7254
B. Zhang, T. D. Lash / Tetrahedron Letters 44 (2003) 7253–7256
MS and NOE difference proton NMR spectroscopy,
and its major component shown to be the nickel(II)
pentamethyl DPEP-type porphyrin 2a.16 This nickel
complex is accompanied by a small amount of the C30
analogue 2b. To date, no syntheses of this unique
organic mineral have been disclosed. In relation to our
efforts to develop improved routes to sedimentary por-
phyrins, we now report the first total synthesis of
abelsonite. In addition, an improved methodology for
the synthesis of DPEP-type petroporphyrins is
demonstrated.
Scheme 2.
DPEP has been synthesized from chlorophyll a,9 and by
total synthesis.10–14 Only the latter allows for the possi-
bility of synthesizing related DPEP-type porphyrins. In
many syntheses, the five-membered ring is introduced
after the porphyrin macrocycle has been generated, but
overall yields are low.11 Flaugh and Rapoport intro-
duced the five-membered carbocyclic ring prior to
cyclization of a b-bilene intermediate 3, but this unit
had a disastrous effect on porphyrin formation and
DPEP was isolated in only 6% yield (Scheme 1).12 In
our studies, we have developed routes for the synthesis
of cyclopenta[b]pyrroles, and related dipyrroles, and
have made use of these compounds in the synthesis of
naturally occurring porphyrins. The MacDonald ‘2+2’
condensation was found to give good yields of the type
II DPEPs 4 (Scheme 2),13,17 but attempts to carry out
Scheme 3.
gave the acetoxy derivative 8 (quantitative) and this
was condensed with a-unsubstituted pyrrole 9 in the
presence of p-toluenesulfonic acid (p-TSA) in acetic
acid to give the dipyrrole 10. Hydrogenolysis over 10%
Pd/C gave the corresponding carboxylic acid 11. Decar-
boxylation with 2 equiv. of p-TSA, followed by formyl-
ation with benzoyl chloride–DMF under Clezy’s
modified Vilsmeier formylation conditions,21 gave the
aldehyde 12. This unit represents the ‘southern’ half of
petroporphyrins 1 and 2. The dipyrrolic precursors 13
corresponding to the ‘northern’ half were prepared by
condensing acetoxymethylpyrroles 14 with a-free
pyrrole 15 in the presence of Montmorillonite clay.22
Deprotection of the benzyl esters (H2–Pd/C) gave the
related carboxylic acids 16, and subsequent conversion
to the corresponding aldehydes 17 could then be carried
out under the conditions used to prepare 12. Condensa-
tion of 11 with 17a in the presence of p-TSA, followed
by treatment with HCl, gave the required b-bilene 18a.
Alternatively, 12 could be reacted with 16a to produce
the same tetrapyrrolic product. The latter route proved
to give superior yields and is the recommended proce-
dure for these syntheses. The b-bilene was reasonable
stable and could be recrystallized from ether and fully
characterized. Cyclization with TFA–trimethyl ortho-
formate, followed by air oxidation in the presence of
zinc acetate, gave DPEP in 30% yield. Only minimal
chromatography to remove polar impurities was neces-
sary. Reaction of 12 with 16b gave the pentamethyl
b-bilene 18b, and this was cyclized as previously
described to give porphyrin 19 in 30% yield. Porphyrin
19 is the demetallated form of abelsonite and the
synthesis was easily completed by reacting 19 with
nickel(II) acetate in refluxing chloroform–methanol to
give 2a in 89% yield. The related 3-unsubstituted por-
phyrin 20 was prepared similarly (35%).
cyclizations of a,c-biladiene
5 were unsuccessful
(Scheme 1).13 One factor that appears to aid the forma-
tion of porphyrins 4 by the ‘2+2’ route is that the
cyclizations occur while the carbon bridge linking the
carbocyclic ring is sp3 hybridized and this decreases
deleterious steric interactions at this critical stage. How-
ever, the MacDonald condensation cannot be used to
prepare DPEP because one of the two condensing
dipyrrolic units must be symmetrical in order to avoid
the formation of two isomeric porphyrin products. In
an attempt to overcome this problem by carrying out a
stepwise MacDonald condensation, b-bilenes 6 were
generated and cyclized under mild conditions (Scheme
3).14,18 Although this strategy was successful,14 and a
series of bacteriopetroporphyrins could be synthesized
in addition to DPEP itself,19 significant difficulties were
encountered. The b-bilene intermediates proved to be
very unstable and could not be purified. Instead, the
crude material was directly converted to porphyrin.
Unfortunately, etioporphyrin by-products were also
formed in these reactions and it was necessary to con-
vert the porphyrin mixtures into their nickel(II) chelates
and carry out extensive purification by flash chro-
matography. This severely limits the chemistry and does
not allow the synthesis of less soluble petroporphyrins.
The five-membered carbocyclic ring appears to be
responsible for destabilizing the b-bilene intermediates
and we hypothesized that this problem could be cir-
cumvented by constructing b-bilenes where the conju-
gated portion of the molecule was not directly
connected to this unit.
The cyclopenta[b]pyrrole tert-butyl ester 7 was pre-
pared from cyclopentanone using Knorr-type
chemistry20 (Scheme 4). Reaction with lead tetraacetate