I .. 9|
3 on (I...) (“111,00 \ . vs

\I . I l‘uf‘l‘ c0

1" it \(xclni! 1‘] It‘llplll‘lll ‘II -

1' LIIIODGDI‘ . .III‘III ll 3| 151% :1. it”

ll‘l IIIJ ‘1!!! f4. .‘

II. H‘nglfl'u I‘LHIHOIfiI‘l-vloul‘ {[1].}; .I 600””

II I“?! I‘lulfluflufll “gt ‘t’O-“rll

I l' I v'. . .IIIIJ

      

l ‘ ,

  

[11
d
I

 

‘7

1H (
I.
n ' f ‘1
I o I I l ' ‘ . ‘ ‘ 1'.
' .. " u‘n'nfl';3.v-‘"Eh,:p. [sz . 3-: ~ :
ll l i . 11H 'l“; ' ' ’l ,1 |,. ‘ . 1 . .
1 -$ v“ ‘1 ’ '._ ‘g _ u. “I
.r ‘ ‘.fl‘n 1 'l , ‘ ”A, .‘ ' ‘L‘ ' ‘1
' . "1 ' ', .; .1} ‘H '. ‘J ‘, A d A; in ' I' In t
.I $I‘ 45~ I {4 _. .'

 

Ci

II flu .‘AI

‘E
I).

 

W
5
£1
a | c . I ' ‘ I“
0. I I o 13 ~ ‘ .V‘ :(I; {

' c ' ‘ 3: ' ‘ gl

‘ ‘ | 1 ‘3! H 4 Ex 'I |

3 I s“. ‘ .f 1 5 I

v ‘ t 13 >\ ' IV
1'. “u . "I 4". I I '-
. - ‘ 3.! .'. 3 “‘ v ‘
. “3 .
1
$

 
  
  

mat-518

 
  

I. YEMRY
Miehigan State
University

 

This is to certify that the

dissertation entitled
THE SYNTHESIS OF OCTAMETHYLU ,5,'| ,5]PLATYRIN
presented by
0THA GRAY WEAVER, JR.

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Qhemi stry

/ Magor professor E E

MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

Date {fly/”2—

 

 

 

)V1ESI_J RETURNING MATERIALS:
Place in book drop to
“saunas remove this checkout from
“ your record. FINES will

 

 

be charged if book is
returned after the date
stamped below.

 

1

 

 

 

THE SYNTHESIS or OCTAMETHYLE1,5,1,51PLATYRIN

Otha Gray Weaver, Jr.

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1982

w

/ / /’. “/
\;’ s" ’

C"

ABSTRACT
THE SYNTHESIS OF OCTAMETHYL[1,5,1,SJPLATYRIN
By
Otha Gray Weaver, Jr.

The synthesis of the 26n-electron tetrapyrrolic macro-
cycle octamethyl[1,5,1,SJPlatyrin is described. The con-
densation of 2,7-bis(3,A-dimethylpyrrol-Z-yl)-2,3,A,5,6,9-
hexahydronaphthalene with 2,7-bis(5-formyl-3,N-dimethyl-
pyrrol-2-yl)-2,3,A,5,6,9-hexahydronaphtha1ene followed by
air oxidation gave octamethyl[1,5,1,SJplatyrin, I; Both
I and its diprotonated salt were deep violet and exhibited
an intense Soret-like absorption at 563 nm. The nmr Spec-
trum indicated that g was diatropic as expected in a
[26]annulene. Spectral data are presented for the struc-
tural proof of I and its precursors.

Various substituted 2,5—dipyrrol-2-ylpyrrolidines were
prepared in connection with the attempted synthesis of a
[l,3,1,3]heteroplatyrin. N-Protected 2,5-dipyrrol-2-yl-
pyrrolidines were efficiently produced by treating 2,5-
dipyrrol-2-ylpyrrolidine with the apprOpriate electrophile.

Otha Gray Weaver, Jr.

Several l-trifluoroacetyl-Z,5-dipyrrol-2-ylpyrrolidines
were prepared by the Mannich and Vilsmeier reactions. De-

protection of these substituted pyrrolidines led to several

new pyrrolidines.

 

To Laura

To Otha and Jewel Weaver

ii

E
Ll

ACKNOWLEDGMENTS

I would like to thank the Department of Chemistry at
Michigan State University for providing financial support
in the form of teaching assistantships for the past five
years. I would also like to express my appreciation to
Professor Eugene LeGoff for his guidance, patience, and

for arranging financial support.

iii

TABLE OF CONTENTS

Chapter

LIST OF TABLES. . . . . . . . . . .
LIST OF FIGURES .
INTRODUCTION. . . . . . . . . . . . . .
SYNTHESIS OF OCTAMETHYLE1,5,1,SJPLATYRIN. .
ATTEMPTED SYNTHESIS OF NEW PLATYRINS.
EXPERIMENTAL. . . . . . . . . . . .

General Procedure. . . . . . . . .

1,2,3,“,5,6,7,9—Octahydronaphthalene-
2,7-d10ne (%é) o o o o o o o o o o

2,7-Dimethoxy-l,A,5,8-tetrahydro-
naphthalene (l1) . . . . . . . . . . .

2-[7-(3,A-Ilmethyl-ZH-pyrrol-2-ylinene)-
2,3,A,5,6,9-hexahydronaphth-2-yl1-3sh-
dimethyl-lH-pyrrole tetrafluoroborate

an...“

2,7-Bis(3,A-dimethylpyrrol-Z-yl)-
2,3,h,5,6,9-hexahydronaphthalene (IQ).

2,7-Bis(5-formyl-3,A-dimethylpyrrol-
2-yl)-2,3,A,5,6,9-hexahydronaphthalene

(g9)

Octamethyl[1,5,l,51platyrin (gl) . . .

5,5'-Diformyl-3,h,3',A'-tetramethy1-
2,2'-d1pyrry]methane (£3) . . . .

5,5'-Bis(3-oxo-butenyl)-3,3',u,u'-
tetramethyl-2,2'-dipyrrylmethane (gé).

2-(3-Oxo-propenyl)pyrrole (@1) -

iv

Page
. vi

.vii

17
31
50
50

50

50

51

51

52

53

5A

5A
55

Chapter Page

2-(3-Oxo-butenyl)pyrrole (33) . . . . . . . . 55
1, 5- Di- 2-pyrryl- l ,3 -pentadien-

5- -one (3A) . . . . . . . . . . . . . . . . 56
3, 3' ,A ,A'-Tetramethyl- 2 ,2'-dipyrryl-

methene hydrobromide (36). . . . . . . . . 56
3, 3' ,A ,A'-Tetramethyl- 2 ,2'-dipyrryl-

methane (31) . . . . . . . . . . . . . . . . 56
2,5-Dipyrrol-2—ylpyrrolidine (Aé). . . . . . . 57
l- (p- Tosyl)- 2 ,L .dipyrrol- 2-ylpyrrolidine

(Ag) . . . . . . . . . . . . . . . . 57
l- -Acetyl- 2, 5— dipyrrol- 2-ylpyrrolidine

(AZ) . . . . . . . . . . . . . . . . 57

l- -Acetyl- 2, 5- bis(5- formylpyrrol- 2- yl)-
pyrrolidine (Ag) . . . . . . . . . . . . . . 58

l- -Trifluoroacetyl- 2, 5- -dipyrrol- 2- yl-
pyrrolidine (“2) . . . . . . . . . . . . 59

l- -Trifluoroacetyl- 2- (5— —formyl-
pyrrol- 2—yl)- L pyrrol- 2—ylpyrrolidine

(gown... 59

2-(5-Formylpyrrol-2-yl)-5-pyrrol-2-
ylpyrrolidine (fig). . . . . . . . . . . . . . 6O

l-Trifluoroacetyl-2,5-bis(5-dimethyl-
aminomethylpyrrol-2-y1)pyrrolidine (Q2). . . . 61

2,5-Bis(5-dimethylaminomethylpyrrol-2-
yl)pyrrolidine (33) . . . . . . . . . . . . . 61

APPENDIX. . . . . . . . . . . . . . . . . . . . . . 63
BIBLIOGRAPHY. . . . . . . . . . . . . . . . . . . . 77

Table

LIST OF TABLES

Page
Nuclear Magnetic Resonance Ab-
sorption of Octamethyl[1,5,l,5]-
platyrin and Related Compounds . . . . . . 25
Electronic Absorption of Octa-
methyl[1,5,l,5]platyrin and
Related Compounds. . . . . . . . . . . . . 28

vi

Figure

LIST OF FIGURES

Page
Graph of delocalization energy vs
number of n-electrons. . . . . . . . . . . . A
Illustration of diamagnetic ring
current effect . . . . . . . . . . . . . . 7
Infrared spectrum of 2,7-Bis(3,A-
dimethylpyrrol-Z-yl)—2,3,A,5,6,9-
hexahydronaphthalene (l2). . . . . . . . . 63
Infrared spectrum of 2,7-Bis(5-formyl-
3,A-dimethylpyrrol-2-yl)-2,3,A,5,6,9-
hexahydronaphthalene (29). . . . . . . . . 6A

Infrared spectrum of l-Imetyl-2,5-
bis(5—formy1pyrrol-2-yl)pyrrolidine

(fig) . . . . . . . . . . . . . . . . . . . 65
Infrared spectrum of l—Trifluoroacetyl-
2,5-dipyrrol-2-ylpyrrolidine (Ag). . . . . 66
Infrared spectrum of l-Trifluoro-
acetyl-2-(5-formylpyrrol-2-yl)-5-
pyrrol-2-ylpyrrolidine (£9) . . . . . . . 67
Infrared spectrum of lJTrifluoroacetyl-
2,5-bis(5-dimethylaminomethylpyrrol-
2-yl)pYrrolidine (ég). . . . . . . . . . . 68

vii

Figure Page

9 PMR spectrum of 2-[7-(3,A-Dimethyl-
2H-pyrrol-2-ylinene)-2,3,A,5,6,9-
hexahydronaphth-Z-yIJ-3,A-dimethyl-
lH-pyrrole tetrafluoroborate (lg). . . . 69

10 PMR spectrum of 2,7—Bis(3,A-dimethyl-
pyrrol-2-yl)-2,3,A,5,6,9-hexahydro-
naphthalene (l9) . . . . . . . . . . . . 69

ll PMR spectrum of 2,7-Bis(5-formyl-3,A-
dimethylpyrrol-2-yl)-2,3,A,5,6,9-
hexahydronaphthalene (2Q). . . . . . . . 7O

12 PMR spectrum of 5,5'-Bis(3-oxo-butenyl)-

3,3',A,A'-tetramethyl-2,2'-d1pyrry1-

methane (25) . . . . . . . . . . . . . . 7O
13 PMR spectrum of 2-(3-Oxo-propenyl)-

pyrrole (3%) . . . . . . . . . . . . . . 71
IA PMR spectrum of 2-(3-Oxo-butenyl)-

pyrrole (33) . . . . . . . . . . . . . . 71
15 PMR spectrum of l,5-Di-2-pyrryl-l,3-

pentadien-S-one (3A) . . . . . . . . . . 72
16 PMR spectrum of l-(p-Tosyl)-2,5-

dipyrrol-2-ylpyrrolidine (A6). . . . . . 72
17 PMR spectrum of l-Acetyl-2,5-dipyrrol-

2-ylpyrrolidine (Ax) . . . . . . . . . . 73
18 PMR spectrum of l-Trifluoroacetyl-2,5-

dipyrrol-2—ylpyrrolidine (Ag). . . . . . 73

viii

Figure

19

2O

21

22

23

PMR spectrum of l—Acetyl-2,5-bis-
(5-formylpyrrol-2-yl)pyrrolidine (A9).
PMR spectrum of l-Trifluoroacetyl-
2-(5—formylpyrrol-2-yl)-5-pYrrol-
2-ylpyrrolidine (5Q)

PMR spectrum of 2-(5-Formylpyrrol-2-
yl)-5-pyrrol-2-ylpyrrolidine (5T).

PMR spectrum of l-Trifluoroacetyl-
2,5-bis(5-dimethylaminomethylpyrrol-2-
ylpyrrolidine (52)

PMR spectrum of 2,5-Bis(5-dimethyl-

aminomethylpyrrol-2-yl)pyrrolidine

(as)

ix

Page

7A

7A

75

75

76

 

INTRODUCTION

Since Kekule1 first proposed a structure for benzene
in 1865, there has been an outpouring of theories and
predictions regarding the characteristic properties of
certain annulenes (monocyclic conjugated polyenes), of
which benzene is the most notable. For example, the ex-
ceptional thermodynamic stability of benzene and its ten-
dency to undergo substitution reactions instead of addition
to its double bonds contrasts markedly with the correspond-
ing behavior of cyclooctatetraene. From 1865 to the 1920's
numerous structural explanations were put forth by Baeyer,2

A and others but it was not until the develop-

Claus,3 Meyer,
ment of quantum mechanics in the 1920's that convincing
explanations were proposed. In 1927 Heitler-London5
introduced the valence bond theory and in 1931 Hfickel6
published his molecular orbital theory. Both theories

try to explain aromaticity on the basis of the ground state
of the molecule rather than some transition state property.
The different assumptions in these theories resulted in

one being more accurate in its predictions and thus more
useful to chemists.

Valence bond theory is based on the premise that an

atom retains much of its individual character even when

 

——4—__’—_———__— _.__._.....___.

part of a covalent bond. Bonding is characterized by an
overlap of atomic orbitals and a sharing of an electron
pair. In order to explain the bonding in the symmetrical
benzene molecule, valence bond theory introduced the con-
cept of "resonance" and "resonance energy".

The valence bond treatment starts with individual
atoms and considers the interaction between them. The
resulting wave functions may be regarded as an approxima-
tion to the wave function for the ground state.23 A
suitably-chosen combination of these wave functions will
then provide the wave functions which correspond to the
two Kekule1 structures of benzene which are the major
contributing wave functions to the combination. Further—
more, an estimate of the energy based on the mixture of
the functions will be lower than an estimate based on any
of the individual functions making up the combination.

The difference between the theoretical energy of a par-
ticular function or resonance structure and the experi-
mentally determined energy is called the resonance energy.
For benzene, the.resonance energy is determined by comparing
its energy to that of three isolated olefinic bonds.

The problem with the valence bond theory is that it
predicts a sizeable resonance energy for such compounds
as cyclobutadiene and planar cyclooctatetraene.7 This
has not been observed and consequently its use as a pre-
dictive tool has been diminished.

In molecular orbital theory orbitals on atoms are

brought together and a linear combination of these atomic
orbitals produces a series of molecular orbitals of vary-
ing energy. Electrons are placed in these molecular or-
bitals, starting with the lowest energy orbital and fol-
lowing Hund's Rule and the Pauli Exclusion principle, until
all the electrons have been placed in orbitals. From this
theory evolved a general rule which became known as Hfickel's
Rule. It states that "amongst fully conjugated, planar
monocyclic polyolefins only those possessing (An+2) n-
electrons, where a is an integer, will have special aro-

matic stability."6

As a corollary, all similar systems
containing An n—electrons will not possess any aromatic
stability.

However, with the larger monocyclic ring systems
Hfickel's Rule and HMO theory begin to show inconsistencies.
For example, HMO theory predicts that the energy dif-
ference between An and (An+2) n-electron systems approaches
zero at higher p values and that the resonance energy of
both An and (An+2) n-electron systems increases as ring
size increases (see Figure l). The former statement
has been shown to be correct with the higher annulenes but
the latter statement does not agree with the observed
facts. In addition to this latter criticism, it is evi-
dent from Figure 1 that HMO theory predicts resonance
stabilization for even the An series. This is not the

case. The lower members of the An series actually show

OHIO

'0!
I

9y (eV)
.0 t—
o: o

 

AA
v VV\/\

.6
o:

 

”l
4 6

Resonance Ener
o
o

 

 

{2 IS. 20' 2'41 ée‘
7T-Electrons

Figure 1. Graph of delocalization energy vs. number of
n-electrons. ‘

a sizeable destablilization - a negative delocalization
energy. Thus Hfickel's Rule does give a good qualitative
prediction of delocalized ring systems, but gives very poor
quantitative results for delocalization energies.

In order to remove the quantitative discrepancies of
the HMO method, many theoreticians since 1931 have sought
better quantum-mechanical methods, using fewer or more
logical approximations in the calculations. Dewarll
published some of his work in 1965. By using accurate

l2

geometries and bond lengths and following Pople's scheme

for closed shell molecules he calculated resonance energies

for the annulenes as plotted in Figure l. The An annu-
lenes have negative resonance energies and localized
double bonds whereas (An+2) annulenes possess positive
resonance energies and are delocalized systems. The value
of the resonance energy decreases with increasing ring
size to a point (zero resonance energy) at which a (An+2)
system becomes a simple polyolefin. This occurs somewhere
between [22] and [26]annulene.

Based on these results Dewar proposed the following
definition for aromaticity:l3 Cyclic conjugated systems
are considered aromatic if cyclic delocalization of elec-
trons make a negative contribution to their heats of
formation. It should be mentioned here that even Dewar's
calculations show too rapid of a decrease in resonance
energies for the (An+2) n-electron systems, but they have
produced the closest agreement with experimental data for
compounds having large n values. In 1970 FigeysllI con-
firmed Dewar's results through another molecular orbital
method.

Both Dewar's and Hfickel's works were concerned with
the resonance energy of cyclic conjugated molecules. De-
war's calculations can be and in most cases have been
verified by experimentally determined heats of combustion
and hydrogenation. Unfortunately, these procedures are
tedious to perform. However, there are other measurable
physical prOperties of annulenes which have been used to

categorize these cyclic conjugated systems. Two of the

most commonly used methods are electronic absorption spec-
troscopy and nmr spectroscopy.15

Electronic absorption spectroscopy is based on the
fact that cyclic, delocalized systems absorb light in the
visible and ultraviolet region of the spectrum. This
absorption corresponds to electronic transitions from
lower to higher energy states. While this transition
takes place between the occupied bonding n-molecular or-
bital and the corresponding antibonding orbital in both
simple olefins and conjugated systems the energy required
for the electronic transition will be less for conjugated
systems and absorption bands will be observed at longer
wavelength. The highest-filled molecular orbitals of
cyclic, fully conjugated systems are degenerate so that a
number of different transitions are possible. For ex—
ample, benzene has three absorption regions called the
8, para, and a-bands. Another phenomenon of these systems
is that for a given series of compounds the ultraviolet-
visible spectra can possess the same general shape.16
Consequently, the identification of new compounds is greatly
facilitated.

The delocalization of n—electrons in these conjugated
systems has a further advantage in that the electron flow,
similar to a current loop, gives these molecules a diamag-
l7

netic ring current. This ring current sets up a small

magnetic field, H', which Opposes any external magnetic

field, H inside the ring and strengthens it outside the

o’
ring as shown in Figure 2. When placed within the magnetic
field of an nmr spectrometer the protons of the compound
which are inside the ring will be shielded by the induced
magnetic field and appear at a higher field than its non-

aromatic counterpart. Protons external to the ring will

Induced
current

 

 

 

v \/

Induced }
fund

Figure 2. Illustration of diamagnetic ring current effect.

be shifted to a lower field (deshielded) in the nmr spec-
trum. Molecules showing this effect in the nmr spectrum
are named diatropic.“7 Conversely, molecules containing

An n-electrons should sustain a paramagnetic ring current,

a consequence of which will be to deshield the inner protons
and shield the outer ring protons. Substances showing this
effect are conveniently named para’cropic.’47 Compounds

which show no ring current are called atropic.“7 The ease
with which nmr spectra can be obtained makes the use of

nmr spectroscopy a convenient and qualitative method for
determining paramagnetic or diamagnetic effects in molecules.

18 was the first to use nmr spectroscopy in

Sondheimer
the analysis of annulenes and dehydroannulenes. These
compounds are well suited for this technique for they can
possess both internal and external protons. In order to
prevent interchange of interior and exterior protons the
nmr spectra of annulenes must be taken at low temperatures.
The acetylenic bonds in the dehydroannulenes are present to
make the molecules more rigid and less likely to rotate
into non-equivalent conformers. In fact, the nmr spectra
of 1,7,l3—tridehydro[18]annu1ene is essentially unchanged
upon heating to 150°C with inner and outer protons still
appearing as fairly sharp peaks.19

Of the (An+2) annulenes synthesized by Sondheimer

18 and [22]annu1ene20 (l) show

[lA]annulene,18 [18]annulene
a definite ring current with inner proton peaks upfield from
tetramethylsilane at60.6, -3.0 and -0.A to -l.2 ppm, respec—
tively, and outer proton peaksat 67.9, 9.3 and 8.5 to 9.65
ppm, respectively. Both [26] and [30]annulene have been

prepared and by nmr analysis seem to be atropic for there

are no discreet inner and outer protons.18 The nmr spectra
of the (An+2) dehydroannulenes from [1A] to [22]dehydroannu-

18 show a similar effect as the [1A] to [22]annulenes

lene
and the [26] (g) and [30]dehydroannulenes failed to show a
ring current even at temperatures as low as minus 60°C.

Only a broad multipletat 65.5 to 8.0 ppm was obtained for

[26]dehydroannulene.

_\

10

The major drawback in Sondheimer's work is that the
geometry of the molecules vary as the ring size is increased.
Thus, any change in the nmr spectra may be attributable to
some geometrical variation and not the number of n-electrons.

In an attempt to eliminate this problem Nakagawa22

synthe-
sized an entire series of didehydro[An+2]annulenes and two
tetradehydro[An+2]annu1enes.

The nmr spectra of the tetra-t-butyltetradehydro[18]

and [22]annulenes (3) are similar in that the outer protons

\

of both systems occur around 610 ppm and the inner protons
differ by only 1.A ppm with the 18 n-electron system having
the higher field peak at G-A.92 ppm.22 This is an upfield
shift of 2.0 ppm from that of [18]annu1ene and 2.7 ppm for
[22]annulene.

The results obtained from the nmr's of the didehydro-
[An+2]annulenes indicate that the difference in chemical
shift for the inner and outer protons (do-61) progressively

narrows as the ring size increases. A maximum value of

11

13.8 ppm occurs with tetra-t-butyldidehydro[1AJannulene and
a minimum value of A.0 ppm is reached with tetra-t-butyldi—

dehydro[30]annulene.22

The theoretical prediction that bond
alternation increases in the [An+2]annulenes with ring size
is demonstrated in the nmr's of this series of [An+2]annu-
lenes. However, assessing the question of polyene vs. aro-
matic character in these series and comparison of these
dehydroannulenes to Sondheimer's annulenes is difficult
because of the unknown acetylene-cumulene anisotropy effects
as well as the unknown effect of having shortened fixed
bonds in the annulene perimeter.

A series of compounds which contain no acetylenic bonds
are the methano[An+2]annulenes of Vogel. l,6-Methano[10]-

annulene (A) exhibits a ring current effect with the ole-

finic protons appearingat 66.95 and 7.3 ppm and the methylene

 

protons at -0.52 ppm.23 In contrast to Nakagawa's systems
the next higher homologue syn-1,6:8,l3-bismethano[lA]annu-
lene (5) shows a stronger ring current with the methylene

protons resonatingan;50.9 (endo) and -l.2 ppm (exo) and the

outer protonsen;67.A to 7.9 ppm.27 Because of the steric

l2

interaction between endo protons on the two methylene bridges

the ring tends to pucker which prevents maximum orbital over-

lap. This steric interaction is magnified in the 18n-

electron system and could be the culpret causing a reduced

ring current effect relative to the lAtt-electron system.25

A family of [An+2]annulenes which contain no acetylenic
bonds and have fewer steric problems than Vogel's compounds

26

are the bridged [An+2]annulenes of Boekelheide. The syn-

thesis of the lAn and le—electron systems have been ac-

complished and work is progressing on a 22n-electron system.

The highly planar perimeter of these molecules gives trans-

15,l6-didehydropyrene (6) and hexahydrocoronene (7) some of

 

27

13

the highest upfield peaks known for diamagnetic type com-

pounds. The cavity protons in 6 occuran:5-5.A9 ppm and hexa-
hydrocoronene exhibits peaksafi16-6.5A, -6.9A and -7.96 ppm.26
An increase in ring current is observed as the ring size is

increased.

O
@181

3e

/

If Boekelheide's 22n-electron system exhibits a further
upfield shift of its protons relative to hexahydrocoronene
then this would be the first series of compounds which
consist of three or more members that shows an increase in
ring current with an increase in ring size. This would
agree quite well with the theoretical conclusions drawn

by R. c. Haddon.28

He demonstrated for the first time that
there is a direct mathematical dependence between the reson-
ance energy of a molecule and its ring current as it applies
to (An+2) systems. Specifically, Haddon concludes that the

ring current increases linearly with n (number of atoms in

1A

conjugation) whereas the resonance energy is inversely
pr0portional to 8.28

Although the brilliant work of Sondheimer, Nakagawa,
and others has provided numerous annulenes which disagree
with Haddon's theory these compounds are either conforma-
tionally mobile, contain acetylene linkages or are nonplaner.
Therefore, these compounds should not be considered in the
evaluation of Haddon's theory.

The ideal molecules for evaluating ring currents are
those having rigid, planar perimeters. These criteria
are presently best met in the hetero-annulenes. These are
species in which one or more of the carbon atoms in the
annulene ring are replaced by a hetero-atom such as nitrogen
or oxygen. Perhaps the best known hetero[An+2]annulene is

the porphyrin molecule 8. Porphyrins are stable 18n-e1ec-
’b

tron systems and show a substantial ring current effect.

15

The prospect looked good that a 22n—electron system based
on the structure of the porphyrin molecule would also
exhibit a ring current. Such a system could be synthesized
by expanding two opposite meso-bridges from one carbon to
three carbon atoms. An expansion to five carbon atoms would
give a 26n-electron system.

Indeed, in 1978 an expanded porphyrin molecule (9) was

prepared by E. LeGoff and R. Berger29

and was given the
name platyrin from the Greek word "platys", meaning broad

or wide. These diatropic 22r-e1ectron systems exhibit a

 

larger ring current than octamethylporphyrin.3O It's shift
values correspond closely with decamethylsapphyrin30 IQ

which is also a 22n-electron system.

 

 

‘N u \Ntd O‘i‘
A1. .1. .n..
nC Oh . t s
"an e

0‘,
a 169

16

 

The success of this synthesis prompted an interest in
the synthesis of the next higher homolog. In accord with
Haddon's theory, this [l,5,l,5]platyrin (numbers designating
bridging carbon atoms between each pair of pyrroles) should
exhibit a larger ring current than Berger's [1,3,l,3]platyrin
but have a smaller resonance energy. The nmr should re-
flect this with upfield peaks shifted further upfield from
tetramethylsilane than in [1,3,l,3]platyrin. Because the
literature contains so few 26n-electron systems the suc-
cessful synthesis of such a platyrin would not only sub-
stantiate Haddon's theory, but could possibly determine the

point at which (An+2) systems become simple polyenes.

SYNTHESIS OF OCTAMETHYL[1,5,1,5]PLATYRIN

The methods used for the linkage of two pyrrolic units
through a single carbon bridge are based on the fact that
pyrroles are very susceptible toward electrophilic attack
at an a-position. Of the many reactions of this type used
to construct the basic classes of dipyrrolic units (pyrro-
methanes, pyrromethenes, and pyrroketones) the classical
technique32 is the acid catalyzed condensation of an a-formyl-
pyrrole, for example ll, with a pyrrole having a free a-
position to provide a dipyrromethene (12) (Scheme 1). This
reaction is generally facile and affords fairly high yields.
However, for each individual case extensive experimentation

must be conducted to establish the best reaction conditions.

 

H’ow

   

Scheme 1

17

18

 

Scheme 2

19

A successful extension of this reaction to the prepara-
tion of [l,3,1,3]p1atyrin 9 has been developed by Berger.33
Condensation of a 5,5'-diformyldipyrrotrimethine with a
5,5'-unsubstituted dipyrrotrimethine under acidic conditions,
followed by air oxidation of the intermediate cyclic product
gave a 19% Yield of 9 (Scheme 2).

The synthetic approach adopted in our quest for a 26n-
electron platyrin was modeled after Berger's work, making
use of the aldehyde coupling reaction. Thus, a 5,5'—un-
substituted dipyrropentamethine (13) when condensed with
a 5,5'-diformyl dipyrropentamethine (15) should afford a
rectangularly shaped [l,5,l,5]platyrin such as 16 (Scheme
3).

 

Scheme 3

The compound chosen as the pentamethine unit must be
capable of holding the platyrin in a planar, conformationally—

rigid orientation. A completely planar molecule maximizes

20

orbital overlap, and without rigidity the molecule would
assume lower-energy conformations to minimize crowding
between N-H and interior hydrogens. By using the decalin
system 15 as the pentamethine bridge, the desired rigidity

can be achieved.

I! I C)

15

10

The insertion of 15 between two pyrroles was accomplished
by treating a refluxing solution of IX and 3,A-dimethy1-
pyrrole with tetrafluoroboric acid. Under the acidic condi-
tions of this reaction, compound 11 rearranges to 15, which
immediately reacts with 3,A-dimethylpyrrole to afford the
dark green, powdery solid $8 in quantitative yield as shown
in Scheme A. The absorption maximum at 658 nm in the ultra—
violet-visible spectrum of 18 correlates well with Berger's
tetramethyltrimethine (Amax 580 nm) and 3,3',A,A'-tetra-

methyl—2,2'-dipyrrylmethene hydrobromide (Ama A90 nm) -

x
an average increase of 85 nm for every double bond added to
the system. The mass spectrum of 18 was characteristic of
such salts, with loss of tetrafluoroboric acid from the

parent compound being the dominant peak. For 18 this peak

was at m/e 318 (C22H26N2).

21

18 '3“

Scheme A

Reactivation of the pyrrole 2-positions toward electro-
philic attack was effected by treatment of l8 with sodium
borohydride, furnishing the conjugated dienyl bis-pyrrole l9.

Compound T9 was found to be extremely sensitive to air

 

19

22

oxidation and had to be handled under an inert atmosphere to
prevent decomposition. With rigorous exclusion of oxygen,
19 could be successfully formylated under standard Vilsmeier
conditions to the corresponding a-formyl pyrrole 20 in 67%
yield. The nmr spectrum of 29 showed two equal singlets

at 69.30 and 9.25 ppm for the aldehydes. This was expected,

20

since the dienyl system in 29 is so constituted that the
two aldehydes are not equivalent. The methyl groups on the
two pyrrole units are also nonequivalent and gave distinct
signals in the nmr spectrum.

With the two halves of the target platyrin in hand, at-
tention was focused on the final condensation. The union
of IQ and 20 took place in about 3% yield by the action of
aqueous hydrobromic acid in methanol. Air was bubbled
through the medium to effect an in gitg oxidation and
chromatography of the crude reaction mixture on neutral

alumina afforded a violet material presumed to be

23

[l,5,l,5]platyrin gl-

 

The most convincing piece of data in support of both
the structure and the diatropic nature of the [l,5,l,5]-
platyrin system was the 250 MHz nmr spectrum of a trifluoro-
acetic acid/dimethylsulfoxide “d6 solution of the diprotonated
[l,5,l,5]platyrin 22. This spectrum displayed a signal at
511.75 ppm. (protons a in 2%), generated by the exterior
meso protons; two sharp singletsat 6A.51 and A.A3 ppm (1:1
ratio), assigned to the eight methyl groups on the pyrrole
rings; a series of broad peaksat 61.0-2.6 ppm, attributed to
the aliphatic ring protons; and finally two broadened sing-
lets upfield from tetramethylsilaneat 6-10.58 (protons b)
and -1A.26 ppm (protons c) in a ratio of 1:1, associated

with the imino and interior meso protons, respectively.

2A

Pa

 

The disappearance of the broad peak at 6—10.58 ppm upon ad-
dition of deuterium oxide permitted assignment of this peak
to the four exchangeable imino protons. For comparative
purposes, the nmr data for a porphyrin (an an-electron
system), sapphyrin (a 22n-electron system), [l,3,1,3]platyrin,
and 22 are presented in Table 1. Each of these compounds
displays the same general spectral characteristics, con-
sisting of well separated single peaks with no observable
coupling between the individual protons.

An interesting solvent effect was noted in the 1H-

nmr spectra of 22. In acetonitrile-d3 the imino and

25

Table 1. Nuclear Magnetic Resonance Absorption of Octa-
methy1[1,5,l,5]platyrin and Related Compounds.

 

 

 

Compound CH3 Meso CH NH
OctamethylEl,5,1,5]platy-b u.51 11.75d -1o.58
rin bis-trifluoroacetate A.A3 -1A.26e
salt
Octamethyl[l,5,1,5]p1atyrinc u.u1 11.71: -6.67
bis-trifluoroacetate salt A.33 -10.Al
Octamethylporphyrinf -
bistrifluoroacetate 3.62 10.53 -A.52
salt
Decamethylsapphyrinf u.22,h.19 11.71 —u.8u
bistrifluoroacetate A.08,A.0A 11.80 —5.00
salt l:2:1:l 1:1 -5.A6

2:1:2
Bis(trimethylene)octa-c A.22 11.6Ag
methylplatyrin bis-tri- A.l7 -8.97 -5.6

fluoroacetate salt

 

 

aAs d-values referred to internal tetramethylsilane.

bTaken in dimethylsulfoxide-d6.

0Taken in deuterochloroform with trifluoroacetic acid.

dExterior of ring.

6Interior of ring.

fTaken in deuterochloroform.

26

interior meso protons displayed signals at 6-8.80 and -12.82
ppm, respectively, whereas in deuterochloroform these sig-
nals appeared at 6-6.67 and -lO.A1 ppm. This characteris-
tic shift in going from solvents of high to low dielectric
constant has also been observed in a diprotonated deca-
methylsapphyrin and octamethylporphyrin (Table 1).

Both dimethylsulfoxide and acetonitrile contain func-
tionality which generate a small magnetic field when placed
in the magnetic field of an nmr spectrophotometer. A
platyrin molecule in either of these solvents will orient
itself so that the magnetic fields of solvent and platyrin
molecules are aligned. Platyrin protons which are simul-
taneously within the magnetic fields of both solvent and
platyrin experience a stronger magnetic field than either
platyrin or solvent is capable of generating separately.
The result is additional shielding of interior and de-
shielding of exterior ring protons. In acetonitrile the
shielding effect of the solvent is less than in dimethyl-
sulfoxide and smaller upfield shifts of interior protons
are observed. Since deuterochloroform molecules do not
generate an electric field the chemical shifts of the
platyrin protons are solely attributable to the diamagnetic
ring current of the macrocycle.

Fully-conjugated monocyclic ring systems exhibit an
extremely intense absorption band, the Soret band, in their

ultraviolet-visible spectra. The absorption maximum of

27

the Soret band occurs around A00-A20 nm for porphyrins,
A55 nm for sapphyrins, and A77 nm for [l,3,l,3]platyrins
(Table 2). The Soret band in the spectra of 22 has a wave-
length maximum of 538 nm (in chloroform) which is a bath-
ochromic shift of 61 nm from the Soret band of [l,3,l,3]-
platyrin (2). In going from diprotonated octamethylpor-
phyrin to diprotonated [l,3,l,3]platyrin (Table 2) a shift
to longer wavelength by 63 nm is observed. A calculation
of the extinction coefficient for the Soret band of 22
determined it to be 376,000, a value which compares favor-
ably with the extinction coefficients of [l,3,l,3]platyrin
(398,000) and octamethylporphyrin dihydrochloride (266,700).
A further study of the visible spectrum of 22 under
acidic conditions revealed another chemical property pos-
sessed in common with the porphyrins. The Soret band of the
dication 22 was considerably more intense than that of the
neutral platyrin. This difference can be explained by con—
sidering the symmetry of the dication verses the neutral
platyrin. In the dication all four nitrogen atoms are
equivalent, but in the neutral molecule this is not the
case, as only two of them are bonded to hydrogens. This
equivalency increases the symmetry of the molecule by a
factor of two and causes a loss of degeneracy in the
n-n* transitions. The consequence is a more intense Soret
band.

Another characteristic of [l,5,l,5]platyrin, which

28

Table 2.

Electronic Absorption of 0ctamethyl[l,5,l,5]-

platyrin and Related Compounds.

 

 

Compound

xmax nm (e)

 

OctamethylEl,5,l,5]platyrina

OctamethylEl,5,l,5]platyrina
bis-trifluoroacetate salt

Decamethylsapphyrinb

dihydrochloride

Decamethylsapphyrinb

Bis(trimethylene)octa-a
methylplatyrin

Bis(trimethylene)octa-a
methylplatyrin bis-
trifluoroacetate salt

"95(123,000), 536(1uu,000),
651(15,700), 705(12 300),
718(13,500), 780(9,5oo).

“95(98,000), 536(376,ooo).
651(u7,000). 703(32,000),

719(3u,ooo), 780(2u,ooo),

830(19,000).

u3l<56,ooo>, u56.5(59u,000),
579(3,uoo), 625(1h.000).
677(20,000), 689(17,200).

“55(329,000), 533, 59", 6H3,
660sh, 720, 730.

“53sh(60.u00). u77<398.000).
607(11.800). 6u9(9.3u0).
7u7(2,100). 767sh(1,52o).
8u6(1,850).

“538h(53,200), ”77(398,000),
6258h(9,030), 637(10,700),
6u7sh(9,130), 672(53190):
688(6,6N0), 705(7,990):
717Sh(6,220)3 73uSh(33630)3
788(6,220).

 

 

aMeasured in methylene chloride.

bMeasured in chloroform.

29

limited the methods used in its identification, was its

poor chemical stability. For example, the electronic
spectrum of a concentrated sample displayed a greatly re-
duced Soret band on standing, as compared with a freshly
purified sample. The possibility that 22 might be light
sensitive was studied. When the reaction of 22 with 29 was
performed in covered reaction vessels under subdued lighting,
a slight increase in yield was noted, but solvent removal
(in the dark) again led to rapid deterioration (ultraviolet—
visible analysis). Samples kept at room temperature for
varying lengths of time, protected as above from light,

also showed a marked decrease in the intensity of the

Soret band.

This decomposition has also been observed by nmr spec-
trometry. Spectra of 22 measured eight hours apart were
considerably different. No discernible peaks associated
with the platyrin system were observed in the latter nmr
and the electronic spectrum of recovered material contained
a broad peak from “70-550 nm with no distinct Soret band.

The pronounced instability exhibited by this platyrin

28 theory that the resonance energy of

agrees with Haddon's
a fully conjugated ring system is inversely proportional
to ring size. As one goes from [l,5,l,5]platyrin 22 to
[l,3,l,3]platyrin g to the porphyrins the chemical stability

of the heteroannulenes increases. Dewarl3 predicted that

a 26n-electron system would possess a small negative

30

resonance energy (Figure 1). If that is the case, dis-
ruption of the electron rich platyrin system (22) by addi-
tion of a reactive impurity would actually increase the
stability of the molecule.

Haddon28 also concluded that the ring current increases
linearly with the number of atoms in conjugation. The
linear upfield progression of interior imino protons as one
goes from the dications of octamethylporphyrin (an-elec-
trons; G-u.82 ppm), octamethylEl,3,1,3]platyrin g (22n-
electrons; 6-5.60 ppm) and octamethyl[l,5,l,5]platyrin
(26n-electrons;6-6.62 ppm) supports this theoretical pre-
diction. Thus, the goal of synthesizing a 26n-electron
platyrin which would verify Haddon's unified theory was
achieved. Furthermore, the large ring current in this
special 26n-electron ring system, wherein nitrogen atoms
are incorporated, indicates that the ring size at which

these larger molecules behave simply as polyenes lies

beyond 26 membered ring systems.

ATTEMPTED SYNTHESIS OF NEW PLATYRINS

The instability of the [l,5,l,5]platyrin (22) was an-
ticipated upon consideration of the resonance energy ex-
pected for 26n-electron systems. Despite reasonable evi—
dence supporting the structure assigned to compound 22,
its lack of stability creates problems in unequivocally
establishing the structure. Therefore, a second synthesis
of the platyrin system was investigated in order to supply
quantities of related and possibly more stable compounds,
using routes outlined in Scheme 5. A platyrin synthesized
by either of these routes, and whose structure was verified,
would help confirm the nature of the previously synthesized
[l,5,l,5]platyrin.

The synthetic approaches in Scheme 5 were suggested by
the observations38 that a,B-unsaturated carbonyl compounds
react readily and regiospecifically at the a-position of
pyrroles. This reaction allows the introduction of three
or four carbons of the five carbon bridge to the a-positions
of a dipyrrylmethane in a single step. The carbonyl groups
at the terminus of the chain should then allow the intro-
duction of a properly substituted dipyrrylmethane to provide
the desired platyrin skeleton. Subsequent oxidation and
enolization of this intermediate would then give a di-

hydroxyplatyrin (2g).

31

32

 

 

 

Scheme 5

33

In order to test this plan a model system was investi-
gated, starting with four different a-substituted pyrroles.
Such a study should allow us to ascertain the Optimum con-
densation conditions. Model compounds 22 and 22 were
synthesized by the addition of propynal and 3-butyn-2-one
respectively, to pyrrole. Pyrroles 22 and 22 were com—
mercially available.

Reaction of these compounds, as shown in Scheme 6,
and using a wide variety of bases and solvents, did not
provide reasonable quantities of aldol condensation-de-
hydration products. The only conditions which gave any
of the desired aldol products involved treatment of 22
and 22 with an ethanolic sodium hydroxide solution at room
temperature. Using these conditions, the previously un-
known unsaturated ketone 22 was obtained in 17% yield.
Similarly, the condensation of 22 and 22 under identical
conditions produced ketone 25 in approximately 17% yield.

With conditions in hand for preparing enones 22 and
25, albiet in modest yields at best, an examination of the
dipyrrylmethane forming step was then undertaken. The
synthesis of substituted dipyrrylmethane 22 has been pre-
viously accomplished by several methods.39 The method“0
chosen for this investigation is illustrated in Scheme
7. Thus the reaction of 3,h-dimethylpyrrole (22) with 2-
formyl-3,h-dimethylpyrrole in aqueous acidic methanol

resulted in the precipitation of 26 as bright red needles.

31+

:\ \ ”5
33
O
\
\\ fi‘ 35
.i o
\.
31

34

Scheme 6

35

 

Scheme 7

Reduction of this salt with sodium borohydride yielded 22
in 73% yield. Because of its extreme air sensitivity 22
was always freshly prepared before use. The drop-wise ad-

dition of 3-butyn-2-one to an oxygen free alcoholic solution

36

of 22 under nitrogen gave the bisfunctionalized dipyrryl-
methane 25 as a pale yellow powder (53% yield). The re-
maining dipyrrylmethane, 22, was prepared from 22 using
Vilsmeier-Haack formylation in 72% yield.

With intermediates 22 and 25 in hand, the final step
in the synthesis of platyrin 22% was attempted as described
in route 1 (Scheme 5). Treatment of 22 and 25 under the
conditions used in the model system did not afford the desired
macrocycle. Preliminary investigation indicated that intra-
molecular cyclization and intermolecular condensation of
25 has occurred to the exclusion of macrocycle formation.
Although the self condensation was not unexpected, we were
dismayed that numerous attempts under various conditions
failed to provide even trace amounts of macrocyclic pro-
ducts. As a result the synthesis outlined in route 2
was not attempted.

One last series of reactions, employing unsaturated
ketone precursors, was investigated in an effort to form
a platyrin. This synthetic plan is outlined in Scheme
8. When 2,2'-dipyrrylmethane (29) was reacted with two
equivalents of ethyl magnesium bromide the dipyrryl mag—
nesium bromide 22 was obtained. This magnesium salt was
slowly added to one equivalent of malonyl dichloride, but
this reaction yielded only unidentified, intractable
products. Reverse addition of the magnesium salt to the

dichloride also failed to give the desired macrocyclic

37

‘~. a” 4"
\\ F“ N \\ ”V :N 4/
H H N m mwb
40 41

 

Scheme 8

product 55. In contrast, it has been observed that reac-
tion of pyrrole magnesium bromide (33; Scheme 9) with
malonyl dichloride, under identical conditions, yielded
the diketone 3% in 35% yield. The inability to obtain fig
may be attributed to an expected linear condensation of
the dipyrrylmethane units.

Having attempted the construction of [l,5,l,5] and

[l,3,l,3]platyrin systems with the one carbon bridge in

38

 

Scheme 9

place, we next examined the possibility of synthesizing

a [l,3,l,3]platyrin having a cavity capable of containing
a transition metal. Up to this point, both attempted and
successful syntheses of [l,3,l,3] and [l,5,l,5]platyrin
have incorporated carbon bridges bearing hydrogen atoms
that protruded into the interior of the ring. In fact,
these hydrogens occupied most of the cavity, leaving room
for only a very small atom such as boron?9 Replacement
of the central carbon atom in each extended bridge with
nitrogen eliminates these interior hydrogens and greatly

increases the cavity size. Removal of the acidic hydrogen

39

atoms from the pyrrole units Creates enough room within

the cavity for a transition metal ion. A search of the
literature disclosed the pyrrolidine fié: which incorporates
a nitrogen atom into the bridge. A synthetic plan to
produce a [l,3,l,3]heteroplatyrin involving this trimer is
outlined in Scheme 10.

Dennstedtul

was the first to synthesize flé by simply
allowing pyrrole to stir in an aqueous hydrochloric acid
solution at 0°C for one minute. Neutralization and extrac-
tion provided the trimer Qé in 30% yield. Attempted
formylation of this trimer provided polymeric materials
as the major products. Since the secondary amine could
react with the Vilsmeier complex, this function needed to
be protected prior to the formylation attempt.

A number of protecting groups were examined, including

tosylamide fig, amide £1 and trifluoroacetamide Qfi. However,

the high insolubility of fig in all attempted formylations

46

led to its rejection as a suitable protecting group. The

acetylated trimer 51 was then submitted to the Vilsmeier

 

60
Scheme 10

#1

formylation and yielded the bis—d-formyl compound $2 in

60% yield. Unfortunately, the bis-a—formyl trimer could

W
a u
pk,”

47

not be deprotected. All efforts to remove the acetyl group

resulted in complete destruction of the molecule.

:2

" N
H 0*01, H

49

In the quest for a more easily removed protecting group,
the trifluoroacetyl derivative fig was prepared readily in
53% yield by reaction of fig with trifluoroacetic anhydride
and triethylamine. Formylation of fig provided the mono-
aldehyde éQ in almost quantitative yield, and none of the

expected dialdehyde. Even when fig was treated with an

42

excess of Vilsmeier reagent the only product obtained was

monoaldehyde £9 in 9U% yield. This may not be as puzzling

N
a " A
A H
0 er,

a result as it seems at first. In both the cis and trans
trimer the trifluoromethyl group acts as an umbrella and
protects one of the pyrrole rings from attack. The other
pyrrole ring is Open to attack by the Vilsmeier reagent.
No further attempts were made to obtain the diformylated
product, because the monoformylated compound is actually
a more desirable intermediate in our synthesis of the
heterOplatyrin system than is the diformylated compound.

Removal of the trifluoroacetyl protecting group proceeded

43

under mildly basic conditions to provide the monoaldehyde

i% (81% yield).

\ /\ °

12
12
22

51

The stage was now set for the important cyclization
step. Treatment of él with methanolic hydrobromic acid
resulted in loss of the starting material and precipitation
of a brown solid after fourteen hours. The solid was in-
soluble in any solvent system and was assumed to be a
polymer. In acetic and aqueous hydroiodic acid the start-
ing material was rapidly consumed, but again, an insoluble
orange-brown solid was obtained. Even reaction in very
dilute solution did not avoid this apparent polymer forma-
tion. Obviously, linear condensation is favored over cyclic
condensation. To increase the possibility of cyclization
zinc chloride and mercuric acetate were employed in an
attempt to orient the desired cyclization by chelation
with the amine nitrogens. No change in product formation
was observed.

We then turned our attention to the use of quaternary
ammonium salts which could undergo cyclization when treated
with an a-unsubstituted pyrrole. The Mannichu2 reaction

worked effectively on the trifluoroacetyl protected trimer (Q§),

44

N N N
F! €§L‘~ H
CH,/ \CH, 0 CF, CH3 \CH3
52

yielding quantitatively the Mannich product éé. Again,
deprotection proceeded smoothly to give the unprotected

compound §% (81% yield). The amine salt ééa was reacted

12
12
:2

N
CH3/N\CH, CH:/ _\CH,

53

in a Mannich type sequence with 3,A-dimethylpyrrole in order

to determine its reactivity toward pyrroles. In acetic acid

1+5

solution no reaction occurred at room temperature, and at

'0'
Al
“1

cu,’ |\CH, C": L

53a

steam bath temperatures the product was a black tar. No
reaction took place when other solvent systems were used.
Due to the possibility that protonation of the pyrrolidine
ring was the cause of difficulty, an N-protected compound
was examined. Quaternary salt ég was obtained in high

yield from gg by treatment with methyl iodide. Unfortunately,

 

46

3,A-dimethylpyrrole failed to react with éfl from 25 to
100°C, and starting materials were always recovered. When
the smaller and more reactive nucleophile cyanide was used
only about 20% of the cyano substituted product éé was
formed. We speculate that steric hindrance may be pre-

venting 3,u-dimethylpyrrole from reacting efficiently

ON C"
65‘

with the quaternary ammonium salt §3- The recovery of £3
from the reaction tends to rule out a unimolecular forma-
tion of the pyrryl cation as a reactive intermediate. The
low yield of éé might be attributable to steric problems
similar to those observed in the_formation of the mono-
aldehyde 28.

Because of these difficulties a more direct cyclization
procedure, based on the knowledge that thiophosgene reacts
readily with pyrroleu3 (see Scheme 11), was explored as
a possible route to the dithioketone éé. Unfortunately,

thiOphosgene did not react with trimer 3% in the desired

47

 

fashion, but instead deposited an insoluble polymer film
in the reaction flask. Repeated attempts at cyclization
produced no identifiable products, and this was again

attributed to linear condensation.

 

Scheme 11

#8

The inability to form any cyclic products so far has
been attributed to linear condensation of the intermediates
and/or steric constraints. However, if two trimer mole-
cules could be held together by a removable nitrogen-
nitrogen bridge, then the proximity of the remaining re-
active sites should reduce polymer formation and increase
the probability of ring formation. Oxalyl chloride was
chosen as the bridging unit. Mass spectra and nmr analysis
of the reaction of trimer gé with oxalyl chloride suggest

that one of the products was the cyclized material QZ, and

/\ \
\

(:_—.C

,// \\

57

not the desired bridged structure §§° If the pyrrole nitro-
gens are being attacked by oxalyl chloride then a success-
ful synthesis of QQ is unlikely. Further attempts to
achieve this synthesis were consequently abandoned.

Although a heteroplatyrin was never synthesized by
these routes, the chemistry of 2,5-dipyrr-2-ylpyrrolidine
(fié) has been explored. It is hoped that the information

obtained will be of help in planning future work on the

49

synthesis of a heterobridged platyrin system. The ready
availability of QQ makes it an attractive starting point
for the synthesis of a heterOplatyrin, and the appropriate

conditions may one day be found for such a synthesis.

EXPERIMENTAL

General Procedure

 

The melting points were determined on a Thomas Hoover
Uni-melt melting point apparatus and are uncorrected.

The infrared spectra were recorded on a Perkin—Elmer
Model 237B spectrometer. The PMR spectra were obtained on
a Varian T-6O or Bruker 250 spectrometer with chemical
shifts reported in 6-units measured from tetramethylsilane
as the internal standard. The UV and visible spectra were
reported on a Unicam SP-BOO or Cary 219 spectrometer using
1 cm quartz cells. A Finnigan “000 mass spectrometer was
used to obtain the mass spectra. Microanalyses were not

obtained due to the chemical instability of the compounds.

1,2,3,u,5,6,7,9—Octahydronaphthalene-2,7-dione (lg)

 

The procedure of Marshalluu was used and the product

recrystallized from ethyl acetate, mp l73-l75°C.

g,7—Dimethoxy-l,U,5,8-tetrahydronaphthalene (l1)

h35 was followed in the prepara-

The procedure of Radlis
tion of ll, mp 63.5-6U.5°C. The product was slightly

moisture sensitive.

50

51

2-[7-(3,M-Dimethyl-2H-pyrrol-2-ylinene)-2,3,u,5,6,9-hexa-

hydronaphth-2-yl]-3,N-dimethyl-lH-pyrrole tetrafluoroborate
$l§l

3,A-Dimethylpyrrole (0.20 g) and compound II (0.20 g)
were dissolved in 60 ml ethanol. To this stirred solution
tetrafluoroboric acid (0.6 ml) was added dropwise. After
stirring for ten minutes the green solid was collected by
filtration and washed with ethyl ether until the ether
washings were clear. There was obtained 0.Nl g (96.9%

yield) of l8: mp > 300°C; 1

H NMR (DMSO-d6): 611.73
(broad, 2H), 7.20 (m, 2H), 6.7“ (s, 2H), 3.2-1.“ (m, 9H),
2.18 (s, 6H), 1.85 (s, 6H); UV-Vis (CHCl3) Amax: 658 nm;
mass spectrum, m/e (relative intensity) 318 (2A, M+-HBFu),

9” (100).

2,7-Bis(3,u-dimethylpyrrol-2-yl)-2,3,H,5,6,9—hexahydro-

naphthalene (lg)

 

A solution of IQ (0.5 g) and sodium borohydride (0.“ g)
in acetonitrite (100 ml) was refluxed on a steam bath for
5 minutes. The solvent was removed under reduced pressure and
the residue washed with an ether-water mixture. The ether
layer was dried (potassium carbonate), filtered, and con-
densed to give 0.39 g of *3 as an oil in quantitative yield.
Exposure to air caused rapid decomposition: IR (CHCl3):

3&55 (N—H), 3025 and 298 (C-H), 1u12 (=C-H) cm'l; 1H NMR

52
(CDC13): 5'7.66 (broad, 1H), 7.uu (broad, 1H), 6.35 (d, 2H),
6.10 (s, 1H), 5.u2 (d, 1H), 3.58 (0, 1H), 2.7—1.1 (9H),

2.13 (s, 3H), 2.02 (s, 9H).

2,7-Bis(5-formy1-3,4-dimethy1pyrrol-2-yl)-2,3,fl,5,6,9-

 

hexahydronaphthalene (20)

 

To a magnetically stirred solution of dry dimethyl-
formamide (0.2 m1) under a nitrogen atmosphere at 0°C was
added phosphorous oxychloride (0.25 ml) dropwise. The
resulting solution solidified after 10 minutes and was
dissolved in dichloroethane (10 ml). Compound 19 (0.5 g)
was dissolved in dichloroethane (10 ml) and added dropwise
to the stirred solution. Stirring was continued for 15
minutes at 0°C and then refluxed for 20 minutes. After
cooling the reaction mixture was added to 200 ml water con-
taining 35 g of sodium acetate which was then refluxed for
AS minutes. The solution was cooled and extracted with
methylene chloride (3X). The methylene chloride extracts
were combined, washed with saturated sodium chloride (2X)
and dried over potassium carbonate. Filtration and solvent
removal yielded 20 as an oil. Chromatography of the oil on
neutral alumina with methylene chloride followed by 1%
methanol in methylene chloride gave 28 (0.“2 g) as the
second band (yield 71%): IR (CHCl3k 1710 (C=0), 1630 (C=C)
cm’l; 1H NMR (CDC13): «59.60 (broad, 2H), 9.30 (s, 1H),
9.25 (s, 1H), 5.98 (s, 1H), 5.U2 (d, 1H), 3.6M (d, 1H),

53

2.24 (s, 6H), 2.05 (s, 3H), 1.98 (s, 3H); mass spectrum,

m/e (relative intensity} 376 (u, M+), 137 (19), 117 (100).

Octamethy1[1,5,1,5]platyrin (21)

 

Compound 19 (0.5 g) was dissolved in 50 m1 of methanol.
Compound 20 (0.588 g) was dissolved in 40 m1 of methanol
and 10 ml dichloromethane. The two solutions were added
dropwise from separate addition funnels to a refluxing
solution of 20 ml 48% HBr in 700 ml of oxygen purged methanol
under nitrogen in a foil covered flask. The solution was
then refluxed for 10 minutes under nitrogen and then for 20
minutes with oxygen bubbling through the mixture. After
sitting in the dark for 36 hours 1 liter of water was added
along with 20 m1 of 30% sodium hydroxide and the solution
extracted (3X) with dichloromethane. The dichloromethane
extracts were dried (potassium carbonate) and the solvent
removed. The residue was chromatographed on neutral
alumina (Activity I) with dichloromethane followed by
1% to h% methanol in dichloromethane in 0.5% increments.
The purple band (Xmax 538 nm) was collected and the sol—
vent removed to give 12 mg of 21 (yield 2%): mp > 350°C;
1H NMR (DMSO-d6-CF3C02H): 611.75 (s, 2H, mesa—9g), n.51
(s, 12H), D.U3 (s, 12H), 2.7-1.0 (m, 18H), -10.58 (3, NH),

-1u.26 (8, NH); UV-Vis (CH2C12) A (em)(bis-tetraf1uoro-

max
borate salt): H95 (171,000), 536 (376,000), 651 (“7,000),

703 (32,000), 719 (3“,000), 780 (28,000), 830 (19,000)

54

UV-Vis (CH2C12) Amax (em): M95 (123,000), 536 (1uu,000),
651 (15,700). 705 (12,300). 718 (13.500), 780 (9,000).

5,5'-Diformyl-3,M,3',h'-tetramethyl-2,2'-dipyrrylmethane
$301

The procedure of Clezy39 was followed in the preparation

of 25, mp 283-285°C. The product was recrystallized from

chloroform.

5,5'—Bis(3-oxo-butenyl)-3,3',U,U'-tetramethyl-2,2'-di-
pyrrylmethane (gé)38

To compound 31 (0.5 g) in oxygen free methanol (60
m1) under nitrogen was added 3-butyn-2-one (0.28 ml) drop—
wise. The solution was refluxed for 2 days and water added
to the cooled solution. The yellow solid was collected by
filtration and recrystallized from chloroform/ether to give
. g5 (0.h5 g) in 53% yield: mp 208-210°C; IR (NuJol)
(NH), 1670 (00), 1590 and 1550 (C=C) cm'l; 1H NMR (00013)
610.2 (s, 2H), 7.u5 (d, 2H), 6.30 (d, 2H), 3.95 (s, 2H),
2.38 (s, 6H), 2.1“ (s, 6H), 2.08 (s, 6H); UV-Vis (EtOH)
A : A04 nm; mass spectrum, m/e (relative intensity) 338

max
(100).

55

2-(3-0xo-prgpeny1)pyrrole (3})38

A solution of pyrrole (2.0 g), propargyl aldehyde
(1.6 g) and oxygen free MeOH (10 ml) was stirred at room
temperature for 12 h under a nitrogen atmosphere. The
MeOH was removed and the residue recrystallized from CH2012/
CClu affording 3.2Sg;(93%) of 31: mp 108-110°C; IR (Nujolk
3160 (NH), 1615 (00), 1595 (C=C) cm'lg 1H NMR (00013):
69.55 (d, 1H), 7.32 (d, 1H), 7.02 (m, 1H), 6.60 (m, 1H),
6.u1 (m, 1H), 6.26 (m, 1H); mass spectrum, m/e (relative
intensity) 121 (100, M+), 93 (A7), 67 (70).

2-(3-0xo-buteny1)pyrrole(33)38

 

A solution of pyrrole (5 ml), 3-butyn-2-one (5.6“ m1)
and oxygen free EtOH (30 ml) was stirred under a nitrogen
atmosphere at room temperature for 12 h. The EtOH was
removed and the residue recrystallized from CH2012/CClu
to yield 10.01 g (96%) of 33: mp 109-110°c; IR (Nujol):
3212 (NH), 1660 and 1635 (C=C), 1605 (00) cm'l; 1H NMR
(CDC13):67.95 (d, 1H), 7.02 (m, 1H), 6.60 (m, 1H), 6.95
(d, 1H), 6.28 (m, 1H), 2.U2 (s, 3H); mass spectrum, m/e
(relative intensity) 135 (100, M+), 120 (78), 92 (7h),

65 (33).

56

1,5-Di-2gpyrry111,3:pentadien-5-one (35)

To a stirred 15% NaOH solution (12 ml) containing
EtOH (12 m1) and 31 (0.2 g) was added 30 (0.18 g) over a
6 hour period. The solution was stirred for 12 hrs, poured
into water (100 ml) containing ho m1 of a saturated NaCl
solution, and thoroughly extracted with CH2C12. The organic
layer was dried over anhydrous MgSOu and chromatographed on
neutral alumina using CH2C12/1% MeOH. The second band was

collected and reduced to give 60 mg of 3A (16%): l

H NMR
(DMSO-d6): 67.1-6.7 (m, NH), 6.80 (m, 2H), 6.25-5.97
(m, DH); mass spectrum, m/e (relative intensity) 212 (78,

+ o I
M ), 178 (31), 118 (100), UV-Vis (CHC13) Amaf 399 nm.

333',u,u'-Tetramethy1-2,2'—dipyrry1methene hydrobromide
13%)u0

The procedure of Atkinson, Johnson, and Raudenbusch

 

was followed except for the following changes: 1) the
reaction was carried out at 0°C. 2) No glacial acetic acid
was used. Typical yields were 75%. Recrystallization

from chloroform gave red-orange prisms, mp 218-220°C.

3,3',fl,“'-Tetramethyl-2,2'-dipyrry1methane (3Z)u5

 

A solution of 36 (0.5 g) and sodium borohydride (0.07 g)
in acetonitrile (50 ml) was refluxed until the yellow color

had disappeared (~20 minutes). The solvent was removed

57

and the residue dissolved in ether. The ether was ex—
tracted with water, dried (potassium carbonate) and con-
densed to give a brown oil of 37 which was identical to
that reported by Clezy. Recrystallization from light
petroleum gave colorless plates: mp 86-87°C (lit. mp 87-
8800).”3

2,5-Dipyrrol-2-ylpyrrolidine(IQ/5fl6

 

The procedure of Potts was followed in the preparation
of 35 except the solution was stirred for 1 minute; mp

99-100°C.

1-(p-Tosyl)-2,5-dipyrrol-2-y1pyrro1idine (36)

A solution of 35 (A g), Et3N (6 m1) and tosyl chloride
(3.8 g) in dry benzene (100 ml) was stirred for 12 h under
a nitrogen atmosphere. Filtration of the reaction mixture
afforded 6.8 g (97%) of 36 as cream colored crystals:
mp 286°C; IR (Nujol): 3350(NH), 1300 and llUO (S=O), 1075
(C-N) cm‘l; 1H NMR (DMSO-d6): 610.25 (d, 2H), 6.85 (s,
UH), 6.90 (m, 2H), 5.6“ (m, NH), 2.20 (s, 3H), 1.75-1.95
(m, AH); mass spectrum, m/e (relative intensity) 355 (10,
M+), 200 (hi), 18a (60), 133 (100).

1-Acetyl-2,5-dipyrrol-2-y1pyrrolidine_(£Z)

A solution of 35 (8 g), (Et)3N (12 m1) and acetic

58

anhydride (8 ml) in dry benzene (125 m1) under a nitrogen
atmosphere was stirred at 25°C for 2 days. The reaction
mixture was filtered to give 4.0 g of £1. The filtrate

was reduced to half volume and allowed to stand for 1 day.
Filtration of the solution afforded another 1.5 g of £1 for
an overall yield of 56%: mp 186-187°C; IR (CHCl3): 3450
(NH), 1625 (00), 1080 and 1025 (CN) cm'l; 1H NMR (00013):
56.60 (m, 2H), 5.95 (m, 4H), 5.u3 (m, 1H), 9.99 (m, 1H),
2.45-1.80 (m, 4H), 1.82 (s, 3H); mass spectrum, m/e (rela-

tive intensity) 2H3 (24, M+), 184 (20), 150 (100), 93 (67).

1-Trif1uoroacetyle2,5-dipyrrol-2-y1pyrrolidine ($8)

 

To a stirred solution of 45 (6 g) and Et3N in benzene
(150 ml) under a nitrogen atmosphere at 0°C was added tri-
fluoroacetic anhydride (4.2 ml) in benzene (25 m1) over a
1 h period. The solution was then stirred at 25°C for 12 h
and the solvent removed. The residue was dissolved in
ether (100 m1) and extracted with water. The organic layer
was dried (MgSOu) and condensed to give a pale yellow oil.
The oil was recrystallized from CH2C12 yielding 4.7 g (53%)
of 58: mp 181-182°c; IR (Nujol): 3340(NH), 1650 (00),

1180 (CF) cm‘l; 1H NMR (CDC13/DMSO-d6); 610.15 (m, 1H),
9.72 (m, 1H), 6.55 (m, 2H), 5.93 (m, 4H), 5.43 (t, 2H),
2.40-1.75 (m, 4H); mass spectrum, m/e (relative intensity)
297 (100, M+), 230 (21), 20a (21), 191 (25).

59

1-Acetyl-2,5-bis(5-formylpyrrol-2—yl)pyrrolidine (49)

To a stirred solution of 47 (1 g) and dry dimethyl-
formamide (8 m1) under nitrogen at 0°C was added benzoyl
chloride (0.96 ml) dropwise. The solution was stirred for
2 h at 0°C and then 2 h at 25°C. Benzene (20 ml) was added
slowly to precipitate the imine salt.r After 20 min the
salt was collected by filtration, washed with benzene, and
dissolved immediately in water (30 ml) containing sodium
acetate (3 g). The mixture was warmed for 15 min on the
steambath, diluted with water (20 ml), the product collected
and recrystallized from ethanol yielding 0.73 g (60%) of
88‘ IR (NuJolk 3230 (NH), 16u5 and 1630 (00), 1025 (0N)cm'1;
1H NMR (DMSO-d6): 59.35 (s, 2H), 6.59 (m, 2H), 5.95 (m, 2H),
5.43 (m, 1H), 4.98 (m, 1H), 2.40-1.80 (m, 4H), 1.85 (s,
3H); mass spectrum, m/e (relative intensity) 299 (5, M+),
178 (72), 149 (26), 136 (67), 44 (100).

1—Trif1uoroacetyl-2-(5-formylpyrrol-2-yl)-5:pyrrol-2-yl-

pyrrolidine (50)

To a stirred solution of 39 (0.5 g) and dimethyl-
formamide (1 m1) under a nitrogen atmosphere at 0°C was
added benzoyl chloride (0.21 ml) dropwise. The solution
was stirred at 0°C for 2 h and then at 25°C for 2 h. Water
(10 ml) containing sodium acetate (1.2 g) was added and
the solution stirred for 5 min. CH2C12 (20 ml) was added

and the aqueous layer thoroughly extracted. The CH2C12

60

layers were combined, washed with water, dried over MgSOu
and reduced. The residue was chromatographed on silica

gel with CH2C12/5% ether. The ether concentration was
increased until 50 was eluted from the column (first band).
Recrystallization from CHCl3 gave 0.51 g (93%) of 59 as
colorless crystals: mp l61-162°C; IR (NuJol):3312 and 3000
(NH), 1660 (00), 1623 (00), 1170 (CF) cm‘l; 1H NMR (00013):
611.35 (br s, 1H), 10.75 (m, 1H), 9.30 (s, 1H), 6.87

(m, 1H), 6.65 (m, 1H), 6.05 (m, 3H), 5.58 (m, 2H), 2.70-
1.90 (m, 4H); mass spectrum, m/e (relative intensity)

325 (26. M+>. 297 (5), 258 (u), 20“ (loo).

 

2-(5-Formy1pyrrol-2-yl)-Sgpyrrol-2-y1pyrrolidine (51)

A solution of 50 (0.15 g) and potassium carbonate
(0.6 g) in aqueous MeOH (10 ml; 70%) was stirred at 25°C
for 24 h. The mixture was added to saturated sodium chloride
solution (50 m1) and extracted with CH2C12. The CH2C12
layer was washed with water, dried (MgSOu) and reduced to

afford 0.85 g (81%) of 51: 1

H NMR (CDCl3/DMSO-d6) 68.88
(s, 1H), 6.82 (d, 1H), 6.60 (m, 1H), 6.10 (d, 1H), 5.99

(m, 2H), 4.30 (m, 2H), 2.40-1.55 (m, 4H); mass spectrum,
m/e (relative intensity) 229 (100, M+), 212 (10), 201 (12),

184 (60), 162 (an).

61

1-Trif1uoroacetyl-2,5-bis(5-dimethy1aminomethylpyrrol-2-y1)-
pyrrolidine (52)

To a stirred solution of 49 (l g) in MeOH (50 m1)
under a nitrogen atmosphere at -20°C was added a mixture
of 37% formaldehyde (0.59 g), potassium acetate (0.68 g),
dimethylamine hydrochloride (0.56 g) and water (4 m1) drOp—
wise. The solution was stirred at -10°C for 30 min and then
at 25°C for 15 h. 5% HCl (12 ml) was added and the solu-
tion extracted with ether. The aqueous layer was cooled to
0°C and 2N NaOH added to a pH of 12. The basic solution was
extracted repeatedly with ether. The combined organic
fractions were dried over K2CO3 and condensed to give a
yellow oil. The oil was recrystallized from ether to yield
1.36 g (98%) of 52: mp l46-l47°C; IR (NuJol):l670 (00),
1175 (CF), 1005 (CN) cm‘l; 1H NMR (00013): 89.95 (br s,
1H), 9.65 (br s, 1H), 5.90 (m, 3H), 5.72 (m, 1H), 5.45
(m, 1H), 5.25 (m, 1H), 3.41 (br s, 4H), 2.28 (d, 12H),
2.45-1.60 (m, 4H); mass spectrum, m/e (relative intensity)
366 (17, M+—HN(Me)2), 321 (35, M+-2HN(Me)2), 209 (82),
106 (100).

2,5-Bis(5—dimethylaminomethylpyrrol-Z-yl)pyrrolidine (53)

A solution of 52 (0.3 g) and potassium carbonate (0.9 g)
in aqueous MeOH (10 ml; 70%) was stirred under a nitrogen

atmosphere for 48 h. The solution was added to water (50 m1)

62

and thoroughly extracted with CH 01 The CH2C12 extracts

2 2'
were combined, dried over K2C03 and condensed. Recrystal-
lization of the solid residue from ether gave 0.21 g (91%)
of 53: mp 138-139°C; IR (Nujol): 3025 (NH), 2725 and 1150
(NC) cm'l; 1H NMR (00013): 89.37 (br s, 2H), 5.83 (d, 4H),
4.12 (m, 2H), 3.25 (s, 4H), 2.17 (s, 12H); mass spectrum,
m/e (relative intensity) 315 (2, M+), 270 (29, M+-HN(Me)2),

225 (47, M+-2HN(Me)2), 209 (33), an (100).

APPENDIX

63

 

Microns
25 30 315 «10 $0 60 80
loo .. I I 1 I I '00
80- -80

 

 

 

1..
7 fl

 

 

 

 

 

 

 

 

20. K 520
0 j ‘ 1 l J l 0
4KXK> 3600 3000 2500 2000 ”00

Frequency (001")

 

 

 

 

 

 

 

 

 

 

Microns
:10 6.0 7.0 0.0 (0.0 ".0 l2.0 06.0

'00 I I 1 j I I

B

.. 0 ..
g 40 - 40
.3

20- - 20

O l J 1 l‘ j IR-Lo

2000 IOOO 600 I400 yaw) IOOO 800

Frequencykm’ )

Figure 3. Infrared spectrum of 2,7-Bis(3,4—dimethylpyrrol-
2-yl)-2,3,4,5,6,9-hexahydronaphthalene (19).

64

 

 

 

 

 

 

 

 

 

 

 

 

 

Micron: .
2.5 3.0 3.5 4.0 5.0 6.0 8.0
'00 . I ’ I fi I I l00
80 '- 80
8 so I-
‘07. ' 4°
.-
20 " all 20
o 4 l 1 _l J o
4000 3500 3000 2500 2000 I500
Frequency km")
Microne
5.0 6.0 7.0 8.0 l0.0 ll .0 l2.0 l5.0
'00 V I I I U I m
AA:
8 ' H 30
8 - 1
60 - 60
g 40 ‘ 40
E
20 b - 20
L 1 l J j 4 o

 

 

 

3000 1000 600 I400 .200 IOOO 000
Fremencflcm' )

Figure 4. Infrared spectrum of 2,7-Bis(5-formy1-3,4-di-
methylpyrrol-2-yl)-2,3,4,5,6,9-hexahydronaphtha-
lene (20).

65

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Microns
2.5 3.0 5.5 4.0 5.0 6.0
'00 i r I j I 1 I00
80 I- n . so
3 60- «so
0 I I\ ,0
. II " II
40 - O‘KQI' N . 40
g:
20 .. g 4 20
o J 1 LN l A l o
4000 3500 3000 2500 2000 I500
Frequency km")
Microns .
5.0 6.0 7.0 6.0 I00 ".0 l2.0 I60
'00 V W I I I I '00
60 60
60 s 60
.1
40 3 40
E
20 ' - 20
o 1 l 1 i l J o
2000 I600 I600 I400 I200 I000 600
Frequency (crn’ ' I
Figure 5. Infrared spectrum of l—Acetyl-2,5-bis(5-formyl-

pyrrol-2-ylpyrrolidene (£8).

66

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Inknmm
2.5 3.0 3. 5 4.0 5.0 6.0 0.0
'00 . I I I 1 T [00
ear so
I \ I
40 - a O‘LCI' " 40
3
p. z
20 .. U ‘ 20
o L J l _J l o
4000 3500 3000 2500 2000 I500
Frequency Ian")
Nmmuu
5.0 6.0 7.0 8.0 l0.0 ".0 I20 I60
'00 I l I T I I m
e 80
8 so .60
g 40 4 4o
.3
20 - 20
o J J 1 1 1 4 a
2000 l000 600 I400 'l200 I000 000
fimnmumykmi)

Figure 6. Infrared spectrum of 1—Tr1fluoroacetyl-2,5-
d1pyrrol-2-y1pyrr011dine (fig).

67

 

hMmpns
25 3.0 3.5 4.0 5.0 6.0 80
'00 f 1’ I i T .00

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

60 60
8 60
4o - 40
" S
20 z -' 20
O J 1 1 11 J U
4000 3500 5000 2500 2000 I500
thmmmy knfU
Micm
5.0 6.0 7.0 6.0 I00 ".0 l2.0 I60
'00 I j I U 1 1 m
60
8 so . 60
{E40 3 Ia:
E
20 20
o 1 1 1 1 1 l o
2000 I800 I600 I400 'I200 I000 600
Fflnuuwykni)
Figure 7. Infrared spectrum of l-Trifluoroacetyl-24

(S-formylpyrrol-Z-yl)-5-pyrrol-2—yl-pyrrolidine

(g9).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Microns
2.5 30 35 4.0 50 6.0
'00,. T I I i j '00
I \ II I
/ a . J\ a k
30- a; \cu. ° Chgc’ ca, 00
a so. 1
40- ‘40
20- g d -20
a
z
o 1 4 1 _l L o
4000 _ 3500 3000 2500 2000 1500
Hequency Ian")
Micron:
5.0 6.0 7.0 8.0 l0.0 I I0 I20 I6.0
'00 I I I I I I no
6
géso-
40T
E
20’
o 1 _ 1 n 1 n n *0
2000 I800 I600 I400 ‘ I200 I000 600
FrequencyIcrri')
Figure 8. Infrared spectrum of l-Trifluoroacetyl-2,5-

bis(S-dimethylaminomethylpyrrol-2-yl)pyrrolidine

(gag).

 

 

rijwfvleijvawVJfiwfifiIvvjv-vvlvvvw]v1vw]v+vva
v 1 v v W 1

in d n n

 

 

 

 

 

 

A
w

A i l I

1
4 AAAL AA 4—AA ‘AJIIJJAJ‘LJAAJ

 

 

 

 

Figure 9. PMR spectrum of 2-[7-(3,h-Dimethyl-ZH-pyrrol-2-
ylinene)-2,3,H,5,6,9-hexahydronaphth—2-y1]-3,u-
dimethyl-lH-pyrrole tetrafluoroborate (IQ).

 

 

 

 

 

 

 

'Vfil'wfifiI"j'IfVfi']'"'iffifii__lfi"j]‘vfifil""l'
Tfi ‘V I V I v j v v 1 I
w - IA n II On

u

. é”! u'é

F
I
i
D
AL: A 4 A A J 4 u...
*4 Aj A44 A114 A A A14¥44 A 14 A4 A‘ILLgA AJ A A A A1 A A A A 14 A A A .
. - : - ..-.- . ‘ a i V

Figure 10. PMR spectrum of 2,7-bis(3,N-Dimethylpyrrol-Z-
yl)-2,3,H,5,6,9-hexahydronaphthalene ($2).

7O

 

 

 

 

 

 

 

 

 

Figure 11. PMR spectrum of 2,7-Bis(5-formyl-3,u-dimethyl-
pyrrol-Z-y1)-2,3,h,5,6,9-hexahydronaphthalene

 

 

 

 

(39>.
'uifiiij'fififil"';lj"‘l'"'jj"'l"fili'fi'l'
1" j I fl I V j‘ *1 fl I
K ‘ u I“ U 0"
I”
d
I
:I
1:
% m I
I
II... I. I - - 1

 

 

Figure 12. PMR spectrum of 5,5'—Bis(3-oxo-butenyl)-3,3'-
h,H'-tetramethyl-2,2'-dipyrry1methane (gé).

71

 

 

 

 

 

 

Figure 13. PMR spectrum of 2-(3-0xo-propenyl)pyrrole
(a).
A.

 

 

fit 'J I f‘fi ' ' V J f‘ ' '1 fl ' ‘Ji ' ' I I ‘ I I Iii ‘ 1* V ' I
fl I ' I ' I ' I fl I W I I
ID at an n m

 

 

 

 

 

m
1 ‘1 Al 4 - J A 1 1
--il ---IA111] --il - -l 1111 i A [iii

 

 

Figure I“. PMR spectrum of 2—(3-Oxo-butenyl)pyrrole (gé).

72

 

 

 

 

 

 

 

 

 

I l 'ffij"“1 ffi'Ji"'l fj'l
I ‘ ' T a 7 ‘ I ' I I ‘ I
”9 II x n '0: on.
a”
N
I
J A A 1 A l A 1 ‘II A I A
AAIAAAAIA- A11 AIAAAAJAAAAIAAAAj AA IA AA
5‘» i~ u s new. 0 a ..

 

Figure 15. PMR spectrum of l,5-Di-2-pyrryl-l,3-pentadien-
5-one (fig).

.1

 

 

fifi'l"'*]""lfiIfiIjI‘IfiJfi‘IJI"‘j*“‘]‘*fi‘]'
v Vj * I a I v V I v I
d I In M I-

W

I

f..- f“

 

 

 

 

 

 

 

 

 

Figure 16. PMR spectrum of l—(p-Tosyl)—2,S-dipyrrol-Z-
ylpyrrolidine (£6).

73

 

 

 

 

 

 

 

Iii fi'ijIVIJ'fi'JWj‘fiJj Val ‘ijvvvj VIII
Ifi fl I fl 1 ‘ Ifi fi‘I * I
w I! an )0 wt «-
.o.

I T I

0
d !
I
I
L

A A A 1 “‘3 1*

AA JA AAjAAAA]AAAAIAAAAjAAIAAjAAAAT A
00 I. u so any. a: q

 

Figure 17. PMR spectrum of l-Acetyl-Z,S-dipyrrol-Z-yl-
pyrrolidine (51).

A.

 

 

'IJ'IF'JI‘ I "1" I1 *“I*II‘]I‘I‘I“I*TI

II *1 ' j " j'

IO ‘ II n " .~
'u

 

 

 

 

 

Figure 18. PMR spectrum of l-Trifluoroacetyl-Z,5-dipyrrol-
‘ 2-ylpyrrolidine (fiQ).

7h

 

 

 

 

 

 

 

 

A.
}14 *' 1‘; *'1 '1 '1'*',“II'*II*g"‘!' "{I
U ‘ 3 u W 00"
- i
f
O 0
II 1 II
II I.
A A1 4A 1 A 1A A] A
rrti‘“‘l*r"1*“*1‘;:‘344‘4?:‘4*j"J‘Trr‘rj‘
Figure 19. PMR spectrum of l-Acetyl-2,5-bis(5-formyl—
pyrrol-Z-yl)pyrrolidine (32).
.A
"'I""J"fi'_I'—I"T‘
I' fiT ' 1

 

AD.

2;“!

A A A A A A A A A A A A A A A A A A A A
00 1' n u u n. | o u to. .

 

 

 

 

RAW“. A _l

Figure 20. PMR spectrum of l-Trifluoroacety1-2-(S-formyl-
pyrrol-Z—yl)—5-pyrrol-2—y1pyrrolidine (fig).

75

 

 

 

I
I
I
A l A A 1 A L
1 AA A A14 A A A I A A A A I A A_AA A.
e I. . j

A

fr—II-
I

A l A
.
AAJAAAAjAAAA
I, 0

Figure 21. PMR spectrum of 2-(5-Formylpyrrol-2-yl)-5-
pyrrol-2-y1pyrrolidine (2%).

 

 

 

 

 

 

 

V‘I‘Ifigffi 7J7 ‘J *III‘ ‘ T ' ‘177 71“ 1‘
l I ' F I I I I
“I n '. ‘1
u I I
u ‘4‘: u I
_/‘ °"’ " °" l
1M F; Jr
,I
I
la
'.
A A A I A A 1 A
AA IA A I A 1A AA1 A 1A AjAA IAA TAAAA
n .0 u a. c a; :0 '0 7

PMR spectrum of l—Trifluoroacetyl—Z,5-bis(5-

Figure 22.
dimethylaminomethylpyrrol-2-ylpyrr011dine (ég)-

76

 

 

‘j fl 1‘ T * '1fi *]*W I fi'l‘fif‘ T‘
Ifi I f j I f 1 a
)9.
u I I I
a I I
01. DO.-

 

 

 

 

 

Figure 23. PMR spectrum of 2,5-B1s(S-dimethylaminomethyl—
pyrrol-2-y1)pyrrolidine (éi).

REFERENCES

NH

10.

ll.

l2.

13.

l”.

15.
16.

REFERENCES

F. A. Kekule, Bull. Chim. Soc. France, 1865, g, 98.

 

A. Baeyer, Ann., 1888, £52, 106.

A. Claus, "Theoretische Betrachtungen und deren An-
wendung zur Systematik der organische Chemie." p. 208
(O. Hallander, Freiberg, 1867).

L. Meyer, "Modern Theorien inder Chemie," 2nd ed.,

1872.
H. Heitler and F. London, Z. Phys., 1927, 3%, “55.

E. Hfickel, z. Physik, 1931, 19, 20A; 1931, 72, 310.

 

D. P. Craig, Non-Benzenoid Aromatic Compounds, D.
Ginsberg, ed., (New York: Interscience Publishers,
Inc., 1959), p. 1“.

 

H. C. Longuet-Higgins and L. Salem, Proc. Roy; Soc.

 

Ser. A, 1959, 591, 172.

H. C. Longuet-Higgins and L. Salem, Proc. Roy. Soc.
Ser. A., 1960, ééz, HHS.

C. A. Coulson and W. T. Dixon, Tetrahedron, 1962,17,
215.

 

 

M. J. S. Dewar and G. J. Gleicher, J. Am. Chem. Soc.,
1965, Q1, 685.

A. Brickstock and J. A. Pople, Trans. Faradax Soc.,
196M, @8: 901.

M. J. S. Dewar, in Aromaticity Special Publication No.
21 of the Chemical Societx, London, pp. 177-198 (1967).

 

H. P. Figeys, Tetrahedron, 1970, 26, 5225.

 

J. A. Pople, J. Chem. Phys., 1956, 35, 1111.

 

J. Griffiths, Colour and Constitutiongof Qrganic Mole-
cules, (New York: Academic Press, Inc., 1976), p. 226.

77

17.
18.

19.
20.

21.

22.

23.
2H.

25.
26.

27.

28.
29.

30.
31.

32.

33.

3n.
35.
36.

37.

78

L. Pauling, J. Chem. Phys., 1936, Q, 673.

 

 

F. Sondheimer, Proc. Roy. Soc;, Ser. A., 1967, 297,
173 and references cited therein.

F. Sondheimer, Chem. Soc. Spec. Publ., 1967, 21, 75.

 

F. Sondheimer, Chem. Comm., 1971, 338.

 

C. C. Leznoff and F. Sondheimer, J. Amer. Chem. Soc.,
1967, 89, H2U7.

 

M. Nakagawa, Pure Applied Chem., 1975, 5%, 885 and
references therein.

 

E. Vogel and H. D. Roth, Angew Chem., 1981, 76, HUS.

 

 

E. Vogel, et. a1., Angew Chem., 1975, 87, 591.

E. Vogel, et a1., Angew Chem. Int. Ed., 1978, 17, 956.

 

T. Otsubo, B. Gray, and V. Boekelheide, J. Amer. Chem.
Soc., 1978, 100, 2uu9.

 

T. Otsubo, D. Stusche, and V. Boekelheide, J. Org.
Chem., 1978, $3, 3U66.

R. C. Haddon, J. Amer. Chem. Soc., 1979, 101, 1722.

 

R. g. Berger, Ph.D. Thesis, Michigan State University,
197 .

M. M. King, Ph.D. Thesis, Harvard University, 1970.

D. Dolphin, The Porphyrins, Vol. 1, (New York: Academic
Press, Inc., 1978) pp. 85-23u.

 

A. Markovac and S. F. MacDonald, Can. J. Chem., 1965,
5;, 336a.

R. A. Berger, Ph.D. Thesis, Michigan State University,
1978.

 

A. J. Birch, J. Chem. Soc., 19uu, M30.

 

P. Radlich, J. Org. Chem., 1965, 29, 3208.

 

E. J. H. Chu and T. O. Chu, J. Org. Chem., 195“, 12,
266.

G. P. Arsenault, E. B. Bullock, and S. F. McDonald,
J. Am. Chem. Soc., 1960, 82, N38“.

 

38.
39.

“0.

U1.
U2.

“3.

AA.

“5.

A6.

79

A. w. Johnson, J. Chem. Soc., 19u7, 1626.

 

P. S. Clezy and A. J. Liepa, Aust. J. Chem., 1970,
23, 2493.

J. H. Atkinson, A. w. Johnson, and w. Raudenbusch,
J. Chem. Soc. (C), 1966, 1155.

 

 

Dennstedt and Zimmermann, Ber., 1887, 20, 857.

w. Bachman and L. Heisey, J. Amer. Chem. Soc., 1946,
68 2996.
3

P. S. Clezy and G. A. Smythe, Aust. J. Chem., 1969,

22, 2A7.

J. A. Marshall and N. Andersen, J. Org: Chem., 1965,
99, 1292.

 

 

 

C. Jutz, A. F. Kisschner, and R. M. Wagner, Chem. Ber.,

 

1977, 119, 1259.

H. A. Potts and G. F. Smith, J. Chem. Soc., 1957,
A018.