sample workup was
performed in duplicate for each reaction. Agreement between
the duplicate samples was consistently good.
(7.3 mmol) in 98 mL of CHCl3 (ethanol-free, stabilized with
amylenes) was treated with ethanol (10 mmol) and BF3-
etherate (1.8 mmol) at room temperature. The solution was
stirred for 3 h, and then DDQ (5.7 mmol) was added followed
a few minutes later by triethylamine (14.3 mmol). The reaction
mixture was concentrated and purified by column chromatog-
raphy (identically as described for the isolation of TPP and
NC-TPP). TMP was isolated in 25% yield (350 mg), and NC-
TMP was isolated in 3% yield (45 mg). TMP was recrystallized
from methanol and CH2Cl2. NC-TMP was recrystallized from
methanol/CH2Cl2 and from methanol/benzene. In both cases,
the recovery of NC-TMP from the recrystallization was low
(50%) and did not improve the purity. The analytical data for
TMP were identical to published data.11 NC-TMP : UV-vis
λmax (CH2Cl2, nm, log ꢀ) 435 (5.15), 532 (4.01), 571 (3.83), 721
(3.91), λem (CH2Cl2, λexc ) 532 nm) 733, 811 nm; 1H NMR
(CDCl3, 300 MHz) δ -4.92 (s, 1 H), -2.07 (bs, 2 H), 1.91 (s, 12
H), 1.92 (s, 6 H), 1.94 (s, 6 H), 2.57 (s, 3 H), 2.59 (s, 3 H), 2.61
(s, 6 H), 7.24 (s, 4 H), 7.27 (s, 4 H), 8.33-8.41 (m, 4 H), 8.56
(d, 1 H, J ) 4.5 Hz), 8.58 (d, 1 H, J ) 4.5 Hz), 8.82 (s, 1 H);
13C NMR (CDCl3, 300 MHz) δ 21.6, 21.8, 21.9, 29.9, 101.3,
115.1, 116.8, 122.7, 124.4, 125.3, 127.5, 128.1, 134.0, 134.3,
134.6, 136.1, 136.6, 136.9, 137.1, 137.9, 138.3, 138.8, 139.1,
139.2, 139.5, 139.6, 139.9, 150.2, 156.3, 157.4, 157.6; C56H54N4
calcd exact mass 783.4427, obsd m/z 783.1 (LD-MS), obsd m/z
783.4437 (FAB-MS).
An a lysis. HPLC analysis was performed using a Hewlett-
Packard 1100 series HPLC with a quaternary pump, an
autosampler, thermostated column compartment at 25 °C, and
a diode array UV-vis detector. A silica gel analytical column
was used (Alltech, Altima, 4.6 mm × 250 mm) with an isocratic
solvent mixture of 92.5% hexanes and 7.5% acetone. The
hexanes solvent was 50% water saturated by mixing equal
volumes of dry hexanes and hexanes stored over water.9 The
solvent flow rate was controlled as follows: T ) 0-5 min, 1
mL/min; T ) 5-7 min, linear increase-2.5 mL/min; T )
7-10.5 min, 2.5 mL/min; T ) 10.5-11.5 min, linear decrease
to 1 mL/min; and T ) 11.5-13 min, 1 mL/min. The solvent
front occurred at 2.5 min, TPP eluted at 5.0 min, TPS eluted
at 7.9 min, and NC-TPP eluted at 9.9 min. Detection was
performed at the following wavelengths: TPP, 417 and 590
nm; TPS, 490 nm; and NC-TPP, 438 nm. Further HPLC
method development information and representative chro-
matograms are provided in the Supporting Information. The
detection limits for TPP, TPS, and NC-TPP were 3, 6, and 3
pmol, respectively. (For 10 mM reactions, these detection limits
correspond to the following yields: TPP ) 0.2%, TPS ) 0.4%,
and NC-TPP ) 0.2%. For 100 mM reactions, these detection
limits correspond to the following yields: TPP ) 0.2%, TPS )
0.3%, and NC-TPP ) 0.2%.) To provide a comparison for the
HPLC yield of TPP, samples were examined by UV-vis
absorption spectroscopy as described previously.5
An a lysis Ca libr a tion s. The HPLC was calibrated using
stock solutions of TPP, NC-TPP, and TPS. Retention times and
peak areas were very stable over multiple analyses. The
calibration was repeated periodically, and no significant
change was observed. For accurate TPP quantification, injec-
tion volumes needed to be 3 µL or less in order to avoid
splitting of the TPP peak due to the injection solvent. Reaction
workup calibration (correction for workup dilution) was done
by comparing HPLC analyses of a mixture of TPP and NC-
TPP for which a portion of the mixture was directly analyzed
and a second portion was subjected to the workup conditions
(vide supra). This calibration was performed multiple times
with excellent repeatability (2% variance). The ratio of TPP
and NC-TPP was not altered in the workup.
Resu lts a n d Discu ssion
Distr ibu tion of TP P , TP S, a n d NC-TP P Yield s.
The yields of TPP, TPS, and NC-TPP were examined as
a function of six reaction parameters: (1) TFA and BF3-
etherate, (2) acid concentration, (3) pyrrole and benzal-
dehyde (reactants) concentration, (4) pyrrole:benzalde-
hyde ratio, (5) salt, and (6) time. These parameters are
the most frequently examined variables in the optimiza-
tion of one-flask room-temperature porphyrin syntheses.
All reactions were performed in CH2Cl2 at room temper-
ature at 10-mL scale, and yields were determined by
HPLC. Selected results are presented here with results
for all experiments provided in the Supporting Informa-
tion.
The reaction of benzaldehyde and pyrrole was per-
formed under representative conditions reported in the
literature, and the yields of TPP, NC-TPP, and TPS were
determined (Table 1). The method of Latos-Grazynski,
using excess pyrrole, produced TPP, NC-TPP, and TPS
as expected (entries 1 and 2). The use of equimolar
pyrrole and benzaldehyde also produced appreciable
TMP a n d NC-TMP . Pyrrole and mesitaldehyde were
reacted following a refined method for the preparation of
TMP.10 A solution of pyrrole (7.3 mmol) and mesitaldehyde
(10) Wagner, R. W.; Li, F.; Du, H.; Lindsey, J . S. Org. Proc. Res.
(9) Snyder, L. R.; Kirkland, J . J .; Glajch, J . L. Practical HPLC
Method Development, Second Edition; J ohn Wiley and Sons: New York,
1997; pp 744-745.
Dev. 1993, 3, 28-37.
(11) Lindsey, J . S.; Wagner, R. W. J . Org. Chem. 1989, 54, 828-
836.

TPS and NC-TPP Formation in Syntheses of TPP
J . Org. Chem., Vol. 64, No. 5, 1999 1599
F igu r e 2. Yields of TPP and NC-TPP as a function of reactant
F igu r e 1. Yields of TPP and NC-TPP as a function of acid
concentration. All reactions were performed for 1 h.
concentration. All reactions were performed for 1 h.
amounts of NC-TPP (up to 5.8%) but little or no TPS
(entries 3-10). The ratio of TPP:NC-TPP ranged from
3:1 to 20:1. Larger amounts of NC-TPP were generally
observed with BF3-etherate than with TFA catalysis. NC-
TPP is clearly a byproduct of the acid-catalyzed conden-
sation of pyrrole and benzaldehyde under the usual
porphyrin-forming conditions, not just under the condi-
tions of excess pyrrole reported in the literature.1,3
The concentration of acid catalyst has a significant
effect on the rate and course of reaction.5,12 Thus, the
yields of TPP, NC-TPP, and TPS were measured as a
function of TFA or BF3-etherate concentration with 10
mM reactants. The yield of NC-TPP generally tracked
that of TPP (Figure 1). For the optimal concentrations
of acid, higher yields of NC-TPP were obtained with BF3-
etherate than with TFA. The TPS yield was less than
1% under all conditions examined (not shown). An
identical study at 100 mM reactants gave similar results
(see Supporting Information).
The yield of TPP depends on the concentration of
benzaldehyde and pyrrole, with maximal yields generally
observed at 10 mM.5,12 Accordingly, the yields of TPP,
NC-TPP, and TPS were measured as a function of
benzaldehyde and pyrrole concentrations with fixed
concentrations of TFA (20 mM) or BF3-etherate (1 mM).
Again, the yield of NC-TPP generally tracked that of TPP
but at a diminished level (Figure 2). The TPS yield was
less than 1% under all conditions examined (not shown).
An identical study with higher acid concentrations gave
similar results (see Supporting Information).
The reports by Latos-Grazynski of the formation of
N-confused tetra-p-tolylporphyrin1 and TPS3 with the use
of excess pyrrole prompted us to examine the yields of
TPP, NC-TPP, and TPS as a function of the pyrrole:
benzaldehyde ratio. For pyrrole at 100 mM, benzaldehyde
was varied from 12.5 to 800 mM with BF3-etherate or
TFA catalysis. The same experiments were performed
with benzaldehyde at 100 mM, and pyrrole varied from
12.5 to 800 mM. The results are displayed in Figure 3.
In all cases examined, the yield of TPP was peaked at
equimolar pyrrole and benzaldehyde. In general, the
distribution of NC-TPP was slightly skewed toward
excess pyrrole, but the peak remained at or near equimo-
lar pyrrole and benzaldehyde. This result shows that
excess pyrrole is not required for the formation of NC-
TPP. The TPS yield was less than 1% under all conditions
examined (not shown). A study with one reactant fixed
at 10 mM and the other reactant spanning the range of
1.25-80 mM gave similar results (see Supporting Infor-
mation).
Recently, the beneficial effects of added salt on the
yield of porphyrin were reported.13 The salt effect can
nearly double the yield of TPP with BF3-etherate as the
catalyst. The effect of salt on the yields of NC-TPP and
TPS was examined by performing reactions at 10 and
100 mM in the presence of NaCl (Figure 4). Although
(12) Lindsey, J . S.; MacCrum, K. A.; Tyhonas, J . S.; Chuang, Y. Y.
J . Org. Chem. 1994, 59, 579-587.
(13) Li, F.; Yang, K.; Tyhonas, J . S.; MacCrum, K. A.; Lindsey, J . S.
Tetrahedron 1997, 53, 12339-12360.

1600 J . Org. Chem., Vol. 64, No. 5, 1999
Geier and Lindsey
F igu r e 3. Yields of TPP and NC-TPP as a function of benzaldehyde:pyrrole ratio. All reactions were performed for 1 h.
the expected increase in TPP yield was observed in the
BF3-etherate catalyzed reactions, very little enhancement
of the NC-TPP yield occurred. With TFA catalysis, no salt
effect was observed for either the TPP or NC-TPP
products. The yield of TPS was <1% regardless of the
presence of salt with either catalyst (data not shown).
The extensive experimentation concerning the forma-
tion of NC-TPP allowed identification of conditions lead-
ing to 5.5-7.5% NC-TPP. The conditions giving the
highest yields are listed in Table 2. A wide variety of
catalyst concentrations and reactant concentrations can
be employed, but the pyrrole:benzaldehyde ratio is con-
strained to the range of 0.5-2. For the synthesis of TPS,
the best conditions found were 100 mM reactants with
64 mM TFA in CH2Cl2 at room temperature for 30 s prior
to oxidation, which gave TPS in 1.2% yield.
The yields of TPP, NC-TPP, and TPS were monitored
over time for eight variations of the one-flask room-
temperature porphyrin synthesis (10 and 100 mM reac-
tants for BF3-etherate, TFA catalysis, salt or no salt).5,12,13
Time points were obtained from 15 s to 30 h. Figure 5
summarizes the results of the BF3-etherate experiments.
In all cases, the yield of NC-TPP over time mirrored that
of TPP. Both species reached a maximum at a similar
point and showed similar decreases in yield at longer
time points. Adding NaCl increased the rate of formation
for both species, but their kinetic profiles remained
similar. Very little TPS (<0.2%) was detected at any time
under BF3-etherate catalysis. The results from the TFA
experiments (see Supporting Information) are very simi-
lar with only two exceptions: (1) No salt effect was
observed. (2) TPS was readily detected at early times but
not after 1 h. At 10 mM reactants, the yield of TPS
peaked at 0.5% between 8 and 15 min and was no longer
detected by 30 min. At 100 mM reactants, the yield of
TPS peaked at 1.1% between 30 s and 1 min and was no
longer detected by 15 min. This observation of rapid TPS
formation and decline explains the general absence of
TPS in the majority of experiments using a 1-h reaction
time.
Ch r om atogr aph ic P r oper ties of NC-TP P an d TP S.
Chromatography of NC-TPP and TPS was initially
performed to provide materials for HPLC method devel-
opment and calibration. The chromatographic properties
of NC-TPP acquired more significance when NC-TPP was
observed under a wide range of reaction conditions in
yields up to 7.5%, as reported in the previous section.
Our initial attempts to isolate NC-TPP and TPS, using
the published procedures of both Latos-Grazynski1,3 and
Furuta2 were unsuccessful, with only TPP being isolated.
The activity I grade basic alumina and silica gel used
initially were found to strongly adsorb NC-TPP and TPS.
In fact, the chromatographic behavior of NC-TPP and
TPS more closely resembles the black byproducts of the
reaction than TPP. This conclusion is supported by a
comparison of TLC retention characteristics of the three
macrocycles on silica and alumina with a variety of
developing solvents (Table 3). On silica, NC-TPP and TPS
do not elute at all with the usual development solvents
for TPP. On alumina, NC-TPP and TPS elute with

TPS and NC-TPP Formation in Syntheses of TPP
J . Org. Chem., Vol. 64, No. 5, 1999 1601
F igu r e 4. Yields of TPP and NC-TPP as a function of amount of NaCl. All reactions were performed for 1 h.
slightly higher Rf than the black byproducts and with
lower Rf than TPP when developing with CH2Cl2 and less
polar solvents. Chromatography, as described in the
Experimental Section, allowed isolation of NC-TPP and
TPS in about 95% purity with little difficulty. Given the
substantial polarity differences of TPP and NC-TPP, TPP
purified by chromatography is likely not contaminated
with NC-TPP. The removal of NC-TPP is especially likely
for samples that have not been neutralized prior to
chromatography because the NC-TPP is known to be
more basic than TPP.1
Obser va tion of Ad d ition a l P igm en ts. A key ques-
tion concerns whether other porphyrinic macrocycles are
present in significant abundance in the crude reaction
mixture. Such macrocycles could include corrole,14 ruby-
rin, pentaphyrin, hexaphyrin, and larger expanded por-
phyrins, multiply N-confused porphyrins, as well as yet
uncharacterized structures. In our attempts to isolate
NC-TPP (and NC-TMP, vide infra), about 4-6 additional
pigments (distinct from the nondescript black background
materials) were detected which eluted later than the
porphyrin and earlier than the N-confused porphyrin.15
The yield of those pigments, as estimated by gravimetric
analysis, was uniformly low (less than that of TPS).
Likewise, in the HPLC analysis of the reaction mixtures,
a handful of peaks in addition to TPP, TPS, and NC-TPP
were detected. No single peak appeared consistently
under a wide range of conditions, and very few of the
peaks were of significant area, indicating low yield if the
molar extinction coefficient is similar to TPP, NC-TPP,
or TPS. Although the nature of these pigments is
unknown, none is apparently produced in significant
yield. The yield of cyclic products (of polarity bracketed
by TPP and NC-TPP) can be summarized as follows: TPP
> NC-TPP . TPS ≈ other pigments. Thus, within the
range of conditions examined, the one-flask synthetic
methodology involving equimolar reactants5 appears best
suited for the preparation of porphyrin and N-confused
porphyrin.
Su r vey of NC-TP P a n d TP S in th e Ad ler Rea c-
tion . We examined the reaction of benzaldehyde and
pyrrole (0.27 M each) in refluxing propionic acid exposed
to air (Adler reaction)4 to determine if NC-TPP forms
under these conditions and if NC-TPP contaminates the
isolated crystals of TPP. The formation of TPS in the
Adler reaction has been reported.3 The TPP crystals and
filtrate were examined by HPLC. The crystalline material
was free of NC-TPP and TPS. The filtrate contained NC-
TPP in about 1% overall yield and TPS in about 0.5%
yield. Thus, it appears that crystallization effectively
removes NC-TPP and TPS from the TPP.
(14) A report has appeared concerning the formation of tris(2,6-
dinitro-4-tert-butylphenyl)corrole, but yield and complete analytical
data have not been published. See: Rose, E.; Kossanyi, A.; Quelquejeu,
M.; Soleilhavoup, M.; Duwavarn, F.; Bernard, N.; Lecas, A. J . Am.
Chem. Soc. 1996, 118, 1567-1568.
(15) After elution of NC-TPP, additional dark material remained
on the column (largely, but not entirely, quinone and hydroquinone
materials). No attempt was made to further fractionate that material
to look for additional polar macrocyclic pigments.
N-Con fu sed Tetr a m esitylp or p h yr in . The reaction
of pyrrole and mesitaldehyde with BF3-ethanol cocataly-

1602 J . Org. Chem., Vol. 64, No. 5, 1999
Geier and Lindsey
F igu r e 5. Kinetic profiles of TPP and NC-TPP yields for BF3-etherate catalyzed reactions. Note the log scale used for displaying
the time axis.
Ta ble 2. Su m m a r y of Con d ition s P r od u cin g High est
Yield s of NC-TP P a
Ta ble 3. Com p a r ison of TLC Rf Va lu es for TP P , NC-TP P ,
a n d TP S
BF3-etherate, [PhCHO], [pyrrole], [NaCl], %NC-
silica
alumina
entry
mM
mM
mM
equivb TPP %TPP
solvent
toluene
1:1 hexanes/CH2Cl2 0.51
TPP NC-TPP TPS TPP NC-TPP TPS
1
2
3
4
5
6
7
8
9
3.2
3.2
3.2
1
10
8
32
10
32
10
50
40
3.2
100
50
32
10
32
5
100
70
7.5
7.0
7.0
6.3
6.3
6.1
6.0
5.8
5.8
5.6
32
16
41
30
20
17
41
42
14
24
0.80
0.0
0.0
0.0
0.05
0.33
0.84
0.85
0.0
0.0
0.82
0.77
0.09
0.17
0.65
0.74
0.97
0.85
0.89
0.27
0.30
0.83
0.91
0.97
0.87
0.89
3.2
1
CH2Cl2
CH2Cl2, 1% TEA
0.88
0.90
0.00 0.92
0.05 0.94
0.40 0.97
0.84 0.88
0.86 0.89
CH2Cl2, 5% MeOH 0.98
1
3.2
78
1
THF
acetone
0.84
0.88
10
10
10
100
100
100
10
100
NC-TPP, indicating that the facial encumbrance of the
mesityl groups did not significantly alter the chromato-
graphic behavior. These results provide further evidence
for the generality of formation of N-confused porphyrins
during pyrrole-aldehyde condensations, as well as the
facile removal of the N-confused porphyrin via chroma-
tography.
a
All reactions were performed in CH2Cl2 at room temperature
for 1 h, followed by DDQ oxidation. The yields were determined
b
by HPLC. Equivalents relative to pyrrole.
sis smoothly affords TMP.11 These condensation condi-
tions provide the basis for the synthesis of a wide variety
of facially encumbered porphyrins.16 The reaction of
pyrrole and mesitaldehyde by means of a refined method10
was examined for the presence of the N-confused tet-
ramesitylporphyrin (NC-TMP). Column chromatography
of the crude reaction mixture afforded TMP (25%) and
NC-TMP (3%). NC-TMP was isolated in about 95% purity
using chromatographic conditions identical to those for
(16) Lindsey, J . S. In Metalloporphyrin-Catalyzed Oxidations; Mon-
tanari, F., Casella, L., Eds.; Kluwer Academic Publishers: The
Netherlands, 1994; pp 49-86.

TPS and NC-TPP Formation in Syntheses of TPP
J . Org. Chem., Vol. 64, No. 5, 1999 1603
Mech a n istic Im p lica tion s. The generality of the
production of NC-TPP under a variety of porphyrin-
forming conditions shows that the formation of â-pyrrole
linkages is not restricted to the unusual conditions
reported earlier.1,2,3 Rather, â-pyrrole linkages provided
a route for the diversion of starting materials under most
conditions that produced porphyrin in this study. The
distribution of NC-TPP relative to TPP suggests that the
two macrocycles share a similar reaction pathway. Detec-
tion of TPS shows that oligomers larger than tetramers
are also able to cyclize. The reduced precursor of TPS is
distinguished from those of TPP and NC-TPP by the fact
that it rapidly forms in low yield and then rapidly
decreases in yield. Differences between TFA and BF3-
etherate catalysis were also observed, with BF3-etherate
producing more NC-TPP and TFA leading to more TPS.
Whether factors other than acid strength are also in-
volved is unknown. The exact distribution of all products
(porphyrin, dipyrrin shunt products,11 â-pyrrole linked
products, bipyrrole-containing products, other macro-
cycles, and linear oligomers), the effects of changes in
reaction conditions, and the way in which the distribution
changes over time remain to be determined.
of 7.5%. The distribution of NC-TPP mirrored that of
TPP, and higher yields were obtained under BF3-etherate
catalysis. TPS was formed in higher yields under TFA
catalysis, but all yields were quite low (generally <1%).
Clearly, special conditions are not required for the
formation of NC-TPP. The chromatographic properties
of the porphyrins and N-confused porphyrins examined
are substantially different, making for straightforward
isolation of the purified porphyrin devoid of contaminat-
ing N-confused porphyrin. A handful of pigments have
been detected with chromatographic properties spanning
the range from TPP to NC-TPP, albeit in low yield.
Ack n ow led gm en t. Financial support was provided
by the NIH (GM36238). We thank Prof. Lechoslaw
Latos-Grazynski and Dr. Piotr J . Chmielewski for a
generous gift of an authentic sample of NC-TPP. We
thank Dr. Sankar Sabapathy for assistance with the 1H
and 13C NMR spectra for NC-TMP. Mass spectra were
obtained at the Mass Spectrometry Laboratory for
Biotechnology. Partial funding for the facility was
obtained from the North Carolina Biotechnology Center
and the National Science Foundation.
Con clu sion s
Su p p or t in g In for m a t ion Ava ila b le: 1H and 13C NMR
spectra for NC-TMP, HPLC method development information,
representative HPLC chromatograms, method verification and
repeatability data, and complete summaries of results from
all experiments. This material is available free of charge via
the Internet at http://pubs.acs.org.
The yields of TPP, NC-TPP, and TPS formed under a
variety of conditions used in one-flask porphyrin-forming
reactions have been determined. TPP was the major
macrocyclic product formed under all conditions exam-
ined. NC-TPP was present under a wide range of condi-
tions (including the Adler method), with a maximal yield
J O982068R