One-Pot Approach to Chlorins, Isobacteriochlorins, Bacteriochlorins, and Pyrrocorphins

Article abstract of DOI:10.1021/acs.orglett.8b03433

A Diels-Alder strategy is reported to synthesize the complete set of hydroporphyrins: chlorins, bacteriochlorins, isobacteriochlorins, and pyrrocorphins. Porphyrins and Ni-porphyrins react with isobenzofuran in very high yields at 70 °C to form the corres

Full text of DOI:10.1021/acs.orglett.8b03433

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
One-Pot Approach to Chlorins, Isobacteriochlorins, Bacteriochlorins,
and Pyrrocorphins
†
†
‡
Morten K. Peters, Fynn Rohricht, Christian Nather, and Rainer Herges*,†
̈
̈
†Otto Diels-Institute of Organic Chemistry, Christian Albrechts University Kiel, Otto Hahn Platz 4, 24118 Kiel, Germany
‡Institut fu
̈
r Anorganische Chemie, Christian Albrechts University Kiel, Max-Eyth-Strasse 2, 24118 Kiel, Germany
S
* Supporting Information
ABSTRACT: A Diels−Alder strategy is reported to synthesize
the complete set of hydroporphyrins: chlorins, bacteriochlor-
ins, isobacteriochlorins, and pyrrocorphins. Porphyrins and Ni-
porphyrins react with isobenzofuran in very high yields at 70
°C to form the corresponding chlorins. Electron-deficient
porphyrins react with a second equivalent of isobenzofuran
yielding exclusively bacteriochlorin (82%), and Ni-porphyrin
gives only isobacteriochlorin (99%). All cycloadditions are
completely regio- and stereoselective. The regiochemistry is
correctly predicted using the ACID method.
orphyrins and their reduced derivatives (chlorins,
Scheme 2. [4 + 2] Cycloaddition of Isobenzofuran 4 with
Free-Base H2TPPF20 (1), NiTPPF20 (2), and TPP (11)
Forming the Corresponding Chlorins
P
bacteriochlorins, isobacteriochlorins, and pyrrocorphins;
Scheme 1) are ubiquitous in nature, and more than 50 000
synthetic analogues have been synthesized so far.1 Besides
stability and catalytic activity, their photophysical properties,2
found a wide range of applications. Porphyrin and hydro-
porphyrin derivatives have been used as photosensitizers for
photodynamic therapy,3−5 as light-harvesting antennae in
artificial photosynthesis,6,7 as chromophores in dye-sensitized
solar cells,8 for multiphoton microscopy9 and for the design of
a number of molecular photonic devices and materials.10
Even though hydroporphyrins such as chlorins and
bacteriochlorins are the most important pigments in the
photosynthetic systems of green plants and bacteria, their
favorable absorption maxima in the visible and NIR range
notwithstanding, examples of synthetic hydroporpyrins de-
crease with increasing hydrogenation state (porphyrins:
45 000, chlorins: 5100, bacteriochlorins: 576, isobacteriochlor-
ins: 180, pyrrocorphins: 68).1 Most of the bacterio- and
isobacteriochlorins are semi-synthetic derivatives of natural
products11 because the number of fully synthetic methods is
limited.12 There are three general synthetic approaches to
chlorins and (iso)bacteriochlorins starting from porphyrins.13
The conversion of β,β′ pyrrolic double bonds can be achieved
by (1) direct reduction, (2) oxidation, and (3) cycloadditions.
Direct reduction (1) leads to chlorins that rapidly oxidize back
to the porphyrins. Oxidation (2) is usually performed with
OsO4 and leads to the dihydroxylated chlorins, which are
further derivatized. Cycloadditions (3), while providing a
convenient, one-step access to chlorins, suffer from the low
reactivity of the β,β′ pyrrolic double bond. Only very reactive
1,3-dipoles, such as azomethine ylides,14−16 nitrile oxides,17
and nitronates18 react via 1,3-dipolar cycloadditions. Diels−
Alder reactions are restricted to highly reactive quinodi-
methanes as dienes, which have to be generated at very high
temperatures in situ, leading to chlorins and (iso)-
bacteriochlorins that are susceptible to oxidation to porphyrins
under air, limiting applications in photodynamic therapy.19,20
Scheme 1. Schematic Structures of Tetrapyrrolic
Compounds Including 18 Electron Annulene
a
Substructures
a
Received: October 26, 2018
Hydrogenated pyrrole rings are shown in gray.
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b03433
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Synthesis of Different Porphyrin Derivatives via
[4 + 2] Cycloadditions
Table 1. Diels-Alder Reactions of Different Porphyrins with
a
IBF at 70 °C
yield (%)
remaining
eq of
IBF
starting
material
tetrapyrrole
H2TPPF20
H2TPPF20
H2TPCF20
Ni-TPPF20
Ni-TPPF20
H2P
monoadduct
bisadduct
1
1
5
2
5
1
1
5
2
−
(6)
(9)
(2)
(1)
5
5
5
6
−
(88)
(26)
(16)
(14)
7
(6)
15
10
5
10
10
10
10
7
7
8
8
(65)
(82)
(83)
(99)
b
c
d
Figure 2. Electronic absorption spectra of chlorin 5 and
bacteriochlorin 7 and their reduced derivatives 6, 8, 9, and 10 at 25
°C in acetonitrile (tetrapyrrole concentration of 10 μM).
d
2
13
11
12
not isolated
11
not isolated
−
−
−
d
H2TPP
Ni-TPP
(21)
14
(72)
the meso position of the porphyrin. It was therefore
straightforward to use the parent isobenzofuran 4 (IBF) as
the diene in the Diels−Alder reaction with porphyrins. IBF 4 is
considerably more reactive than DBPF and concomitantly
sterically less hindered. On the downside, IBF is less stable
than DBPF and has to be freshly prepared before use. We
recently published a convenient synthesis of pure IBF 4, which
can be stored indefinitely as a solid at −20 °C.21 We here
report on the high reactivity and selectivity of IBF 4 in Diels−
Alder reactions with free-base porphyrins and Ni-porphyrins,
forming chlorins (1:1 addition products) and (iso)-
bacteriochlorins (2:1 addition) with complete regio and
stereoselectivity.
a
Reaction time was 36 h in all cases; anhydrous benzene was used as a
solvent.; and in all experiments, 50 mg of porphyrin was used (except
b
H2P). Some product was lost through retro Diels−Alder reactions
c
d
exo 53%, endo 19% Some product was lost during column
chromatography
We selected tetrakis(pentafluorophenyl)porphyrin
(H2TPPF20) 1 and Ni-tetrakis(pentafluorophenyl)porphyrin
(Ni-TPPF20) 2 as dienophiles because they are readily
accessible and have been previously subjected to cyclo-
additions. We reacted both porphyrins with freshly prepared
IBF 4 for 36 h at 70 °C in benzene. After workup, 1:1
(chlorins), and 2:1 cycloaddition adducts (bacterio- and
isobacteriochlorins) were isolated (Schemes 2 and 3). The
electron-rich IBF and electron-deficient porphyrins form
charge-transfer complexes in a pre-equilibrium preceding the
cycloaddition. The formation of porphyrin charge-transfer
complexes is typically indicated by the quenching of the
fluorescence of the porphyrin.22,23 The fluorescence of
H2TPPF20 1 (emissions at 643 and 702 nm) is indeed
quenched upon addition of IBF 4 (see the Supporting
Figure 1. Crystal structure of endo free-base chlorin 5 and the anti−
endo free-base bacteriochlorin 7.
Driving force for oxidation is the rearomatization of the
newly formed cyclohexene ring. In principle, re-oxidation could
be prevented if stable, cyclic quinodimethanes such as the
commercially available 1,3-diphenylisobenzofuran (DBPF)
would be applied because a bridged cyclohexene would be
formed that is not susceptible to rearomatization. Unfortu-
nately, in our hands, DBPF does not react even with electron-
poor porphyrins such as tetrakis(pentafluorophenyl)porphyrin
1. Model calculations reveal that the phenyl substituents in
DBPF impose severe sterical hindrance with the aryl groups at
B
DOI: 10.1021/acs.orglett.8b03433
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Figure 3. ACID plots predicting the regiochemistry of the Diels−
Alder reaction of free base chlorin 5′ and the Ni-chlorin 6′ with
isobenzofuran (IBF, 4). Both chlorins include an aromatic 18annulene
subunit (highlighted with bold lines) and are represented as
contiguous isosurfaces in the ACID plot, leaving one double bond
isolated, i.e., olefinic in nature and, thus, more reactive toward a
cycloaddition (indicated with red arrows). For the sake of simplicity
and to save computer time, the pentafluorophenyl substituents are
replaced by fluorine, and the 1,3-dihydroisobenzofuran unit is
replaced by two hydrogen atoms. Critical isosurface values (CIV)
are given for all conjugated C−C and C−N bonds. Large CIV values
represent bonds with strong conjugation.
Information).24 The formation of stable charge-transfer
complexes is further corroborated by density functional theory
calculations (see the vide infra results).
Figure 4. (a) DFT calculations of the Diels−Alder reactions of the
free base porphyrin TPPF20 1 with isobenzofuran (IBF, 4) yielding
the corresponding chlorins 5 in endo and exo configurations and the
subsequent Diels−Alder reactions with a second molecule of IBF (4)
to form all 12 conceivable isomers of the 2:1 addition products
(bacterio- and isobacteriochlorins) in exo, endo, anti, and syn
configurations. (b) DFT calculations of the Diels−Alder reactions
of the Ni-porphyrin Ni-TPPF20 2 analogous to that shown in panel a).
All reaction energies and activation barriers are given relative to the
corresponding charge-transfer (van der Waals) complex from which
the cycloaddition starts. The experimentally observed pathways and
products are highlighted. Predicted and observed products are
identical with one exception. According to experiment, the cyclo-
addition of IBF to the free base chlorin 5 gives anti, endo, and endo
bacteriochlorin 7. DFT calculations favor the formation of the syn,
endo, and endo isomers based on kinetic as well as thermodynamic
data.
Ni-porphyrin 2 is more reactive than the corresponding free-
base porphyrin 1 and forms higher amounts of 2:1 products
(see Table 1). NMR spectra and a crystal structure analysis
(Figure 1) reveal that only endo products are formed.
Remaining starting material (porphyrin), 1:1 (chlorins), and
2:1 products can be easily separated by column chromatog-
raphy (see the Supporting Information). We attribute the
strong endo stereospecifity to the pre-orientation of IBF 4 in
the charge-transfer complex (Scheme 2). Electron-rich H2TPP
11 forms a much-weaker complex with IBF 4, and therefore,
both endo and exo products are formed.
With a higher excess (10−15 equiv) of IBF 4 (at 70 °C),
predominantly 2:1 products are formed. Free-base porphyrin 1
exclusively forms the corresponding bacteriochlorin 7 with an
82% yield, and Ni-porphyrin 2 gives 99% Ni-isobacteriochlorin
8. Compared to previously published cycloadditions, both
yields are exceptional. Moreover, both cycloadditions are
completely regio- and stereoselective. Among the 12
conceivable isomers (iso/bacteriochlorin, endo/exo, and syn/
anti), only one isomer is formed. Nuclear magnetic resonance
and a crystal-structure analysis (Figure 1) prove that only the
anti and endo−endo products of the free base bacterio- and Ni-
isobacteriochlorin are isolated. To gain access to the complete
series of Ni-tetrapyrroles: Ni-porphyrin, Ni-chlorin, Ni-
isobacteriochlorin, Ni-bacteriochlorin, and Ni-pyrrocor-
phin,25,26 we inserted Ni into bacteriochlorin 7 (Scheme 3;
for UV−vis spectra, see Figure 2).
Ni2+ insertion into 7 following conventionally procedures
such as treatment with Ni(OAc)2 in DMF at 150 °C were
unsuccessful because cycloreversion took place at high
temperatures. We therefore used a previously published
method that allows Ni2+ insertion under mild conditions.
Treatment of 7 with Ni(COD)2 at 20 °C yielded the Ni-
C
DOI: 10.1021/acs.orglett.8b03433
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
bacteriochlorin 9 and Ni-pyrrocorphin 10.27 The Ni-
pyrrocorphin 10 can be converted to the Ni-bacteriochlorin
9 using trifluoroacetic acid under air (see the Supporting
Information).
To explore the scope and limitation of our new method, we
subjected tetraphenylporphyrin 11 (TPP), Ni-tetraphenylpor-
phyrin 12 (Ni-TPP), and parent porphin 13 (H2P) to
cycloadditions with IBF 4. TPP gives chlorin 14 with endo
and exo configurations that can be separated by column
chromatography. However, no 2:1 products are formed. Ni-
TPP and parent porphin also give chlorins that have not been
further characterized because of very poor solubility (Table 1).
The Diels−Alder reactions also proceed at room temperature;
however, reaction times until complete conversion are several
days.
Figure 4. Thermodynamic and kinetic data are within the
expected range for Diels−Alder reactions proceeding at room
temperature or slightly elevated temperatures. Our calculations
correctly predict the stereo- and regiochemistry except for the
experimentally observed formation of anti, endo, endo
bacteriochlorin 7. According to the DFT calculations, the
formation of the corresponding syn product should be
preferred (Figure 4 a). This discrepancy can be explained by
the fact that our calculations do not include entropy. The
formation of the syn isomer is entropically disfavored because
the attack of IBF at the porphyrin π hemisphere, which is
already shielded by the first IBF addition, is disfavored.
Cycloaddition therefore takes place from the less-hindered
side.
In summary, we present a convenient, high-yielding, regio-
and stereoselective synthesis of chlorins, bacteriochlorins,
isobacteriochlorins, and pyrrocorphins (hexahydroporphyrins).
With Ni2+ in the tetrapyrrolic core, the whole series of
hydroporphyrins has been prepared and characterized. In
particular, bacteriochlorins are highly interesting chromo-
phores because they exhibit strong absorption bands in the
near-infrared. At these wavelengths, blood-supported tissue is
relatively transparent. Free-base bacteriochlorin 7 is strongly
fluorescent and stable toward oxidation, and with a strong
absorption band at 733 nm, it is a good candidate for
applications in photodynamic therapy. The corresponding Ni-
bacteriochlorin 9 is non-fluorescent and a stable NIR dye with
a strong absorbance at 757 nm. Both compounds are good
starting points for further derivatization because the penta-
fluorophenyl rings are susceptible to aromatic substitution with
nucleophiles, including water-solubilizing groups.32
A similar regioselectivity in the cycloadditions of electron-
deficient free-base chlorins yielding bacteriochlorins and of the
corresponding Ni-chlorins forming isobacteriochlorins has
been previously observed upon reaction of TPPF20 and Ni-
TPPF20 with two equivalents of azomethine ylides, quinodi-
methanes, and hydrogenation.19,25,28 Bruhn and Bru
̈
ckner
explain this regioselectivity based on their average ionization
energy (ALIEs) procedure.28 We here apply our ACID method
that is known to reliably predict stereoelectronic effects.29,30
The ACID scalar field is interpreted as the density of
delocalized electrons, which is represented as an isosurface
(just as the total electron density). The degree of conjugation
between atoms or bonds can be quantified by the critical
isosurface value (CIV). CIV is defined as the ACID value at
points where the gradient of the ACID scalar field is zero and
the Laplacian has two negative and one positive eigenvalues.
High CIV values indicate strong conjugation. ACID applied to
the chlorins 5′ and 6′ gives an intuitive and conclusive picture.
Both chlorins include 18annulene subunits (bold lines in Figure
3, top), which are aromatic with 4n + 2 delocalized electrons.
The ACID plot of free base chlorin 5′ clearly reflects this
classical view. The aromatic subsystem is readily identifiable as
a contiguous ACID isosurface (yellow surface in Figure 3,
bottom, left). Separate and isolated from this 18annulene unit is
the β12,β13 double bond, opposite to the hydrogenated C−C
bond (indicated with a red arrow, Figure 3). This double bond
is rather olefinic (CIV: 0.055) and, hence, more reactive as a
dienophile in Diels−Alder reactions or any other reaction
converting this C=C double bond into a single bond. The C=C
bonds participating in the aromatic system (CIV: 0.083) are
less reactive, which explains the experimental finding that
exclusively the bacteriochlorin is formed. The situation is more
complicated in Ni-chlorin 6′. A pair of degenerate, symmetry-
equivalent valence bond substructures can be drawn to
describe the aromatic 18annulene subunit (one of them is
shown in Figure 3 top, right). Here the β7,β8 and β17,β18
double bonds (indicated with red arrows, Figure 3 bottom
right) are more isolated (CIV: 0.069) compared to the β12,β13
double bond (CIV: 0.073), which explains the experimentally
observed formation of the isobacteriochlorins.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b03433.
Experimental procedures and spectral data, details of
computational studies, and crystallographic data (PDF)
Accession Codes
CCDC entries nos. 1851406 and 1851407 contain the
supplementary crystallographic data for this paper. These
data can be obtained free of charge via www.ccdc.cam.ac.uk/
data_request/cif, by e-mailing data_request@ccdc.cam.ac.uk,
or by contacting The Cambridge Crystallographic Data
Centre, 12 Union Road, Cambridge CB2 1EZ, U.K.; fax:
+44 1223 336033.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: rherges@oc.uni-kiel.de.
ORCID
To further elucidate the reason for the remarkable stereo-
and regioselectivity of our cycloadditions, we performed
calculations at the B3LYP-D3/6-31G* level of density
functional theory (DFT), including dispersion.31 Reaction
enthalpies and activation barriers were determined relative to
the charge-transfer (van der Waals) complex from which the
cycloaddition starts. These complexes are stable minima on the
potential energy hypersurface. The results are summarized in
̈
Christian Nather: 0000-0001-8741-6508
Rainer Herges: 0000-0002-6396-6991
Notes
The authors declare no competing financial interest.
D
DOI: 10.1021/acs.orglett.8b03433
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(32) Dommaschk, M.; Gutzeit, F.; Boretius, S.; Haag, R.; Herges, R.
Chem. Commun. 2014, 50, 12476−12478.
ACKNOWLEDGMENTS
■
The authors gratefully acknowledge financial support by the
Deutsche Forschungsgesellschaft (DFG) within the Sonderfor-
schungsbereich 677, “Function by Switching”.
REFERENCES
(1) Reaxys database May 2018.
■
(2) Porphyrins: Optical spectra and electronic structure of porphyrins
and related rings; Gouterman, M.; Dolphin, D., Eds.; Academic Press:
Boston, 1978; Vol 3 (Part A), pp 1−165.
(3) Bonnett, R. Chem. Soc. Rev. 1995, 24, 19−33.
(4) Sternberg, E. D.; Dolphin, D.; Bruckner, C. Tetrahedron 1998,
54, 4151−4202.
(5) Mazor, O.; Brandis, I.; Plaks, V.; Neumark, E.; Rosenbach-
Belkin, V.; Salomon, Y.; Scherz, A. Photochem. Photobiol. 2005, 81,
342−351.
(6) Gust, D.; Moore, T. A.; Moore, A. L. Acc. Chem. Res. 2009, 42,
1890−1898.
(7) Esswein, A. J.; Nocera, D. G. Chem. Rev. 2007, 107, 4022−4047.
(8) Li, L.-L.; Diau, E. W.-G. Chem. Soc. Rev. 2013, 42, 291−304.
(9) Konig, K. J. Microsc. 2000, 200, 83−104.
(10) Holten, D.; Bocian, D. F.; Lindsey, J. S. Acc. Chem. Res. 2002,
35, 57−69.
(11) Scheer, H. An Overview of Chlorophylls and Bacteriochlor-
ophylls: Biochemistry, Biophysics, Functions and Applications.
Advances in Photosynthesis and Respiration 2006, 25, 1−26.
(12) Pandey, R. K.; Zheng, G. Porphyrin Handbook (Porphyrins as
photosensitizers in photodynamic therapy), Kadish, K. M.; Smith, K. M.;
Guilard, R., Eds; Academic Press: Boston; Vol 6, Chapter 43, pp
157−230.
(13) Taniguchi, M.; Lindsey, J. S. Chem. Rev. 2017, 117, 344−535.
́
(14) Silva, A. M. G.; Tome, A. C.; Neves, M. G. P. M. S.; Silva, A. M.
S.; Cavaleiro, J. A. S. J. Org. Chem. 2005, 70, 2306−2314.
́
(15) Silva, A. M. G.; Tome, A. C.; Neves, M. G. P. M. S.; Silva, A. M.
S.; Cavaleiro, J. A. S. Chem. Commun. 1999, 1767−1768.
́
(16) Silva, A. M. G.; Tome, A. C.; Neves, M. G. P. M. S.; Silva, A. M.
S.; Cavaleiro, J. A. S. J. Org. Chem. 2002, 67, 726.
(17) Liu, X.; Li, C.; Peng, X.; Zhou, Y.; Zeng, Z.; Li, Y.; Zhang, T.;
Zhang, B.; Dong, Y.; Sun, D.; Cheng, P.; Feng, Y. Dyes Pigm. 2013,
98, 181−189.
(18) Ebrahim, M. M.; Moreau, M.; Senge, M. O. Heterocycles 2011,
83, 627−630.
́
(19) Tome, A. C.; Lacerda, P. S. S.; Neves, M. G. P. M. S.; Cavaleiro,
J. A. S. Chem. Commun. 1997, 1199−1200.
(20) Vicente, M. G. H.; Cancilla, M. T.; Lebrilla, C. B.; Smith, K. M.
Chem. Commun. 1998, 2355−2356.
(21) Peters, M. K.; Herges, R. Beilstein J. Org. Chem. 2017, 13,
2659−2662.
(22) Gouterman, M.; Stevenson, P. E. J. Chem. Phys. 1962, 37,
2266−2269.
(23) Aoki, T.; Sakai, H.; Ohkubo, K.; Sakanoue, T.; Takenobu, T.;
Fukuzumi, S.; Hasobe, T. Chem. Sci. 2015, 6, 1498−1509.
(24) Rahman; Harmon, H. J. Spectrochim. Acta, Part A 2006, 65,
901−906.
(25) Ide, Y.; Kuwahara, T.; Takeshita, S.; Fujishiro, R.; Suzuki, M.;
Mori, S.; Shinokubo, H.; Nakamura, M.; Yoshino, K.; Ikeue, T. J.
Inorg. Biochem. 2018, 178, 115−124.
(26) Sehwesinger, R.; Waditschatka, R.; Rigby, J.; Nordmann, R.;
Schweizer, W. B.; Zass, E.; Eschenmoser, A. Helv. Chim. Acta 1982,
65, 600−610.
(27) Peters, M. K.; Herges, R. Inorg. Chem. 2018, 57, 3177−3182.
(28) Bruhn, T.; Bruckner, C. J. Org. Chem. 2015, 80, 4861−4868.
̈
̈
(29) Geuenich, D.; Hess, K.; Kohler, F.; Herges, R. Chem. Rev. 2005,
105, 3758−3772.
(30) Herges, R.; Geuenich, D. J. Phys. Chem. A 2001, 105, 3214−
3220.
(31) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys.
2010, 132, 154104.
E
DOI: 10.1021/acs.orglett.8b03433
Org. Lett. XXXX, XXX, XXX−XXX