Refined synthesis of 2,3,4,5-tetrahydro-1,3,3-trimethyldipyrrin, a deceptively simple precursor to hydroporphyrins

Article abstract of DOI:10.1021/op050087e

2,3,4,5-Tetrahydro-1,3,3-trimethyldipyrrin (1) is a crucial building block in the rational synthesis of chlorins and oxochlorins. The prior five-step synthesis of 1 from pyrrole-2-carboxaldehyde (2) employed relatively simple and well-known reactions yet suffered from several drawbacks, including limited scale (≤0.5 g of 1 per run). A streamlined preparation of 1 has been developed that entails four steps: (i) nitro-aldol condensation of 2 and nitromethane under neat conditions to give 2-(2-nitrovinyl)pyrrole (3), (ii) reduction of 3 with NaBH₄ to give 2-(2-nitroethyl)pyrrole (4), (iii) Michael addition of 4 with mesityl oxide under neat conditions or at high concentration to give γ-nitrohexanone-pyrrole 5, and (iv) reductive cyclization of 5 with zinc/ammonium formate to give 1. Several multistep transformations have been established, including the direct conversion of 2 → 1. The advantages of the new procedures include (1) fewer steps, (2) avoidance of several problematic reagents, (3) diminished consumption of solvents and reagents, (4) lessened reliance on chromatography, and (5) scalability. The new procedures facilitate the preparation of 1 at the multigram scale.

Full text of DOI:10.1021/op050087e

Organic Process Research & Development 2005, 9, 651−659
Refined Synthesis of 2,3,4,5-Tetrahydro-1,3,3-trimethyldipyrrin, a Deceptively
Simple Precursor to Hydroporphyrins
Marcin Ptaszek, Jayeeta Bhaumik, Han-Je Kim, Masahiko Taniguchi, and Jonathan S. Lindsey*
Department of Chemistry, North Carolina State UniVersity, Raleigh, North Carolina 27695-8204
Scheme 1
Abstract:
2,3,4,5-Tetrahydro-1,3,3-trimethyldipyrrin (1) is a crucial build-
ing block in the rational synthesis of chlorins and oxochlorins.
The prior five-step synthesis of 1 from pyrrole-2-carboxalde-
hyde (2) employed relatively simple and well-known reactions
yet suffered from several drawbacks, including limited scale
(e0.5 g of 1 per run). A streamlined preparation of 1 has been
developed that entails four steps: (i) nitro-aldol condensation
of 2 and nitromethane under neat conditions to give 2-(2-
nitrovinyl)pyrrole (3), (ii) reduction of 3 with NaBH4 to give
2-(2-nitroethyl)pyrrole (4), (iii) Michael addition of 4 with
mesityl oxide under neat conditions or at high concentration
to give γ-nitrohexanone-pyrrole 5, and (iv) reductive cyclization
of 5 with zinc/ammonium formate to give 1. Several multistep
transformations have been established, including the direct
conversion of 2 f 1. The advantages of the new procedures
include (1) fewer steps, (2) avoidance of several problematic
reagents, (3) diminished consumption of solvents and reagents,
(4) lessened reliance on chromatography, and (5) scalability.
The new procedures facilitate the preparation of 1 at the
multigram scale.
able.2,4,7 However, the synthesis of the Western half has
presented a number of challenges.
Our previous route to 1, which entailed five steps, is
summarized in Scheme 2. The synthesis involved (i) nitro-
aldol condensation of pyrrole-2-carboxaldehyde (2) with
nitromethane,2 (ii) NaBH4 reduction of nitrovinylpyrrole 3,2
(iii) Michael addition of nitroethylpyrrole 4 to mesityl oxide,2
(iv) reductive cyclization of nitrohexanone 5 by Zn/AcOH,4
and (v) deoxygenation of N-oxide 6 using Ti(0) prepared in
situ.2
Compound 1 appears quite simple, but the appearance is
deceptive because the two different heterocycles (pyrrole,
pyrroline) present quite distinct reactivity. The complexity
of the chemistry presented by 1 is as follows: (1) the pyrrole
contains three open sites for electrophilic substitution, (2)
the pyrrole is activated toward electrophiles by the 2-alkyl
substituent, (3) the imine is susceptible to reduction and
addition, (4) the imine nitrogen can coordinate to metals,
(5) the pyrrole is a weak acid whereas the pyrroline is a
weak base, and (6) the R-pyrrolic methylene is susceptible
to oxidation. In addition, tetrahydrodipyrrin 1 undergoes
intramolecular cyclization in the presence of acid, forming
a pyrrolo[3.2.1]azabicyclooctane byproduct (7) (eq 1).4
Precursors 2-6 also are sensitive to a number of reaction
conditions. Thus, the scope of suitable synthetic methods is
limited.
Introduction
Hydroporphyrins (e.g., chlorins, bacteriochlorins, isobac-
teriochlorins) are naturally occurring pigments with diverse
functions. Methods for preparing chlorins have been aimed
at total syntheses of naturally occurring compounds as well
as more simple syntheses of chlorins for use in a variety of
applications.1 We recently developed rational syntheses of
non-natural chlorins2-4 (Scheme 1) and oxochlorins5 by
extending routes developed by Battersby for the synthesis
of naturally occurring hydroporphyrins.6 The synthesis entails
reaction of an Eastern half (A) with a Western half (1) to
form a tetrahydrobilene-a, which upon oxidative cyclization
affords the zinc chlorin. The Eastern half is easily avail-
* Corresponding author. E-mail: jlindsey@ncsu.edu.
(1) (a) Montforts, F.-P.; Gerlach, B.; Ho¨per, F. Chem. ReV. 1994, 94, 327-
347. (b) Chen, Y.; Li, G.; Pandey, R. K. Curr. Org. Chem. 2004, 8, 1105-
1134.
(2) Strachan, J.-P.; O’Shea, D. F.; Balasubramanian, T.; Lindsey, J. S. J. Org.
Chem. 2000, 65, 3160-3172.
(3) Balasubramanian, T.; Strachan, J. P.; Boyle, P. D.; Lindsey, J. S. J. Org.
Chem. 2000, 65, 7919-7929.
(4) Taniguchi, M.; Ra, D.; Mo, G.; Balasubramanian, T.; Lindsey, J. S. J. Org.
Chem. 2001, 66, 7342-7354.
(5) Taniguchi, M.; Kim, H.-J.; Ra, D.; Schwartz, J. K.; Kirmaier, C.; Hindin,
E.; Diers, J. R.; Prathapan, S.; Bocian, D. F.; Holten, D.; Lindsey, J. S. J.
Org. Chem. 2002, 67, 7329-7342.
(6) Battersby, A. R.; Fookes, C. J. R.; Snow, R. J. J. Chem. Soc., Perkin Trans.
1 1984, 2725-2732.
(7) (a) Rao, P. D.; Littler, B. J.; Geier, G. R., III; Lindsey, J. S. J. Org. Chem.
2000, 65, 1084-1092. (b) Muthukumaran, K.; Ptaszek, M.; Noll, B.; Scheidt,
W. R.; Lindsey, J. S. J. Org. Chem. 2004, 69, 5354-5364.
10.1021/op050087e CCC: $30.25 © 2005 American Chemical Society
Published on Web 08/10/2005
Vol. 9, No. 5, 2005 / Organic Process Research & Development
•
651

Scheme 2
Chart 1
room temperature yields the decarboxylated tetrahydrodipyr-
rin, which is stable to the strong acid,6 whereas 1 undergoes
slow cyclization in the presence of dilute TFA to afford the
byproduct 74 (eq 1). In addition, the byproduct 7 was
observed upon reductive cyclization of nitrohexanone 5 in
AcOH and was the main product upon attempted deoxygen-
ation of 5 using Ti(III) at pH 6.4 Thus, deoxygenation
methods developed for â-substituted pyrrolic derivatives
provided a low yield of 1. Regardless, each tetrahydrodipyrrin
was prepared at relatively small scale (448 mg of 1,4 139
mg of B,6 914 mg of C,6 228 mg of D5). Attempts to increase
the scale of synthesis of 1 met with several obstacles. First,
each intermediate was purified by chromatography. Some
synthetic steps required low concentrations and a large excess
of reagents, hence increasing the scale was prohibitive in
terms of solvent and materials consumption. Moreover, the
yields of most of the synthetic steps decreased substantially
upon increasing the scale, which we attributed to the
instability of the various reactants under the conditions
employed.
In this paper we report a refined preparation of 1 that
proceeds in reasonable yield, with limited chromatography,
and diminished consumption of solvents and reactants. The
number of steps in the synthesis has been reduced from five
to four, two or one. Taken together, the improvements have
enabled the routine preparation of 1 at 60-mmol scale to
afford 2.5 g of product in shorter time (3 days) or at medium
scale (0.4 mol) to give 11 g of 1 (22 times more than reported
previously). Application of this methodology to substituted
analogues of 1 will be described elsewhere.
One of the significant structural differences between 1
and tetrahydrodipyrrinic analogues reported earlier by Bat-
tersby (Chart 1, compounds B and C)6 is that the latter have
two â-substituents in the pyrrolic ring, whereas 1 and
derivatives thereof (D)5 typically have no pyrrolic â-sub-
stituents. In general, tetrahydrodipyrrins lacking â-pyrrolic
substituents have been prepared in lower overall yield and
have required milder reaction conditions than the more stable
â-substituted species. Indeed, treatment of C in neat TFA at
Results and Discussion
I. Refinement of Individual Transformations. A. Nitro-
aldol Condensation (2 f 3). The typical conditions for the
nitro-aldol condensation of pyrrole-2-carboxaldehyde require
the presence of an ammonium salt and/or weak base. The
condensation of pyrrole-2-carboxaldehyde with a nitroalkane
typically is done in the presence of one of the following:
652
•
Vol. 9, No. 5, 2005 / Organic Process Research & Development

ammonium acetate,8 ammonium acetate with microwave
irradiation,9 ethylenediamine,10 sodium acetate,11 or a mixture
of methylamine hydrochloride and sodium acetate.2-6 The
latter reagent affords superior results for N-unprotected
pyrrole-2-carboxaldehydes, whereas the former reagents are
more suitable for N-substituted pyrrolic derivatives. However,
methylamine hydrochloride is scrutinized by U.S. drug
enforcement agencies owing to use in methamphetamine
syntheses; hence, we searched for a comparable reagent for
the first step of the synthesis. The amine must be easily
removed from the reaction mixture (to proceed to the next
step without laborious workup) and enable reaction at high
concentration or even under solventless conditions (to
diminish the use of solvents in large-scale preparations).
Various catalytic systems for the nitro-aldol condensation
were examined to identify conditions that meet three objec-
tives: (1) complete consumption of 2, (2) good yield of
product 3, and (3) little or no black byproducts. The results
of a survey of different amines and ammonium salts and the
influence of a water scavenger (molecular sieves) are listed
in the Supporting Information. The cleanest reaction and the
mildest condition were obtained with 2 dissolved in a 3-fold
excess of nitromethane (no other solvent) containing AcOH/
n-PrNH2 (0.60 equiv/0.55 equiv) at room temperature
(Scheme 3). The reaction is complete within 2-3 h. The
ratio of AcOH/n-PrNH2 is critical. A slight excess of
n-propylamine caused formation of a significant amount of
dark byproducts, which interfered with the workup and
resulted in a lower yield. On the other hand, a slight excess
(0.57-0.60 equiv) of acetic acid provided a relatively clean
product with minor amounts of dark byproducts. The small
excess of acid also diminished the Michael addition of
nitromethane to the 2-(2-nitrovinyl)pyrrole (3), which forms
the ostensible byproduct 2-(2-pyrrolyl)-1,3-dinitropropane
(8). An analogous byproduct was observed upon similar
Scheme 3
somewhat unstable; therefore for this reason alone, chroma-
tography should be avoided. Treatment of the crude reaction
mixture with NaBH4 (in an appropriate solvent) did not prove
fruitful, as 2-(2-nitroethyl)pyrrole (4) was obtained in low
yield accompanied by several byproducts. An effective
workup was achieved by washing the crude mixture contain-
ing 3 with water and then removing excess MeNO2 under
high vacuum.
B. Reduction of 2-(2-Nitrovinyl)pyrrole (3 f 4). The
previous synthetic route utilized NaBH4 in DMF/THF or
MeOH/THF.2,4 This procedure required a large excess of
NaBH4 (3.5 mol equiv) and gave a nonreproducible yield at
larger scale owing to the formation of byproducts. Moreover,
the abundant boron salts caused difficulties in purification
of the final product. Use of a different solvent system
(CHCl3/2-propanol, 3:1) in the presence of silica12 enabled
use of a lesser quantity of NaBH4 (2 mol equiv), diminished
the amount of byproducts, and gave the desired product in
76% yield (Scheme 3). Additionally, the presence of silica
in the reaction mixture facilitated purification. Most of the
impurities were absorbed on the silica, and filtration through
a pad of silica afforded the pure 4.
reaction of a 3-(4-iodophenyl)-2-(2-nitrovinyl)pyrrole.3 A
small amount of 8, which was detected in the crude reaction
mixture, can be easily separated by precipitation of 3 from
CH2Cl2/hexanes. Another problem encountered was an
exothermic reaction between AcOH and n-PrNH2. Therefore,
the salt was prepared in a small amount of methanol in a
separate flask and then transferred to the solution of pyrrole-
2-carboxaldehyde.
Another issue was the choice of workup procedure
suitable for the next step. 2-(2-Nitrovinyl)pyrrole (3) is
(8) (a) Hamdan, A.; Wasley, J. W. F. Synth. Commun. 1985, 15, 71-74. (b)
Dumoulin, H.; Rault, S.; Robba, M. J. Heterocycl. Chem. 1997, 34, 13-
16. (c) Fuji, M.; Muratake, H.; Natsume, M. Chem. Pharm. Bull. 1992, 40,
2344-2352.
(9) Kuster, G. J. T.; Steeghs, R. H. J.; Scheeren, H. W. Eur. J. Org. Chem.
2001, 553-560.
(10) Aleksiev, D. I.; Ivanova, S. M. Russ. J. Org. Chem. 1993, 29, 1845-1847.
(11) Khorana, N.; Smith, C.; Herrick-Davis, K.; Purohit, A.; Teitler, M.; Grella,
B.; Dukat, M.; Glennon, R. A. J. Med. Chem. 2003, 46, 3930-3937.
C. Michael Addition with Mesityl Oxide (4 f 5). The
previously reported procedures for Michael addition of
(12) Sinhababu, A. K.; Borchardt, R. T. Tetrahedron Lett. 1983, 24, 227-230.
Vol. 9, No. 5, 2005 / Organic Process Research & Development
•
653

Table 1. Conditions for the Michael addition (4 f 5)a
temperature; entry 10) with 10.0 mmol of 4 but only 1.1
equiv of mesityl oxide for 24 h afforded 5 in 78% yield.
D. Reductive Cyclization (5 f 1). The previous reduc-
tion of nitrohexanone 5 to the tetrahydrodipyrrin 1 was
performed in two-steps: cyclization to the N-oxide 6
followed by deoxygenation of 6. This procedure suffers from
several problems: (1) the yield of both steps decreased
substantially with increasing scale, (2) both steps required a
large excess of metal reagents (25 equiv of zinc, 7 equiv of
TiCl4), and (3) the deoxygenation step was done at relatively
low concentration (35 mM). To scale-up the synthesis of 1
necessitated a significant improvement in each step.
Several mild methods are known for direct reductive
cyclization of γ-nitro-ketones to the corresponding cyclic
imines,16 reduction of nitroalkanes to the corresponding
amines,17 or the deoxygenation of N-oxides.18 Some recent
pertinent examples include the following:
4: equiv, solvent
temp time yield
(°C) (h) of 5b
entry base (equiv)
1c CsF (5.7)
(concentrated)
5.0, CH3CN (150 mM)d 70
10, CH3CN (350 mM)f 80
10, neat
10, neat
1.0, neat
16
18
14
18
14
65%e
61%e
noneg
traceg
none
50%
2
3
4
5
6
7
8
9
CsF (3.0)
CsF (3.0)
CsF (3.0)
Al2O3
60
rt
rt
DBU (1.0) 2.0, CH3CN (200 mM) reflux 4
DBU (1.0) 1.1, THF (500 mM)
DBU (1.0) 1.1, THF (500 mM)
DBU (1.0) 10, neat
rt
3
trace
trace
75%
reflux 1
rt
rt
rt
rt
14
10 DBU (3.0) 10, neat
11 TMG (0.25) 1.1, THF (250 mM)
12 TMG (1.25) 1.1, THF (250 mM)
16
14
1
78%e
none
trace
trace
13 TMG (0.25) 1.1, CH3CN (250 mM) rt
14
a The reactions were carried out with 1 mmol of 4 unless noted otherwise.
b Yields were estimated on the basis of GC. c Conditions employed previously.2
d 11.4 mmol scale. e Isolated yield. f 25 mmol scale. g Extensive byproducts.
(a) γ-nitro-ketones f cyclic imines: Zn/NH4Cl,16a
Fe/NH4Cl,16b NiCl2‚H2O/NaBH4/NH2NH2‚H2O,16c Pd-C/
HCOONH4,16d and Zn/HCOOH;16e
mesityl oxide and 2-(2-nitroethyl)pyrroles utilized fluoride
anion (e.g., TBAF13 or CsF2-5) to effect reaction (entry 1,
Table 1). We examined CsF and other reagents in the reaction
of 4 with excess mesityl oxide (10 equiv) that have been
reported to be effective for the Michael addition of R,â-
enones and nitroalkanes (Al2O3,14 DBU,15 and TMG16). The
ideal reagent should give a clean reaction, a good yield at
high concentration (or even neat conditions), and be easily
removed from the reaction mixture. The reaction with CsF
in refluxing CH3CN gave 61% yield (entry 2), but reaction
under neat conditions either at 60 °C (entry 3) or room
temperature (entry 4) gave no product. Reaction with alumina
under neat conditions also gave no product (entry 5).
Replacement of CsF with DBU in refluxing CH3CN gave
50% yield (entry 6). Replacing CH3CN with THF gave only
a trace of product (entries 7 and 8). On the other hand, the
reaction with DBU under neat conditions at room temperature
gave 5 in high yield (entries 9 and 10). Attempts to use
tetramethylguanidine (TMG) in place of DBU gave little or
no product (entries 11-13).
The advantages of DBU versus CsF include slightly
higher yield, lower cost, absence of solvent, and greater ease
of handling (no requirement for desiccation). The results
shown in Table 1 employ 10 equiv of mesityl oxide and were
performed using 1 mmol of 4. Although mesityl oxide is
inexpensive, an excess is unattractive for applications envis-
aged with more valuable enones. We found that the Michael
addition also proceeded well with lesser quantities of mesityl
oxide as long as the reaction time was prolonged. Thus,
application of the best conditions (DBU, solventless, room
(b) nitroalkanes f amines: NH2NH2‚H2O/graphite,17a
NH2NH2‚H2O/Raney Ni,17b Zn/HCOONH4,17c Mg/
HCOONH4,17d Mg/NH2NH2,17e Zn/NH2NH2/HCOOH,17f and
ZrCl4/NaBH4;17g
(c) deoxygenation of N-oxides: Pd-C/HCOONH4,18a Zn/
HCOONH4,18b In/NH4Cl,18c InCl3,18d Ga/H2O,18e and formic
pivalic anhydride.18f Earlier methods have been reviewed.18g
A survey of reducing agents and conditions was carried
out for the direct conversion of the γ-nitro-ketone 5 to the
tetrahydrodipyrrin 1 (Scheme 4). Generally, we searched for
mild conditions with an aim toward application at >10-g
scale. Selected results are shown in Table 2 (for a broader
survey, see the Supporting Information). The use of Zn/
HCOONH4 gave preferentially the desired target 1 ac-
companied by limited amounts of N-oxide 6 (entries 1-9).
The choice of solvent was critical. Although methanol
appeared to provide a good conversion of 5 to 1, the isolated
yield was very low (∼10%). We reasoned that the low yield
might stem from the formation of a complex between zinc-
(II) and 1, as zinc(II) is known to coordinate with the related
dipyrrin species forming bis(dipyrrinato)zinc(II) complexes.19
Accordingly, THF was used, and the isolated yield increased
to 60%. The reaction was clean, proceeding at relatively high
concentration, and workup was straightforward. Large-scale
(17) (a) Han, B. H.; Shin, D. H.; Cho, S. Y. Tetrahedron Lett. 1985, 26, 6233-
6234. (b) Gowda, S.; Gowda, D. C. Tetrahedron 2002, 58, 2211-2213. (c)
Gowda, D. C.; Mahesh, B.; Gowda, S. Ind. J. Chem. 2001, 40B, 75-77.
(d) Srinivasa, G. R.; Abiraj, K.; Gowda, D. C. Ind. J. Chem. 2003, 42B,
2882-2884. (e) Srinivasa, G. R.; Abiraj, K.; Gowda, D. C. Ind. J. Chem.
Sec. 2003, 42B, 2885-2887. (f) Gowda, S.; Gowda, B. K. K.; Gowda, D.
C. Synth. Commun. 2003, 33, 281-289. (g) Chary, K. P.; Ram, S. R.;
Iyengar, D. S. Synlett 2000, 683-685.
(13) Harrison, P. J.; Sheng, Z.-C.; Fookes, C. J. R.; Battersby, A. R. J. Chem.
Soc., Perkin Trans. 1 1987, 1667-1678.
(14) Ranu, B. C.; Bhar, S.; Sarkar, D. C. Tetrahedron Lett. 1991, 32, 2811-
2812.
(18) (a) Balicki, R. Synthesis 1989, 645-646. (b) Balicki, R.; Cybulski, M.;
Maciejewski, G. Synth. Commun. 2003, 33, 4137-4141. (c) Yadav, J. S.;
Reddy, B. V. S.; Reddy, M. M. Tetrahedron Lett. 2000, 41, 2663-2665.
(d) Ilias, M.; Barman, D. C.; Prajapati, D.; Sandhu, J. S. Tetrahedron Lett.
2002, 43, 1877-1879. (e) Han, J. H.; Choi, K. I.; Kim, J. H.; Yoo, B. W.
Synth. Commun. 2004, 34, 3197-3201. (f) Rosenau, T.; Potthast, A.; Ebner,
G.; Kosma, P. Synlett 1999, 623-625. (g) Albini, A.; Pietra, S. Heterocyclic
N-Oxides; CRC Press: Boca Raton, FL, 1991; pp 120-134.
(15) Ballini, R.; Bosica, G.; Fiorini, D.; Palmieri, A.; Petrini, M. Chem. ReV.
2005, 105, 933-971 and references therein.
(16) (a) Black, D. S.; Edwards, G. L.; Evans, R. H.; Keller, P. A.; Laaman, S.
M. Tetrahedron 2000, 56, 1889-1897. (b) Take, K.; Okumara, K.; Tsubaki,
K.; Terai, T.; Shiokawa, Y. Chem. Pharm. Bull. 1993, 41, 507-515. (c)
Urrutia, A.; Rodriguez, J. G. Tetrahedron 1999, 55, 11095-11108. (d)
Benchekroun-Mounir, N.; Dugat, D.; Gramain, J.-C.; Husson, H.-P. J. Org.
Chem. 1993, 58, 6457-6465. (e) Cheruku, S. R.; Padmanilayam, M. P.;
Vennerstrom, J. L. Tetrahedron Lett. 2003, 44, 3701-3703.
(19) Yu, L.; Muthukumaran, K.; Sazanovich, I. V.; Kirmaier, C.; Hindin, E.;
Diers, J. R.; Boyle, P. D.; Bocian, D. F.; Holten, D.; Lindsey, J. S. Inorg.
Chem. 2003, 42, 6629-6647.
654
•
Vol. 9, No. 5, 2005 / Organic Process Research & Development

Scheme 4
Workup entailed diluting the crude mixture with CH2Cl2,
removing the ammonium salt by washing with water, and
removing excess nitromethane under high vacuum. The crude
3 was dissolved in CHCl3/2-propanol (3:1) and treated with
2.0 mol equiv of NaBH4 in the presence of silica, affording
2-(2-nitroethyl)pyrrole (4) in 40% yield (Scheme 5).
B. Conversion of 4 f 1. The conversion of 2-(2-
nitroethyl)pyrrole (4) to the desired tetrahydrodipyrrin 1 was
prompted by the observation that GC and TLC analyses of
the crude reaction mixture following Michael addition of 4
and mesityl oxide did not show any significant side products.
Therefore we employed a two-step, one-flask procedure
entailing Michael addition followed by reduction of the crude
nitrohexanone 5. The crude reaction mixture was diluted with
ethyl acetate and washed with water; the organic layer was
dried; excess mesityl oxide was removed under high vacuum;
and the resulting oil was subjected to reductive cyclization
using Zn/HCOONH4. The yield of 1 was 53% (Scheme 5).
C. Conversion of 2 f 5. The conversion of pyrrole-2-
carboxaldehyde (2) to the nitrohexanone-pyrrole (5) was
carried out by solventless condensation with nitromethane,
reduction with NaBH4, Michael addition with mesityl oxide,
and workup including chromatography. The desired 5 was
obtained in 29% yield (Scheme 5).
D. Conversion of 2 f 1. Taking together all improve-
ments in each of the four steps, the direct conversion of
pyrrole-2-carboxaldehyde (2) to 1 was investigated (Scheme
5). Treatment of a solution of 2 in MeNO2 with propylam-
monium acetate afforded crude 2-(2-nitrovinyl)pyrrole 3.
After workup (dilution with CH2Cl2, washing with water,
removal of excess MeNO2 under high vacuum), the crude 3
was dissolved in CHCl3/2-propanol, to which silica was
added followed by NaBH4 (2 mol equiv). After workup
(filtration, concentration of the filtrate), the crude 4 (brown-
greenish oil) was dissolved in excess mesityl oxide (1.5
equiv), excess DBU (3 equiv) was added, and the Michael
addition was performed overnight at room temperature. After
workup (dilution with ethyl acetate, washing with water and
brine, removal of solvent and excess mesityl oxide under
reduced pressure), the crude 5 was dissolved in THF (ACS
grade) and treated with excess Zn/HCOONH4 for 2 h.
Workup (filtration, concentration of the filtrate, and chro-
matography on silica) afforded 1 as a light-brown solid in
22% overall yield. The tetrahydrodipyrrin obtained in this
manner from 2 was slightly darker than upon synthesis from
a pure sample of the immediate precursor 5, although the
1H NMR spectrum did not show any noticeable impurities.
The entire synthesis at the 60-mmol scale afforded 2.5 g
of 1 (22% yield) in a 3-day period. The procedure at the 0.4
mol-scale provided ∼11 g of 1 (14.5% yield) in one batch,
which is 22 times more than reported previously. Compound
1 gave satisfactory elemental analysis data in only one
instance (upon preparation from 6); in all other cases, all
other characterization data were satisfactory, and samples
of 1 were successfully employed in several syntheses of
chlorins.
applications appear viable because the yield remains high
upon increasing the scale (up to 0.4 mol), inexpensive off-
the-shelf reagents (zinc dust and ammonium formate) are
employed, and the reaction is quite robust in not requiring
dry solvents or special precautions.
E. Reductive Cyclization and Deoxygenation (5 f 6
f 1). In the previous synthesis,1,2 nitrohexanone 5 was
converted to N-oxide 6 using Zn in the presence of acetic
acid. Such conditions are incompatible with a number of
functional groups. In the synthesis of 1, we were content to
bypass the preparation of the N-oxide 6. However, N-oxides
are valuable intermediates in the syntheses of functionalized
tetrahydrodipyrrins.20 Accordingly, we investigated alterna-
tives for the preparation and deoxygenation of N-oxide 6.
During the survey of conditions for conversion of 5 f 1,
we found that Zn/NH4Cl in THF/H2O preferentially gave
the N-oxide 6 (Table 2, entries 10-17). For the preparative
scale, nitrohexanone 5 was dissolved in THF/H2O (1:1) and
treated with Zn/NH4Cl at 0 °C for 15 min. GC analysis
showed 70% of N-oxide 6 and 19% of 1. Chromatography
afforded 6 in 66% yield. For the direct deoxygenation of
N-oxide 6, we examined a number of reagents (InCl3, PPh3/
toluene, (MeO)3P/toluene, CS2, HCOONH4/pivaloyl chloride,
and Zn/HCOONH4). Most of the methods caused extensive
decomposition of 6 or no reaction. Again, the best result
was obtained using Zn/HCOONH4 in THF, affording 1 in
81% yield (Scheme 4).
II. Investigation of Multistep Transformations. Here
we investigated a number of approaches to obtain intermedi-
ates or the target compound 1 in an expeditious manner.
A. Conversion of 2 f 4. Pyrrole-2-carboxaldehyde (2)
in excess nitromethane was treated with n-propylammonium
acetate, affording the crude 2-(2-nitrovinyl)pyrrole (3).
In summary, the valuable tetrahydrodipyrrin 1 can be
obtained in a stepwise manner via streamlined procedures.
(20) Kim, H.-J. Ph.D. Thesis, North Carolina State University, 2005.
Vol. 9, No. 5, 2005 / Organic Process Research & Development
•
655

Table 2. Conditions for the Reductive Cyclization (5 f 1)
Zn
NH4X
entry
(equiv)
(equiv)
solvent
MeOH
MeOH
MeOH
MeOH
MeOH
MeCN
THF
temp, time
5:6:1a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
15
2
HCOONH4 (3)
HCOONH4 (15)
HCOONH4 (15)
HCOONH4 (15)
HCOONH4 (15)
HCOONH4 (15)
HCOONH4 (15)
HCOONH4 (15)
HCOONH4 (15)
NH4Cl (3)
NH4Cl (3)
NH4Cl (6)
NH4Cl (3)
NH4Cl (3)
rt, 30 min
rt, 30 min
rt, 30 min
rt, 2 h
reflux, 20 min
rt, 2 h
rt, 2 h
rt, 2 h
rt, 2 h
rt, 30 min
rt, 1 h
rt, 30 min
rt, 30 min
reflux, 30 min
rt, 30 min
rt, 30 min
rt, 30 min
0:44:34
0:42:40
0:27:63
0:5:68
0:0:51
0:4:64
15
15
15
15
15
15
15
15
15
15
25
15
15
5
1:0:76
ethyl acetate
CH2Cl2
0:18:54
0:37:48
0:61:21
0:65:22
2:70:23
2:68:17
<1:66:25
1:72:20
4:72:17
34:33:9
THF/H2O, 1:1
THF/H2O, 1:1
THF/H2O, 1:1
THF/H2O, 1:1
THF/H2O, 1:1
THF/ H2O, 4:1
THF/H2O, 1:1
MeOH/THF, 1:1
NH4Cl (3)
NH4Cl (3)
NH4Cl (3)
15
a Ratio determined by GC analysis. The data do not sum to 100% because of side products and/or impurities.
Scheme 5
Table 3. Comparison of routes for converting 2 f 1
quantities of 1 should facilitate entry into numerous hydro-
durationa
(day)
yieldb
(%)
porphyrin architectures.
number
of steps
isolated
intermediates
Experimental Section
5
4
3
2
1
3, 4, 5, 6
3, 4, 5
4, 5
10
9
7
11
17
19
General. 1H NMR (400 MHz) and 13C NMR (100 MHz)
spectra were collected at room temperature in CDCl3. Melting
points are uncorrected. Column chromatography was per-
formed with flash silica or alumina (80-200 mesh). The
CHCl3 contained 0.8% ethanol. All solvents and reagents
were used as received. Zinc was employed in the form of
zinc dust (<10 micron). In all cases when a salt (e.g.,
HCOONH4) and zinc dust were added to a reaction mixture,
the salt was added first. All procedures were performed at
room temperature unless noted otherwise.
4
5
21
3,c 4d
22,c 14.5d
a Estimated working days. b Overall yield based on the amount of starting
material 2. c 60 mmol scale. d 0.40 mol scale.
In addition, a straightforward method has been developed
for the direct conversion of pyrrole-2-carboxaldehyde (2) to
1 that is relatively fast, affords reasonable overall yield,
employs simple reagents and methods, and has limited
reliance on chromatography. A comparison of the various
routes for converting 2 f 1 is provided in Table 3. Taken
together, the availability of expedient routes to multigram
2-(2-Nitrovinyl)pyrrole (3). A stirred solution of acetic
acid (2.00 mL, 35.0 mmol) in methanol (5 mL) under argon
at 0 °C was treated dropwise with n-propylamine (2.71 mL,
33.0 mmol). The resulting n-propylammonium acetate solu-
tion was stirred at 0 °C for 5 min, then added dropwise to a
656
•
Vol. 9, No. 5, 2005 / Organic Process Research & Development

stirred solution of 2 (5.70 g, 60.0 mmol) in nitromethane
(9.66 mL, 180 mmol) at 0 °C. The resulting mixture was
stirred at 0 °C. After 15 min, the cooling bath was removed,
and stirring was continued at room temperature. The color
changed from yellow to dark orange during the course of
reaction. After 2 h, CH2Cl2 (100 mL) was added, and the
organic phase was washed with water and brine. The organic
layer was dried (Na2SO4) and concentrated under high
vacuum to afford an orange-brown solid. Filtration through
a silica pad (CH2Cl2) afforded an orange solid, which was
dissolved in a minimal volume of CH2Cl2 and precipitated
with hexanes to afford orange crystals (3.93 g, 47%): mp
110-112 °C (lit.1 112-114 °C); 1H NMR δ 6.38-6.40 (m,
1H), 6.81-6.83 (m, 1H), 7.12-7.13 (m, 1H), 7.48 (d, 5J )
3.8 Hz, 3J ) 13.6 Hz, 1H), 7.99 (d, J ) 13.6 Hz, 1H), 9.04-
9.15 (br. s, 1H); 13C NMR δ 112.9, 119.4, 124.1, 126.5,
129.9, 131.0. Anal. Calcd for C6H6N2O2: C, 52.17; H, 4.38;
N, 20.28. Found: C, 52.31; H, 4.40; N, 20.22. FAB-MS obsd
139.0500, calcd 139.0508 [(M + H)+, M ) C6H6N2O2].
2-(2-Nitroethyl)pyrrole (4). Following a published pro-
cedure,12 a solution of 3 (0.138 g, 1.00 mmol) in CHCl3 (9.0
mL) and 2-propanol (3.0 mL) was treated with silica (1.2
g). The resulting suspension was treated in one portion with
NaBH4 (0.0760 g, 2.00 mmol) under vigorous stirring. The
mixture was stirred at room temperature for 20 min,
accompanied by a color change from deep yellow to pale
brown. The mixture was filtered. The filter cake was washed
with CH2Cl2. The combined filtrate was concentrated. The
resulting oil was dissolved in CH2Cl2. The organic solution
was washed with water and brine. The organic layer was
dried (Na2SO4), concentrated, and subjected to high vacuum
to remove traces of 2-propanol. The residue was dissolved
in a small quantity of CH2Cl2 and filtered through a silica
pad (CH2Cl2) to afford a pale yellow oil (0.106 g, 76%):
1H NMR data were consistent with previously reported
values;2 1H NMR δ 3.31 (t, J ) 6.8 Hz, 2H), 4.60 (t, J )
6.8 Hz, 2H), 6.00-6.09 (m, 1H), 6.14-6.18 (m, 1H), 6.71-
6.73 (m, 1H), 8.20-8.24 (br, 1H); 13C NMR δ 25.6, 75.5,
107.0, 108.9, 118.0, 126.1. Anal. Calcd for C6H8N2O2: C,
51.42; H, 5.75; N, 19.99. Found: C, 51.54; H, 5.88; N, 19.84.
4,4-Dimethyl-5-nitro-6-(1H-pyrrol-2-yl)hexan-2-one (5).
A mixture of 4 (1.40 g, 10.0 mmol) and mesityl oxide (1.26
mL, 11.0 mmol) was treated with DBU (4.57 g, 30.0 mmol).
The temperature rose immediately, and the reaction mixture
darkened. The reaction mixture was stirred at room temper-
ature for 24 h, diluted with ethyl acetate (100 mL), and
washed with water and brine. The organic layer was dried
(Na2SO4) and concentrated. Excess mesityl oxide was
removed under high vacuum. The resulting oil was dissolved
in a minimal amount of CH2Cl2 and filtered through a silica
pad [ethyl acetate/hexanes (1:3)] to afford a light brown oil
which solidified to a light-brown solid (1.85 g, 78%): mp
s, 1H); 13C NMR δ 24.3, 24.6, 26.8, 32.1, 36.9, 51.6, 94.9,
107.5, 108.9, 118.0, 126.2, 207.2. Anal. Calcd for
C12H18N2O3: C, 60.49; H, 7.61; N, 11.76. Found: C, 60.42;
H, 7.61; N, 11.48.
Screening Protocol for Reductive Cyclization of 5.
Reactions were done at the 0.1-mmol scale. Compound 5
was stirred in an appropriate solvent (0.2 mL) in the presence
of the reagents for reductive cyclization for the given time.
The reaction mixture was diluted with ethyl acetate (2 mL),
filtered through a plug of cotton, and analyzed by GC.
2,3,4,5-Tetrahydro-1,3,3-trimethyldipyrrin N-Oxide (6).
A mixture of 5 (2.38 g, 10.0 mmol), THF (25 mL), and water
(25 mL) was treated with NH4Cl (1.60 g, 30.0 mmol) and
zinc dust (9.80 g, 150 mmol) at 0 °C. The resulting
suspension was stirred vigorously at 0 °C for 15 min and
then was diluted with ethyl acetate and filtered. The filter
cake was washed with ethyl acetate. The combined filtrate
was washed with water and brine. The organic phase was
dried (Na2SO4) and concentrated under vacuum. The result-
ing oil was chromatographed [silica, ethyl acetate then CH2-
Cl2/MeOH (10:1)] to afford a pale yellow oil which solidified
to a white solid (1.36 g, 66%): mp 99-102 °C (lit.4 85-87
1
°C); H NMR (CDCl3) δ 1.12 (s, 3H), 1.19 (s, 3H), 2.05-
2.06 (m, 3H), 2.27-2.32 (m, 1H), 2.42-2.48 (m, 1H), 2.98-
3.08 (m, 2H), 3.88-3.91 (m, 1H), 5.93-5.94 (m, 1H), 6.06-
6.08 (m, 1H), 6.69-6.70 (m, 1H), 10.55-10.65 (br, 1H);
13C NMR δ 13.5, 23.0, 25.8, 27.9, 37.3, 47.2, 81.3, 106.3,
107.4, 117.6, 128.9; FAB-MS obsd 207.1479, calcd 207.1497
[(M + H)+, M ) C12H18N2O].
2-(1,3-Dinitroprop-2-yl)pyrrole (8). A small sample was
isolated from the reaction of pyrrole-2-carboxaldehyde with
excess nitromethane, as an orange oil. Limited data were
obtained: 1H NMR δ 4.34-4.41 (m, 1H), 4.70-4.79 (m,
5H), 6.05-6.07 (m, 1H), 6.15-6.17 (m, 1H), 6.73-6.75 (m,
1H), 8.30-836 (br s, 1H); 13C NMR δ 35.4, 76.4, 106.9,
109.5, 119.3, 124.5.
2,3,4,5-Tetrahydro-1,3,3-trimethyldipyrrin (5 f 1). A
stirred suspension of HCOONH4 (9.45 g, 150 mmol) and
nitrohexanone 5 (2.38 g, 10.0 mmol) in THF (40 mL) was
treated in one portion with zinc dust (9.76 g, 150 mmol).
The resulting mixture was stirred vigorously at room
temperature. After 2 h, ethyl acetate (40 mL) was added,
and the mixture was filtered. The filter cake was washed
with ethyl acetate. The filtrate was washed with water and
brine and dried (Na2SO4). The solvent was concentrated to
afford a light brown oil, which slowly solidified upon
standing. The crude product (1.65 g) was chromatographed
(silica, ethyl acetate) to afford a light brown oil which
solidified to a pale yellow-orange solid (1.15 g, 60%): mp
1
55-56 °C (lit.2 53-54 °C); H NMR δ 0.96 (s, 3H), 1.13
(s, 3H), 2.05-2.06 (m, 3H), 2.31 (AB, J ) 16.8 Hz, 1H),
3
2
2.39 (AB, J ) 16.8 Hz, 1H), 2.62 (ABX, J ) 11.6 Hz, J
1
3
2
59-61 °C (lit.2 54-55 °C); H NMR (CDCl3) δ 1.12 (s,
) 14.8 Hz, 1H), 2.80 (ABX, J ) 3.2 Hz, J ) 14.8 Hz,
1H), 3.64-3.67 (m, 1H), 5.95-5.97 (m, 1H), 6.11-6.13 (m,
1H), 6.70-6.72 (m, 1H), 9.75-9.83 (br. s, 1H); 13C NMR
δ 20.6, 23.0, 27.4, 28.1, 42.0, 54.4, 80.4, 105.3, 107.4, 116.5,
131.8, 174.6; FAB-MS obsd 191.1534, calcd 191.1548 [(M
+ H)+, M ) C12H18N2].
3H), 1.25 (s, 3H), 2.14 (s, 3H), 2.41 (AB, J ) 17.4 Hz, 1H),
2.59 (AB, J ) 17.4 Hz, 1H), 3.04 (ABX, 3J ) 2.4 Hz, 2J )
15.6 Hz, 1H), 3.35 (ABX, 3J ) 11.6 Hz, 2J ) 15.6 Hz, 1H),
5.13 (ABX, 3J ) 2.4 Hz, 2J ) 11.6 Hz, 1H), 5.95-5.99 (m,
1H), 6.08-6.10 (m, 1H), 6.65-6.67 (m, 1H), 8.10-8.13 (br
Vol. 9, No. 5, 2005 / Organic Process Research & Development
•
657

2,3,4,5-Tetrahydro-1,3,3-trimethyldipyrrin (6 f 1). A
solution of 6 (1.03 g, 4.99 mmol) in THF (20 mL) was
treated with HCOONH4 (4.78 g, 75.0 mmol) and zinc dust
(4.88 g, 74.7 mmol). The resulting suspension was stirred
at room temperature. After 2 h the mixture was diluted with
ethyl acetate and filtered. The filter cake was washed with
ethyl acetate. The combined filtrate was washed with water
and brine. The organic phase was dried (Na2SO4) and
concentrated. The resulting oil was chromatographed (silica,
ethyl acetate) to afford a yellow oil which solidified to white
mmol) was added in one batch. After 1 h, TLC analysis
showed the presence of starting material. Hence, another
batch of NaBH4 (2.80 g, 74.0 mmol) was added, and stirring
was continued. After 2 h, TLC analysis showed the complete
consumption of starting material, and GC analysis showed
the formation of 4. The mixture was filtered, and the filter
cake was washed with CH2Cl2. The filtrate was concentrated.
The resulting dark oil was dissolved in mesityl oxide (20
mL), and then DBU (13.7 g, 90.0 mmol) was added. The
resulting solution was stirred for 13 h. The crude product
was then filtered through a silica pad [ethyl acetate/hexanes
(1:3)] to afford a light brown oil (4.2 g, 29%): 1H NMR
data were consistent with values reported above. Anal. Calcd
for C12H18N2O3: C, 60.49; H, 7.61; N, 11.76. Found: C,
60.73; H, 7.79; N, 11.83.
1
crystals (0.62 g, 81%): the H and 13C NMR spectra were
identical as described above; mp 52-53 °C (lit.4 53-54 °C).
Anal. Calcd for C12H18N2: C, 75.74; H 9.53; N, 14.72.
Found: C, 75.61; H, 9.71; N, 14.63.
Direct Conversion of 2 f 4. A stirred solution of acetic
acid (2.00 mL, 35.0 mmol) in methanol (5 mL) under argon
at 0 °C was treated dropwise with n-propylamine (2.71 mL,
33.0 mmol). The resulting n-propylammonium acetate mix-
ture was stirred at 0 °C for 5 min, then added dropwise to a
stirred solution of 2 (5.70 g, 60.0 mmol) in nitromethane
(9.66 mL, 180 mmol) at 0 °C. The resulting mixture was
stirred at 0 °C. After 15 min, the cooling bath was removed,
and stirring was continued at room temperature. The color
changed from yellow to dark orange during the course of
reaction. After 2 h (see note below), CH2Cl2 (100 mL) was
added, and the organic phase was washed with water and
brine. The organic layer was dried (Na2SO4) and concentrated
under high vacuum to afford an orange-brown solid. The
crude 2-(2-nitrovinyl)pyrrole was dissolved in a mixture of
CHCl3 (300 mL) and 2-propanol (100 mL), to which silica
(72 g) was added. The mixture was stirred vigorously, and
NaBH4 (4.54 g, 120 mmol) was added in one batch. After
30 min, TLC analysis showed complete consumption of the
starting 2-(2-nitrovinylpyrrole). The mixture was filtered. The
filter cake was washed with CH2Cl2. The filtrate was
concentrated, and the resulting dark oil was filtered through
a silica pad (CH2Cl2) to afford an orange oil (3.52 g, 40%).
Direct Conversion of 4 f 1. A solution of 4 (2.10 g,
15.0 mmol) in mesityl oxide (2.83 mL, 22.5 mmol) was
treated with DBU (6.86 g, 45.0 mmol). The resulting dark
mixture was stirred for 14 h. The reaction mixture was diluted
with ethyl acetate (50 mL) and washed with water and brine.
The organic phase was dried (Na2SO4) and concentrated.
Excess mesityl oxide was removed under high vacuum. The
resulting brown oil was dissolved in THF (40 mL), and then
HCOONH4 (14.2 g, 225 mmol) and zinc dust (17.7 g, 225
mmol) were added. The resulting mixture was stirred
vigorously at room temperature. After 2 h, GC analysis
showed complete consumption of hexanone 5 and N-oxide
6. Ethyl acetate (50 mL) was added, and the mixture was
filtered through filter paper. The filter cake was washed with
ethyl acetate. The combined filtrate was washed with water
and brine. The organic phase was dried (Na2SO4) and
concentrated. The resulting oil was chromatographed (silica,
ethyl acetate) to afford a brown oil which solidified to a pale
brown solid (1.51 g, 53%). The characterization data (mp,
1H NMR, 13C NMR, and FAB-MS) were consistent with
values reported above.
Direct Conversion of 2 f 1 (60 mmol Scale). A stirred
solution of acetic acid (2.00 mL, 35.0 mmol) in methanol
(5 mL) under argon at 0 °C was treated dropwise with
n-propylamine (2.71 mL, 33.0 mmol). The resulting n-
propylammonium acetate solution was stirred at 0 °C for 5
min and then added dropwise to a stirred solution of 2 (5.70
g, 60.0 mmol) in nitromethane (9.66 mL, 180 mmol) at 0
°C. The resulting mixture was stirred at 0 °C. After 15 min,
the cooling bath was removed, and stirring was continued
at room temperature. The color changed from yellow to dark
orange during the course of the reaction. After 2 h (Note 1),
CH2Cl2 (100 mL) was added, and the organic phase was
washed with water and brine. The organic layer was dried
(Na2SO4) and concentrated under high vacuum to afford an
orange oil that solidified to give an orange-brown solid. The
crude 2-(2-nitrovinyl)pyrrole was dissolved in a mixture of
CHCl3 (300 mL) and 2-propanol (100 mL), to which silica
(72 g) was added. The mixture was stirred vigorously, and
NaBH4 (4.54 g, 120 mmol) was added in one batch. After
30 min, TLC analysis showed complete consumption of the
starting 2-(2-nitrovinyl)pyrrole. The mixture was filtered. The
filter cake was washed with CH2Cl2. The filtrate was
1
The H NMR data were consistent with reported values.2
Note: The time for completion of the reaction varied from
1 to 3 h. It is recommended to monitor the progress of the
reaction with TLC and carry out the workup immediately
upon disappearance of the starting material.
Direct Conversion of 2 f 5. A stirred solution of acetic
acid (2.00 mL, 35.0 mmol) in methanol (5 mL) under argon
at 0 °C was treated dropwise with n-propylamine (2.71 mL,
33.0 mmol). The resulting n-propylammonium acetate mix-
ture was stirred at 0 °C for 5 min and then added dropwise
to a stirred solution of 2 (5.70 g, 60.0 mmol) in nitromethane
(9.66 mL, 180 mmol) at 0 °C in a 100 mL flask. The
resulting mixture was stirred at 0 °C. After 15 min, the
cooling bath was removed, and the stirring was continued
at room temperature for 2 h. CH2Cl2 (100 mL) was added,
and the organic phase was washed with water and brine. The
organic layer was dried (Na2SO4) and concentrated under
high vacuum to afford a brownish oil. The brownish oil was
placed in a 1 L flask, and a mixture of CHCl3 (300 mL) and
2-propanol (100 mL) was added, followed by silica (72 g).
The mixture was stirred vigorously, and NaBH4 (4.54 g, 120
658
•
Vol. 9, No. 5, 2005 / Organic Process Research & Development

concentrated. The resulting dark oil was dissolved in mesityl
oxide (20 mL), and then DBU (13.7 g, 90.0 mmol) was
added. The resulting dark solution was stirred for 14 h. The
reaction mixture was diluted with ethyl acetate (150 mL)
and washed with water and brine. The organic phase was
dried (Na2SO4) and concentrated. Excess mesityl oxide was
removed under high vacuum, followed by entrainment with
hexanes (2×). The resulting brown oil was dissolved in THF
(160 mL), and then HCOONH4 (37.8 g, 600 mmol) and zinc
dust (39.2 g, 600 mmol) were added. The resulting mixture
was stirred vigorously at room temperature. After 3 h, GC
analysis did not show the presence of hexanone 5 and
N-oxide 6. Ethyl acetate (200 mL) was added, and the
mixture was filtered through Celite. The filter cake was
washed with ethyl acetate. The combined filtrate was washed
with water and brine. The organic phase was dried (Na2SO4)
and concentrated. The resulting oil was chromatographed
(silica, ethyl acetate) to afford a brown oil which solidified
to a pale brown solid (2.54 g, 22%). The characterization
stirred using a mechanical stirrer (see note iv below). After
2 h, GC analysis did not show any starting hexanone (5) or
intermediate N-oxide (6). The mixture was filtered through
filter paper, and the filter cake was washed with ethyl acetate
(2500 mL). The filtrate was concentrated to afford a brown
solid. The brown solid was chromatographed (6 cm × 22
cm, 260 g of silica) using ethyl acetate (3000 mL, ∼8 h),
affording a brown oil which slowly crystallized to a pale
brown solid (8.80 g). The mixed fractions were rechromato-
graphed on a small silica column to afford an additional 2.26
g of title compound (total yield 11.06 g, 14.5%; see note v
1
below). The characterization data (mp, H NMR, FAB-MS
data) were consistent with those described above.
Notes:
(i) Excess acetic acid is necessary because the use of
equimolar amounts of AcOH and propylamine in some
experiments causes extensive decomposition of starting
material.
(ii) The time for completion of the reaction varied from
4 to 7 h. It is recommended to monitor the progress of the
reaction with TLC and carry out the workup immediately
upon disappearance of the starting material.
1
data (mp, H NMR, FAB-MS data) were consistent with
those described above.
Direct Conversion of 2 f 1 (0.40 mol Scale). A mixture
of acetic acid (12.9 mL, 225 mmol) in MeOH (20 mL) was
treated dropwise with n-propylamine (16.44 mL, 200.0
mmol) at 0 °C (see note i below). The resulting mixture was
stirred at room temperature for 15 min and then transferred
to the solution of pyrrole-2-carboxaldehyde (38.0 g, 0.400
mol) in MeNO2 (64.5 mL, 1.20 mol) under argon at 0 °C.
The mixture was stirred at room temperature until TLC
showed complete consumption of the starting aldehyde (∼5
h, see note ii below). Excess MeNO2 was removed under
high vacuum (water bath, 40 °C). The resulting oil was
triturated with hexanes (50 mL), and the volatile components
were evaporated. This procedure was repeated three times.
The resulting brown oil was dissolved in CHCl3 and filtered
through a pad of silica (3 cm × 7 cm). The filter cake was
washed with CHCl3 (the total volume of solvent used was
2000 mL, see note iii below). The filtrate was transferred to
a 5 L flask. 2-Propanol (667 mL) and silica (480 g) were
added. Under vigorous stirring, NaBH4 (38 g, 1.0 mmol) was
added over the course of 1 min. The mixture was stirred for
45 min, then another batch of NaBH4 (10.0 g, 264 mmol)
was added and stirring was continued for 75 min. TLC
showed complete consumption of 2-(2-nitrovinyl)pyrrole (3).
The mixture was filtered through filter paper and the filter
cake was washed with CH2Cl2 (∼4000 mL). The filtrate was
concentrated to afford a dark brown oil. The resulting crude
2-(2-nitroethyl)pyrrole (4) was dissolved in mesityl oxide
(130 mL, 1.13 mol). DBU (91.0 g, 598 mmol) was added,
and the mixture was stirred overnight. Ethyl acetate (500
mL) was added, and the mixture was washed with water (3
× 200 mL) and brine (100 mL). The organic phase was dried
(Na2SO4). Removal of the solvent and excess mesityl oxide
under high vacuum afforded crude 5 as a brown oil. The
crude oil was dissolved in THF (1000 mL), and the resulting
solution was transferred to a three-necked 2-L flask.
HCOONH4 (252 g, 3.90 mol) and zinc dust (261 g, 3.90
mol) were added, and the resulting suspension was vigorously
(iii) Alternatively, the reaction mixture was diluted with
CH2Cl2 (∼300 mL) and washed with water and brine. The
organic phase was dried (Na2SO4). The CH2Cl2 was removed
under low vacuum, and then excess MeNO2 was removed
under high vacuum (water bath, 40 °C). The resulting oil
was triturated with hexanes (50 mL), and the volatile
components were removed. This procedure was repeated
three times. The resulting brown-orange solid was subjected
to reduction as described above.
(iv) The reductive cyclization was done under heteroge-
neous conditions and required a large amount of solid
material (Zn/HCOONH4). The heterogeneous mixture caused
difficulties in stirring, which affected the reaction time
required for completion. A mechanical stirrer is recom-
mended for effective stirring.
(v) Most intermediates (especially 3 and 4) were relatively
unstable upon standing in the crude reaction mixtures;
therefore, all operations as well as subsequent reaction steps
should be carried out promptly.
Acknowledgment
This work was funded by the NIH (GM36238). 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.
Supporting Information Available
Tables of data for survey reactions; 1H NMR spectral data
for 1 and 6. This material is available free of charge via the
Internet at http://pubs.acs.org.
Received for review June 2, 2005.
OP050087E
Vol. 9, No. 5, 2005 / Organic Process Research & Development
•
659